key: cord-025358-nlmta9po authors: costa-font, joan title: review of ‘knowledge resistance. how avoid insight from others’ by mikael klintman, manchester university press, 2019, 256 pages, manchester, retail price 12.99£ date: 2020-04-17 journal: j behav exp econ doi: 10.1016/j.socec.2020.101540 sha: doc_id: 25358 cord_uid: nlmta9po nan in an ideal world, we would only absorb valid knowledge claims and resist false or unproven ones. however, knowledge acquisition is far from the simple process of information transmission described by 'rational expectations' models. this is why the study of how individuals learn is at the core of behavioral economics. understanding learning entails a identifying the root causes of 'knowledge resistance'. this book is a step in that direction. mikael klintman in this book discusses several different explanations for knowledge resistance drawing on evidence from several social sciences, and especially, on behavioral economics. borrowing from nietzsche (1956) , the author distinguishes between 'dionysian' tendencies in knowledge acquisition, which are driven by passion and groupcenteredness, and 'apollonian', or rational and fact-oriented ones. a classification that compares to kahneman's thinking fast and slow approach (kahneman, 2011) . accordingly, knowledge's resistance can be explained by dionysian tendencies of group loyalty that encourage us not to deviate from our local culture (and its dominant ideology). another explanation, at the core of behavioural economics, lies in the influence of 'negativity bias' (an evolutionary tendency to pay more attention to negative information to increase our chances of survival) on our beliefs, which makes us pay more attention to the risks of certain behaviors, and resist knowledge about their associated benefits. central to the book is the claim that when individuals differ in their cultural values, they naturally exhibit social incentives to knowledge claims. this phenomenon responds to a clear evolutionary advantage, namely the fact that a better adaptation to the local cultural norms increases the chances of survival and reproduction. hence, an utter motivaion of knowledge resistance is that of strengthening the bonds within groups', thus enhancing collaboration. the logic in such a setting (according to the author), is that that the more the beliefs of one group deviate from that of others, the stronger their group cohesion. in other words, there is warmth in sensing something 'unique' that others do not sense. this explains the creation of 'knowledge tribes', which are instrumental in explaining knowledge resistance. the problem lies when knowledge and moral claims overlap. this is typically the what happens when individuals form beliefs on contentious issues such as global warming, or child smacking, which are divisive issues in society. however, such divisions are mostly a reflection of people's moral values, especially at the extremes of the ideological debate. for instance, progressives are more likely to neglect findings that challenge organic agriculture, whilst conservatives are more likely to challenge evidence that 'smacking' has adverse effects on children's mental wellbeing. knowledge resistance comes from how people frame such morally divisive problems. a common behavioral mechanism is the use of backward induction in learning. that is, individuals tend to make a choice of their 'desired conclusion', and work out their arguments backwards. people anchor their views on priors, and search for arguments that confirm and strengthen their priors (which we define as 'confirmation bias'). for instance, some research shows that individuals anchor their attitudes to nuclear power on political values (costafont, rudisill & mossialos, 2008) . others, in the context of covid 19, anchor their values on the economic consequences of government decisions such as lockdowns alone. the author strongest claim is to argue that 'knowledge resistance is universal'. although, in the thurst of a public discussion, individuals commonly tend to accuse each other of being 'knowledge resistant'the truth is that, when social interests are in conflict with factual knowledge, they tend to choose to satisfy their our social interests. contrary to comon beliefs, human beings have not evolved to be 'knowledge maximisers'. far from it, we are sensitive to the social incentives such as in the form of status that come from group collaboration. this can be explained by some form of 'behavioural learning' such as loss-aversion. indeed, lose averse individuals are more likely to weigh less the costs from following knowledge facts, and betraying their groups loyalties. another important point that the book makes is that intelligence ('defined as the ability to accomplish complex goals'), contrary to one would expect, makes people more (rather than less) 'competent knowledge resisters'. this is epitomized by the goal of 'winning an argument', which often overrides 'getting the facts right'. ignorance can be purposefully chosen. for instance, individuals might prefer not know the gender of their children before birth, or their personal genetic predisposition to some life threatening condition. ignorance in this context, is consistent with a growing literature about why people prefer to follow non-knowledgeable beliefs, or what is known as 'rational irrationality' (caplan, 2001a) . when learning does not bring expected pay-offs (e.g., kills the hopes of a preferred child gender or, of a disease free future), it might be rational to develop some form of non-rational beliefs. this can explain the way people vote (caplan, 2001b) , or the formation of beliefs about new technologies individuals don't know much about (costafont & mossialos, 2006) . in the context of life under covid-19, it can explain the spread of fake news on the effects of the pandemic, or the consequences on the shortages of essential products. finally, an important lesson the book refers to recurrently refers to people's sensitivity to framing biases. for instance, when knowledge is specifically pitched to different cultural values (e.g., climate change can be presented as way to keep tradition among conservatives, and as an intergenerational equity matter among progressives), the message has more chances to be internalised . hence, framing can help avoiding conflict between what the author calls 'social', and 'substantive' rationality. however, the book does not discuss how such reframing and cultural adaptation should take place, and perhaps does not stress enough its complexities. i would argue, it is the role of a behavioral economist to study what choice architecture supports knowledge acquisition, given individual specific contextual constraints. rational irrationality and the microfoundations of political failure rational ignorance versus rational irrationality the public as a limit to technology transfer: the influence of knowledge and beliefs in attitudes towards biotechnology in the uk attitudes as an expression of knowledge and "political anchoring": the case of nuclear power in the united kingdom thinking, fast and slow the birth of tragedy and the genealogy of morals 81 doubleday although the book does not attempt to have the last word of what stands behind knowledge resistance, it portrays a number of implicit policy recommendations. i would highlight three which are naturally presented in the last chapter, namely:• knowledge is socially produced, and collaboration between 'moral tribes' is needed to minimize knowledge resistance.• ignorance can be creative, the views of the 'newcomers' to a field can enhance knowledge building, and ultimately expand our freedom.• as a society, we should reward. both socially. and financially, the collaboration between individuals upholding different social values and cultures, especially in the process of knowledge production.this presents a 'bigger picture' about why, and how human beings 'resist knowledge', and what to do about it. the interdisciplinary nature of the book inevitably comes with limited precision, and will not satisfy disciplinary scholars . but it is an excellent contribution to the understanding of the social barriers to knowledge transfer, which is a question no discipline on its own can solve. supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.socec.2020.101540. key: cord-350270-rcft3xfh authors: hulme, mike; lidskog, rolf; white, james m.; standring, adam title: social scientific knowledge in times of crisis: what climate change can learn from coronavirus (and vice versa) date: 2020-05-28 journal: wiley interdiscip rev clim change doi: 10.1002/wcc.656 sha: doc_id: 350270 cord_uid: rcft3xfh nan of knowledge (e.g., certain disciplines within natural sciences) are seen as inherently superior, rather than as complimentary. such a diversification of knowledge would benefit both the effectiveness of decisions made, as well as the legitimacy of those decisions among publics. there is widespread misunderstanding on the part of publics, propagated by dominant political (and scientific) discourses, that scientific and technical knowledge can provide clear and unequivocal answers to policy problems. this obscures the reality that political decision-making is rooted in careful consideration of a variety of options and tradeoffs-in jasanoff's terms, "science will not come on a white horse with a solution". 1 these discourses are already prominent in normal periods, but become even more so in times of crisis when uncertainty, the risk of policy failure and associated blame are at stake. by obscuring the "messy" nature of knowledge production and its relation to policy-making, the danger is that the expertise on which policies are based will become delegitimized whenever those policies are seen to fail. this is not a call to abandon or dilute expertise. rather it is a call to strengthen the public standing of expertise through pluralizing it. the most recent report from the intergovernmental panel on climate change (ipcc) claims that rising global temperature changes the likelihood and distribution of local, regional and global disease outbreaks. this is due to a number of factors, including the expanding geographical range of some disease vectors (ipcc, 2018:180) and the changing seasonality and transmission intensity of some infectious diseases (semenza & menne, 2009 ). medical, public health and epidemiological knowledge needs to be complemented with rigorous and varied social scientific knowledge on the fundamental social causes behind mobilities and movements of populations and products that act as vectors of disease. social science has the analytical capacity to show why the risks associated with these movements are harder to govern, the multilevel and transboundary nature of the risks, and the strengths and weaknesses of global institutions of governance that deal with risk. national and local responses to pandemic disease similarly vary widely. 2 so do the vulnerabilities of populations and their faculties for resilience, adaptation and mitigation of risk. social scientific knowledge adds to our understanding of these differences. it also grants decision-makers, as well as publics, critical capacity to evaluate the trade-offs, whether political, economic or social, that responses will entail. numbers should not be allowed to substitute for or obscure political judgment. the covid-19 pandemic has shone light on a scientific discipline that, in an age of improving sanitation and public health systems, we rarely see entering the public sphere, at least in the global north: epidemiology. in particular, epidemiological mapping and statistical modeling have ameliorated an unknown and unpredictable situation for governments across the world (rhodes, lancaster, & rosengarten, 2020) . governments always have to weigh up different forms of knowledge and expertise, from scientific and technical knowledge to policy and political considerations. in normal times, this is stage-managed to show a consolidated and consistent position. but in periods of crisis-when information is more uncertain, susceptible to rapid change and when more attention is given to decision-making processes-it becomes increasingly difficult to ensure the effectiveness of government interventions. this was witnessed recently in the cases of both the uk 3 and sweden, 4 where the respective governments' positions and advice have been subjected to increased scrutiny and criticism. climate change too has seen mathematical modeling take a prime position in the search for authoritative knowledge in the context of deep uncertainty (wynne, 2010) . in the epistemic battle fought in the early days of the ipcc's scientific assessments, it was mathematical models of the climate system that won out over more experimental or observational methods of generating relevant knowledge (coen, 2018) . it has been argued that such models provided the objectivity and authority that policymakers craved in order to legitimatize decisions (oppenheimer et al., 2019) . such mathematical models of reality-whether of the climate system or a pandemic-are always partial. not all physical processes are known or simulated. and many of the social systems which condition their predictions are either excluded or else the assumptions upon which human behavior is simulated remain hidden from view. apart from offering inevitably uncertain predictions, mathematical models also obscure the social nature of the climate risk being faced (wynne, 2010) . this has also been the case in the initial stages of covid-19 in which uncertain or incomplete data, habitual thinking and the desire to conform to political sensibilities has led to diverging estimates of the risk the virus posed (qi, du, liu, zhao, & dong, 2020) . parallels between the framing of scales for covid-19 and climate change have already been drawn. the claim of dr michael mcbride, the chief medical officer of northern ireland, that "viruses don't recognise borders" 5 echoes a similar refrain about air pollution and greenhouse gases from those who wish to emphasize the global character of climate change. the spread of the pandemic, and the spatially differentiated responses, emphasizes the importance of local and contextual knowledge. this goes beyond comparative analysis of health systems-although this is of great importance. it embraces questions of state power, organizational uncertainty and trust in institutions in order to better understand the relative effectiveness of specific interventions in different circumstances. central to these questions is an appreciation of varying cultural and political values and how they interact with scientific or medical knowledge. equally, pleas to "listen to the experts" have emerged in response to the avalanche of misinformation regarding both the covid-19 pandemic and climate change. but which experts and what advice? when does valid dissent become obstructionist denial? (nemeth, 2018) . social science helps illuminate the problems inherent to knowledge legitimacy, epistemic authority and competing truth claims. as with climate change, the impact of covid-19 is likely to be as far-reaching in its secondary consequences as in its primary ones. with stock markets falling, transport links suspended, unemployment reaching record levels 6 and the prospect of global recession looming, the economic and social ramifications appear to be as salient as the medical ones. the pandemic serves to highlight the interrelated nature of environmental, health, social and economic issues. there are multiple dimensions to the risk being faced; trade-offs are inevitable. the pandemic also focuses attention on the variable power available to different states, regions, groups and individuals to adapt and mitigate crises. a pertinent example is the different levels of social and economic capital available to people that are self-isolating. social scientific knowledge deepens our understanding of how perceptions of risk, fear and trust impact on crisis mitigation. it also illuminates the importance of structural factors, social positioning and cultural belongings-such as wealth, race and gender-in developing responses at the individual and institutional level, and how this provides different opportunities for and constraints on action. crises exacerbate existing vulnerabilities and inequalities, both among populations and between states (rhodes et al., 2012) . the covid-19 pandemic has already highlighted an uneven distribution of social and institutional support networks. social scientific knowledge will prove invaluable in understanding the long-term consequences of quarantine, social distancing and isolation on the mental and physical well-being of populations and vulnerable groups. this has proven to be the case in scholarship on previous epidemics, such as hiv/aids (emlet, 2006) . in the current pandemic we can already see that socio-economic class poses different levels of risk, both in terms of contracting the virus in the first place and then the subsequent prognosis for the infected. 7 the search for a technical fix to reduce the numbers infected-typically through various forms of social distancing and quarantine-is rarely sensitive to the (in)capacity of many people to adhere to such requirements, dependent as they are on leaving their house to work. one of the most important lessons to be learned from covid-19 is a fuller appreciation of how framings such as "crisis" and "emergency" mobilize, legitimate and yet also constrain certain forms of action. both carry with them medical connotations. the origin of the word "crisis" can be traced back to classical greek, kρίσις, meaning a moment of decisive intervention, in the medical sense a choice between life and death (koselleck, 2006) . the covid-19 pandemic has seen unprecedented state intervention in society and the economy, from locking down and quarantining entire populations, tenacious surveillance of the individual, banning free assembly and travel, and placing a moratorium on debt repayments. on the one hand, this undermines arguments that have been made in relation to climate change that such decisive interventions are either impractical or impossible. on the other hand, it will be necessary for social scientists to analyze and understand the differing political, social and cultural conditions that made these types of intervention possible, not to mention whether the corporate and state-led incursions into our private lives are necessary or desirable. how did different population groups accede to or resist such impositions? what was the long-term political fall-out of the suspension of normal decision-making mechanisms? how was political accountability either maintained or eroded during the crisis? these questions will play out differently in different political cultures under different types of governance regimes. this again highlights why one-size-fits-all technical solutions are inappropriate and why a more plural knowledge paradigm, inclusive of social-scientific knowledge, is necessary. an important distinction to make between covid-19 and climate change concerns temporality. the impact of the pandemic appears clear and immediate, while the impact of climate change is diffuse, variable and uncertain. the underlying drivers of climate change are much more deeply rooted in global economic, technological, cultural and political structures than are those for covid-19. deploying a crisis/emergency frame can help to attract attention, concentrate resources and provide public legitimacy for action. but it can also serve to obscure the hidden and inertial causes of climate change (asayama, bellamy, geden, pearce, & hulme, 2019) . the scientific expertise deployed in response to the pandemic seeks to return us to the status quo ante, whereas climate knowledge recognizes the inherent unsustainability of the underlying conditions, suggesting that things can never be the same again. crises highlight how important knowledge is to politicians and policymakers. but they also bring into focus some of the underlying tensions in this relationship between science and policy. technical and scientific knowledge is always partial, uncertain and often contradictory-as we see particularly well in the case of mathematical modeling. that is not to say that such knowledge is not valuable. it is rather to say that to effectively deal with crises, multiple forms of knowledge and expertise are required and political judgment is then necessary to sort, select and present it to publics. resolving a crisis is never about just getting the numbers right. not for climate-it is not just about net-zero -and nor for covid-19-it's not just about minimizing some arbitrary mortality statistic. it is about providing effective and ultimately trustworthy transformative change that is grounded on the broadest base of knowledge and to which normative judgments are applied. the effectiveness and legitimacy of interventions in a crisis is reliant on well-informed, transparent but subjective processes of decision-making. it is not sufficient to present the interpretation of knowledge as automatic or to restrict the expertise that presents, interprets and discusses relevant knowledge. decision-makers must be attuned to the insights that social sciences can reveal about the reasons for different individual and collective human behaviors in the face of a threat-their beliefs, values, cultures, norms, expectations and trust. it is vital for the public to retain its trust in the expertise which informs decisions, but also to understand the normative judgments which then guide future policy. this is true for dealing with covid-19 and it is true for dealing with climate change. 4 the guardian, march 30, 2020. https://www.theguardian.com/world/2020/mar/30/catastrophe-sweden-coronavirusstoicism-lockdown-europe [accessed on may 8, 2020]. 5 https://www.standard.co.uk/news/uk/northern-ireland-confirms-first-case-of-coronavirus-a4373751.html [accessed on may 8, 2020]. 6 https://news.un.org/en/story/2020/04/1061322 [accessed on may 8, 2020]. 7 brookings institute march 27, 2020. https://www.brookings.edu/blog/up-front/2020/03/27/class-and-covid-how-theless-affluent-face-double-risks/ [accessed on may 8, 2020]. why setting a climate deadline is dangerous changing the intellectual climate climate change and the quest for understanding an examination of the social networks and social isolation in older and younger adults living with hiv/aids meet the humanities special report: global warming of 1.5 c who speaks for the future of earth? how critical social science can extend the conversation on the anthropocene in defense of troublemakers: the power of dissent in life and business discerning experts: the practices of scientific assessment for environmental policy the variability of critical care bed numbers in europe a model society: maths, models and expertise in viral outbreaks climate change and infectious diseases in europe strange weather, again. theory, culture & society experts' conservative judgment and containment of covid-19 in early outbreak the authors have declared no conflicts of interest for this article. adam standring: initiator of the idea for this commentary, intellectual lead and corresponding author; rolf lidskog: contribution to drafting; james white: contribution to drafting; mike hulme: contribution to drafting and final editing. key: cord-301085-tk2vvxj7 authors: askarian, mehrdad; mclaws, mary-louise; meylan, marysia title: knowledge, attitude, and practices related to standard precautions of surgeons and physicians in university-affiliated hospitals of shiraz, iran date: 2006-07-11 journal: int j infect dis doi: 10.1016/j.ijid.2006.01.006 sha: doc_id: 301085 cord_uid: tk2vvxj7 objective: to measure levels of knowledge, attitudes, and practice toward standard precautions (sp) in medical practitioners of shiraz university of medical sciences affiliated hospitals in iran. method: in this cross-sectional study, knowledge, attitude, and practice related to sp among four medical staff groups – surgeons, surgical residents, physicians and medical residents – were assessed using a questionnaire. results: across the four medical staffing groups the median levels of knowledge ranged from 6 to 7 (maximum score 9), median attitude scores were high ranging from 35 to 36 (maximum score 45), while median practice scores were low, ranging from 2 to 3 (maximum score 9). a moderate relationship between knowledge and attitudes was found in surgical residents and medical residents (r = 0.397, p = 0.030 and r = 0.554, p = 0.006, respectively). no significant correlation was found between knowledge and practice between the groups. a significant but poor (r = 0.399, p = 0.029) relationship between attitude and practice was found in surgical residents. conclusion: specific training programs may have to target newly graduated medical practitioners to establish acceptance of appropriate practices that will enable them to adopt and adhere to sp while their older counterparts may require more intense continuous assistance. objective: to measure levels of knowledge, attitudes, and practice toward standard precautions (sp) in medical practitioners of shiraz university of medical sciences affiliated hospitals in iran. method: in this cross-sectional study, knowledge, attitude, and practice related to sp among four medical staff groups -surgeons, surgical residents, physicians and medical residents -were assessed using a questionnaire. results: across the four medical staffing groups the median levels of knowledge ranged from 6 to 7 (maximum score 9), median attitude scores were high ranging from 35 to 36 (maximum score 45), while median practice scores were low, ranging from 2 to 3 (maximum score 9). a moderate relationship between knowledge and attitudes was found in surgical residents and medical residents (r = 0.397, p = 0.030 and r = 0.554, p = 0.006, respectively). no significant correlation was found between knowledge and practice between the groups. a significant but poor (r = 0.399, p = 0.029) relationship between attitude and practice was found in surgical residents. conclusion: specific training programs may have to target newly graduated medical practitioners to establish acceptance of appropriate practices that will enable them to adopt and adhere to sp while their older counterparts may require more intense continuous assistance. # 2006 international society for infectious diseases. published by elsevier ltd. all rights reserved. a heightened understanding of transmission of blood-borne diseases in the mid-1980s [1] [2] [3] [4] [5] [6] to healthcare workers (hcws), including surgeons, physicians, and residents in training, and the importance of adherence to standard precautions (sp) is well accepted. adherence to sp is even more important with the emergence of infectious diseases, such as avian influenza, severe acute respiratory syndrome, and the threat of bioterrorism. 7 the problems of containing drugresistant organisms such as methicillin-resistant staphylococcus 8-10 and vancomycin-resistant enterococci from colonizing patients give a continuous reminder to hcws that adherence to sp is also pivotal to patient safety in terms of healthcare-associated infections. in 1996, the centers for disease control proposed guidelines for isolation precautions in hospitals, as new, two-tiered best practice of infection control precautions that are standard for all patients who are to be regarded as potential carriers of pathogenic microorganisms. 1,10 strict adherence to sp guidelines is necessary to prevent exposure to potentially life-threatening infections, 3,11-15 yet a high level of compliance with sp has been reported to be problematic worldwide. 3, 4, [16] [17] [18] [19] [20] medical practitioners, especially surgeons, are among high-risk healthcare workers for exposure to blood-borne or other infections during direct patient contact. [2] [3] [4] the purpose of our study was to measure the level of knowledge, attitude, and practice in surgeons and physicians in shiraz university of medical sciences affiliated hospitals. a cross-sectional survey was conducted in shiraz university of medical sciences, shiraz, iran between may and november 2003 of four groups of medical staff; the questionnaire was to be answered by physicians, surgeons, surgical residents, and medical residents. the questionnaire was prepared by an infection control expert, a pediatrician certified in infectious diseases, and a psychiatrist, and reviewed by experts from the iranian national expert group of infection control specialists. it consisted of questions on knowledge, attitude, and practice of the guidelines with respect to standard isolation precautions as described by the cdc. the questionnaire was pre-tested on a random sample of participants to ensure practicability, validity, and interpretation of responses. the validity of the questionnaire was assessed using the kuder-richardson test for reliability and cronbach's alpha internal consistency coefficient. items in the questionnaire included demographic data, specialty and status of medical practitioner (surgeon, physician, surgical, or medical resident), previous sp education, willingness to be trained, and nine questions pertaining to hand-washing, personal protective equipment, m. askarian et al. use of antiseptic solution, and disposal method for used syringes. responses to items for knowledge were ''yes'', ''no'', or ''don't know''. the questions used to assess attitude were in the format of the likert scale with responses that included ''very strong'', ''strong'', ''considerable'', ''weak'', or ''null'' and the five-point likert scale response for practice questions (always, often, sometimes, seldom, never). all responses in accordance with cdc guidelines 1,10 were given a score value of 1 for correct answers to the knowledge questions and when answers for practice questions were ''always'', while a score zero was assigned to all other answers. the total scores ranged from zero to 9. for attitude questions, a score of 5 was equivalent to the answer ''very strong'' and a score of 1 to ''nil'', therefore, the total score ranged from 9 to 45 ( table 1 ). the questionnaire was pre-tested on 21 randomly-selected members from the target population with high test-retest reliability (alpha = 0.73). descriptive and inferential statistics including significance tests, wilcoxon rank-sum test, kruskall-wallis test, and spearman correlation coefficient were performed using spss version 10.0. alpha was set at the 5% level. of the 250 questionnaires distributed, 155 (62%) were returned completed. participants included 78 senior medical staff (42 surgeons and 36 physicians) and 77 residents (41 surgical residents and 36 medical residents) ( table 2 ). there were more male, 73.8% ( p = 0.002) than female surgeons, while there was no significant ( p = 0.061) gender difference within the physicians, 54.8% males. neither were there significant differences in the proportion of male surgical residents (56.1%, p = 0.435) and male medical residents (58.3%, p = 0.317). the median age of all senior medical staff was 34 years (range 26-70 years) and 31 years (range 24-39 years) for all residents ( table 2) . regardless of the status of medical practitioners, the majority (85.9% senior practitioners and 87.0% residents) reported no previous formal sp education and most (87.2% senior practitioners and 88.3% residents) were willing to receive sp training ( table 2) . the median scores for sp knowledge for all medical practitioners ranged from 6 to 7, while the range of median scores for attitudes was from 35 to 36 and that for practices from 2 to 3 ( table 3 ). the median scores for knowledge and attitude were not significantly different ( p = 0.077 and p = 0.653, respectively) between surgeons (knowledge median 7.0 and attitudes median score 36) and physicians (knowledge median score 6.0 and attitudes median score 35); median scores for practices were less than half the possible total score of nine, although the median scores for surgeons, 3.0, and physicians, 3.0, were equal but differed significantly ( p = 0.036) ( table 3) . although median scores for knowledge and attitudes were moderate to high, surgeons were the only group where a knowledge, attitude, and practices related to standard precautions 215 moderate to strong (r = 0.748) relationship between knowledge and attitudes was significant ( p < 0.0001) while for other medical groups this relationship was poor ( table 4 ). the relationship between knowledge and practices for all practitioners was not significant, as was the case for attitudes and practices, where correlations were not significant for all practitioners except surgical residents, where the relationship was significant but not strong, r = 0.399 ( p = 0.029) ( table 4 ). all medical practitioners' levels of knowledge were high for the same six items that were answered correctly by at least three quarters of practitioners within each group (table 5) . few medical practitioners answered correctly that needles should not be bent before disposal with correct knowledge ranging from 27.8% (physicians) to 55.6% (medical residents). as few as 9.5% of surgeons and 9.8% of surgical residents correctly believed that betadine was not used as a hand washing antiseptic solution. less than three quarters of practitioners correctly knew that they had to wash hands before glove use with proportions of practitioners answering correctly ranging from 52.8% (physicians) to 70.7% (surgical residents). only two of nine attitudinal items were answered as ''very much'' by more than 75% of practitioners (table 6 ). these two items indicating good attitudes related to sp activities do not require effort or much persuasion; the proportion of practitioners washing hands after touching blood, body fluids, excretions and contaminated items ranged from 91.7% to 95.2%, and glove use for touching mucous membranes, proportions ranged from 69.4% to 95.2%. only one sp practice item, hand washing after touching contaminated items, was always practiced by 75.6% to 100% of practitioners (table 5) . a medical resident's knowledge towards sp was related to willingness to be trained ( p = 0.013) and a surgeon's practice of sp was related only to female gender ( p = 0.006). no other significant predictive relationships were found. m. askarian et al. adherence to standard precaution guidelines has been problematic for hcws universally. [2] [3] [4] this study was limited by its reliance on self-reporting rather than by observing compliance with sp by practitioners. however, self-reported attitudes towards sp and compliance were low. our poor compliance with many sp practices is not unique to our teaching hospitals, with similar results published in other centers. 3, 4, [17] [18] [19] [20] [21] while principles of sp should be strictly observed by surgeons and physicians for their own safety as well as setting leadership roles for their residents, success in the implementation of sp guidelines depends on many factors such as optimal awareness and a positive attitude in all hcws, and these goals are not reached without qualified personnel receiving continuing regular education, as has been repeatedly demonstrated in the literature. 4, [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] our study revealed that more than 80% of all medical practitioners had not received previous sp education, and that more than 80% were willing to be trained. these findings illustrate that sp practices are not behaviors readily adopted, even by those aware of the sp issue and moderate to low levels of attitudes. a revision of current medical curricula offered in iran may be required that mandates all medical students and hospital trainees attend infection control courses specific to their clinical terms. although all our medical practitioners reported a willingness to be trained, compliance behavior is complex. 33 some barriers to adherence observed internationally include inadequacy of equipment and facilities and difficulties of access to equipment, stressful working conditions, and the belief that practice of sp may interfere with patient care. 2, 23, [34] [35] [36] [37] [38] [39] [40] [41] influencing the iranian senior practitioner may require elements of other successful programs knowledge, attitude, and practices related to standard precautions 217 and innovative approaches, such as improved hospital infrastructure that offers an infection control awareness week with infection control auditors to inform about correct practices at the time incorrect practices are observed, freely available personal protective equipment, such as gowns, gloves and masks, with reinforcing posters but also staff seminars launched on a platform of the importance of consistent sp practice for personal protection in times of emerging diseases. systematic review of adherence to infection control guidelines in dentistry noncompliance with universal precautions and the associated risk of mucocutaneous blood exposure among danish physicians nurses' knowledge of and compliance with universal precautions in an acute care hospital knowledge and performance of the universal precautions by nursing and medical students in korea management of healthcare workers infected with hepatitis b virus, hepatitis c virus, human immunodeficiency virus, or other bloodborne pathogens. aids/tb committee of the society for healthcare epidemiology of america frequency of glove perforations and subsequent blood contact in association with selected obstetric surgical procedures knowledge, perceptions, and practice of nurses toward hiv+/ aids patients diagnosed with tuberculosis can antibioticresistant nosocomial infections be controlled? revised uk guidelines for the control of methicillin-resistant staphylococcus aureus in hospitals standard precaution--a new approach to reducing infection transmission in the hospital setting double gloving protecting surgeons from blood contamination in the operating room efficacy of double gloving as a protection against blood exposure in dermatologic surgery use of double gloves to protect the surgeon from blood contact during aesthetic procedures double gloving as self protection in abdominal surgery subjective effects of double gloves on surgical performance orthopedic trauma surgeons' attitudes and practices towards bloodborne pathogens effect of education on the use of universal precautions in a university hospital emergency department compliance with universal precautions in a university hospital emergency department survey on adoption of measures to prevent nosocomial infection by anesthesia personnel compliance with universal precautions among emergency department personnel: implications for prevention programs factors associated with compliance of critical care nurses with universal precautions: a pilot study compliance with universal precautions among physicians the infection control practices of general dental practitioners nurses' compliance with universal precautions before and after implementation of osha regulations universal precautions training of preclinical students: impact on knowledge, attitudes, and compliance multidrug-resistant bacteria infection control: study of compliance with isolation precautions in a paris university hospital universal precautions: improving the knowledge of trained nurses orthopedic surgery residents and the cdc and aaos hiv precautionary measures hiv-related knowledge and precautions among michigan nurses a training program in universal precautions for second-year medical students improved compliance with universal precaution in the operating room following an educational intervention compliance with recommended infection control procedures among canadian dentists: results of a national survey the prevalence of and factors related to, compliance with glove utilization among nurses in hospital univeriti sains malaysia testing the reliability and validity of a measure of safety climate compliance with universal precautions: knowledge and behavior of residents and students in a department of obstetrics and gynecology compliance with universal precautions by emergency room nurses at maharaj nakorn chiang mai hospital risk taking by healthcare workers variables influencing worker compliance with universal precautions in the emergency department the use and failure rates of protective equipment to prevent blood and bodily fluid contamination in the obstetric healthcare worker compliance with handwashing and barrier precautions this article is the result of a research project (81-1412) approved by the deputy of research of shiraz university of medical sciences and health services. we want to express our special appreciation to the respected research deputy who not only provided financial support, but also showed his interest in solving difficulties.conflict of interest: no competing interest is declared. key: cord-321258-fj9nel2z authors: huynh, giao; nguyen, minh quan; tran, thien thuan; nguyen, van tap; nguyen, truong vien; do, thi hoai thuong; nguyen, phi hong ngan; phan, thi hoai yen; vu, thanh thuy; nguyen, thi ngoc han title: knowledge, attitude, and practices regarding covid-19 among chronic illness patients at outpatient departments in ho chi minh city, vietnam date: 2020-09-14 journal: risk manag healthc policy doi: 10.2147/rmhp.s268876 sha: doc_id: 321258 cord_uid: fj9nel2z background: the novel coronavirus disease (covid-19) has become a major threat to human life around the world. this study aims to assess the knowledge, attitude, and practices regarding covid-19 among people with chronic diseases at the outpatient departments in ho chi minh city. methods: a cross-sectional study was carried out between february and march 2020 using a convenience sampling strategy in three hospitals in ho chi minh city (hcmc) via the use of a structured self-administered questionnaire. factors relating to practices, prevalence ratio (pr), and 95% confidence interval were estimated by using the poisson regression with robust options. p-value <0.05 was considered as statistically different. results: a total of 522 participants had a mean age of 51.5 ± 10.6 years. most of them reported seeing information regarding the covid-19 pandemic (93.7%) via television and social media (72.8% and 62.1%, respectively). just over two-thirds of the participants (68.4%) answered with sufficient knowledge of covid-19. most respondents had a positive attitude toward covid-19 (90.8%), although some misconceptions existed. almost over three-fourths of them (77.2%) maintained good practices for prevention. the rate of good practices in those who had sufficient knowledge was 1.24 times greater than that among those who had insufficient knowledge (pr 1.24, 95% ci: 1.10–1.41, p<0.05). also, the rate of good practices in males was lower than that of females (pr: 0.91, 95% ci: 0.83–0.99, p<0.05). conclusion: there still exists an amount of insufficient knowledge and negative attitude regarding covid-19, which may be barriers to good prevention practices among chronic illness patients. education programs need to continue via television and social media and emphasize that people with chronic diseases are more likely to experience severe symptoms, including death from covid-19. additionally, management authorities should prolong specific policies to protect the more vulnerable in our community. the covid-19 was first reported in wuhan, china in december 2019, which has been reported as an emerging respiratory infection, with the number deaths exceeding other well-known viruses such as severe acute respiratory syndrome (sars) and middle east respiratory syndrome (mers). the impact of the covid-19 pandemic has caused an excessive loss of life and hardship, and become a major threat to human life around the world. [1] [2] [3] in the first half of 2020, the number of cases grew exponentially in many countries with infected cases having been recorded almost everywhere, including hospitals and in the community. 4 by 21 july 2020, the world health organization (who) recorded 14,562,550 confirmed cases leading to 607,781 deaths worldwide, with the highest number of confirmed cases and deaths recorded in the americas and europe. 1 clinical symptoms of infected persons have been reported to include cough, fever, and shortness of breath, with the average incubation period of 5-6 days, but can be up to 14 days. 5 in terms of transmission, who has stated that touching contaminated droplets on surfaces or objects, or close contact with an infected person, within about 1 m (3 feet), can facilitate transmission between humans. 6 the challenge with this virus is that many victims have reported mostly mild symptoms. this has contributed to the threat to public health because the detection of infected persons can be very difficult in the early stages. some sufferers could develop more serious symptoms with a small number resulting in death. since the first case was reported at the end of january 2020, vietnam has recorded 396 confirmed cases across the country but most of the cases seemed to be centered around the two big cities of hanoi and ho chi minh city. 7 due to the management of the virus, vietnam is becoming a noted country having successfully isolated the covid-19 pandemic via rapid response, clear leadership, and support by clinical care and public health response. 8 at the beginning of the pandemic, all the suspected cases were isolated for 2 weeks. all were tested and followed, as well as the country having mobilized teams to respond and control the spread of covid-19 in the community. in particular, people with chronic diseases were advised to stay home and practice social distancing. in some big cities including ho chi minh city, doctors and nurses checked the condition of these patients at their homes and promoted online medical advice for older patients or they were examined and received quality treatment at the hospital once every 2 months instead of once a month. 9 it is widely noted that people over 60 years old with underlying noncommunicable diseases (ncds) like diabetes, heart, or lung disease, were more likely to be a high risk of severe disease and death. 10 currently, there is no recommended effective treatment with vaccine development still in progress. thus, it is necessary to perform health education interventions aimed towards preventive measures and how to avoid being exposed to the virus in the community, especially for people with chronic diseases. 11 some previous studies about infectious diseases such as sars, mers showed knowledge and attitudes were associated with the level of practices to prevent the spread of disease. 12, 13 however, some findings showed a lack of understanding about this pandemic and available processes to prevent transmission. 14,15 therefore, assessing the knowledge, attitude and practices toward covid-19 plays an important role, especially in patients with a chronic illness, which will help to develop the effective educational interventions to ensure safe and quality care while stopping onwards viral transmission. a cross-sectional study was conducted between february and march 2020. a total of 522 people with chronic diseases such as diabetes and hypertension at outpatient departments in three hospitals in hcmc including district 2, thu duc, and binh thanh were recruited for this study. a convenience sampling strategy was considered on all the people with chronic diseases who have been diagnosed by a physician and more than 18 years old. all participants were informed regarding the objective of the study and completed the consent form before participating. the questionnaire was designed by using frequently asked questions regarding covid-19 from the who webpage, and ahmed m. asaad's questionnaire about the middle east respiratory syndrome coronavirus (mers-cov), and our questionnaire about covid-19 among healthcare workers. [16] [17] [18] firstly, a pilot study was done to test 10 subjects at the outpatient department at district 2 hospital to assess comprehension and accessibility of the questionnaire, with no necessary reformation recognized from participants. the final structured questionnaire was defined by the authors. data were collected via the use of a structured self-administered questionnaire, which included four main domains. the first section elicited information on demographic characteristics including age, gender, occupation, education, and the source information of covid-19 knowledge including types of face masks. the second section was designed to assess participants' knowledge of covid-19 including a set of 9 yes/ to assess the knowledge score, nine items of knowledge on the questionnaire were measured, each correct answer was counted as 1 point with no points allotted for an incorrect answer. the general knowledge score was recorded from 0 (no correct answers) to 9 (all correct answers), and the cut-off level of ≥6 was set as the level of sufficient knowledge. 19 to assess 6 items of attitude, the scores were based on 5 point likert scales, ranged from strongly agree to strongly disagree. for each attitude item, the point was assigned into 1 point (strongly agree and agree) and 0 points (strongly disagree, disagree, undecided), the cut-off level of total attitude items ≥4 indicated as a positive general attitude. response questions for 3 items of the practices ranged from 0 (no correct answers) to 3 points (all correct answers), a total of ≥2 points for good practices. the data were calculated using stata 13 and epidata 3.1 software. factors with practices, prevalence ratio (pr), and 95% confidence interval (95% ci) were estimated by poisson regression with robust options. all factors which had a significance level p-value <0.20 in the bivariate analysis were included in the multivariate analysis by the poisson regression model. p-value <0.05 was considered as statistically different. our research complied with the declaration of helsinki. all the eligible participants were informed about the aims of the study, they signed a consent form before participation. the protocols of the study were approved by the ethics council of university of medicine and pharmacy at ho chi minh city (protocol number 162/ump -board). as shown in table 1 , a total of 522 (87.0%) participants completed and returned the questionnaire from the outpatient department of the three hospitals. some of the main characteristics of participants were mainly in the mean age of 51. 5 table 3 that 77.2% of participants reported having good practices. however, only 40.4% of respondents knew that the average time for washing hands was at least 20 s. as table 4 showed, there was a significant correlation between knowledge, gender, and practices, whereby the rate of good practices in those who had sufficient knowledge was 1.24 times greater than that among those who had insufficient knowledge (pr 1.24, 95% ci: 1.10-1.41, p<0.05). also, the study showed that the rate of good practices in males was lower than that of females (pr: 0.91, 95% ci: 0.83-0.99, p<0.05). this is the first study of its type to assess the knowledge, attitudes, and practices of people with chronic diseases in vietnam, and it was performed at the first stage of the pandemic. in the beginning, vietnam applied a multitude of strict measures to prevent covid-19, this included installing strict social distancing in the first half of april 2020. 8 as of july 2020, the spread of covid-19 is continuing to rapidly increase in many countries around the world and vietnam cases are continuing to grow with the support of the healthcare environments in vietnam. 1, 8 therefore, there is an immediate need for effective prevention through enhancing knowledge, attitude, and practices by high-risk persons in the community. 8 the characteristics of participants in our study were mainly female (55.7%) and educated (61.9%) to high school level. we found most of the participants reported seeing information regarding the covid-19 being a global outbreak (93.7%) with the main sources being television and social media (72.8% and 62.1%) while only 34.3% of information was attained via hospital websites or the health ministry. these results were also described in our previous 12, 20 because our survey was performed in the early period of the pandemic in ho chi minh city, numbers of reported cases were still very low in the local community. this meant that educational programs were not highly developed, to offer information on the dangers of the covid-19. however, china had been much further into the pandemic, having stated in 2019, so the rate of sufficient knowledge was higher in our findings compared to studies of other viruses. our study found that only 62.1% and 60.9% of respondents knew the signs and symptoms as well as the prevention measures through washing hands and wearing a face mask while 44.1% of participants had good knowledge of the required treatment for the virus. therefore, as highlighted by the cdc, the provision of information to the community about the virus, its transmission modes, and necessary protective measures played a vital role in controlling the pandemic. 11 additionally, there were only 43.5% and 63.2% of respondents that were aware of the isolation period and the high risk of severe infection among people with underlying chronic diseases. people with chronic diseases in vietnam are often examined and received quality treatment once a month at the hospital. these patients may be at risk of contracting the virus in health-care environments, with their lack of knowledge on the risk of infection and severe symptoms. therefore, the government needs to reconsider isolating at-risk patients at home, which seems a much safer process, rather than exposing them, on a daily basis, to the hospital environment, as well as maintaining social distancing by people with chronic diseases as per who's recommendation. 7, 18 according to the findings, there are the changes of policy for patients with chronic diseases, such as having medical consultations at home by doctors and nurses, encouraging them to obtain online medical advice, and changes to hospital policy which permits them to receive medication once every 2 months instead of once a month. 9 additionally, community knowledge was improved by increasing the frequency of media messages via social media and television every day to encourage people to perform preventive measures. these have successfully contributed to the control of covid-19 transmission in the community in vietnam with no confirmed cases in the community for the past 99 days as of july 23, 2020. 7 surprisingly, the vast majority of participants had a positive attitude. these results corroborated the findings of bao-liang z.'s study, 20 nevertheless, there were still 66.5% and 64.8% of participants worried about the likelihood of getting the virus or having a family member contract the virus. these findings were also similar to our study among healthcare workers toward covid-19. 18 these results may be explained by the lack of knowledge about the spread of the disease so they did not believe their family and close associates could get the disease, this was dangerous if they contracted covid-19 and became a source of transmission in the community. we found that only 40.4% of participants had good practices relating to the average time for washing hands. this also accords with akalu et al (2020), which showed a low level of knowledge and practice with only 33.9% and 47.3%, respectively. 15 the finding was likely to be related to a lack of knowledge and awareness about the importance of prevention. it was observed that knowledge was significantly associated with the practice, with participants having a good knowledge showing a higher probability of good practices. another important finding was that females had better practices than males. it seems possible that these results are due to some researches showed that males are more severely affected and have higher mortality rate from covid-19. 21, 22 therefore, educational interventions should continue and make a priority to focus on those at a greater risk, to reduce exposure in the community. another interesting finding was the vast majority of participants wearing face masks accounted for 98.3%. the majority of them wore face masks in public places (83.3%). these findings showed that participants had a positive behavior toward prevention and this is one of the important measures that may prevent the spread of covid-19 as cdc and who's recommendations. 23, 24 limitation some limitations of our study in interpreting the results because no previous studies have been performed to assess prevention practices toward covid-19; in this time, we just may collect the study samples at three hospitals in hcmc and the convenience sampling strategy. these may limit the generalization of the results with respect to the outpatient population at all the hospitals of hcmc. insufficient knowledge and negative attitudes regarding covid-19 were still found in some participants in the community, which may be barriers to good prevention practices. additional education programs need to continue via television and social media and must emphasize that people with chronic diseases are more likely to be a high risk of developing severe symptoms and death. besides, management authorities need to maintain current policies to protect them from covid-19 infection in the community and at healthcare facilities. available upon request to the first author. covid-19) pandemic summary of probable sars cases with onset of illness from 1 who mers global summary and assessment of risk clinical characteristics of hospitalized patients with sars-cov-2 infection: a single arm meta-analysis clinical management of covid-19. world health organization getting your workplace ready for covid-19 ministry of health in vietnam combating the covid-19 epidemic: experiences from vietnam social responses for older people in covid-19 pandemic: experience from vietnam covid-19 and ncds how to protect yourself attitude and practice towards sars middle east respiratory syndrome (mers): comparing the knowledge, attitude and practices of different health care workers public knowledge, attitudes and practices towards covid-19: a cross-sectional study in malaysia knowledge, attitude and practice towards covid-19 among chronic disease patients at addis zemen hospital, northwest ethiopia who. q&a on coronaviruses (covid-19). 2020. available from knowledge and attitudes towards middle east respiratory syndrome-coronavirus (mers-cov) among health care workers in south-western saudi arabia knowledge and attitude toward covid-19 among healthcare workers at district 2 hospital knowledge and attitude of healthcare workers about middle east respiratory syndrome in multispecialty hospitals of qassim, saudi arabia knowledge, attitudes, and practices towards covid-19 among chinese residents during the rapid rise period of the covid-19 outbreak: a quick online cross-sectional survey gender differences in patients with covid-19: focus on severity and mortality. front public health sex and gender differences in the outcome of patients with covid-19 interim infection prevention and control recommendations for patients with suspected or confirmed coronavirus disease 2019 (covid-19) in healthcare settings covid-19) advice for the public: when and how to use masks we wish to acknowledge the cooperation and support of outpatients at the hospital of district 2, thu duc, and binh thanh for the time and effort that they devoted to the study. all authors made substantial contributions to conception and design, acquisition of data, analysis and interpretation of data, took part in revising the article critically for important intellectual content, gave final approval of the version to be published, and agree to be accountable for all aspects of the work. there is no funding to report. the authors declare that they have no competing interests for this work. key: cord-302595-t6yd3znu authors: mechessa, desalegn feyissa; ejeta, fikadu; abebe, lemi; henok, andualem; nigussie, tadesse; kebede, oliyad; mamo, yitagesu title: community’s knowledge of covid-19 and its associated factors in mizan-aman town, southwest ethiopia, 2020 date: 2020-08-17 journal: int j gen med doi: 10.2147/ijgm.s263665 sha: doc_id: 302595 cord_uid: t6yd3znu background: coronaviruses are a large group of viruses that are common throughout the community. they are associated with mortality, hospitalization, substantial extra costs and lower patient’s quality of life. thus, this study aimed to assess the community’s knowledge of covid-19 and associated factors in mizan-aman town, southwest ethiopia. methods: community-based cross-sectional study design was conducted among the community of mizan-aman from april 14 to may 14, 2020. a systematic sampling technique was used to collect data from selected households. data were entered into epi data version 4.0.2.101 and then exported to spss version 24.0 for analysis. to identify the predictors of knowledge of covid-19, multiple backward logistic regression analysis was used. to show the accuracy of data analysis, 95% ci was used, and statistical significance was considered at p <0.05. results: from 423 sampled population, 393 (92.9%) of them responded to the questionnaire. of these, 233 (59.3%) were male, 225 (57.3%) were 18–34 years old, and 250 (63.6%) were married. the overall correct rate of the knowledge questionnaire was 74.75%. more than 85% of respondents were well aware of the main clinical symptoms of covid-19, its transmission by close contact, its prevention by not going to crowded places and isolation of infected persons. male gender (aor=3.74, ci: 1.87–7.49), age (35–54 years) (aor=3.81, ci: 1.35–10.70), age ≥55 (aor=2.97, ci: 1.16–7.62), lack of formal education (aor=6.0, ci: 1.54–23.40), farmer (aor=8.72, ci: 2.08–35.53), daily laborer (aor=7.57, ci: 2.28–25.15), merchant (aor=6.34, ci: 2.06–19.43), house wife (aor=11.59, ci: 2.91–46.23) were significantly associated with poor knowledge, whereas single marital status was less likely associated with poor knowledge of covid-19. conclusion: one-third of the study participants had poor knowledge regarding covid-19. male gender, age above thirty-five years, lack of formal education, being farmer, daily laborer, merchant and house wife were significantly associated with poor knowledge. therefore, awareness creation should be given. coronavirus (cov) infections are emerging respiratory viruses and are known to cause illnesses ranging from the common cold to severe acute respiratory syndrome (sars). 1 its clinical presentations are characterized by acute respiratory distress syndrome, septic shock, metabolic acidosis, bleeding and coagulation dysfunction. 2 the previous outbreaks of coronaviruses such as sars-cov and middle east respiratory syndrome (mers-cov) in 2003 and 2015 showed similarities to the novel coronavirus, which was first reported by chinese authorities in wuhan city, the capital of hubei province in china at the end of december 2019. it is currently the disease resulting in a worldwide outbreak. 3, 4 the virus is transmitted from humans to human via droplets coughed or exhaled by infected persons, and also by touching droplet-contaminated surfaces or objects and then touching the eyes, nose or mouth. 5, 6 the most commonly reported clinical symptom in laboratory-confirmed cases is fever, followed by dry cough, fatigue, sputum production, dyspnea, sore throat, headache with myalgia or arthralgia. 7 despite a global health crisis, there is currently no proven prophylaxis for those who have been exposed to covid-19 and treatment for those who go on to develop covid-19. however, many of the symptoms can be treated and getting early care from a healthcare provider can make the disease less dangerous. there are several clinical trials that are being conducted to evaluate potential therapeutics for covid-19. 8, 9 the battle against covid-19 is still continuing all over the world. 10 the world health organization (who) set strategies to prevent and decline the transmission of covid-19 such as frequent hand-washing with soap, avoiding close contact, staying home, covering mouth and nose with flexed elbow and maintaining social distancing. to guarantee the final success, people's adherence to these control measures is essential, which is largely affected by their knowledge of covid-19. 11, 12 to facilitate outbreak management of covid-19 in ethiopia, there is an urgent need to assess the public's awareness of covid-19 at this critical moment. therefore, this study aimed to assess the knowledge of covid-19 and its associated factors among the community of mizan-aman town, southwest ethiopia. the findings of this study will help the responsible body organize the necessary interventional programs (education, demonstration) in order to provide up-to-date information to control covid-19 disease. this study was conducted in mizan-aman town, southwest ethiopia, from april 14 to may 14, 2020. mizan-aman is a zonal town of the benchi-sheko zone. it is one of the zones in southern nation, nationalities, and people's region. it is 583 kilometers from addis ababa, the capital city of ethiopia, to the southwest direction. mizan-tepi university, mizan-aman college of health science and mizan-aman town health office disseminate health information to the residents face to face while fana fm and debub fm radios were delivering information using their stations. a community-based cross-sectional study was conducted in mizan-aman town to assess the community's knowledge of covid-19. all adults of mizan-aman town were the source population and all selected individuals who fulfilled the eligibility criteria were the study population. adults (age≥18 years) and who were permanent residents (greater than 6 months) of mizan-aman town were included in the study, whereas adults who were unwilling to respond and houses closed during data collection were excluded from the study. the sample size was determined based on the single population proportion formula: n= [(z α/2 ) 2 p (1-p)]/d 2 with the assumption of a 95% confidence interval (zα/ 2 = 1.96), marginal error (d) of 5% and p=50% with a 10% non-response rate, the required total sample size was 423. systematic random sampling was used to select sample households that represent the entire town. the interval, k, was calculated by dividing total households by 423. the sample household was selected systematically by jumping every 30 households. the first household was selected by the lottery method. from the selected household, the study participant was chosen by the lottery method if more than one eligible individual present. in this study, the dependent variable was knowledge, whereas independent variables included sex, age, marital status, ethnicity, religion, educational status, occupational status and presence of health professionals in the household. a structured questionnaire was adapted from similar literature, which contains 2 parts. part i includes respondent's background data and part ii includes knowledgerelated questions. the questionnaire was translated into the local language (amharic) by persons who were proficient in both languages. then, the questionnaire was pre-tested on a 5% of the total sample size in a kite town, which is 8 km away from mizan-aman town, and modified based on the pretest. five experienced bsc public health and two supervisors were recruited and trained for data collection and supervision, respectively. the training was given for two days on how to ensure confidentiality, tool, how to protect themselves and community from infection and interview techniques. additionally, the data collectors and supervisors adhered to the who and ethiopian ministry of health guidelines on covid-19 prevention. 13, 14 accordingly, they maintained social distance, worn personal protective equipment and used an alcohol-based hand sanitizer during the data collection period. data were collected through face-to-face interview method. the supervisors and principal investigators supervised the process of data collection on a daily basis. a knowledge questionnaire was developed from different studies. [15] [16] [17] the questionnaire had 12 questions with 3 regarding clinical presentations, 4 regarding transmission routes, and 5 regarding prevention and control of covid-19. these questions were answered on a true/false basis with an additional "i don't know" option. a correct answer was assigned 1 point and an incorrect/unknown answer was assigned 0 point. during the analysis, negatively worded items of knowledge (number 7 and 10) were reversely scored. the total knowledge score ranged from 0 to 12, with a higher score denoting a better knowledge of covid-19. the cronbach's alpha coefficient of the knowledge questionnaire was calculated to the check internal consistency of the tool, which was 0.78. data quality was assured by careful selection and collection of complete and appropriate data. the clarity and completeness of data collection formats were checked before the actual data collection. data were cleared before entered into epi data manager version 4.0.2.101. the collected data were processed and retained cautiously in line with its objective. a 5% sample pretest was performed on randomly selected residents before the beginning of the study. the data on questionnaire were entered into epi data manager version 4.0.2.101 and double entry verification was made. the data were exported to spss version 24 statistical packages for analysis. the data were explored to check for outliers, missing data and assumptions. during analysis, frequencies and percentages were used to describe categorical variables, while means and standard deviations were used to describe continuous variables. bivariate analysis was performed to select variables for multivariate analysis. variables with a p-value <0.25 in the bivariate analysis were taken as candidates for multivariable analysis. finally, a multivariable logistic model was created to predict the determinants of knowledge. p-value less than 0.05 was accepted as statistically significant variables. knowledge each correct answer in relation to the knowledge of covid-19 was given one point. the total knowledge score varied between 0 (with no correct answer) and 12 (for all correct answers), and a cut off level of <8 was evaluated as poor knowledge, and ≥8 indicated good knowledge. 15 ethical clearance was obtained from mizan-tepi university ethics review board. the mizan-aman town administration office was informed about the purpose of the study to get permission. both data collectors and supervisors adhered to the who and ethiopian ministry of health covid-19 prevention guidelines during data collection. confidentiality of the respondents was secured by excluding respondent's identifiers like name and house number from the data collection format. informed consent was obtained from the respondents before conducting the study. from a total of 423 sampled populations, 393 (92.9%) of the study population responded to the questionnaire. the mean age of the study population was 33.85 (sd±13.79) years. two hundred thirty-three (59.3%) were male, 250 (63.6%) were married, 158 (40.2%) were orthodox and 133 (33.8%) were bench in ethnicity. eighteen percent of the participants had health care professionals in their homes (table 1) . mixed responses were obtained for 12 knowledge items. the mean covid-19 knowledge score was 8.97±1.7. more than 85% of respondents were well aware of the main clinical symptoms of covid-19, its transmission by close contact, its prevention by not going to crowded places and isolation of infected persons. additionally, 84.2% of the participants responded correctly for not all persons with covid-19 will develop to severe cases, and (77.4%) of them knew as covid-19 virus spreads via respiratory droplets of infected individuals. furthermore, two-thirds of participants knew that eating or contacting wild animals would result in infection and ordinary residents can wear general medical masks to prevent infection. however, when respondents asked questions related to transmission of covid 19 in the absence of fever in infected persons, the importance of taking measures for infants and young children to prevent the infection and uncommon symptoms of covid 19 compared to common cold, 66.4%, 54.7% and 49.1% of them were unable to identify correct answers, respectively. generally, 35.4% of the population had poor knowledge (table 2) . differences in knowledge scores among different demographic characteristics were assessed using t-tests. the results show that knowledge scores were significantly different across age categories, educational status, presence of health professionals in households, occupation status and marital status ( table 3 ). sex (male), age (35-54 and >55 years), marital status (single), educational status (no formal education), occupational status (farmer, student, unemployed/housewife) were significantly associated with poor knowledge of covid-19 (table 4 ). currently, covid-19 is a global discussion topic in the media and among the public. until now, the outbreak is considered an emergency and the community has an increased risk of infection. however, up to our knowledge, there is no investigation of the community's knowledge of covid-19 and its associated factors in this study area. the findings in our study showed that the overall correct rate of the knowledge questionnaire was 74.75%. this is lower than the study done in china where the overall correct rate of the knowledge questionnaire was 90%. 15 this discrepancy might be due to the difference in technology access and educational level between the study populations. in general, in our study, 35.4% of participants had poor knowledge about the disease, its mode of transmission, prevention and control. in this study, 87.5% of the respondents knew that the main clinical symptoms of covid-19 were fever, fatigue, dry cough, and myalgia and 73.8% of them knew as currently, there is no effective cure for covid-2019, but early symptomatic and supportive treatment can help most patients recover from the infection. this is similar to the finding reported by kebede et al 18 where 83% of the respondents knew that the main clinical symptoms of covid-19 are fever, fatigue, dry cough, and myalgia and 72.1% of respondents knew that currently, there is no effective cure for covid-2019, but early symptomatic multiple logistic regression analysis of this finding showed that being male was significantly associated with poor knowledge of covid-19 as compared to females. this is similar to a study done in china and saudi arabia, where male was significantly associated with low knowledge scores. 15, 19 in our finding age above 35 years was significantly associated with poor knowledge. this is in agreement with the study done in egypt, where respondents' age older than 50 years were significantly associated with poor knowledge. 16 however, this is in contrast to a study done in china where age above 30 years was associated with good knowledge. 15 lack of formal education was significantly associated with poor knowledge. this is similar to a previous study. 15 the possible justification might be formal education providing basic infectious disease (such as the sars-cov) concept, mode of transmission and prevention. in this study, being unemployed was significantly associated with poor knowledge of covid-19. this is in agreement with a previous report finding. 15 in addition to this being farmer, daily laborer and merchants were significantly associated with poor knowledge of covid-19. however, being single in marital status was less likely associated with poor knowledge. similarly, other previous study also demonstrated that being unmarried was associated with good knowledge. 20 the possible reason might be that unmarried persons have less social and family responsibility, which reduces psychosocial burden, which in turn increases their knowledge. one-third of the respondents had poor knowledge of the covid-19 concept, transmission, prevention and control. being male, age above 35 years, lack of formal education, being farmer, daily laborer, merchant and unemployed/ house wife were significantly associated with poor knowledge, whereas being single was less likely associated with poor knowledge of covid-19. syndrome; sars, severe acute respiratory syndrome; who, world health organization. all relevant data are within the paper. the international journal of general medicine is an international, peer-reviewed open-access journal that focuses on general and internal medicine, pathogenesis, epidemiology, diagnosis, monitoring and treatment protocols. the journal is characterized by the rapid reporting of reviews, original research and clinical studies across all disease areas. the manuscript management system is completely online and includes a very quick and fair peer-review system, which is all easy to use. visit http://www.dovepress.com/ testimonials.php to read real quotes from published authors. submit your manuscript here: https://www.dovepress.com/international-journal-of-general-medicine-journal sars and other coronaviruses as causes of pneumonia novel coronavirus pneumonia emergency response epidemiology team. the epidemiological characteristics of an outbreak of novel coronavirus diseases (covid-19) in china the neuroinvasive potential of sars-cov2 may play a role in the respiratory failure of covid-19 patients public responses to the novel 2019 coronavirus (2019-ncov) in japan: mental health consequences and target populations european centre for disease prevention and control. factsheet for health professionals on coronaviruses european centre for disease prevention and control. event background covid-19 report of the who-china joint mission on coronavirus disease (covid-19). available from coronavirus covid-19 global cases by the center for systems science and engineering at johns hopkins key-messages-and-actions-for-covid-19-prevention-and-control-in -schools-march-2020.pdf?sfvrsn=baf81d52_. accessed covid-19) situation report 35 coronavirus disease 2019 (covid-19). available from who director-general's opening remarks at the mission briefing on covid-19 who infection prevention and control guidance -(covid-19) ethiopian fedral ministry of health. national comprehensive covid19 management handbook knowledge, attitudes, and practices towards covid-19 among chinese residents during the rapid rise period of the covid-19 outbreak: a quick online cross-sectional survey knowledge, perceptions, and attitude of egyptians towards the novel disease (covid-19) knowledge, attitudes, and practices towards covid-19 among nepalese residents: a quick online cross-sectional survey knowledge, perceptions and preventive practices towards covid-19 early in the outbreak among jimma university medical center visitors, southwest ethiopia mers-cov infection: mind the public knowledge gap knowledge, attitude, and preventive practices toward covid-19 among bangladeshi internet users we are also extended our gratitude to all individuals who were involved in data collection and also the participants who genuinely provided us with all necessary information. all authors made substantial contributions to the conception and design, acquisition of data, analysis and interpretation of data, took part in revising the article, gave final approval of the version to be published and agreed to be accountable for all parts of the work. not funded by any organization. the authors declare that they have no competing interests. key: cord-029769-1c4wl7xy authors: amoo-adare, epifania akosua title: the art of (un)thinking: when hyper productivity says ‘enough!’, is a feast date: 2020-07-27 journal: postdigit sci educ doi: 10.1007/s42438-020-00162-z sha: doc_id: 29769 cord_uid: 1c4wl7xy nan in january 2017, i decided to drop out of a 'westernized university' (see grosfoguel 2013) . i had to leave that never-ending scholarly tale of competition, hyper productivity, over exhaustion, and constant intellectual insecurity. i abandoned the academic track, so i could take the time to deeply reflect on the state of my intellectual affairs. i wanted to decolonize my intellectual existence, in ways that would contribute (in small part) to a long overdue paradigm shift-a radical transformation in the way we think and act in the name of civilizational progress. in my mind, the academy served as a microcosm of the dire state of neoliberalism across the globe. i found it impossible to continue my unquestioned participation in academic normativity without perpetuating its ideological system of power-knowledge that did absolutely nothing for the many struggles for social and cognitive justice and for eradicating the root causes of the world's numerous 'wicked problems' (conklin 2005) . through the production of knowledge for a rather linear narrative of growth and civilizational progress, we-as scholars-are significantly implicated in today's world of competitive neoliberalism. and it is this narrow notion of development that is taking us headlong over the edge of our planetary boundary, through extractivism, exploitation, and the mismanagement of the natural environment. we have pushed so much so that nature has seen fit to strike back. and we are now in the midst of a global pandemic-a self-orchestrated danger-that has put us all on notice. here in ghana, the coronavirus pandemic has resulted in the infection of 12,929 people, plus the death of 66 individuals-by the time of writing this article (ghana health service 2020). the virus has greatly slowed down the economy, with adverse effects on agriculture, industry, and services. consequently, the ghana government is faced with the insurmountable challenge of enforcing social distancing measures and meeting daily survival needs of numerous individuals working in its informal sector 1 (knott 2020) , all during a politically charged election year. accordingly, in march 2020, the country was put on partial lockdown-with the closing of mosques, churches, and schools, as well as its borders. the lockdown has subsequently revealed stark inequalities, especially in the country's education system (mohamed 2020) , including in private schools (quartey 2020 ). ghana's coronavirus-challenged existence is no anomaly. countries around the globe struggle in varying degrees to maintain their economies, keep their citizenry alive, and mediate gaping social inequalities. this global crisis is a significant opportunity to stop and engage in an (un)thinking of the science that makes the world churn. the moment for making this long overdue paradigm shift is right here and right now. i use the word (un)thinking-versus say, rethinking-to stress the need for moving beyond scientific approaches mired in a language of hierarchical cartesian binaries (functowicz and pereira 2015) constructed between minds and bodies, culture and nature, theory and practice, physical and metaphysical, men and women, self and other, and so on-ad infinitum. we need to consider transgressive practices that enable us to gain understandings, albeit partial ones, of contemporary phenomena through knowledge production processes that are not reductionist, exclusionary, elitist, and/or westernized. i argue for a decolonization of current modes of academic knowledge production, especially in the social and hard sciences. i propose that we seek ways to become reflexive and embodied agents for social and cognitive justice within our institutions, communities, societies, and the world-at-large. this becoming needs to include the development of critical pedagogies (darder 1992 (darder /2009 freire 1973; mclaren 2001 mclaren , 2007 , which enable us to teach ourselves (and others) how to transgress established knowledge boundaries and move beyond cartesian thinking. we can do this by inhabiting the ambiguous and frustrating interstices of such false hierarchical binaries, with all of our bodies, minds, hearts, and souls. we need to find ways to link our multiple and intersectional selves into our work, leaving nothing by the wayside-not especially the quirky aspects of our identities. if we are aware that reality is socially constructed and that much of our everyday is about performativity (butler 1988) , then why do we see fit to unquestionably apportion some parts of ourselves (e.g., emotion, artistry, activism, or spirituality) out from 'objective' scientific work? to engage in (un)thinking, we need to think with more than our minds. we need to also think with our hearts, since, after all, heart cells also have memory (pearsall 1999; pearsall, schwartz and russek 1999) . paul pearsall (1999: 65) describes the power of the heart, its cellular memory and ability to transmit energy into the human body. he tells us that 'the heart is 5000 times more electromagnetically powerful than the brain,' thus giving us an inkling of the real potential in the idea of 'thinking from the heart.' unsurprisingly, this is already well known within the indigenous epistemology of the aymara, a native american population living in the andes (burman 2012) . additionally, we must consider and experience knowledges that are not only read through the textual but also expressed through the visual, heard through ears, smelt with the nose, tasted by the tongue and the skin, intuited in our bones, foretold in our dreams, and sensed on both physical and metaphysical planes. (un)thinking, then, is a wallowing, and perhaps for some, even a drowning in changeable seas of knowledge (im)possibilities, which are at times fresh and soothing in their familiarity, but often rough, salty, and terrifying in their chaotic inundations-as they wrench us through many unknown conceptions of existence. this notion of (un)thinking science can be situated within other current scientific debates that recognize the need for renewed intersectional, embodied, and transgressive approaches to knowledge production. such approaches are necessary because of the complex, dynamic, 'multiplicitous,' and uncertain nature of contemporary phenomena. they are also crucial for the decolonization of the westernized university and its many modes of internationalized knowledge production, associated with the occlusion and/or eradication of other ways of seeing, knowing, sensing, and becoming in and of this world (asante 2012; grosfoguel 2013; mignolo 2009; ndlovu-gatsheni 2015) . (un)thinking, ultimately, is also an art form that contributes to the long overdue paradigm shift in knowledge production. as one example of (un)thinking, i have become entangled in an uncharted exploration of various embodied modes of un-learning, which include my continuous development of critical spatial literacy (amoo-adare 2013). 2 in doing so, i have been formulating, for myself (and potentially others), some kind of embodied womanistfeminist intellectualism (maparyan 2012 ) that is not only organic and spiritual but also purposeful, measured, and yet too rhizomatic in its creative and sacred pursuit of learning by/through living without many of the boundaries that matter to (a successful, ibid career-oriented) existence in today's material world. this intellectualism is aimed towards the co-construction of diverse theories and transformations that highlight the diversity of epistemologies, ontologies, cosmologies, and philosophies of life that inhabit our multiverse. i am still new to this radical practice that began three years ago, with my decision to leave the fast-paced machinations of productive academic life, in order to sit, listen, wait, and see. more importantly, i have taken up key spiritual questions that in fact drive much of scientific and even personal inquiry around the globe: who am i? what do i want in life? how do i relate (myself and what i want) back to others (gangaji 2013) ? i do so, while recognizing how my economic privilege enables this luxury of self-inquiry. in this way, i have been attempting to inhabit a kind of 'in-between existence'; one which attempts to strike a balance between an artless meandering through subjects i claim to know and the vast unknown. this existence counters the idea of being a knowledgeable expert on any aspect of life in its sacred complexity and dynamism. many years before, i engaged in this kind of query, however, through strong identification with my work in higher education in the global north and international development in the global south. at that time, any aspect of me that may have been categorized as artistic flair was required to take a back seat within the hierarchical segregation of everyday academic life. here, i speak of academia's inherent 'disciplinary decadence' (gordon 2006: 8) that affixes scholars in academic niches, funding cycles, and/or set career pathways, while requiring pain-of-death allegiances-simply because many of these 'disciplines lose sight of themselves as efforts to understand the world and have collapsed into the hubris of asserting themselves as the world'. this is yet another false splitting of the intellectual, embodied, experiential, and spiritual self, which according to some cultures is seen as a collective made up of millions of cells, microbiomes, memories, and much more than we are yet to fathom. i have found that taking up those disciplinary distinctions undermines my understanding of a true pursuit of knowledge. from my experience of deep inquiry, this pursuit is often a matter of sitting with many uncomfortable and pain-filled days of 'doing nothing' purposefully, 'waiting and seeing,' plus raising questions into a mind quieted through meditation, or other quiet contemplation. within rhizomatic configurations of knowledge, this approach of 'laissez faire meets some modicum of discipline' leads one into, through, around, and/or beyond various knowledge boundaries that attempt to contain academic, and other equally relevant, ways of knowing, sensing, or being. i advocate this type of coddiwomple into the in-betweens of diverse knowledges and their modes of production because traveling purposefully towards an unknown destination develops a type of 'border consciousness' (anzaldua 1987 (anzaldua /2012 . it also creates a borderland, which is fertile ground for the development of 'threshold theories' (keating 2013) . and it is here also that we find a liminal space of yet-to-be known change and infinite possibility. finally, it must become a form of knowledge production that is additionally intent on (un)learning in order to re-learn a sociality that is geared towards social and cognitive justice (santos 2014) , plus the reconstitution of 'worlds and knowledges otherwise' (escobar 2007) . this is a non-dualistic practice in which 'i don't know' takes central place. in its practice, the experiential must also receive greater attention, as we relearn how to listen to our bodies keenly. the same applies to our ability to truly listen deeply to others, especially of the flora and fauna variety-since here too lies the teeming question of justice. and in this age of the anthropocene (haraway et al. 2015) , we can quickly bear the dire consequences of our own stubborn inability to be attentive to nature and its other inhabitants. in today's multifaceted, dynamic, ambiguous, and uncertain world, we find that certain linear modes of thought-such as ideas of modernity, progress, and development-are now null and void. it is why ziauddin sardar (2010 sardar ( , 2015 speaks of these as 'postnormal times,' in which we need to think and act-differently-in non-linear terms. more specifically, sardar tells us that 'postnormal times' are subject to the 3cs of 'chaos, complexity and contradiction,' which not only challenge the very notion of life as we assume it to be but also how we imagine our common futures. and this also includes how we conduct the scientific research that informs it all. it is for reasons such as these that reil miller (2013 reminds us of the need 'to use a freer anticipatory approachthe emerging discipline of anticipationor futures literacy, which can help reduce fear of the unknown' (2013: p. 107). for miller, this is a more systematic way of using the future to inform the present, through scientific inquiry, rather than the typical focus on predictive science. similarly, sardar and john sweeney (2016) ask us to make uncertainty central to research, as it is often not catered for in most kinds of analysis in the social sciences. more specifically, they propose a new method for foresight and futures researchers (and practitioners), for which they outline 'the three tomorrows of postnormal times.' in this, sardar and sweeney define the future as being comprised of (1) the surface uncertainty of an 'extended present' that has largely already been colonized; (2) the shallow uncertainty of 'familiar future(s),' which are mediated by existing imaginings of the future; and (3) the deep uncertainty of 'unthought future(s).' in light of this understanding of what constitutes as the future, they argue that only collaborative creativity and ethical imagination can be used as tools to help us construct scenarios for this interactive terrain, which is mediated by pure possibility and many 'unknown unknowns.' covid-19 is very much about the abrupt encounter of our 'extended present' with an 'unthought future,' which is dragging us kicking and screaming into many 'unknown unknowns.' basically, we have crashed into the hard edge of our planetary boundary, through the onslaught of the coronavirus. and we are now faced with it, as a dire consequence of our overexploitation and mismanagement of the natural environment. this is epitomized by the creation of the virus's-still contested-epicenter, a wet market in wuhan, china (maron 2020; morrison 2020) . it is also safe to say that our deep exploitation and practice of extractivism is a direct result of a western paradigm that sees a great binary separation between culture and nature, with the latter being at the service (and disposal) of the former. 3 moreover, the coronavirus is clearly a negative effect of global capitalism in its increased flows; that is, the fluid epigenetic entanglements (lock 2015) , or networks, of people, fauna, flora, places, ideas, things, etc. (appadurai 1999; urry 2007; latour 2011) . it is also an example of how the flows of capital manifest differentially, often, in uneven spatial development (massey 1991 (massey , 1994 . this is something we have come to learn the hard way, through covid-19, as we witness many go hungry, lose jobs, become homeless, and mostly die as a consequence of stark social inequality and the ineptitude of governance without foresight. this all being found even within the best examples of democracy. indeed, covid-19 is here to remind us that 'enough' is actually a feast that we have chosen to ignore for far too long. it is why we now also find ourselves in the uproar of many past social struggles (countryman 2020) ; struggles that-as angela davis (2018) says-do not simply emerge spontaneously but rather come as deeply rooted and connected to several heinous past events, with continued nefarious current effects. and while we human beings reactivate these past contestations over geography and knowledge (said 1977 (said /1995 , through moderate protest and marches, nature instead is more brutal and to the point. it leaves us with no doubt that we cannot continue with business-as-usual. it forces us into deep self-inquiry about who we are, what we want, and how we need to relate that back into (and with) the natural world-especially if we care to live beyond the inevitability of the next global crisis of our making. ultimately, this harsh reality demands innovative forms of knowledge production about the many dynamic (non)human interaction processes. these can only be barely understood through multiple disciplinary perspectives, 4 as well as through a critical reflection on the positionality 5 of every researcher enacting these open-ended processes of inquiry. spatial literacy: contemporary asante women's place-making borderlands la frontera: the new mestiza disjuncture and difference in the global cultural economy reconstituting curricula in african universities: in search of an afro-centric design places to think with, books to think about: words, experience and the decolonization of knowledge in the bolivian andes performative acts and gender constitution: an essay in phenomenology and feminist theory turning in upon ourselves: positionality, subjectivity and reflexivity in case study and ethnographic research dialogue mapping: building shared understanding of wicked problems 2020 uprisings, unprecedented in scope, join a long river of struggle in america. the conversation epilogue: teaching as an act of love: reflections on paulo freire and his contributions to our lives and our work angela davis on feminism, communism and being a black panther during the civil rights movement worlds and knowledges otherwise education: the practice of freedom. london: writers and readers publishing cooperative cartesian dreams updates on coronavirus disease (covid-19) in ghana as they happen disciplinary decadence: living thought for trying times the structure of knowledge in westernized universities: epistemic racism/sexism and the four genocides/epistemicides of the long 16 th century anthropologists are talking -about the anthropocene transformation now! toward a post-oppositional politics of change ghana's decision to lift partial covid-19 lockdown criticized by some networks, societies, spheres: reflections of an actor-network theorist comprehending the body in the era of epigenome the womanist idea wet markets' likely launched the coronavirus. here's what you need to know a global sense of space space, place and gender che guevara, paulo freire, and the politics of hope: reclaiming critical pedagogy life in schools: an introduction to critical pedagogy in the foundations of education epistemic disobedience, independent thought and de-colonial freedom changing the conditions of change by learning to use the future differently learning, the future, and complexity. an essay on the emergence of futures literacy what covid-19 reveals about educational inequality in ghana now china says wuhan wet market was not the origin of the coronavirus pandemic but it may have been the site of a 'super-spreader' event. daily mail, 29 may genealogies of coloniality and implications for africa's development the informal sector in ghana. unpublished manuscript the heart's code: tapping the wisdom and power of our heart energy changes in heart transplant recipients that parallel the personalities of donors covid-19 and the plight of private school teachers in ghana /1995). orientalism epistemologies of the south: justice against epistemicide welcome to postnormal times postnormal times revisited the three tomorrows of postnormal times key: cord-330474-c6eq1djd authors: fox, j; khan, o; curtis, h.; wright, a; pal, c; cockburn, n; cooper, j; chandan, j.s.; nirantharakumar, k title: rapid translation of clinical guidelines into executable knowledge: a case study of covid‐19 and on‐line demonstration date: 2020-06-18 journal: learn health syst doi: 10.1002/lrh2.10236 sha: doc_id: 330474 cord_uid: c6eq1djd the polyphony programme is a rapidly established collaboration whose aim is to build and maintain a collection of current healthcare knowledge about detection, diagnosis and treatment of covid‐19 infections, and use artificial intelligence (knowledge engineering) techniques to apply the results in patient care. the initial goal is to assess whether the platform is adequate for rapidly building executable models of clinical expertise, while the longer term goal is to use the resulting covid‐19 knowledge model as a reference and resource for medical training, research and, with partners, develop products and services for better patient care. in this polyphony progress‐report we describe the first prototype of a care pathway and decision support system that is accessible on openclinical.net, a knowledge sharing repository. pathfinder 1 demonstrates services including situation assessment and inference, decision making, outcome prediction and workflow management. pathfinder 1 represents encouraging evidence that it is possible to rapidly develop and deploy practical clinical services for patient care and we hope to validate an advanced version in a collaborative internet trial. finally, we discuss wider implications of the polyphony framework for developing rapid learning systems in healthcare, and how we may prepare for using ai in future public health emergencies. this article is protected by copyright. all rights reserved. the covid-19 emergency is a massive challenge to human expertise and organisation, but it is also widely recognised as an opportunity to demonstrate, test and improve medical technologies, including ai techniques for delivering rapid learning systems. over recent years we have been developing a flexible methodology for creating executable models of specialist clinical expertise and a platform for sharing these models called openclinical (www.openclinical.net; fox et al, 2013) . openclinical is one of a number of efforts in recent years to use knowledge engineering and other techniques to formalise clinical guidelines as "executable knowledge (e.g. peleg et al, 2010; friedman 2019) . although systematic reviews, clinical guidelines etc have been important tools of the evidence based medicine movement their impact in improving consistency and quality of care has been less than hoped for because they are disseminated purely as human readable content (e.g. text, diagrams) in a traditional, often slow, publication and revision cycle. openclinical has made use of one particular approach to formalising clinical guidelines based on a specialised modelling language called proforma (fox and das, 2000; sutton and fox 2003) ; we adopted this approach because it has been used and trialled successfully in many medical applications (fox et al, in press) and we have wide experience using it. we see openclinical as the basis of a rapid learning system as illustrated in figure 1. this shows a knowledge life cycle for creating and maintaining executable models of care using the openclinical.net knowledge modelling and publishing platform. the top arrow of the cycle represents creation and testing of models using proforma authoring software; the next step on the right is to publish them on the openclinical.net repository (which currently carries 50+ examples of executable models for many clinical settings and specialties https://dev.openclinical.net/index.php?id=69). figure 1 : the openclinical knowledge-to-data cycle for rapid learning systems this lifecycle overlaps with other rls proposals, notably friedman et al's mcbk proposals (2019) but is distinctive in using proforma, a specialised ai language, for modelling and to support open source sharing of models. although proforma has been successful in these roles the goal of openclinical is primarily to provide a platform for crowdsourcing, validating and disseminating models not to deliver clinical services directly. a key aim of the openclinical project is to "close the loop" (left arrow) by incorporating software for acquiring clinical data from clinical implementations and trials of the models. this will permit use of data mining, machine learning and other techniques to update the evidence base and refine clinical guidelines and other care quality standards. the polyphony project was initiated on 18 march 2020 with the following mission to create, validate, publish and maintain knowledge of best medical practice regarding the detection, diagnosis and management of covid-19 infections, in a computer executable form. the purpose is to provide a resource for clinicians and researchers, healthcare provider organisations, technology developers and other users, to (1) develop point of care products and services which (2) embody best clinical practice in decision-making, workflow, data analysis and other "intelligent" services across the covid patient journey. 2. how useful is the openclinical knowledge sharing framework for empowering clinicians to critique and improve models of decision-making and care across the patient journey? 3. is it possible to adapt components of the model for use in different clinical settings or in local variants of care pathways? 4. what is the potential for combining knowledge engineering methods with techniques from data science (e.g. statistical analysis, data mining and machine learning)? this article is a progress report on our first prototype, pathfinder 1, and the design and engineering framework we have developed to support the polyphony mission. the on-line demonstrator is not complete or "clinical strength" but this is the goal of future cycles of figure 1. 1 however sufficient progress has been made that it may be of interest to the rls community. the proforma language is based on a general framework for modelling tasks, including reasoning, decision-making and planning. task models can be applied to knowledge about a particular medical domain such as the diagnosis and treatment of covid-19 patients, formalised in proforma or available in external resources. the openclinical knowledge model is illustrated schematically in figure 2 . at the bottom of the ladder are symbols (eg "fever", '38.6') which can be combined to represent data (e.g. presenting symptoms include fever) and concepts (e.g. diagnosis is a kind of decision; pathway is a kind of plan) and descriptions (e.g. patient histories). proforma can be used to model knowledge as rules for inference or action. where inference is uncertain rules can participate in complex decisions as a basis for constructing arguments for and against competing decision options. finally, decisions, actions and enquiries (actions that acquire information) can be composed into plans to achieve particular objectives. in principle plans can be encapsulated as agents that can carry out complex behaviour autonomously, though agent modelling is not within the scope of the polyphony program. proforma can be used to model clinical guidelines of various types including medical logic and recommendations, decision trees, clinical algorithms etc. early development of the covid-19 knowledge base in pathfinder 1 relied primarily on documentation published by bmj best practice in march 2020 2 . the openclinical modelling and testing platform was used by an experienced proforma modeller advised by clinicians with specialist knowledge in public health and primary care. in pathfinder 1 the concepts and components of the bmj best practice guidance were mapped directly to levels of the knowledge ladder. the resulting covid-19 model consists of five main sections: the data model; clinical contexts (data sets and scenarios); rules (inference and alerts); decisions (arguments and evidence); and care pathways. in the following paragraphs we summarise how these are modelled in the various modules of pathfinder 1. a module overview can also be seen in the graphic on the pathfinder page on openclinical. this article is protected by copyright. all rights reserved. the openclinical platform includes a tool to create data definitions for relevant clinical and other parameters, their types (strings, integers, date-times etc) and other properties 3 . the full model for pathfinder 1 consists of 46 data definitions based on the bmj best practice documentation (march 2020). these are at the symbol level of the knowledge ladder in figure 2 . a clinical context is typically a scenario on the patient journey in which one or more decisions may be taken and for which subsets of the data model are relevant. one scenario is "patient triage" where a handful of questions may be asked relevant to a decision whether to do nothing, advise self-isolation or book an ambulance. another context is "hospital work-up" which covers a detailed patient history to inform a provisional diagnosis and initial selection of investigations. based on the bmj guidance (op cit) ten contexts were identified and modelled. pathfinder 1 emphasises prehospital scenarios with less emphasis on hospital and acute care, though by the time of publication bmj best practice has added more guidance on the latter stages of the journey. as shown in the knowledge ladder proforma supports several knowledge types, of which the simplest is the if…then… rule. rules respond to situations or events expressed in boolean logic. for example, if a patient's temperature is > 38.6 pathfinder infers there is a fever; when hazardous situations are detected "red flags" are raised. other uses of rules are to alert the user if a task has not been carried out or is overdue, or when a patient is eligible for inclusion in an open research trial (e.g. patkar et al, 2013) . a proforma decision model consists of a set of options and a set of "arguments" (pros and cons) associated with each option. if an argument expression is evaluated against current patient data or other information and found to be valid, it is recorded as a reason for (or against) the relevant option with an explanation and, if required, supporting evidence that justifies the argument. all options in an active decision context accumulate patient-specific pros-and cons as data are acquired, and the decision engine aggregates these to provide a continuously updated measure of confidence in each option. in the initial modelling stage arguments are usually modelled qualitatively but if statistical or other data are available they can be assigned quantitative weights which can be aggregated using various possible decision algorithms. eight main decisions have been modelled for the covid-19 patient journey, including triage, diagnosis, prognosis, prediction of complications and choice of management plan. all decisions are concurrently active in pathfinder 1 but they can also be deployed at specific points or in particular scenarios in the care pathway. a "pathway" is a network of decisions and tasks for acquiring data and carrying out plans in a sequential and/or conditional way as illustrated in figure 3 . in the first version there is an enquiry (green diamond) about a small number of key data relevant to an initial assessment (e.g. presenting complaints, age of patient and whether patient seems ill). if required a more detailed history is taken but either way an escalation decision follows. a later version shown in the second panel is based on more detailed guidance to gps published later by bmj. pathfinder 1 can be accessed via the link https://www.openclinical.net/index.php?id=68, where there are instructions for running the demonstrator against example cases provided or against the user's own cases. figure 4 shows three screenshots from a typical run against example case 1. when running the on-line demonstrator the following points should be borne in mind: (1) the model is developed as a "standalone" resource for testing and validation and is not intended to be deployed directly into clinical use. the model is incomplete with respect to current covid-19 guidance which has evolved significantly since completion of the first modelling cycle; we expect to make significant revisions to the data model and knowledge content in light of experience and feedback before starting on the next modelling cycle (pathfinder 2). assessment against project objectives 1. creation of an executable model of best practice for supporting care of covid-19 patients. proforma is capable of modelling the main decisions and workflows across the covid-19 patient journey. pathfinder was developed in about three weeks but is incomplete with respect to published medical knowledge and recommended practice and the data, decision and pathway models all require improvement and validation by qualified clinicians. our provisional assessment is that the model could be completed and maintained on a rapid timescale. a complementary pathway this article is protected by copyright. all rights reserved. for hospital care of covid-19 patients has also been rapidly developed in proforma by deontics ltd 4 . the focus of pathfinder 2 is on extending the pathways for additional clinical contexts (e.g. comorbidity management). appraisal of the openclinical platform as a practical knowledge sharing infrastructure. pathfinder 1 was deployed on openclinical.net as a globally accessible, open access demonstrator in early april 2020. a modified version was made in response to comments and suggestions of the clinical authors and made privately available for testing by them shortly after. the incorporation of example test cases provided by clinical collaborators and independent reviewers helps to quickly familiarise users with the demonstration and to critique the decision models and pathways against realistic patient data. reusability of the data and knowledge models at different points in the care journey the "escalation decision" in the reference model was initially used in a self-triage pathway and reused in the residential care triage pathway of pathfinder 1, but was replaced with three different decisions in the gp consultation pathway in light of clinical comments and new published guidance. in the latter pathway the diagnosis decision from the pathfinder 1 model was reused without change, but significant changes were made to the initial assessment scenario and additional decisions about diagnosis and appropriate actions were added to the later pathway model. at this early stage of the project we have not been able to progress this question but are seeking to collaborate with data science specialists in the next cycle of the project. initial discussions suggest however that once case records for a population of patients are available it will be practical to exploit well established statistical and machine learning methods to calibrate argument weights for patient populations for example, and there is also scope for symbolic machine learning methods to suggest extensions to the logical knowledge model (e.g. rule induction methods). as explained above the main goal of the polyphony project is to develop a "reference" data and knowledge model, not to deploy the model directly but to be a resource for others to use in developing clinical services. this is in part due to the complexities of integrating decision support and other services with existing it infrastructure. as mentioned above deontics ltd have developed a decision support application for hospital patient assessment and deployed it in the emergency department of the liverpool and broadgreen university hospital. integration with the hospital emr and infrastructure has been achieved but than was significantly more complex than the proforma modelling step. readers interested in deployment of services will ask how proforma engages with relevant technical standards, such as medical coding and terminology standards (e.g. icd10; loinc), ontologies (e.g. snomed ct, uml), logical expressions (e.g. gello) and rules (e.g. arden syntax; cql). proforma overlaps conceptually but does not comply with these standards. the emphasis in proforma is on capturing knowledge of these and other kinds in a single language for which it offers a unified syntax and execution semantics (sutton and fox 2003) . this is highly preferable to modelling a complex body of knowledge with ad hoc combinations of notations . the development of hl7 fhir resources has a similar motivation and we are interested in exploring whether proforma can implement certain classes of fhir resources, such as the fhir plan definition (https://www.hl7.org/fhir/plandefinition.html) and fhir executable knowledge artifact (https://www.hl7.org/fhir/clinicalreasoning-knowledge-artifact-representation.html). we think that this may make integration and deployment easier without losing the simplicity and intuitiveness of the proforma and polyphony framework. if the polyphony model is to have a sustainable future it will need to be repeatedly updated as knowledge of covid-19 and current clinical practice change. maintenance can take place at two levels, the implementation (e.g. software) level and the practice (knowledge) level. we anticipate that the maintenance of the implementation can be managed through standard software engineering and version control methods. this allows messages and documentation outlining the rationale for the change to be associated with each iteration of the implementation. present version control tools also keep a history of all changes, including highlighting differences between versions for easy review/audit. classical version control would, for example, be appropriate at the level of hl7 resources and their mappings to the knowledge model. maintenance at the knowledge level is likely to be very different because the principle stakeholders here are professional clinicians, researchers and other subject matter experts. polyphony was established under the auspices of the openclinical knowledge sharing project which seeks to establish a framework for knowledge dissemination analogous to open access publishing. key here are the need for models to be intuitive and open to review and criticism by professional who have little expertise or interest in technicalities. new versions of knowledge models need to specify their provenance, the evidence on which changes are based and in clinical use any unexpected behaviour must be explainable in appropriate medical terms. if covid-19 remains clinically challenging in the medium-term if it would be desirable to recruit experienced healthcare professionals to contribute to modelling the knowledge underlying decisions and pathways. through openclinical we hope to support a sustainable community of practice to promote discussion and debate and to own and maintain the reference knowledge base. a possible organisational structure could be analogous to the "chromosome committees" of the human genome project in that specialist professional groups of gps, emergency medicine clinicians etc. would take responsibility for data and knowledge modelling in specific contexts along the patient journey, but adopting common data and knowledge representation standards. in the longer term, with the hoped-for arrival of an effective vaccine or treatments, the covid-19 emergency may pass or become tolerated as a seasonal burden like flu. these futures are controversial, and we take no position on them, but it is widely accepted that the covid-19 pandemic is only the latest in a series of infections with major consequences for human populations and there will be more to come. it will be important to have "rapid response" as well as "rapid learning" mechanisms in place. polyphony may help to inform the design of policies and mechanisms this article is protected by copyright. all rights reserved. by which expert and experienced healthcare professionals can form rapid response teams to address emerging threats. a longer-term objective is based on the proposition that the polyphony approach is not limited to the covid-19 emergency nor even only to infections. the methods outlined here are applicable to rapid deployment of executable clinical guidelines and quality standards generally. we believe they can be used to create open access and open source models of practice for many conditions whether acute or chronic, commonplace or rare, "from home to hospital to home". implementing nice guidance cognitive systems at the point of care: the credo programme net: a platform for creating and sharing knowledge and promoting best practice in healthcare syntax and semantics of the proforma guideline modelling language delivering clinical decision support services: there is nothing as practical as a good theory using computerised decision support to improve compliance of cancer multidisciplinary meetings with evidence-based guidance comparing computer-interpretable guideline models: a case-study approach thanks to ali rahmanzadeh and david sutton who were the original developers of the tallis authoring suite and proforma execution engine and generously provided technical support during the project. john fox is founder and non-executive director of deontics ltd. he also works pro bono as director of openclinical cic, a non-profit community interest company. joht singh chandan, omar khan, jenny cooper, neil cockburn, krishnarajah nirantharakumar and carla pal state that they have no conflicts of interest. both hywel curtis and andrew wright, who worked in advisory capacities as consultants on the project, also assert they have no conflicts of interest this article is protected by copyright. all rights reserved. key: cord-144033-pmchx05r authors: shin, dongmin; shim, yugeun; yu, hangyeol; lee, seewoo; kim, byungsoo; choi, youngduck title: saint+: integrating temporal features for ednet correctness prediction date: 2020-10-19 journal: nan doi: nan sha: doc_id: 144033 cord_uid: pmchx05r we propose saint+, a successor of saint which is a transformer based knowledge tracing model that separately processes exercise information and student response information. following the architecture of saint, saint+ has an encoder-decoder structure where the encoder applies self-attention layers to a stream of exercise embeddings, and the decoder alternately applies self-attention layers and encoder-decoder attention layers to streams of response embeddings and encoder output. moreover, saint+ incorporates two temporal feature embeddings into the response embeddings: elapsed time, the time taken for a student to answer, and lag time, the time interval between adjacent learning activities. we empirically evaluate the effectiveness of saint+ on ednet, the largest publicly available benchmark dataset in the education domain. experimental results show that saint+ achieves state-of-the-art performance in knowledge tracing with an improvement of 1.25% in area under receiver operating characteristic curve compared to saint, the current state-of-the-art model in ednet dataset. the recent covid-19 pandemic has raised needs for social distancing, leading many organizations to develop virtual and remote services to prevent widespread infection of the disease. accordingly, educational systems have developed remote learning environments including massive open online courses and intelligent tutoring systems (itss). knowledge tracing, modeling a student's knowledge state based on the history of their learning activities records, is a fundamental task for creating its that aims to provide personalized learning experiences to each student. traditionally, bayesian knowledge tracing [4, 5, 10, 15-17, 19, 20, 24, 26] and collaborative filtering based models [11, 22] were common approaches for knowledge tracing. however, with the widespread adoption of deep learning (dl) in many machine learning problems, dl based models [2, 6, 9, 12-14, 18, 21, 23, 25, 27] have become the de facto standard for knowledge tracing by capturing complex relations among interactions in students' learning activities records. the strength of the dl based models has become amplified with the advent of large scale public benchmark dataset in artificial intelligence in education. ednet [3] is such a publicly available dataset which is the largest in scale with more than 131m learning activities records from around 780k students. in this paper, we propose saint+, a successor of saint [2] which enhances knowledge tracing with temporal feature embeddings, and empirically verify the effectiveness of the model on ednet dataset. saint is a transformer [25] based knowledge tracing model that separately processes information of exercise and student response. specifically, a stream of exercises that a student consumes is fed to an encoder, and a decoder gets a corresponding response sequence and encoder output sequence, computing the final output sequence of the model. saint+ augments saint by integrating two temporal feature embeddings: elapsed time, the time taken for a student to answer, and lag time, the time interval between adjacent learning activities. empirical evaluations conducted on ednet dataset show that saint+ improves saint, the current state-of-the-art knowledge tracing model on ednet dataset, by 1.25% in area under receiver operating characteristic curve (auc). also, the experimental results show that incorporating the temporal features into the decoder input achieves the best auc compared to incorporating them into the encoder input, and both the encoder and decoder input, verifying the hypothesis that separately processing exercise information and student response information is appropriate for knowledge tracing. knowledge tracing is a fundamental task for many computer-aided educational applications and has been studied extensively in the field of aied. traditional approaches addressed knowledge tracing based on bayesian knowledge tracing (bkt) [4, 5, 10, 15-17, 19, 20, 24, 26] and collaborative filtering (cf) [11, 22] . basically, bkt is the hidden markov model where a latent variable represents evolving student knowledge. bkt assumes the latent student knowledge as a set of binary variables: either the student mastered the knowledge or not. each latent variable is updated based on observations of the student correctly applying the knowledge which are also binary: either the student correctly or incorrectly answered a given exercise. on the other hand, cf based approaches model students and exercises as low-rank matrices. each vector in the student matrix and exercise matrix represents latent traits of each student and latent knowledge required for each exercise, respectively. the probability of a student correctly answers to an exercise is calculated by applying the sigmoid function to the dot product between the corresponding student and exercise vectors. the advances of deep learning (dl) have given rise to neural network based knowledge tracing models. dkt [18] is the first dl based knowledge tracing model. dkt models students' evolving knowledge state through recurrent neural network (rnn) which compresses their past learning activities in a hidden layer. like many rnn based models that commonly leverage attention mechanism, npa [12] models student knowledge through bidirectional long-short term memory (bi-lstm) network equipped with an attention layer that weighs more importance to relevant parts of their learning history for prediction. ekt [9] is also a bi-lstm knowledge tracing model with an attention layer. however, ekt addresses the cold start problem in knowledge tracing by exploiting not only students' learning activities records but also text descriptions of exercises. not all dl based knowledge tracing models are based on rnn architecture. dkvmn [27] is a memory-augmented neural network knowledge tracing model where the key matrix stores knowledge concepts and the value matrix stores students' mastery levels of corresponding concepts. ckt [21] is a knowledge tracing model that applies hierarchical convolutional operations to extract learning rate features from student's learning activities history. applying self-attention mechanism in transformer [25] architecture, which is de facto standard to many sequential prediction tasks, to knowledge tracing is also an actively studied area. sakt [14] is the first knowledge tracing model with self-attention layers. in each self-attention layer of sakt, each query is an exercise embedding vector, and key and value are interaction embedding vectors. saint [2] is the first transformer based knowledge tracing model which leverages encoder-decoder architecture composed of stacked self-attention layers. unlike sakt, saint gets separated streams of exercises and responses as inputs where a sequence of exercises are fed to the encoder, and a sequence of encoder outputs and responses are fed to the decoder. akt [6] also adopts self-attention layers for knowledge tracing. the attention weights in akt are decayed exponentially based on the context-aware relative distance measure. moreover, akt uses the rasch model based exercise and exercise-response embeddings to avoid overparameterization and overfitting. recently, several works [13, 23] attempt to incorporate graph structure to the knowledge tracing model. [13] formulate knowledge tracing as a time series nodelevel classification task in graph structure and proposes gkt which extracts representation of each knowledge concept by aggregating representations of neighboring concepts. hgkt [23] applies graph neural network to get hierarchical exercise graph which better represent groups of similar exercises. we formulate knowledge tracing as a task of predicting the probability of a student's answer being correct to a particular exercise given their previous interaction histories. formally, the student's learning activity is recorded as an interaction sequence 1 , . . . , . each interaction = ( , ) is a tuple of exercise information , the -th exercise given to the student with related metadata, such as the type of the exercise, and response information , the student's response to the exercise along with related metadata including the correctness of the response, the duration of time the student took to respond and the time interval between the current and previous interactions. the student's response correctness ∈ {0, 1} is equal to 1 if the student answered the -th exercise correctly and 0 if not. thus, knowledge tracing aims to estimate the probability, in this subsection, we give a brief review of saint, a separated self-attentive neural knowledge tracing. we refer the paper [2] for those who want to lean detailed aspects of saint. saint is a knowledge tracing model based on transformer [25] architecture which consists of an encoder and a decoder. it separates a stream of student interactions into two sequences: an exercise sequence and a response sequence. then, the encoder takes a sequence of exercise embeddings = [ 1 , 2 , . . . , ] as input and pass an output sequence = [ 1 , 2 , . . . , ] to the decoder. the decoder additionally takes a shifted response embedding sequence = [ , 1 , 2 , . . . , −1 ] as input, whose first element is a start token embedding, to produce the final output sequenceˆ= [ˆ1,ˆ2, . . . ,ˆ]. eachˆis an estimated probability of the student's answer to the -th exercise being correct given the current exercise information and the past interactions 1 , 2 , . . . , −1 . the most fundamental part of saint is a multi-head attention layer. let ℎ be the number of heads. when input query matrix , key matrix , and value matrix are given, we compute = , = , and = for each 1 ≤ ≤ ℎ, where , , and are weight matrices of query, key and value, respectively. then, the scaled dot-product attention computes each head matrix, and the final output is a linear transformation of concatenated head matrices, where is a dimension of the query and key vectors, and is a weight matrix. note that the masking mechanism overwrites the region above the diagonal of the matrix with −∞ so that the corresponding region of the softmax output becomes zero. this prevents the current position from attending to subsequent positions. in other words, saint uses no future information from the sequence while training. the encoder block consists of sequentially aligned copies of encoder layers. a single encoder layer is a multi-headed self-attention layer with an upper triangular mask followed by a feed forward network (ffn) which is defined by where 1 , 2 and 1 , 2 are weights and biases, respectively. suppose is given as input to an encoder layer. then, the output sequence is computed as follows: here, the input sequence for the foremost encoder layer is an exercise embedding sequence while each subsequent layer takes the feed forward output of the previous layer. note that we apply layer normalization [1] and skip connection [7] to every sub-layer. the decoder is a sequence of identical decoder layers, which consists of two multi-head attention layers with upper triangular masks followed by a feed forward network as well. suppose is an input sequence to a decoder layer. if the layer is the foremost, is the response embedding sequence . otherwise, it is the output sequence from the previous decoder layer. the first layer is a multi-headed self-attention layer which only takes . then, its output 1 serves as queries for the second attention whose keys and values are the encoder output . the computation can be summarized as follows: ( 2 )). the output of the last decoder layer is passed to a fully connected layer to produce the final output of the model. saint takes sequences of exercise embeddings and response embeddings, where each entry in the exercise embedding sequence is the sum of vectors of an exercise id, an exercise category and the position, and each entry in the response embedding sequence is the sum of vectors of a response correctness and the position. saint+ augments response embeddings with two temporal feature embeddings: elapsed time, the duration of time a student took to respond, and lag time, the time interval between the current and previous interactions (figure 1 ). the embedding vectors for elapsed time and lag time are added to the response embeddings ( figure 2 ). in the following subsections, we provide detailed explanations of the two temporal feature embeddings. elapsed time is an amount of time that a student spent on solving a given exercise. for example, in figure 1 , et 1 = 2 − 1 (resp. et 2 = 4 − 3 ) is the elapsed time for the exercise 1 (resp. exercise 2). if the student does not have enough knowledge and skills for the exercise, it would be hard to respond correctly within the recommended time limit. hence, elapsed time provides strong evidence for a student's proficiency in knowledge and skills, and student's understanding of concepts associated with the exercise. we propose two different approaches to embed elapsed times as latent vectors: continuous embedding and categorical embedding. in continuous embedding, a latent embedding vector for an elapsed time et is computed as v et = et · w elapsed_time , where w elapsed_time is a single learnable vector. for categorical embedding, unique latent vectors are assigned to each integer seconds. we set the maximum elapsed time as 300 seconds and any time more than that is capped off to 300 seconds. lag time is the time gap between interactions, an important factor that affects complex phenomena occurring in students' learning process. for instance, students tend to forget what they have learned as time passes. if a lot of time passed after a student answers an exercise about certain concepts, it would be hard to respond to similar exercises correctly, even if they provided the correct answer to the exercise before. on the other hand, students need time to refresh. by taking a rest, their brains organize and arrange what they have learned, and prepare for the next learning session. we define lag time as the time interval between the moment a student encountered the current exercise and the moment the student consumed the previous exercise. for instance, in figure 1 , lag time for the exercise 2 (resp. exercise 3) is lt 2 = 3 − 2 (resp. lt 3 = 5 − 4 ). similar to the elapsed time embedding, we use continuous embedding and categorical embedding for lag time. in continuous embedding, a latent embedding vector for a lag time lt is computed as v lt = lt · w lag_time , where w lag_time is a trainable vector. for categorical embedding, lag times are discretized as integer minutes 0, 1, 2, 3, 4, 5, 10, 20, 30, . . . , 1440. as a result, there are a total of 150 trainable unique latent vectors assigned to each integer minute. ednet [3] is the largest publicly available benchmark dataset in education domain consisting of interaction logs collected by santa 1 . we conduct experiments on an updated version of ednet-kt1, which contains problem-solving logs from january 1st, 2019 to june 1st, 2020. details of the dataset statistics are provided in table 1 . also, the distributions of elapsed time and lag time of the dataset are shown in figure 3 . we use interaction logs of the most recent 100k students as test set and 80% (resp. 20%) of the remaining dataset are used as training (resp. validation) set. # of interactions 60294498 # of students 678128 # of exercises 14418 we compare saint+ against benchmark knowledge tracing models, dkt [18] , dkvmn [27] , sakt [14] and saint [2] : • dkt is a simple rnn-based model that uses concept and response correctness as input features, and models student's knowledge status as rnn's hidden state vectors. we choose an lstm [8] architecture. also, we consider each unique exercise id as a concept associated with the exercise. • dkvmn is a memory augmented neural network based model where the key matrix stores knowledge concepts and the value matrix stores students' mastery levels of the corresponding concepts. we use each exercise id as a concept of the corresponding exercise. • sakt is the first knowledge tracing model that utilizes transformer's self-attention architecture. it is a single encoder-based model where exercise ids are used as queries and interaction ids are used as keys and values. • saint is the first transformer-based knowledge tracing model which leverages an encoderdecoder structure to process information of question and student response separately. the encoder takes a sequence of question embeddings and the decoder gets a sequence of encoder outputs and response embeddings to compute the final output. we use the accuracy (acc) and the area under the receiver operating characteristic curve (auc) as the performance metric. we pick the weight with the best validation auc and evaluate them on the test set. for saint+ and saint, we use adam optimizer with = 0.001, 1 = 0.9, 2 = 0.999 and = 1 −8, and set warmup steps to 4000. the window size, number of layers, dimension of the model, dropout rate and batch size is set to 100, 4, 512, 0.0 and 64, respectively. for other benchmark models, both embedding and hidden dimensions of dkt, dkvmn and sakt are searched over [50, 100, 150, 200, 256, 512] , and the best results are reported on the test dataset. also, we set the number of latent concepts as 64 for dkvmn. table 2 shows the performance comparison of saint+ with the benchmark knowledge tracing models. compared to the benchmark models, saint+ achieves increases of acc and auc maximally up to 2.72% and 3.61%, respectively. also, saint+ improves saint with 1.03% and 1.25% gain in acc and auc, respectively, demonstrating the effectiveness of integrating the temporal features for knowledge tracing. 5.5.1 temporal feature embedding: continuous vs. categorical. we compare different approaches for embedding temporal features: continuous and categorical. modeling elapsed time in continuous (resp. categorical) fashion assumes that the relationship between a student's knowledge status and the difficulty of a question for the student is a smooth (resp. stepwise) function of time. similarly, for lag time, continuous (resp. categorical) modeling addresses various aspects in a students' learning process including forgetting, re-organizing concepts and improvement change smoothly (resp. table 3 show that the best result is obtained when using continuous embedding for elapsed time and categorical embedding for lag time. table 4 . when used alone, lag time increases performance more than elapsed time. moreover, regardless of the temporal features used, utilizing only one feature is not good as using both features together (saint+), while certain improvements are obtained compared to saint. figure 2 . since both elapsed time and lag time arise as a result of student response, this approach is naturally aligned with the core idea of saint that providing exercise information to the encoder and response information to the decoder is appropriate for knowledge tracing. we compare saint+ (decoder only) with other two variants: feeding the temporal features to 1) the encoder only, and 2) both the encoder and the decoder. table 5 summarizes the results. as expected, saint+ shows the best performance among the variants. also, saint+ and the variants show better results than saint, demonstrating that the temporal features provide useful information for estimating knowledge status. in this paper, we proposed saint+, a transformer based knowledge tracing model that processes exercise information and student response information separately, and integrates two temporal feature embeddings into the response embeddings: elapsed time and lag time. experiments conducted on ednet dataset show that saint+ improves saint, the former state-of-the-art knowledge tracing model, in both acc and auc. furthermore, the best result was obtained by incorporating the temporal features into the decoder input, verifying the hypothesis that separately processing exercise information and student response information is appropriate for knowledge tracing. avenues of future work include 1) modeling not only students' problem-solving records, but also various learning activities, such as watching lectures and studying explanations for each exercise, 2) exploring architectures for knowledge tracing models other than transformer based encoder-decoder model that separately processes exercise information and student response information. towards an appropriate query, key, and value computation for knowledge tracing ednet: a large-scale hierarchical dataset in education knowledge tracing: modeling the acquisition of procedural knowledge. user modeling and user-adapted interaction more accurate student modeling through contextual estimation of slip and guess probabilities in bayesian knowledge tracing context-aware attentive knowledge tracing deep residual learning for image recognition long short-term memory ekt: exercise-aware knowledge tracing for student performance prediction estimating programming knowledge with bayesian knowledge tracing machine learning approaches for learning analytics: collaborative filtering or regression with experts creating a neural pedagogical agent by jointly learning to review and assess graph-based knowledge tracing: modeling student proficiency using graph neural network a self-attentive model for knowledge tracing navigating the parameter space of bayesian knowledge tracing models: visualizations of the convergence of the expectation maximization algorithm adapting bayesian knowledge tracing to a massive open online course in edx modeling individualization in a bayesian networks implementation of knowledge tracing deep knowledge tracing does time matter? modeling the effect of time with bayesian knowledge tracing incorporating scaffolding and tutor context into bayesian knowledge tracing to predict inquiry skill acquisition convolutional knowledge tracing: modeling individualization in student learning process recommender system for predicting student performance hgkt: introducing problem schema with hierarchical exercise graph for knowledge tracing properties of the bayesian knowledge tracing model attention is all you need individualized bayesian knowledge tracing models dynamic key-value memory networks for knowledge tracing key: cord-311220-3pn04u32 authors: gaddy, hampton gray title: using local knowledge in emerging infectious disease research date: 2020-06-13 journal: soc sci med doi: 10.1016/j.socscimed.2020.113107 sha: doc_id: 311220 cord_uid: 3pn04u32 emerging infectious diseases (eids) are a growing global health threat. the best research protocol to date on predicting and preventing infectious disease emergence states that urgent research must commence to identify unknown human and animal pathogens. this short communication proposes that the ethnobiological knowledge of indigenous and impoverished communities can be a source of information about some of those unknown pathogens. i present the ecological and anthropological theory behind this proposal, as well as a few case studies that serve as a limited proof of concept. this paper also serves as a call to arms for the medical anthropology community. it gives a brief primer on the eid crisis and how anthropology research may be vital to limiting its havoc on global health. local knowledge is not likely to play a major role in eid research initiatives, but the use of the incorporation of eid awareness into standard medical anthropological practice would have myriad benefits, even if no eids were discovered this way. emerging infectious diseases (eids) present a major threat to global health. they are communicable human diseases that have recently grown in their geographic and/or host range. for example, west never fever, lyme disease, and methicillinresistant staphylococcus aureus (mrsa) are eids that have that spread geographically in recent decades. eids that have recently jumped the species barrier from infecting non-humans to infecting humans are called zoonoses, and they include ebola virus disease, hiv/aids, and covid-19. zoonoses comprise about 75% of all eids that have emerged in recent years, and they are especially feared for the toll they could take on global health over time (world health organization, 2014) . that toll is expected to grow. the 20 th century paradigm of host-pathogen interaction suggests that species barriers should be very difficult for pathogens to overcome, but the stockholm paradigm, a new but well-substantiated understanding of pathogens, challenges this. it proposes that species barriers tend to be much lower than previously thought and will tend to fall significantly in the face of anthropogenic environmental change (brooks et al., 2014 (brooks et al., , 2019 . for example, climate change is already associated with many recent vector-borne disease outbreaks (reisen, 2015) , since the 1980s, eids have been discovered at a rate of more than three per year (woolhouse & gaunt, 2007) , and that rate has been increasing since the 1940s (jones et al., 2008) . as such, researchers from various disciplines view eids as a growing crisis-one that requires interdisciplinary research, especially in order to limit infectious disease emergence (parkes et al., 2005; goodwin et al., 2012; brooks et al., 2019) . this paper aims to contribute to that research agenda. the main research protocol for preventing infectious disease emergence is dama (documentation-assessment-monitoring-action). it proposes an integrated research initiative to isolate and categorize unknown pathogens ('documentation'), identify which are most likely to become eids ('assessment'), develop surveillance networks for those pathogens ('monitoring'), and take proactive steps to minimize human exposure to them ('action') (brooks et al., 2014 (brooks et al., , 2019 . dama is still very much on its first phase of implementation. the largest research program that has studied zoonosis emergence was predict, a ten-year effort funded by the u.s. agency for international development (usaid) and launched in 2009. it focused on isolating pathogens from animals in likely hotspots of zoonotic emergence, i.e. low-income areas of africa and asia (kelly et al., 2017) . as yet, that seems to be the only tested strategy for 'documentation.' this paper suggests that utilizing local knowledge may also be a useful strategy for documenting potential eids. local knowledge refers to "the knowledge that people in a given community have developed over time, and continue to develop. it is based on experience; often tested over centuries of use; adapted to the local culture and environment; embedded in community practices, institutions, relationships and rituals; held by individuals or communities; and dynamic and changing" (food and agriculture organization, 2004). the concept of 'local knowledge' significantly overlaps with those of 'traditional knowledge,' 'indigenous knowledge, 'traditional ecological knowledge,' and 'situated knowledge,' each of which tends to be used in different social sciences. this paper primarily uses the term 'local knowledge,' as it is the broadest of these concepts. however, i primarily use it to refer to the experience-and community-based knowledge of indigenous people and vulnerable communities in low-income countries. borrowing from the idea of local knowledge as 'situated knowledge' (e.g. haraway, 1988; nazarea, 1999; tschakert et al., 2016) , i refer to these communities as 'locally-situated communities.' local knowledge is already used as a source of empirical knowledge to different extents by different disciplines. traditional ecological knowledge is a bountiful source of pharmaceutical drugs, as well as knowledge about ecology, environmental health, and the health of indigenous communities (e.g. alves & rosa, 2007; finn et al., 2017) . moreover, the study of local disease knowledge is well-established within medical anthropology. in this paper, i propose that local knowledge is also a useful source of information about potential eids that are unknown to science. local knowledge can be a problematic source of information. however, it likely contains relevant information about human diseases, animal diseases, and zoonoses that has not yet been empirically documented. i will discuss this possibility with reference to theoretical considerations and some limited case studies. there are theoretical reasons to suspect that local knowledge could be useful for identifying potential eids. as already mentioned, ethnobiological systems can surpass empirical knowledge of certain subjects. those subjects are generally distant from the purview of empirical knowledge gathering systems, e.g. geographically remote or in lowincome settings, and they are often salient to local life. human and animal diseases that exist in locally-situated communities fit both of those criteria. useful information about diseases of which the scientific community was already aware. a study of local knowledge in ghana about buruli ulcers, a poorly understood, necrotic infection caused by mycobacterium ulcerans, revealed information that is likely useful for understanding the etiology and life cycle of the disease (tschakert et al., 2016) . similarly, local knowledge in malaysia and latin america has been helpful in identifying neglected hideouts of the animal vectors that transmit dengue fever (dickin et al., 2014) and chagas disease (abad-franch et al., 2011) , respectively. given that such communities have independent knowledge of diseases already known to science, it is likely that they have knowledge of diseases not known to science. this is not a valid inference if science is assumed to have a complete understanding of disease, but this is not the case. accurate estimates of global pathogen diversity are very difficult to obtain, but only a small fraction of potential pathogens have been empirically documented (pedrós-alió & manrubia, 2016). moreover, there are ecological reasons why locally-situated communities are more likely to be in contact with animal diseases. such communities often live in highly biodiverse environments. cultural diversity correlates with biodiversity on a global scale (gorenflo et al., 2012) , and it is likely that this principle applies to pathogen diversity as well (guernier et al., 2004) . therefore, locally-situated populations are arguably in closer proximity to a higher density of animal disease than any other population just because of their geography-not to mention because of any economic or subsistencedriven needs for such communities to interact with animals. it is also likely that locally-situated communities face relatively high rates of zoonotic infection. sharing an ecosystem with a great diversity of pathogens increases the risk of disease transmission almost in and of itself, given that risk of zoonotic infection is a function of ecological connectivity (brooks et al., 2019) . additionally, ecological degradation is a risk factor for zoonosis (ostfeld, 2009; bonds et al., 2012) , and locally-situated communities are generally the most likely to live in degrading ecosystems. the ecosystems in which such communities live are among the most likely to face damaging anthropogenic pollution, e.g. due to poor regulation and the presence of pollution-prone industries (hoover et al., 2012; sapkota & bastola, 2017) . such communities are also least able to mitigate the effects of climate change in their environments. for arctic communities, this is especially true because warming is happening faster there than any other region of the planet (dai et al., 2019) . therefore, locally-situated communities can be expected to be in especially frequent proximity to human diseases, animal diseases, and zoonoses that are unknown to the scientific community. one can expect those communities to have knowledge of at least some of those diseases. zoonoses, as novel diseases, should always have a high degree of salience within communities. regarding the salience of animal diseases, indigenous communities, in particular, rely heavily on the animals in their environment for subsistence. furthermore, many such groups have a high degree of psychological connection with the other organisms around them (salmón, 2000) . animal diseases can be practically, economically, and spiritually relevant to indigenous and low-income communities. there is the problem that pathogens themselves are too small to be observed without formal training, and the same may apply to certain symptoms of human and animal diseases. ethnobiological theory recognizes that the size of natural features limits the perceptual salience of those features (hunn, 1999) . despite that, locally-situated communities do, in fact, tend to have extensive knowledge about animal disease. therefore, since there is likely a high density of potential eids in those communities and those communities have likely noticed many of those diseases, their local knowledge can be expected to be a useful source of novel knowledge about potential eids. history also tells us that this is likely the case. the most salient human and animal diseases were likely known first by ancient populations, who, of course, did not have modern methods for collecting knowledge (see hoeppli, 1956 ). there are, of course, problems with using local knowledge as a source of information about eids. translating local knowledge into empirically relevant information can be difficult at times. for example, this would be a problem if one encounters a locally described disease for which there is no biomedical name. understanding the local description of the disease may prove challenging, especially as doing so may require anthropological and/or linguistic analysis (kleinman, 1980) . then, empirically confirming the existence and nature of that disease may prove challenging, especially if the disease is an alleged zoonosis. i recognize that interdisciplinary integration is often difficult and costly, but the recruitment of anthropologists, linguists, and local guides can mitigate all of these challenges. furthermore, if there is a paradigm shift towards viewing local knowledge as a valued source of disease surveillance, the opportunity costs of collecting and using it may fall. incorporating an awareness of eids into medical anthropology has benefits even if no eids are discovered this way. in particular, i think that an awareness of eids can produce a more dynamic understanding of health progress and security. for example, the current covid-19 pandemic has shown that although traditional statistics like infant mortality rates and life expectancy are usually good markers of health progress, the fundamental mark of a strong health system is its ability to respond to an emerging threat. recognition of the fact that eid outbreaks are likely to become more common may lend new importance to the study of resilience in health systems. that said, i think that the limited case studies that follow show that there is some real promise to discover eids by using local knowledge. the literature on local knowledge and potential eids is scarce. i will discuss three useful but limited case studies below. however, it does not seem that any anthropological or ethnobiological studies have been conducted with the explicit aim of learning about diseases unknown to science. this is not that surprising. anthropology is the discipline that has studied indigenous medicine and disease to the greatest extent. however, until recent decades, medical anthropology had focused more on the ritual and psychology of local medical systems than any biological or epidemiological knowledge that could be gained from studying them (waldstein & adams, 2006) . there is now a slowly growing recognition within health science literature that non-scientist stakeholders can be vital in the production of empirical knowledge (e.g. catley et al., 2012; quinlan & quinlan, 2016; den broeder et al., 2018) . but, i found no studies of local health knowledge that engaged with the possibility that locally-situated communities might know of human infectious diseases, animal diseases, or zoonoses that are still unknown to science. most studies did not actually seem to account for that possibility in their design. i suspect that this disregard for local disease knowledge is partly because of the popular but incorrect perceptions that infectious disease emergence is an uncommon and unpredictable event (brooks et al. 2019) . that thought presumably leads researchers to assume that indigenous knowledge would not be helpful in discovering new diseases. for those reasons, finding proof of concept for what i propose is difficult. however, i present three somewhat promising case studies. the first comes from a study of disease knowledge held by fula-speaking pastoralists in the far north region of cameroon (moritz et al., 2013) . locals identified two diseases called haahaande and gawyel, which they said could infect both cattle and humans. the authors of this study dismissed this information, saying that heartwater (the scientific name for haahaande) and blackleg (the scientific name for gawyel) are not zoonotic diseases. case reports suggest that the local knowledge may have been correct. heartwater is a systemic cattle disease caused by ehrlichia ruminantium; it is characterized by increased vascular permeability that causes fatal respiratory, cardiovascular, and neurological symptoms. there are reports that e. ruminantium can infect humans (esemu et al., 2011) , in whom it may cause fatal encephalitis (allsopp, 2005) . furthermore, e. ruminantium has been shown to infect a wide variety of wild and domestic animals (peter et al., 2002) , suggesting that it is a plausible zoonotic agent. blackleg is a myonecrotic disease caused by species of clostridium. moritz et al. (2003) suggested that gawyel could refer to infection by c. chauvoei or the less common c. septicum. c. chauvoei is known to have infected humans in at least two cases-though it is unclear whether those cases were zoonotic (nagano et al., 2008; weatherhead & tweardy, 2012 ). an analysis of c. chauvoei genomes concluded that the species has too little genetic diversity to infect non-ruminants (rychener et al., 2017) , but recent paradigms of infectious disease emergence suggest that intraspecific genetic diversity is not necessarily requisite for pathogens to colonize new host species (brooks et al. 2019) . c. septicum is known to infect humans, and some human cases have been presumed to be zoonotic in origin (barnham, 1998) . however, neither species of clostridium has been widely recognized as a zoonotic pathogen (e.g. songer, 2010) . this case study suggests that local knowledge can be a source of information about e. ruminantium and clostridium zoonosis. a second case study comes from research on maa-speaking pastoralists in northern and eastern tanzania (mangesho et al., 2017) . these pastoralists also identified blackleg (locally called emburuo) as a zoonosis. additionally, they identified a respiratory infection called mapafu ya kikohozi ya mbuzi, which they said was contracted from goats, and they identified a fungal infection called ndororo, which they said was contracted by stepping in raw cattle remains or slurry. these two diseases were only mentioned in passing by the authors, as was emburuo, and no attempts were made to identify a biomedical name for them, nor to anthropologically or medically validate their existence. i mention this case because of how frustrating it is. ndororo and mapafu ya kikohozi ya mbuzi may refer to diseases that have already been empirically documented, diseases that are not empirically known, or they may be local constructions that do not correspond to any biomedical illness. ndororo could refer to any number of foot fungi. mapafu ya kikohozi ya mbuzi could be contagious caprine pleuropneumonia, a respiratory disease known to infect goats in tanzania, but this disease has never been reported in humans (iqbal yatoo et al., 2019) . in the absence of more information, there is no way to judge whether the diseases are novel or even medically real. however, there are countless similarly brief and frustrating references in existing literature, e.g. ente kurwaara omutima and okunvara ckine in katungukarwakishaya et al. (2004) , eyaliyal in gradé et al. (2009) , and many diseases in catley & mohammed (1995) . i propose that much of this local knowledge may be useful for identifying potential eids. a third case study comes from camel-herding pastoralists in somalia and northern kenya. surveys of these communities have consistently identified two camel respiratory diseases with distinct symptomatic and epidemiological characteristics (wako et al., 2016) . one is called hergeb in somali; it is characterized by frequent outbreaks that cause nasal discharge and mortality among young camels. the other is called dhuguta in somali; it is characterized by infrequent outbreaks that cause coughing and emaciation. there is no empirical knowledge about these diseases. several respiratory pathogens are known to cause disease in camels, including strains of influenza, respirovirus, mycoplasma sp., streptococcus sp., small ruminant morbillivirus, and infamously, mers-cov. however, none of these have yet been identified as hereb or dhuguta. furthermore, either might be one of the many other locally known camel respiratory diseases that have been described across africa, e.g. mhaz in volpato et al. (2015) , sonbobe in bekele (1999) , and ah and laxawgal in catley & mohammed (1995) . given the limited scientific knowledge about camel respiratory diseases and the abundance of local knowledge about them, it seems that the local knowledge of pastoralists should certainly be studied further in the future. case studies show that local knowledge can be a useful source of new information about human diseases (e.g. buruli ulcers), animal diseases (e.g. camel respiratory infections), and potential zoonoses (e.g. blackleg and heartwater). this is not surprising, given how sophisticated systems of ethnomedical and ethnoveterinary knowledge have been found to be. this is also not surprising given the apparent proximity of locally-situated communities to animal, pathogen, and zoonosis biodiversity. this review is preliminary, but the theoretical considerations and limited case studies i present suggest that locally-situated knowledge can be an important source of information about potential eids. the dama protocol is currently our best defense against the growing eid crisis, and the documentation stage, the 'd' in dama, should not fail to engage with the documentation of local knowledge. the inclusion of anthropologists in diverse public health disciplines is growing (e.g. stellmach et al., 2018) , and the study of eids is another area in which their expertise is needed. community participation in chagas disease vector surveillance: systematic review ehrlichia ruminantium: an emerging human pathogen? biodiversity, traditional medicine and public health: where do they meet? clostridium septicum infection and hemolytic uremic syndrome studies on the respiratory disease 'sonbobe' in camels in the eastern lowlands of ethiopia. tropical animal health and production serologic evidence for human ehrlichiosis in africa participatory epidemiology: approaches, methods, experiences. the veterinary ethnoveterinary knowledge in sanaag region, somaliland (part ii): notes on local methods of treating and preventing livestock disease arctic amplification is caused by sea-ice loss under increasing co2 citizen science for public health mosquitoes & vulnerable spaces: mapping local knowledge of sites for dengue control in seremban and putrajaya malaysia ehrlichia species, probable emerging human pathogens in sub-saharan africa: environmental exacerbation the value of traditional ecological knowledge for the environmental health sciences and biomedical research food and agriculture organization interdisciplinary approaches to zoonotic disease ethnoveterinary knowledge in pastoral karamoja ecology drives the worldwide distribution of human diseases situated knowledges: the science question in feminism and the privilege of partial perspective the knowledge of parasites and parasitic infections from ancient times to the 17th century indigenous peoples of north america: environmental exposures and reproductive justice size as limiting the recognition of biodiversity in folkbiological classifications: one of four factors governing the cultural recognition of biological taxa contagious caprine pleuropneumonia -a comprehensive review global trends in emerging infectious diseases organisation for social science research in eastern and southern africa (ossrea) one health proof of concept: bringing a transdisciplinary approach to surveillance for zoonotic viruses at the human-wild animal interface patients and healers in the context of culture: an exploration of the borderland between anthropology, medicine, and psychiatry exploring local knowledge and perceptions on zoonoses among pastoralists in northern and eastern tanzania on not knowing zoonotic diseases: pastoralists' ethnoveterinary knowledge in the far north region of cameroon human fulminant gas gangrene caused by clostridium chauvoei ethnoecology: situated knowledge/located lives biodiversity loss and the rise of zoonotic pathogens all hands on deck: transdisciplinary approaches to emerging infectious disease the vast unknown microbial biosphere ehrlichia ruminantium infection (heartwater) in wild animals ethnobiology in one health vector-borne diseases clostridium chauvoei, an evolutionary dead-end pathogen kincentric ecology: indigenous perceptions of the human-nature relationship foreign direct investment, income, and environmental pollution in developing countries: panel data analysis of latin america clostridia as agents of zoonotic disease anthropology in public health emergencies: what is anthropology good for? situated knowledge of pathogenic landscapes in ghana: understanding the emergence of buruli ulcer through qualitative analysis ethnoveterinary of sahrawi pastoralists of western sahara: camel diseases and remedies indigenous knowledge of pastoralists on respiratory diseases of camels in northern kenya the interface between medical anthropology and medical ethnobiology key: cord-303000-tmk2c9eh authors: alhaj, ahmad kh.; al-saadi, tariq; mohammad, fadil; alabri, said title: neurosurgery residents perspective on the covid-19: knowledge, readiness, and impact of this pandemic. date: 2020-05-16 journal: world neurosurg doi: 10.1016/j.wneu.2020.05.087 sha: doc_id: 303000 cord_uid: tmk2c9eh abstract background the novel coronavirus disease (covid-19) is a life-threatening illness, which represents a challenge to all the health-care workers. neurosurgeons around the world are being affected in different ways. objectives this is the first study regarding the readiness of neurosurgery residents towards the covid-19 pandemic and its impact. the aim is to identify the level of knowledge, readiness, and the impact of this virus among neurosurgery residents in different programs. methods a cross-sectional analysis in which 52 neurosurgery residents from different centers were selected to complete a questionnaire-based survey. the questionnaire comprised of three sections and 27 questions that ranged from knowledge to impact of the current pandemic on various features. results the median knowledge score was 4 out of 5. the proportion of participants with satisfactory knowledge level was 60%. there was a statistically significant difference between the knowledge score and location of the program. around 48% of the neurosurgery residents dealt directly with covid-19 patients. receiving a session about personal protective equipment (ppe) was reported by 57.7%. the neurosurgery training at the hospital was affected. about 90% believed that this pandemic influenced their mental health. conclusion neurosurgery residents have a relatively good knowledge about covid-19. the location of the program was associated with knowledge level. most of the participants did not receive a sufficient training about ppe. almost all responders agree that their training at the hospital was affected. further studies are needed to study the impact of this pandemic on neurosurgery residents. the novel coronavirus disease (covid-19) is a respiratory tract viral infection, caused by the newly emergent, severe acute respiratory syndrome coronavirus (sars-cov-2). 1, 2 it is a life-threatening viral illness, which represents a challenge to all health-care workers over the globe. the world health organization (who) reports that this viral infection confers a 3% to 4% crude mortality rate. 3 this pandemic has affected everyone in all aspects of daily life, especially in the healthcare. the quality of residency training is negatively influenced as a result of the recent pandemic. as the number of individuals infected with this virus rapidly increases, neurosurgeons from different nations are significantly affected in multiple ways. 4, 5, 6, 7 neurosurgery residents are now facing a major challenge, especially for those who work in hospitals with a high number of covid-19 patients. in addition, some residents are fully responsible for patients with this infection. the strategies to increase the regional intensive care unit (icu) allowance included the reduction of all surgical activities, starting with elective, to ultimately, also, include some urgent cases. 6 currently, most of the neurosurgical centers postponed their elective surgeries due the burden of this infection. 4, 5, 8 furthermore, several programs have reduced the number of residents by 50% of normal, thus keeping the remainder of the residents at home. 7 almost all neurosurgery programs around the world have changed their academic meetings to online communication in an attempt to reduce physical contact. to our best knowledge, this is the first study regarding the readiness of neurosurgery residents towards the covid-19 pandemic and the impact it has on their training. the aim of this study is to identify the level of knowledge, readiness, practices, as well as the impact of this virus among neurosurgery residents in various neurosurgical programs. this cross-sectional study involves the assessment of neurosurgery residents through a questionnaire-based survey. the study was conducted during the pandemic, from the 14 th until the 28 th of april 2020. the sample size "n" is represented by a total of 52 respondents from different neurosurgical programs. they completed the survey (appendix a) on the awareness, knowledge, practices, and safety measures about covid-19. the questionnaire was mainly adapted from the current interim guidance and information for healthcare workers, published by the us centers for disease control and prevention (cdc). 10 several editorial studies published recently about the impact of the virus on neurosurgery residents was also utilized to create the questionnaire. 5, 7, 8 the target population consists roughly of around 300 residents, comprised of neurosurgery residents from various neurosurgical centers that we chose. a representative sample from canada, united stated of america (u.s.a.), kuwait, saudi arabia, serbia and italy were selected. we divided the regions into north america, which includes canada and u.s.a. in addition, saudi arabia and kuwait represent the programs in the arabian gulf cooperation council (gcc) countries. regarding the european programs, we reached out to residents from serbia and italy. residents from different centers where selected from three different regions, with the number of participants from each region being close to one another. moreover, we communicated with senior neurosurgeons from these regions, and they provided us with lists of residents with their contacts, which we selected randomly from. therefore, our sampling procedure comprised of random selection of the participants. in addition, each resident was reached in person via a direct phone call or a text message in order to restrict the data to our inclusion criteria, which involved only neurosurgical residents. moreover, this method of direct contact facilitated a very high response rate. all neurosurgeons who finished their training or were above the sixth year of the program were excluded from our data. an informed consent was obtained from each subject. the study objectives were explained to the residents. they were also assured regarding confidentiality of the collected information, and that they were free to decline participation in the study. one participant who refused to complete the survey was excluded. the questionnaire consisted of three sections and 27 questions. the first section (8 items) involves the baseline information: gender, age, location of the program, year of training, and current health condition. the next section (13 items) contains inquiries about basic biological and microbiological knowledge of this virus 11, 12 , hand hygiene, as well as personal protective equipment (ppe). 10 additionally, we evaluated whether the subjects received any formal training in hand hygiene, ppe, and n-95 mask handling. the final section (6 items) focuses on the impact of this pandemic on the resident in terms of neurosurgical training, studying, mental health, as well as whether their social life was affected or not. 5, 7, 8 convenient sampling method was used for data collection, and the distribution of qualitative responses was presented as frequency and percentages. sub-groups were classified on the basis of gender, age, location of the program, and year of residency training. the statistical package for social sciences (ibm spss statistics 23, ibm corporation, armonk, ny, usa, 2016) was used for data entry and analysis. first, univariate analysis was conducted, and qualitative variables were described by frequency and percentage. the quantitative variable (total knowledge score) was calculated by adding the points for the five knowledge items (each item equals one point). this variable, with a non-normal frequency distribution, was summarized by a median and inter-quartile range. we determined that the cut-off of the satisfactory knowledge level is a total knowledge score ≥ median. also, a logistic regression model was used to identify the determinants of low knowledge level. at that point, p-value ≤ 0.05 was used as the cut-off level for statistical significance. pearson's chi-square test was utilized to assess the association between the qualitative variables. mann-whitney u test was used to compare two groups with a non-normal frequency distribution, while kruskal-wallis one-way analysis of variance test was used to compare more than two groups. we tested the association of our questions in relation to age, gender, location of the program, and year of residency training. in the present cross-sectional sample survey, 53 neurosurgery residents attending different centers around the world were contacted directly from the six countries mentioned earlier. out of this number, 52 participants returned a completed self-administered questionnaire, and hence, the analysis was based on this number (response rate = 98.1%). table 1 depicts the descriptive analysis of self-reported baseline information and the current health status of the residents with regard to covid-19 virus. the majority of the participants were male (73.1%). concerning the age, 69.2% were below 30 year old, and 30.8% were 30 year-old or above. regarding the location of the neurosurgery program, the percentage of residents representing each country in our sample were as follow: canada 36.5%, u.s.a. 9.6%, kuwait 9.6%, saudi arabia 23.1%, and from the european countries (italy and serbia) 21.1%. the frequency of participants from each year of the residency (r) training were: (r1) 26.9%, (r2) 11.5%, (r3) 23.1%, (r4) 17.3%, (r5) 15.4% and (r6) 5.8%. besides, table 1, also shows the current situation of residents in terms of this pandemic: 17.3% were under stay home order by their institution or the government; however, 82.7% are resuming their work at the hospital. in addition, according to our results, 21.2% of the neurosurgery residents were under quarantine or isolation. from our sample, only one resident from europe tested positive for covid-19. furthermore, about 36.5% were negative, the rest, which represent the majority, 61.5%, were not tested for the infection. almost half of the responders, 48.1%, dealt directly with covid-19 patients, while the rest did not. table 2a shows the frequency of correct responses to the five items of the knowledge score about the virus and the safety measures in relation to the location of the program. the number of residents who answered the questions correctly were as follows: 76.9% knew that the virus type, 90.4% knew the main mode of transmission, and 86.5% recognized the most common symptoms. the most accurate estimation of the incubation period of this virus was answered by only 40.4% of participants. concerning the preferred hand hygiene method in the healthcare settings; unexpectedly, only 51.9% knew the correct answer. the first section of the table also displays the responses according to each location of the program in details, and some of the items showed statistically significant results. in table 2b, we recorded the responses about the training of safety measures, the safe practices, and the strategies of infection control. as expected, receiving a formal hand hygiene training was reported by 78.8%. in addition, receiving formal session of the correct sequence of ppe donning and doffing was stated by 57.7%. only 50% of our sample knew their correct size of n-95 mask prior to this pandemic. likewise, only 50% knew how to correctly obtain a nasopharyngeal swab sample. most of the results in the previously mentioned items were significantly associated with the location of the program. the distribution of the total knowledge score was shifted to the right with left skewness (figure 1). the maximum total knowledge score was five, and the median knowledge score of our participants was four out of five, representing a good knowledge level. table 3 demonstrates the association of this score about the virus with gender, age, year of residency training, location of the neurosurgery program. the table also depicts the association between the knowledge score and whether the participant was taking care of covid-19 patients. the median for male residents was four, compared to three in females. there was neither a difference in medians nor statistical significant association of the knowledge score with regard to age and year of residency training. the median in each location was as follow: four, four, and three in north america, gcc countries, and europe region, respectively. the difference in medians was statistically significant between europe and north america, as well as between europe and gcc countries (p-value = 0.049). figure 2 shows the boxplot of the knowledge score based on the region of the program. the cut-off of the satisfactory knowledge level was four, which is the median. in our results, 59.6% had satisfactory knowledge about the virus, while 40.4% had a non-satisfactory level. the impact of this pandemic among neurosurgery residents is shown in table 4. almost all of the residents found that their training at the hospital was affected. when asked about their opinion regarding the neurosurgical procedures during this pandemic, roughly 42% desired to resume their elective surgical procedures. additionally, the daily studying hours was affected by about 80%, while the remaining did not face a change in the studying hours per day. the social life of all residents in our sample was influenced by the current situation. shockingly, this pandemic affected the mental health of 90% of the participants. in table 5, the association of the impact on mental health of covid-19 and year of training revealed that the mental health of all residents in the first, third, and sixth year of training was affected. although the percentage of fifth year residents was also high (62.5%), they were the lowest group in this aspect. this association was significant (p-value 0.006), but other confounders' effect could not be eliminated. our study is the first regarding the readiness of neurosurgery residents towards the covid-19 pandemic and the impact it has on their training. this pandemic is evolving rapidly worldwide, disrupting personal and professional life, including that of neurosurgeons and neurosurgical residents. 4 most programs have seen a significant drop in elective or nonessential surgical volume, impacting the functional neurosurgery cases foremost. 7 regarding surgeries, around 57.7% in our study agreed that elective neurosurgical procedures should not be resumed during this pandemic ( figure 3-a) . in another editorial, authors stated the following "we have halted all elective cases, but will continue to schedule urgent and emergent cases, involving head and spine trauma, cauda equina syndrome, embolic stroke, ruptured aneurysms, and acute hydrocephalus are relatively noncontroversial; however, urgent cases such as malignant brain tumors and progressive cervical spondylotic myelopathy may require a more nuanced discussion." 4 responses from our analysis disclosed that only 23% think that brain tumor or compressive spinal cord tumor surgeries should be postponed. in terms of skillset, a trustworthy neurosurgical team should have the ability to treat patients with infectious diseases who also require emergency operations. 8 at the same time, in our survey, most of the residents (≈ 80%) will only do emergency surgery on a confirmed covid-19 patient if there are appropriate ppe, while a minority (≈ 20%) will perform it regardless the presence or absence of ppe; none of the participants refused to perform this surgery in either way ( figure 3-c) . access to and training on proper ppe use are critical to the safety of workers. 14 when asked about the residents' opinion, if neurosurgical programs should involve a session about ppe every year, around 73% agreed that this session is essential ( figure 3-b) . overall, programs report a significant decrease in the volume of cases. clinic visits have transitioned to telemedicine where possible, decreasing resident exposure to outpatient encounters. 7 similarly, all in-person conferences such as grand rounds, resident education conferences, and multidisciplinary meetings have been replaced by video teleconferences. 4 in concordance with our expectations, almost all of the residents found that their training at the hospital was affected. certainly, once this pandemic has concluded, careful retrospective analysis of its impact on resident case volume will be necessary to ensure we are prepared for any future event. 7 the american board of neurological surgery has postponed both primary and oral examinations. 7, 15 the royal college of physicians and surgeons of canada has also decided to postpone the written exam, while the oral component will no longer be required. 16 in our sample, the daily studying hours was affected in about 80%. nonetheless, the studying hours might be affected positively or negatively. a study, about involving physicians in patients' care during epidemics, advised of possible alternatives to real patient-physician interaction as to avoid placing trainees at risk. 17 less than half of the neurosurgery residents (46.2%) feel competent in taking care of covid-19 patient, most of those who feel capable have already dealt with covid -19 patients (figure 4) . due to increasing number of covid-19 patients who require hospitalization, some radiology residents have been reassigned to internal medicine and icu as to care for the high influx of patients. 17 similarly, almost half of the neurosurgery residents in our sample, 48.1%, dealt directly with covid-19 patients in the previously mentioned settings. on the other hand, some neurosurgery residents might gain intensive care skills. in the end, working temporarily in the icu or internal medicine department will benefit any physcians in one way or another. "social distancing measures have circumvented the traditional trainee-faculty member workstation teaching, which is especially disadvantageous for residents who may be rotating on a service for the first or second time in their training." 14 our survey revealed that 96.2% of the sample followed social distancing in a daily manner, while only two of the participants found it difficult to stick to this practice. "work-related stress is a potential cause of concern for health professionals. it has been associated with anxiety including multiple clinical activities, depression in the face of the coexistence of countless deaths, long work shifts with the most diverse unknowns and demands in the treatment with patients with covid-19." 18 in our data, this pandemic negatively affected the mental health of 90% of the participants. however, this influence is not specific, and might affect the residents mental health either positive or negative way. the median knowledge score about covid-19 pandemic and infection control measures was four out of five (80%) with a range from one to five. the correct answers to the knowledge questions were the following: 76.9% knew that the virus type is an rna virus single-stranded 11 and 90.4% knew that the main mode of transmission is via respiratory droplets. 12 moreover, the most common two symptoms of the virus are fever and cough 12 , which 86.5% got correct. the most accurate estimation of the incubation period of this virus is 5 days 12 , and it was answered by only 40.4% of participants. concerning the preferred hand hygiene method in the healthcare settings, which is hand rub for at least 20 seconds with 70% ethanol 10 ; unexpectedly, only 51.9% knew the correct answer. on the other hand, 42.3% think that hand rub for at least 20 seconds with soap and water is the preferred method. only 50% of the residents knew their correct n-95 mask size, as it is required to safely manage any suspected or confirmed cases. this study has some limitations. temporal association of the knowledge score with the aspects discussed above cannot be definitely established due to the nature of the cross-sectional study design. besides, the pandemic has affected the whole world in diverse degrees at different times. we have reached the neurosurgery residents in only six countries with different academic settings and resources. therefore, the ability to generalize the results of this study to all neurosurgery programs may be limited. our study suggested that neurosurgery residents have as a relatively good level of knowledge about covid-19 pandemic, despite that a significant number of the participants did not reach the satisfactory level of the knowledge score. the location of the program was independently and significantly associated with knowledge score after adjusting for confounding between variables. concerning the infection control, most of the neurosurgery residents received a formal hand hygiene training prior to the pandemic. however, receiving a formal training of the correct sequence of ppe donning and doffing was not sufficient. most of the responders agree that their training at the hospital was affected. larger retrospective studies that include a representative sample of neurosurgical residents with a wide range of regions is essential in order to generalize the results to the target population. it is recommended that health authorities provide infection control sessions to prepare the residents for any future events. neurosurgery program directors may consider sharing experiences with other programs to enhance education and decrease the infection rate among surgeons. regarding surgery, delay of elective procedures but proceeding with semi-elective, urgent, and emergency surgeries is advised. that being said, surgeons should also be provided with sufficient quantities of ppe so as procedures can be performed safely. due to increased stress and decrease studying hours, we urge to provide the residents with adequate time to prepare for any upcoming evaluation. according to the reviewed editorials about this pandemic, we also encourage that the academic training should be continued with social distancing measures with a minimal number of attendees, or even online communication. furthermore, ease of access of residents to mental health professional to prevent any psychological traumatic event, and provide them with stress-management sessions is also suggested. • % = column % • p-values were generated using pearson's chi-square test (≤ 0.05 is statistical significant) figure 1 . frequency distribution of the total knowledge score among the neurosurgery residents in our sample. coronavirus disease (covid-2019) situation report-57 the outbreak of covid-19: an overview coronavirus disease (covid-2019) situation report-46 letter: academic neurosurgery department response to covid-19 pandemic: the university of miami/jackson memorial hospital model neurosurgical priority setting during a pandemic: covid-19 neurosurgery in the storm of covid-19: suggestions from the lombardy region, italy (ex malo bonum) impact of covid-19 on neurosurgery resident training and education response to covid-19 in chinese neurosurgery and beyond and chiocca ea. editorial. covid-19 and academic neurosurgery information for healthcare professionals coronaviruses post-sars: update on replication and pathogenesis q&a on coronaviruses (covid-19) covid-19 awareness among healthcare atudents and professionals in mumbai metropolitan region: a questionnairebased survey the impact of covid-19 on radiology trainees the american board of neurological surgery the royal college of physicians and surgeons of canada involving physicians-in-training in the care of patients during epidemics supporting the health care workforce during the covid-• gcc: arabian gulf cooperation council (gcc) countries • p-values were generated using the mann-whitney u test for comparing two groups, and *kruskal-wallis one-way analysis of variance test for comparing more than two groups & canada gcc: arabian gulf cooperation council countries (kuwait & saudi arabia). european country: data selection from italy & serbia we thank all the participants in this project for their time and effort. we also thank dr. dragan savic and dr. jacquelyn corley, for providing us the contacts of some neurosurgery residents. in addition, we would like to thank all the health-care workers in the front lines against covid-19 pandemic. this is the first study regarding the readiness of neurosurgery residents towards the covid-19 pandemic and its impact. the aim is to identify:• the level of knowledge about the novel coronavirus among neurosurgery residents in different programs. • we also studied and discussed the readiness of the neurosurgery residents.• the impact of this virus on the neurosurgery was illustrated in order to give a general picture of the effect of this pandemic on the training in the hospitals, studying, and the general well-being of the residents. • neurosurgery residents have a relatively good knowledge about covid-19.• the location of the program was associated with knowledge level.• most of the participants did not receive a sufficient training about personal protective equipment (ppe).• almost all responders agree that their training at the hospital was affected.• about 90% believed that this pandemic influenced their mental health.• further studies are needed to study the impact of this pandemic on neurosurgery residents. key: cord-026154-9773qanf authors: rezaei, navid; reformat, marek z.; yager, ronald r. title: image-based world-perceiving knowledge graph (wpkg) with imprecision date: 2020-05-18 journal: information processing and management of uncertainty in knowledge-based systems doi: 10.1007/978-3-030-50146-4_31 sha: doc_id: 26154 cord_uid: 9773qanf knowledge graphs are a data format that enables the representation of semantics. most of the available graphs focus on the representation of facts, their features, and relations between them. however, from the point of view of possible applications of semantically rich data formats in intelligent, real-world scenarios, there is a need for knowledge graphs that describe contextual information regarding realistic and casual relations between items in the real world. in this paper, we present a methodology of generating knowledge graphs addressing such a need. we call them world-perceiving knowledge graphs – wpkg. the process of their construction is based on analyzing images. we apply deep learning image processing methods to extract scene graphs. we combine these graphs, and process the obtained graph to determine importance of relations between items detected on the images. the generated wpkg is used as a basis for constructing possibility graphs. we illustrate the process and show some snippets of the generated knowledge and possibility graphs. knowledge graphs are composed of a set of triple relations, i.e. <subject -predicate -object>, where subjects and objects are items connected via predicates representing relations between them. the graphs are useful in representing data semantics and are employed in different applications, such as common-sense and causal reasoning [1, 2] , question-answering [3] , natural language processing [4] , and recommender systems [5] . some examples of existing knowledge graphs are dbpedia [6] , wikidata [7] , yago [8] , the now-retired freebase [9] , and wordnet [10] . the aforementioned knowledge graphs contain information about facts, their features, and basic relations between them. they focus on people, geographical locations, movies, music, and organizations and institutions. they are missing a piece of information about everyday real-world items, their contexts, and arrangements. from the human perspective, we can state that the visual information plays a significant role in human learning processes [11] . at the same time, the eye's information transfer rate is quite high [12] that makes a visual stimulus to be of significant importance in processes of gaining understanding about different items and how they are related to each other. given the importance of visual data, it is appealing to develop systems that could observe, learn and create knowledge based on such data. additionally, traditional knowledge graphs do not provide any degree of confidence associated with relations. it is assumed that all of them are equally important. in this paper, we look at the task of creating knowledge graphs based on visual data. the idea is to process images, generate scene graphs from them, and aggregate these graphs. graphs constructed in such a way contain knowledge about everyday objects, their contexts and their situational information, as well as information related to the importance of common-sense relations between multiple objects in their natural scenarios. we call such a graph world-perceiving knowledge graph, wpkg in short. the quality and suitability of knowledge we retrieve from images depend on the capability of tools and methods we use for image processing. processing an image means generating a scene graph representing relations between objects/entities present on this image. once numerous images are processed, all scene graphs are aggregated. this alone allows us to treat the process of constructing graphs via aggregation as the human-like process of learning via processing of observed images. we also look at a process of using knowledge graphs -wpkgs -to construct possibility graphs reflecting conditional dependencies between sets of entities as observed in their usual environments. the information about the importance of relations allows us to build possibilistic conditional distributions. they are used for processing and reasoning about entities and relations between them in their own relevant contexts. the included case study shows an application of the presented procedure to visual genome (vg) dataset [13] . extracting information from different media to create a knowledge graph has been examined in the literature. yet, the area of focus of these works has been different: some of them focus on images, some on text, and some on a combination of both. also, the methods used for information retrieval can be differentautomatic or manual. a brief overview is presented in subsect. 2.1. possibilistic knowledge bases and graphs are important forms representing uncertainty of data and information [14] , and [15] . a set of basic definitions is included in the following subsections. there is a number of different knowledge graph generation methods that focus on text as the source of information, such as nell [16] , conceptnet [17] , reverb [18] , and quasimodo [19] . some other published approaches, such as webchild kb [20, 21] or levan [22] , extract knowledge from text and image captions or only from image objects without in-image relations. probably, the most relevant work to our work is neil [23] , which create a knowledge graph directly from images. compared to neil, our proposed automatic approach is capable of extracting much more types of object-to-object relations. compared to conceptnet, which represents an example of a semi-automatic method of retrieving knowledge from text, our proposed approach can extract common-sense relations based on only observing visual data. a possibilistic base is a set of pairs (p, α) where p is a proposition, and α is a degree to which p is true and is in the interval (0, 1) [14] . let ω be a set of interpretations of the real world, and possibilistic distribution π a mapping from ω to the interval (0, 1). an interpretation ω that satisfies p has π(ω) = 1, and 1 − α when ω fails to satisfy p. in summary: from now on, we identify the base as = {(p i , α i ), i = 1, . . . , n}. then all interpretations satisfying propositions in have the possibility degree of 1, while other interpretations are ranked based on the highest values of α associated with proposition they do not satisfy, i.e., ∀ω ∈ ω: in other words, π induces a necessity 'grading' of p i that evaluates to what extent p i is a consequence of the available knowledge. the necessity measure nec is: and nec π (p i ) ≥ α i [24] . a possibility distribution π is normal if there is an interpretation ω that it totally possible, i.e., π (ω) = 1. a possibility graph πg is an acyclic directed graph [14] . the nodes of such a graph are associated with variables a i , each with its domain d i ; while its edges represent dependencies between elements of nodes. for the case of binary variables, i.e., when d i = {a i , ¬a i }, the assignment of value to the variable is called an interpretation ω. let us denote a set of nodes that have edges connecting them to a node a i as its parents: p ar(a i ). possibility degrees π associated with nodes are: for each node a i without a parent p ar(a i ) = 0 prior possibility degrees associated with a single node are π(a) for every value a ∈ d i of the variable a i ; possibilities must satisfy the normalization condition: max a∈di : π(a) = 1. for each node a j with parent(s) p ar(a j ) = 0 possibility degrees are conditional ones π(a|ω p ar(aj ) ) where a ∈ d j , and ω p ar(aj ) is an element of the cartesian product of domains d k of variables a k ∈ p ar(a j ); as above, conditional possibilities must satisfy the normalization condition: max a∈dj : π(a|ω p ar(aj ) ) = 1. in our case, a conditional probability measure is defined using min: and obeys [14] : we introduce a systemic approach to generate knowledge graphs given visual data. such graphs provide us with contextual information about objects present in the world with very limited input from humans. there are unique challenges associated with the generation of this type of graph. first, we need methods able to detect objects in images, and second, we require tools to extract relations between the detected objects. once we have the object recognition and relation extraction processes, we execute them on a set of images. the obtained triples -<entity -relation -entity> are aggregated into a single knowledge graph. the strength of relations is determined by the number of co-occurrences of objects with specific relations. the overall process is shown in fig. 1 . having a trained model, the process is liberated from specific visual data and its annotations. additionally, more visual data can be processed using the proposed methodology and comprehensive context-specific knowledge graphs could be created. to detect objects and their corresponding bounding boxes, we use the faster r-cnn model [25] . in this model, the full image is passed through a convolutional neural network (cnn) to generate image features. to detect image features, usually a pre-trained cnn, such as vgg network [26] , trained on imagenet [27] is used. given the image features as input, another neural network, called region proposal network (rpn), predicts regions that may contain an object and their corresponding bounding boxes. this learning network is the principal contribution of the faster r-cnn model compared to the fast r-cnn model [28] . this results in an improvement of performance in both training and inference. the regions of interest (rois) are then mapped into the image feature tensor, and via application of a process called roi pooling the regions are downsampled to be fed to the next neural network. this allows for the prediction of image classes and their correct bounding boxes. given the error losses from the classification and bounding box predictions, the entire network is trained end-to-end using backpropagation and stochastic gradient descent (sgd) [29] . an illustration of the process can be found in fig. 1 . determining relations between objects is required to generate scene graphs and it can be done in several ways. there has been several publications that propose such methods as iterative message passing [30] , neural motifs [31] , graphical contrastive losses [32] , and factorizable net [33] . in our work, we use the iterative message passing model. the iterative message passing model predicts relations between objects detected by the faster r-cnn model. mathematically, a scene graph generation process means finding the optimal x * = arg max x pr(x|i, b i ) that maximizes the following probability function: where i is an image, b i represents proposed object boxes, x is a set of all variables, including classes, bounding boxes and relations (x = {x cls i , x bbox i , x i→j |i = 1 . . . n, j = 1 . . . n, i = j}), with n representing the number of proposed boxes, x cls i as a class label of the i-th proposed box, x bbox i as the offset of bounding box relative to the i-th proposed box, and x i→j as a predicate between the i-th and j-th proposed boxes. the process of amalgamating generated image scene graphs that results in a single knowledge graph has a number of challenges: 1) establishing a unique identifier for each entity; 2) identifying the importance of connections; 3) dealing with missing values and incorrect data; and 4) keeping the knowledge graph updated in presence of new data. in the specific case of the visual genome dataset, we use synsets from word-net to identify nodes and relations, as well as different meanings of a specific word. there are various methods to identify the uniqueness of words, such as using words occurring in natural language, grouping similar words with the same meaning, or trying to assign words to their specific synsets. yet, another way is to keep words and phrases as they are and let their occurrence numbers show the importance of connections and nodes. such a simple approach provides a good indication which relations are more likely to occur. another challenge is to mitigate missing or incorrect information. for example, the used methods/models could incorrectly label objects/relations and the processes could fail to find unique words or synsets. even the hand-annotated data in the visual genome (vg) dataset [13] , which is used for training, has missing and incorrect data [34] . the unknowns are reduced by relying on the information already present, such as recovering a missing synset based on an already-known name to synset relation or wordnet. the iterative message passing model [30] is trained on the vg dataset. it contains 108,077 images that capture everyday scenarios. for evaluation, only the most common 150 object categories and 50 predicates are used. the faster r-cnn model that is applied to detect objects and their bounding boxes is pre-trained on ms-coco dataset. this dataset has 80 object categories. the training set is of size 80k images. validation and test sets are 40k and 20k images, respectively. around thirty percent of the vg dataset (test set) is used to detect objects and predict predicates. the subset has around 30,000 images. running the process described in sect. 3, a wpkg with 138 nodes and 7,287 relations is generated. neo4j [35] software is used to store and analyze the generated graph. it allows us to store object and relationship names and synsets, as well as occurrence numbers. also, it visualizes a structure composed of triples subject-predicateobject. one advantage of the generated wpkg is the existence of common sense relations occurring in the actual world extracted during the processing of visual data. the most important entities related to the entity of interest can be found by inspecting the strength of connections between them. one way to accomplish this is to measure how often these objects are associated with each other. as an example, the entity plate together with the related entities is shown in fig. 2 (a) . as we can see, removing non-frequent relations leads to identification of tightly related objects relevant to the plate, fig. 2 (b) . a sample of relation occurrence statistics is shown in table 1 . based on the analysis of visual data, we can find out about some common-sense knowledge, such as places where a vase can be placed, and what can be put into it. most of the relations, such as flower-in-vase, make sense and agree with the crowdsourced vg dataset. however, some relations, such as vase-in-vase, may not make sense. this could be a shortcoming of the method/model used for prediction of relations. besides a better model, processing more images and detecting more types of relations and objects may improve the results. the comparison of our method, which is based on image processing, with other relevant automatic and semi-automatic methods is demonstrated in table 2 . the generated wpkgs consist of an enormous amount of nodes and relations. the relations -as built via aggregation of scene graph relations -contain information about the frequency of occurrence. this means that each relation is equipped with a weight indicating its strength and importance. for practical use, wpkg can be further processed and a subset of nodes together with relations between them can be used to construct a possibilistic graph. a wpkg is constructed with no constrains. it contains cycles, very strong and weak relations, as well as erroneous information due to the imperfection of used image processing tools. in that context, a possibilistic graph is more organized and 'clean'. therefore, extracting nodes and edges from wpkg and building a graph that satisfies rules of the possibilistic graph (sect. 2.3) seems important steps in utilizing generated wpkgs. first, a proto-possibilistic graph is constructed. it is free of cycles and contains outwards relations linked to the entity of interest. the procedure used to extract relevant entities and connections is presented as algorithms 1 and 2. the important aspects of this process are: algorithm 1, line 4 the value of depth identifies the allowed length of a 'relation chain' at the process of building a graph; algorithm 2, line 6 the procedure randomize creategroups() is crucial in the construction process: 1) randomization of a sequence of entities allows to generate graphs with different paths; once this is combined with a process presented in line 8 (explained below) it prevents the existence of cycles in the generated graph; 2) grouping of relations/predicates connected to the same object, i.e., prepositions/adjectives playing the role of relations; as illustration, see entities flower, window, table, plant, fig. 3 ; algorithm 2, line 8 this allows to solve an issue of cycles, i.e., relations between pairs of entities flower-vase, plant-vase and table-vase, fig. 3 , would lead to cyclic directed graph; however, if a connection between both entities already exist, a new one -in the opposite direction -is not created. the application of the presented procedure leads to a graph that is acyclic and direct. it also contains occurrences associated with each connection. the last step of constructing a possibilistic graph is to determine possibility degrees. to do so, all input connections to a given node are analyzed. the maximum value is identified and is used for normalization of all other occurrence values associated with inward connections to the node. this ensures satisfaction of the requirement of maximum possibility equal to 1.0 (sect. 2.3). the extracted possibilistic graph allows us to build a possibilistic knowledge base. here, we follow the process presented in [14] . for that purpose, we consider the graph as a set of triples: πg = {(a, p a , α) : π(a|p a ) = α} where a is an instance of a i and p a is the cartesian product of domains d k of variables a k ∈ p ar(a i ). each such triple can be represented as a formula: (¬a ∨ ¬p a , 1 − α) so, following [14] , we have that the possibilistic knowledge base associated with πg defined as: let us illustrate the process of building a simple possibilistic graph and a possibilistic knowledge base. we apply the procedure to build a graph of facts related to the entity vase, and relations between this entity and other entities from the vase's environment. given the adaptability of wpkg to new scenarios, context-aware and even time-variant knowledge graphs can be constructed. for example, processing car images from a specific country will lead to the construction of wpkg representing a very specific information related to cars' details and their contextual settings. another important aspect that can be considered is time. it can affect both occurrences of relations and meanings of words linked to the nodes. as future work, better models can be used to improve the overall construction process, biases can be reduced by implementing procedures to diversify the input images, and prediction of unknown objects can be added. atomic: an atlas of machine commonsense for if-then reasoning comet: commonsense transformers for automatic knowledge graph construction an end-to-end model for question answering over knowledge base with cross-attention combining global knowledge infusing knowledge into the textual entailment task using graph convolutional networks billion-scale commodity embedding for e-commerce recommendation in alibaba dbpedia: a nucleus for a web of open data wikidata: a free collaborative knowledgebase yago: a multilingual knowledge base from wikipedia, wordnet, and geonames freebase: a collaboratively created graph database for structuring human knowledge wordnet: a lexical database for english the importance of vision the informational capacity of the human eye visual genome: connecting language and vision using crowdsourced dense image annotations possibilistic logic bases and possibilistic graphs automated construction of possibilistic networks from data never-ending learning conceptnet 5.5: an open multilingual graph of general knowledge identifying relations for open information extraction commonsense properties from query logs and qa forums webchild: harvesting and organizing commonsense knowledge from the web webchild 2.0 : fine-grained commonsense knowledge distillation learning everything about anything: webly-supervised visual concept learning neil: extracting visual knowledge from web data merging uncertain knowledge bases in a possibilistic logic framework faster r-cnn: towards real-time object detection with region proposal networks very deep convolutional networks for large-scale image recognition imagenet: a largescale hierarchical image database fast r-cnn backpropagation applied to handwritten zip code recognition scene graph generation by iterative message passing neural motifs: scene graph parsing with global context graphical contrastive losses for scene graph parsing factorizable net: an efficient subgraph-based framework for scene graph gen scene graph generation with external knowledge and image reconstruction neo4j inc: neo4j application of algorithm 1 to the generated wpkg allows us to extract entities related to the entity of interest, vase. the neo4j snapshot of wpkg with vase and relations to 'relevant' entities is shown in fig. 3(a) . the version processed by the algorithm is shown in fig. 3(b) . it contains -marked as dashed lines -the pairs flower-vase, plant-vase, and table-vase that could result in different graphs depending on the element of randomness embedded in the procedure randomize creategroups(), algorithm 2.the wpkg with occurrences assigned to connections allows us to determine conditional degrees. we have simplified our graph, i.e, combined all inward connections to a node into a single one, as shown in fig. 3 (c). this graph is further processed -the occurrence numbers are used to determine possibility values. based on the graph in fig. 3 (c), we build conditional possibility degrees. all of them are presented in tables 3, 4, and 5. the paper focuses on the automatic construction of a knowledge graph -called world-perceiving knowledge graph (wpkg) -that contains results of the analysis of multiple images. further, the generated wpkg is processed and multiple possibilistic graphs can be constructed based on it.it is shown that using deep learning models, we can extract common-sense situational information about objects present in visual data. the trained neural networks may already know these relations implicitly, but extracting this knowledge in the form of a knowledge graph provides the ability to have this information explainable and explicit. the strength of the overall procedure depends on the capabilities of the applied learning model as well as the data it has been trained on. by improving the models themselves, the overall procedure can be improved.constructed wpkgs are contextualized by images used as an input to the presented process. a different graph will be obtained when images representing a specific geographical location are used, while a different graph will be built based on images illustrated a specific historical event. also, multiple different possibilistic graphs can be created to reason about the correctness of contextual utilization of specific items and relations between them. key: cord-355850-jgcjscjq authors: elhadi, muhammed; msherghi, ahmed; alkeelani, mohammed; zorgani, abdulaziz; zaid, ahmed; alsuyihili, ali; buzreg, anis; ahmed, hazim; elhadi, ahmed; khaled, ala; boughididah, tariq; khel, samer; abdelkabir, mohammed; gaffaz, rawanda; bahroun, sumayyah; alhashimi, ayiman; biala, marwa; abulmida, siraj; elharb, abdelmunam; abukhashem, mohamed; elgzairi, moutaz; alghanai, esra; khaled, taha; boushi, esra; ben saleim, najah; mughrabi, hamad; alnafati, nafati; alwarfalli, moaz; elmabrouk, amna; alhaddad, sarah; madi, farah; madi, malack; elkhfeefi, fatima; ismaeil, mohamed; faraag, belal; badi, majdi; al-agile, ayman; eisay, mohamed; ahmid, jalal; elmabrouk, ola; bin alshiteewi, fatimah; alameen, hind; bikhayr, hala; aleiyan, tahani; almiqlash, bushray; subhi, malak; fadel, mawada; yahya, hana; alkot, safeya; alhadi, abdulmueti; abdullah, abraar; atewa, abdulrahman; amshai, ala title: assessment of healthcare workers’ levels of preparedness and awareness regarding covid-19 infection in low-resource settings date: 2020-06-18 journal: am j trop med hyg doi: 10.4269/ajtmh.20-0330 sha: doc_id: 355850 cord_uid: jgcjscjq covid-19, caused by the sars-cov-2 virus, is spreading rapidly worldwide, with devastating consequences for patients, healthcare workers, health systems, and economies. as it reaches lowand middle-income countries, the pandemic puts healthcare workers at high risk and challenges the abilities of healthcare systems to respond to the crisis. this study measured levels of knowledge and preparedness regarding covid-19 among physicians and nurses. a cross-sectional survey was conducted among healthcare workers in libya between february 26 and march 10, 2020. we obtained 1,572 valid responses of a possible 2,000 (78.6%) participants from 21 hospitals, of which 65.1% were from physicians and 34.9% from nurses. the majority of participants (70%) used social media as a source of information. a total of 47.3% of doctors and 54.7% of nurses received adequate training on how to effectively use personal protective equipment. low confidence in managing suspected covid-19 patients was reported by 83.8% of participants. furthermore, 43.2% of healthcare workers were aware of proper hand hygiene techniques. less than 7% of participants received training on how to manage covid-19 cases, whereas 20.6% of doctors and 26.3% of nurses felt that they were personally prepared for the outbreak. awareness and preparedness for the pandemic were low among frontline workers during the study. therefore, an effective educational training program should be implemented to ensure maintenance of appropriate practices during the covid-19 pandemic. in late 2019, a novel covid-19 was reported to cause severe viral pneumonia in wuhan, china. it has since spread worldwide, resulting in a pandemic that has now infected more than 1.3 million people, causing more than 80,000 deaths globally. 1, 2 in february 2020, the who named the condition covid-19, which stands for coronavirus disease 2019. 3 the severity of symptomatic infections ranges from mild to critical. approximately 80% of patients have mild symptoms, whereas less than 20% experience severe symptoms such as dyspnea and shock; respiratory failure occurs in less than 5% of patients. [4] [5] [6] [7] elderly patients and/or patients with comorbidities, such as cardiovascular diseases, respiratory diseases, hypertension, diabetes mellitus, and chronic kidney diseases, are at a higher risk for severe illness. they have a higher risk of mortality than younger, or otherwise healthier, individuals. 8, 9 previous reports have confirmed that hospitalized patients had a mean age of 49-55 years. 6, 10 in an earlier report provided by the united states, regarding covid-19 patients treated between february 12, 2020 and march 2020, about 67% of those hospitalized were older than 45 years, which is similar to a prior report from china. 11, 12 acute respiratory distress syndrome (ards) is a significant complication for covid-19 patients; an estimated 20-41% of patients develop ards following a covid-19 infection and require mechanical ventilation. 5, 13 this characteristic of the disease can significantly increase the existing burden on healthcare facilities, and it requires extra resources and appropriate management. covid-19 poses a higher risk for physicians and nurses who work in critical care, emergency medicine, infectious diseases, and pulmonary medicine departments. personal protective equipment (ppe), proper handwashing, and hand hygiene are critical in decreasing the transmission and risk of infection of covid-19 in hospitals. therefore, adequate training, knowledge, and resources are necessary to prevent hospital-acquired infections due to crosscontamination to other patients who receive care in these departments. [14] [15] [16] as of april 12, 2020, the libya national center for disease control (ncdc) reported 524 suspected cases of covid-19, with 25 confirmed diagnoses. 17 however, these numbers raise a question regarding the local authorities' ability to perform adequate testing for covid-19, as the libya ncdc can perform an average of 30 tests per day. furthermore, given the current civil war, limited financial resources, and a shortage of machines and materials, libya is more vulnerable to the covid-19 pandemic. in 2017, the who released a report on the healthcare system in libya, revealing that approximately 20% of hospitals were shut down. in addition, several major hospitals only operate at 80% capacity or less. the report states that chronic respiratory disease readiness is less than 29%, whereas the readiness of emergency services is less than 47% of the target. furthermore, readiness for emergency health services is less than 18% in terms of training and 40% for diagnosis abilities. the report indicates that libya does not have a rapid response team or the ability to diagnose and respond to alerts in an organized manner. 18 this raises concerns about the local healthcare authorities' preparedness and capacity to provide an adequate response to covid-19 infection. therefore, critical preparedness, readiness, and knowledge regarding covid-19 are needed for physicians and nurses on the front line. 16, 19 few researchers have addressed the overall issues of preparedness of healthcare systems for covid-19, especially in developing countries, where resources and facilities are limited. considerable concerns have been raised regarding countries' preparedness for covid-19 and their ability to maintain control. this study aimed to determine the preparedness and knowledge of libyan healthcare workers regarding covid-19, and to develop and validate a measurement tool for estimating healthcare preparedness on a global scale. study design. a cross-sectional survey study was conducted in 21 hospitals in libya between february 26, 2020 and march 10, 2020. setting and participants. the study was conducted among physicians and nurses working in the emergency department, intensive care units (icus), and respiratory and infectious disease departments, all of whom may expect to encounter covid-19 patients. the study methodology was explained to the participants, and they were asked to provide consent before participating in the study. doctors and nurses working in other departments or private clinics were excluded from the study. a total of 2,000 targeted participants were given a paper-based questionnaire at their workplace. instruments. the self-administered and anonymous questionnaire was developed by the authors of the study and was validated in a pilot study of 30 participants, who did not participate in the final analysis. some of the questions were based on a framework similar to that of previous studies on infectious disease outbreaks. [20] [21] [22] the structured questionnaire was validated to address questions about healthcare workers' level of preparedness and knowledge of covid-19 (supplemental files i-ii). the first part of the survey was designed to collect data regarding the background characteristics of participants (age, gender, department, years of experience, information sources, and previous experience with outbreaks). the second part comprised seven items intended to evaluate the knowledge and awareness of physicians and nurses regarding covid-19 infection. the third part comprised 11 items to assess the overall preparedness in terms of managing cases of covid-19 infection. the questionnaire evaluates information sources, training experience with covid-19, diagnosis and management of covid-19 patients, use of ppe, safety precautions, isolation procedures, measures to prevent infection, and reporting procedures. to assess a given participant's knowledge, each correct answer was given a score of 1, and an incorrect answer was given a zero. scores were summed for a total score of 7, ranging from 0 to 7 for knowledge. the preparedness ranges from 0 to 11. those who scored ³ 5 on the knowledge scale were deemed to have a high level of knowledge, whereas those who scored ³ 8 on the preparedness scale had adequate preparedness. the questionnaire was developed in english for doctors and was translated into arabic for nurses using a forwardbackward translation by two qualified, independent linguistic translators at the university of tripoli. each translator independently performed a forward translation of the original questionnaire. backward translation from arabic to english was carried out by another linguistic translator. three researchers then reviewed the translated versions of the questionnaire for discrepancies. pilot tests of both versions were carried out, and cronbach's alpha was used to measure the internal consistency reliability of both versions. participants were recruited in relative proportion to the estimated number of doctors and nurses working in these departments in libya. we assumed that 70% of participants had basic knowledge and preparedness regarding covid-19, and the estimated sample size was calculated with 80% power at a 95% confidence limit, with a design effect equal to 2, and using 20 clusters. statistical analysis. data entry and analyses were performed using spss (ibm spss statistics for windows, version 25.0; ibm corp., armonk, ny). descriptive statistics were used to describe the study variables. frequency, percentage, and mean scores were used to report the descriptive analysis. the chi-square test was used to assess the association difference between the groups. statistical significance was considered for p < 0.05. ethical approval. ethical approval for this study was obtained from the bioethics committee at biotechnology research center in libya [reference number: bec-btrc-109.3-2020]. all participants provided consent before participating in the study. we obtained 1,572 valid responses out of a possible 2,000 (78.6%) participants from 21 hospitals in libya. among the 1,572 total healthcare personnel, 65.1% were physicians and 34.9% were nurses. table 1 summarizes the characteristics of the study population. figure 1 provides the distribution of participants based on departments and profession. the mean age of the population (±sd) was 35.5 ± 7.3 years for doctors and 27.8 ± 5.4 years for nurses, with women making up 64.6% of all respondents. approximately half of the participants had low confidence regarding their ability to manage covid-19 patients. most participants (70% of both doctors and nurses) named social media as a source of information for covid-19, which was statistically significant, whereas fewer than 4% viewed training courses as such. doctors had an average of 7.8 ± 7.6 years of healthcare experience, whereas nurses had an average of 5.8 ± 4.1 years. all participants worked in general public hospitals: 30.7% were in icus, 58.8% in the emergency department, 3.2% in the infectious disease department, and 7.4% in respiratory departments. a total of 47.3% of doctors and 54.7% of nurses did not receive adequate training in the use of ppe. meanwhile, 16.2% of doctors and 13.3% of nurses indicated confidence in their ability to handle suspected covid-19 patients, versus 83.8% of doctors and 86.7% of nurses who did not covid-19 level of preparedness and awareness indicate confidence in this regard. about 45% of doctors and 37% of nurses indicated that they were not prepared to handle cases of infection transmitted at the hospital level. the measurements of knowledge, awareness, and preparedness are presented in table 2 . the reliability of the knowledge and preparedness questionnaires was determined using cronbach's alpha, which revealed scores of 0.72 and 0.68 for the knowledge and preparedness scales, respectively. the results showed that less than half of all participants knew the symptoms of covid-19 infection (44.4% of doctors and 31.7% of nurses). similar results were obtained for case identification. however, less than 70% of the participants were aware of the mode by which covid-19 infection was transmitted. in addition, 52.7% of doctors and 45.3% of nurses knew about ppe. interestingly, only 43.2% of doctors and nurses were aware of proper handwashing and hand hygiene techniques to prevent the transmission of covid-19. in terms of preparedness, less than 7% of both doctors and nurses had taken courses, or training, on covid-19. about 18% of participants reported the presence of an isolation room and the availability of a known protocol for isolation. however, only about 13% of all healthcare workers felt that hospitals were prepared for the covid-19 outbreak. meanwhile, 20.6% of doctors and 26.3% of nurses perceived themselves as personally prepared for covid-19. the majority (65.8%) also reported that they were not prepared to take precautions to prevent aerosol transmission via covid-19 patients. the majority (73.5%) reported an inadequate level of knowledge on covid-19. there was no significant association between hospital department type and level of participant preparedness (p = 0.319) or knowledge (p = 0.166). tables 3 and 4 show the level of preparedness and knowledge among healthcare workers by department. a significant number of healthcare workers expressed low levels of awareness and preparedness regarding covid-19. this raises a concern regarding the ability of the libyan healthcare system and its healthcare workers to combat covid-19 infection. despite these concerns, along with the poor local healthcare infrastructure in libya, healthcare workers continue to work during covid-19, risking their lives to save their patients. meanwhile, no official courses or training programs are available, and healthcare workers have to purchase ppe themselves, as they are not provided by the hospitals in adequate amounts. our study provides considerable insights into the necessity of immediate and determined efforts focused on training programs and providing an adequate supply of ppe to alleviate these challenges during the covid-19 pandemic. inadequate knowledge is a risk factor for disease transmission, as it can lead to low levels of care. our study demonstrated that only 6.7% of doctors and 5.5% of nurses had participated in training courses. our study also reported that about 70% of participants received information from social media, which is lower than previously reported (91.1%). furthermore, our study indicated the lowest rate of knowledge compared with previous studies, where we found that the overall rate of respondents providing correct answers on the knowledge questionnaire was a mere 26.5%, compared with a previous study reporting that 90% of healthcare workers provided correct answers. 23 another recent study demonstrated that 89% of healthcare workers had sufficient knowledge on covid-19. 24 another primary concern emerging from this study is that only 21.2% of doctors and 23.9% of nurses knew the criteria for evaluating persons under investigation for covid-19 infection. in addition, only about 25% of doctors knew how to report potential covid-19 cases, which could prompt an unexpected increase in undiagnosed cases, thereby increasing the burden of infection within the community. moreover, approximately 18% of participants were unaware of isolation room specifications and processes for potential covid-19 patients, which could increase the risk of infection within hospitals. interestingly, only 44.2% of doctors and 33.3% of nurses were aware of the proper handwashing and hand hygiene techniques and disinfectants. in addition, about 45.1% of doctors and 37% of nurses were not prepared to manage a case with signs and symptoms of covid-19 infection. this could highlight the risk of cross-contamination within hospitals and could lead to a higher rate of hospital-acquired infections. training and safety precautions, focused on the direct decontamination of contact points among healthcare workers, are needed to prevent the spread of infection. 25, 26 the majority of healthcare workers (77.4%) felt personally unprepared to address covid-19 infection. a total of 50.1% of participants were uneducated about ppe, whereas about 54.3% were not trained to use it. in addition, ppe is very limited, and hospital workers reported that they independently purchase their own ppe because of the inadequate supply provided by libyan hospitals. furthermore, about 35.1% of 27 our study found that doctors and nurses were buying ppe themselves, prompting questions around the hospitals' inabilities to provide this essential equipment. only a small proportion (13.4%) of participants perceived their hospital as prepared for the outbreak. more training and education are needed on the triage and isolation of suspected cases, as approximately 25% of participants are not prepared, or trained, to conduct these protocols. limitations. although the study provided a large-scale sample from 21 healthcare centers, and used a sample size that was sufficiently large enough to allow the adequate assessment of healthcare workers' knowledge and preparedness regarding covid-19 infection, some limitations should still be clarified. the study was conducted in a single country with low resource levels and a lower number of detected covid-19 infections than other countries, which may have affected the results. future multinational studies, using more extensive and varied populations, are needed to validate these findings. in conclusion, our study has illuminated the current level of knowledge and awareness of covid-19 among doctors and nurses, with special consideration for those working in departments responsible for caring for covid-19 patients. we focused on healthcare workers who may come into direct contact with covid-19 patients, and are thus expected to have adequate knowledge and preparedness. by contrast, other studies have focused on more general populations of healthcare workers. 23, 24, 28, 29 this study provides an overview of healthcare workers' preparedness regarding the current pandemic. the respondents had a lower level of preparedness, which highlights the importance of education and training programs for healthcare workers, to control and prevent infection from covid-19. however, the absence of an organized and effective governmental plan, along with a poor healthcare infrastructure, renders developing countries vulnerable. moreover, educational initiatives, along with more tangible forms of support, such as the provision of ppe, should be carried out to help developing countries improve their abilities to control and prevent covid-19 infection. e-mails: muhammed.elhadi.uot@gmail.com, ahmedmsherghi@gmail.com, albshrimohamed@gmail.com e-mails: tariq.bugadeda@gmail.com, moutazashour@ yahoo.com, fatma.mousa91@gmail.com, mohamed.alfssi@gmail.com, belaladel94@gmail.com, and majdibadi3202@gmail.com. mohammed abdelkabir, faculty of medicine who director-general's opening remarks at the mission briefing on covid-19 -9 available at: https:// www.who.int/dg/speeches/detail/who-director-general-sopening-remarks-at-the-media-briefing-on-covid director-general's remarks at the media briefing on 2019-ncov on 11 a familial cluster of pneumonia associated with the 2019 novel coronavirus indicating person-to-person transmission: a study of a family cluster clinical characteristics of 138 hospitalized patients with 2019 novel coronavirus-infected pneumonia in wuhan, china epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in wuhan, china: a descriptive study persons evaluated for 2019 novel coronavirus -united states clinical course and risk factors for mortality of adult inpatients with covid-19 in wuhan, china: a retrospective cohort study characteristics of and important lessons from the coronavirus disease 2019 (covid-19) outbreak in china: summary of a report of 72,314 cases from the chinese center for disease control and prevention clinical features of patients infected with 2019 novel coronavirus in wuhan severe outcomes among patients with coronavirus disease 2019 (covid-19) -united states di napoli r, 2020. features, evaluation and treatment coronavirus risk factors associated with acute respiratory distress syndrome and death in patients with coronavirus disease 2019 pneumonia in wuhan, china infection prevention in the operating room anesthesia work area covid-19 in turkey: an urgent need for the implementation of preparedness and response strategies critical preparedness, readiness and response actions for covid-19 covid-19 status in april 10 health emergencies and humanitarian response update health-care workers' perspectives on preparedness of health-care facilities for outbreak of communicable diseases in nigeria: a qualitative study knowledge and attitude of healthcare workers about middle east respiratory syndrome in multispecialty hospitals of qassim, saudi arabia middle east respiratory syndrome-related knowledge, preventive behaviours and risk perception among nursing students during outbreak health-care workers' perspectives on preparedness of health-care facilities for outbreak of communicable diseases in nigeria: a qualitative study knowledge, attitudes, and practices towards covid-19 among chinese residents during the rapid rise period of the covid-19 outbreak: a quick online cross-sectional survey knowledge, attitude and practice regarding covid-19 among healthcare workers in henan how to train the health personnel for protecting themselves from novel coronavirus (covid-19) infection during their patient or suspected case care knowledge and attitudes of medical staff in chinese psychiatric hospitals regarding covid-19 covid-19: protecting health-care workers knowledge and attitude toward covid-19 among healthcare workers at district 2 hospital, ho chi minh city is pakistan prepared for the covid-19 epidemic? a questionnaire-based survey covid-19 level of preparedness and awareness acknowledgments: we would like to thank all of the staff who voluntarily participated in this study. publication charges for this article were waived due to the ongoing pandemic of covid-19. this is an open-access article distributed under the terms of the creative commons attribution (cc-by) license, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. key: cord-312986-nz8uc7sl authors: abou-abbas, linda; nasser, zeina; fares, youssef; chahrour, mohammad; el haidari, rana; atoui, rola title: knowledge and practice of physicians during covid-19 pandemic: a cross-sectional study in lebanon date: 2020-09-29 journal: bmc public health doi: 10.1186/s12889-020-09585-6 sha: doc_id: 312986 cord_uid: nz8uc7sl background: as the coronavirus disease 2019 (covid-19) pandemic continues to evolve, physicians must be equipped with adequate knowledge, skills on the prevention measures, and confidence in diagnosing and treating covid-19 patients. therefore, it is of great interest to assess the knowledge and practices of physicians to identify existing gaps and improve occupational safety and viral surveillance. methods: a cross-sectional study was conducted in lebanon between 28th march and 11th april 2020. data was collected through an online survey that included information on socio-demographic characteristics, knowledge, practice, physicians fear towards covid-19 as well as their perceptions regarding actions/policies implemented by the ministry of public health (moph) and their health care facilities. multivariable logistic regression analyses were carried out to identify the factors associated with good knowledge of covid-19 and good practice toward its prevention. adjusted odds ratio and their 95% confidence intervals were reported. results: our survey revealed that the majority of lebanese physicians had good knowledge about the disease (89.5%) while approximately half of the respondents adopted good preventive practices (49.7%). the odds of having good knowledge was 2.16 times higher among physicians aged 40 and above (adjusted or = 2.16 with a 95% confidence interval (ci) of 1.08 to 4.34) compared to their counterparts aged less than 40 years old. our results also showed that the odds of good practice was 2 times higher among frontline compared to the second line workers (adjusted or = 2.01 with 95% ci of 1.21 to 3.34). physicians with an experience of 10 years and above were 3.35 times more likely to have good practice compared to their counterparts (adjusted or = 3.35 with 95% ci of 1.60 to 7.02). finally, participants with good knowledge of covid-19 were 2.04 times more likely to have a good practice (or = 2.04 with 95% ci of 1.01 to 4.12). conclusion: lebanese physicians revealed a good level of knowledge; however, they had limited comprehension of the precautionary measures that protect them from this virus. our findings have important implications for the development of strategies suitable for improving the level of practice among physicians and enhance prevention programs. novel coronavirus disease 2019 (covid-19), which first emerged in china in december 2019, has turned into a worldwide disaster affecting at a rapid pace all the countries over the world [1, 2] . the latest figures, at the time of writing, show more than 17.5 million cases worldwide with a death toll exceeding 680.000 [3] . given the nature and burden of covid-19 pandemic, unprecedented challenges have faced governments, communities, and health care systems. physicians, who are directly engaged in the diagnosis, treatment, and care of covid-19 patients, are exposed to infection from aerosol and droplet contamination [4] and at high risk for nosocomial infections [5] . as covid-19 pandemic continues, the death of physicians has been increasingly reported worldwide. a recent cross-sectional study published in april 2020 showed that 120 medical doctors have died due to covid-19; 67 in italy, 34 in china, 6 in france, 4 in the united kingdome (uk), the united states of america (usa), and spain and 1 in south korea [6] . lebanon, a small arab country, is taking part in the global fight against the covid-19 pandemic. the first case was detected on february 21, 2020, in a traveler who had returned from iran. as per august 1st, 2020, the ministry of public health (moph) announced that the number of cases has reached 4730 [7] . during this outbreak, 207 health care workers (hcws) of which 47 physicians had contracted the infection in their health care facilities. most of the cases occurred in the early period of the outbreak due to misdiagnosis of the cases and the inadequate preventive practices. thus, a preparedness of frontline physicians should be the main priority of all lebanese health care settings to function properly and competently combat covid-19. they should be equipped with adequate knowledge, skills on the prevention measures, and confidence in diagnosing and treating covid-19 patients. as part of an epidemic preparedness plan, it is of great interest to assess the knowledge and practices of physicians to identify existing gaps and improve occupational safety and viral surveillance. thus, this study was conducted in the early stage of the covid-19 outbreak in lebanon to assess the knowledge and practices of physicians regarding covid-19. also, we sought to evaluate their fear towards covid-19 and their perceptions regarding policies/actions implemented by the moph and their health care settings in handling covid-19 pandemic. a cross-sectional study, using an online survey, was conducted during the early phase of the covid-19 epidemic in lebanon between 28th march and 11th april 2020. as the lebanese government recommended the public to minimize face-to-face interaction and isolate themselves at home, potential respondents were electronically invited to participate. thus, an online questionnaire using a google form was distributed through "whatsapp" groups and social media using a snowball technique. all physicians, working in hospitals or medical centers in different regions in lebanon and who agreed to participate in the study, were included. no exclusion criteria were applied. participants were identified via professional groups and academic institutions. the sample size was calculated using the online rao-soft sample size calculator designed specifically for population surveys. based on an estimated population of 10,918 physicians [8] , an anticipated response of 50%, a confidence level of 95% and a 5% margin of error, the required sample size would be at least 372. a structured questionnaire was initially developed and designed by the authors in the english language to cover important aspects of knowledge and practice towards covid-19 among physicians (additional file 1). core dimensions and items content of these two domains were identified through a review of the published literature on middle east respiratory syndrome coronavirus (mers-cov) [9] [10] [11] [12] in addition to the most recent available information on covid-19 from the world health organization (who) and the centers for disease control and prevention (cdc) websites up to 25th march 2020. content validity of the resulting version was assessed by a panel of three experts with expertise in implementing infection control procedures and emergency preparedness. they were asked to evaluate the relevance of the items in assessing the knowledge and practices of physicians towards covid-19. a consensus was reached after omitting four items that were rated irrelevant also minor linguistic edits were made. then, the items were translated and adapted to the arabic language by three translators. a final questionnaire was generated and was divided into five sections: 1-socio-demographic information including age, gender, marital status, specialty, place of work, and clinical experience. participants were also asked whether they were directly engaged in providing care to suspected, probable, or confirmed covid-19 cases. those who responded "yes" were considered as frontline physicians. physicians, who answered "no" were considered as second-line workers. 2-knowledge section: six dimensions with a total of 19 items were designed to measure physician's knowledge about nature of the disease (4 items), the transmission of the disease (3 items), actions in dealing with suspected, probable and confirmed cases (5 items), precautionary measures by health care providers (2 items), and treatment of the disease (5 items). all the items were answered on a true/false basis and an additional "do not know" option. a correct answer was assigned 1 point and an incorrect/ unknown answer was assigned 0 points. the total knowledge score, obtained by the sum of the scores, ranged between 0 and 19. based on bloom's cut off point, physicians' overall knowledge was categorized as good if the score was above 60% (≥12 points) and poor if the score was less than 60% (<12 points) [13] . a question exploring the source of their knowledge concerning covid19 was also added to this section. on this question, multiple responses from the participants were allowed. 3-practice sections: seven questions were used to evaluate the uptake of various preventive measures. the items were answered "always", "occasional" and "never" respectively. the answer (always) was assigned 1 point while answers (occasional and never) were assigned 0 points. the overall practice score, obtained by the sum of the scores, ranged between 0 and 7. practice levels were defined as "good" or "poor" based on bloom's cut off point. physicians with scores ≥80% (≥6 points) were classified as having a good practice, while those with scores < 80% (<6 points) were considered having poor practice [14] . 4-physicians' fear towards covid-19 was assessed by 2 items (i am afraid of working in places where patients suspected of covid-19 are admitted/cared for, i am afraid of treating a patient with covid-19) on a 3-point likert scale (1-agree, 2-neutral, 3-disagree). a point of 1 was given to agree to answer while disagree or neutral responses were given a 0 point. 5-physicians' perceptions regarding policies/actions implemented by the ministry of public health (moph) and their health care settings in handling the covid-19 epidemic (1-appears in disarray/disorganized, 2-insufficient, 3-acceptable/appropriate, 4-excessive and unnecessary. the survey was pilot tested in a sample of 5 physicians to check the clarity and readability of all items. physicians did not report any problems in understanding the questionnaire. on average, the survey was completed within approximately 7 min. the data of the pilot study was removed from the final analysis. data was collected using an online survey. an invitation letter, including a link to the web survey platform, requesting lebanese physicians to participate in this survey was prepared. the invitation letter includes information describing the survey and asking for voluntary participation of physicians as well as declarations of confidentiality and anonymity, and instructions for filling in the questionnaire. this letter was sent to approximately 1000 physicians through professional groups and academic institutions. statistical analysis was carried out using the statistical software spss (statistical package for social sciences), version 22.0. descriptive statistics were reported using means and standard deviations (sd) for continuous variables and frequency with percentages for categorical variables. both bivariate and multivariable logistic regression analyses were performed to identify associated factors of good knowledge and good practice. the variables in bivariate analysis with p-value < 0.2 were entered into multivariable logistic regression. adjusted odds ratio and their 95% confidence intervals were reported. the final logistic regression model was reached after ensuring the adequacy of our data using the hosmer and lemeshow test. the statistical significance level was set at p-value < 0.05 (two-sided). a total of 380 physicians participated in the survey among them 63.2% were males. the mean age was 40.4 (sd = 12.0) ranging from 24 to 80 years. the majority of the participants were specialists (85.5%) and 77.4% of respondents were frontline workers who were directly engaged in taking care of suspected, probable, or confirmed cases of covid19. almost half of the physicians (48.2%) had been practicing medicine for 10 years or longer. the summary of characteristics is shown in table 1 . out of the 380 physicians, the majority of 340 (89.5%) had good knowledge. assessment of physicians' fears towards covid-19 the majority of participants (80.3%) declared that policies/actions implemented by the ministry of public health are adequate (fig. 1) , whereas only 63.9% revealed that the policies/actions implemented by their health care facilities were adequate in fighting covid-19 (fig. 2 ). results of the bivariate and multivariable analysis showed that only age was significantly associated with good knowledge at p-value <0.05. the odds of having good knowledge was 2.16 times higher among physicians aged 40 and above (adjusted or = 2.16 with a 95% ci of 1.08 to 4.34) compared to their counterparts aged less than 40 years old (table 5) . factors associated with good practice toward covid-19 prevention since the declaration of the first case on february 21st,2020, a great public health concern emerged in lebanese public and governmental institutions. until this date, no definitive treatment was recommended, and physicians are expected to play an important role in the detection and management of cases of covid-19. in addition, they are carrying the burden to prevent further spreading of the disease. thus, lack of their knowledge regarding transmission and clinical manifestations of the disease as well as inadequate preventive practices could lead to misdiagnosis of the case and increase the risk of infection. this study was conducted during the early stage of the covid-19 outbreak in lebanon to provide insight into the knowledge and practices of physicians. results of our survey revealed that the majority of lebanese physicians had good knowledge about the disease while only half of the respondents adopted good preventive practices. our results also showed that frontline physicians who had been practicing medicine for more than 10 years, and with a good level of knowledge had good practice compared to their counterparts. our finding of a good level of knowledge among physicians is in line with that of minghe zhou et al., who reported that 89.7% of hcws have sufficient knowledge regarding covid-19 with doctors showing higher scores compared to nurses and paramedics [15] . when looking at the dimensions of knowledge, we found that most participants were aware of the nature of the disease (93.5%) and its treatment (75.6%). however, a poor level of knowledge was clearly shown in response to the questions regarding the transmission of the disease (31.5%), similarly for the actions when dealing with covid-19 cases (37.3%) and precautionary measures by health care providers (42.2%). consistent with our findings, akshaya srikanth bhagavathula et al., [16] reported a poor level of knowledge among hcws concerning the transmission of the disease (39%). this could be attributed to the scientific dilemma proposed by the experts regarding this topic. logistic regression analysis showed that the age of the participants was the only significant predictor of good knowledge. this comes inconsistency with the study conducted in pakistan to evaluate knowledge, attitude, practice, and perceived barriers among hcws regarding covid-19 by saqlain et al. [17] . we also found that physicians used official international and governmental websites such as who (85.0%), moph (70.5%), cdc (41.0%), and (idsa) (31%) as main sources of information about covid-19. this indicates that physicians utilize reliable sources to acquire information regarding covid-19 and reflect their good level of knowledge. it is also worth mentioning that some physicians used tv (18.4%) and facebook (16.3%) as sources of information. although these platforms provide an easy way to get the information, they can also be a source of fake news. thus, it is highly recommended for physicians to seek information from scientific and authentic platforms. concerning practice, approximately half of the respondents (49.7%) followed infection control practices. these include regular hand hygiene (97%), wearing a face mask (89%), and gloves (65%). only half of our participants can maintain a social distancing of at least 1.5 m from colleagues (55%). this could be to overcrowding or small surfaces in health care settings. in addition, only 77.5 and 61.8% of the physicians were aware of the proper donning and doffing ppe. a recent study conducted in pakistan showed that 91.4% of physicians had good practices in following precautions to avoid covid-19 [18] . limited resources in the institution, the lack of experience, the poor level of knowledge regarding mode of transmission of the disease, actions are taken when dealing with cases and precautionary measures could partly explain the poor preventive practices of physicians. the lebanese order of physicians (lop), the syndicate of hospitals, and the scientific societies have conducted many online training sessions for hcws mainly physicians. in addition, several protocols and memo regarding sars-cov-2 were issued. despite all of this, a significant number of hcws have been infected due to misdiagnosis of the cases and inadequate preventive practices. thus, continued professional education and training are advised to empower physicians by supporting their ability to acquire and use evidence-based information. this imposes an action plan from lop and syndicate of hospitals to enhance the actions and preventive measures that should be implemented when confronting a novel virus. similar to the findings of zhang et al., [15] , results of our survey showed that frontline physicians who results are expressed in terms of frequency and percentage had been practicing medicine for more than 10 years had better practice compared to their counterparts. this indicates that frontline physicians' with more than 10 years of experience had skills to deal with public health emergencies and are confident in their ability to defeat the virus. a finding of considerable concern in this survey is that more than 30% of the respondents expressed their fear towards treating a patient with covid-19 which in turn was associated with poor practice. indeed, sars-cov-2 is highly contagious which could explain the reluctance of physicians to treat patients with covid-19. thus, psychological interventions to improve physicians' mental health and to enhance confidence in their ability to treat patients are needed. with a deeper understanding of covid-19, we believe that physicians' fear will decrease and the number of physicians who are willing to treat these patients would gradually increase. interestingly, the majority of participants (80.3%) declared that the policies/actions implemented by the moph are adequate. the lebanese governments have set early lockdown measures such as the closure of all educational institutions, international airport, and its sea borders in addition to the nighttime curfew. all these measures have contributed, till the time of the writing, to the success in slowing the pace of covid-19 progression. however, only 63.9% revealed that the policies/actions implemented by their health care facilities were adequate in fighting covid-19. this could be due to the poor infection control practices implemented in the health care facilities and the shortage of available ppe for all hcws. thus, increasing the preparedness of all health care facilities is vital to increase the confidence of physicians so to improve their work. the findings of the present study should be considered in light of several limitations. firstly, no validated tool for the assessment of the knowledge and practices of hcws was available. we have adapted and modified tools used for the assessment of knowledge, attitude, and practice toward mers-cov [9] [10] [11] [12] in addition some items were formulated from who and cdc guidelines. secondly, due to the lockdown, we did not design the sample to statistically represent the lebanese population of physicians and make rigid extrapolations, but to offer for the first time, useful insights of the knowledge and practices towards covid-19. thirdly, only physicians who publicly shared their phone numbers were eligible to participate; this could have led to selection bias. therefore, assessment of knowledge and practices of a significant proportion of physicians and their opinions might be missed in this analysis. fourth, some participants might have provided socially desirable responses rather than their actual opinions. this study offers useful insights into the knowledge and practices of lebanese physicians towards covid-19. lebanese physicians revealed a good level of knowledge; however, they exhibit poor preventive practices. as the global threat of covid-19 continues to emerge, there is 2.04 (1.01-4.12) † others included single, widowed, and divorced, *p-value<0.05 is considered significant a clear need for further education and training, particularly on disease transmission, actions in dealing with covid-19 cases, and preventive measures. this should, in turn, improve their confidence and relief their fears towards getting infected by covid-19 cases. severe acute respiratory syndrome coronavirus 2; covid-19: coronavirus disease 2019; hcws: health care workers; nhcprc: national health commission of the people's republic of china; cdc: centers of disease controlled and prevention board; zhumc: al zahraa hospital university medical center who: world health organization; spss: statistical package for social sciences; sd: standard deviations; ci: confidence interval; idsa: infectious diseases society of america; tv: television; lop: lebanese order of physicians; ppe: personal protective equipment references 1. (who) who. pneumonia of unknown cause -china emergenciespreparedness, response, disease outbreak news, world health organization (who)-march 30: who the epidemiological characteristics of 2019 novel coronavirus diseases (covid-19 world heath organisation. who coronavirus disease (covid-19) dashboard modes of transmission of virus causing covid-19:implications for ipc precaution recommendations death from covid-19 of 23 health care workers in china characteristics of doctors' fatality due to covid-19 in western europe and asia-pacific countries epidemiological surveillance program. surveillance of covid-19 in lebanon characteristics of physicians practising in lebanon: a survey knowledge and attitude of healthcare workers about middle east respiratory syndrome in multispecialty hospitals of qassim, saudi arabia attitude and practices of healthcare providers towards mers-cov infection at makkah hospitals, ksa knowledge and attitude towards the middle east respiratory syndrome coronavirus among healthcare personnel in the southern region of saudi arabia the perceived effectiveness of mers-cov educational programs and knowledge transfer among primary healthcare workers: a cross-sectional survey knowledge and attitudes towards middle east respiratory sydrome-coronavirus (mers-cov) among health care workers in south-western saudi arabia knowledge, attitude and practice towards covid-19 among chronic disease patients at addis zemen hospital, northwest ethiopia knowledge, attitude, and practice regarding covid-19 among healthcare workers in henan novel coronavirus (covid-19) knowledge and perceptions: a survey of healthcare workers knowledge, attitude, practice and perceived barriers among healthcare professionals regarding covid-19: a cross-sectional survey from pakistan knowledge, attitude, practice and perceived barriers among healthcare workers regarding covid-19: a cross-sectional survey from pakistan publisher's note springer nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations the authors would like to thank dr. mohammad obeid, dr. georges maalouf, and mr. mario baakliny for their help during the data collection. the authors are also grateful to all physicians who accepted to be part of this study. authors' contributions laa and ra developed the project idea. mc and rh formulated the questionnaire. laa and zn organized and analyzed the survey. laa, zn, and ra drafted and critically reviewed the paper. yf reviewed the manuscript for important intellectual content. all authors read and agreed on the final version. no funding was received.availability of data and materials data are available from the corresponding authors upon reasonable request. the study was initially approved by the scientific research committee of the neuroscience research center, faculty of medical sciences at the lebanese university (reference #241/2020), and then by the institutional review board (irb) of al zahraa hospital university medical center zhumc (reference # 2/ 2020). participation in the study was voluntary. all the necessary measures to safeguard participants' anonymity and confidentiality of information were respected. written informed consent was obtained from all the participants. not applicable. the author(s) declare that they have no competing interests.author details key: cord-034133-tx0hciiv authors: engda, tigist title: the contribution of medical educational system of the college of medicine, and health sciences of the university of gondar in ethiopia on the knowledge, attitudes, and practices of graduate students of health sciences in relation to the prevention and control of nosocomial infections during the academic year of 2018 date: 2020-10-22 journal: bmc med educ doi: 10.1186/s12909-020-02271-6 sha: doc_id: 34133 cord_uid: tx0hciiv background: nosocomial infection, also called a hospital-acquired infection, is an infection acquired during admitting patients in health care facilities. nosocomial infection can be prevented and controlled by giving training to those responsible. this study aimed to assess the contribution of the medical education system on the knowledge, attitudes, and practices of the graduate students of health sciences about the prevention and control of nosocomial infection in the college of medicine and health sciences at the university of gondar in the academic year of 2018. method: an institution-based cross-sectional study was conducted among all graduate health science students posted in the different departments at the university of gondar in the college of medicine and health sciences from february to june 2018. a total of 422 study participants were included. data were analyzed using spss version 20. results: out of a total of 422 respondents, only 40% have taken training for infection prevention; out of which 39% had taken the training for a year ago. moreover, only 35.5% have good knowledge of nosocomial infections as a result of the training; and only 32.5% have good understanding of the practical training given on prevention and control. only 36% have good attitude towards its prevention and control. conclusion: the result shows that only a few of the respondents have taken the infection prevention training. yet, a smaller proportion of them had good knowledge, attitude, and practice on nosocomial infections. hence, the medical education system should give more attention to the training of the nosocomial infection control by developing different strategies to prepare the students on these issues before they start their clinical attachment. nosocomial infection is a localized or systemic infection that is acquired in a health care facility that may manifest 48 h after the patient's admission to or discharged from the health care facility [1] . it can be caused by bacteria, viruses, parasites, and fungi that may be present in the air, surfaces, or equipment surrounding the health institutions [2] . it can affect patients of all age groups; however, neonates, immunocompromised adults, and the elders are the most vulnerable ones [3, 4] . nosocomial infections have been recognized as a problem affecting the quality of health care services and are the principal source of adverse healthcare effects. increased hospital stays, increased costs of healthcare, economic hardship to patients, and their families, and even deaths are some of the negative effects [5] [6] [7] . the amount of nosocomial infections of low-and middle-income countries is higher because of the limited knowledge and utilization of post exposure prophylaxis (pep), limited knowledge of professional risks, low adherence to universal precautions (up), and inaccessibility of personal protective equipment (ppe) [2, 3, 5, 8] . findings of several epidemiological studies show that hcws, mainly physicians, dentists, laboratories, and nurses are involved in the transmission of nosocomial infections [8] [9] [10] [11] . it has also been reported that its transmission increases during the performances of medical procedures whenever hcws fail to follow aseptic precautions [8, 10] . the world health organization (who), in conjunction with the cdc, gives high attention to the prevention of nosocomial infections as it has developed a practical manual for the prevention of nosocomial infections globally (who, 2002) . some recommended strategies included in the manual were the use of hand decontamination, personal hygiene, utilization of personal protectives, and proper methods of handling soiled clothing when healthcare workers perform patient care activities. it also recommends methods of preventing environmental contamination including the cleaning of the hospital environment using hot superheated water, sterilizing patient equipment, and preventing the transmission of pathogens like hiv, hepatitis b and c viruses, and tb to the staff [1] . efficient pre-service and in-service training given by incorporating in the medical education system supported with good monitoring and evaluation methods of hcw practices play a pivotal role in the sustainability of the knowledge, attitude, and practice of universal precautions and infection control [12, 13] . this scheme was supported by many studies conducted worldwide. for example, in india, an educational module had effectively elevated the knowledge, attitude, and practice score of hcws from 14% before the intervention to 94% thereafter [14] . in korea, it was also investigated that a group of nurses and medical students who had received education on hais showed high knowledge (p = 0.036) and performance (p < 0.01) levels [15] . similarly, a study at seton hall university in new jersey, usa, indicated that the total score for the knowledge category was 93% [16] . likewise, a study in egypt reported that physicians had the best level of knowledge, but the least in practicing general safety measures than others in the preintervention phase. however, they increased their practice score from 54.3 to 86.1% after receiving continuing education [14] . a study in pakistan reported that the knowledge score was 3.8 with a median of 4. the dispenser had the highest knowledge score while the housekeepers had the lowest. knowledge about the mode of transmission of bloodborne pathogen and the work experience alone significantly predicted the use of universal precaution methods in multiple linear regression models [17] . this principle is also supported by studies conducted in ethiopia. in 2015, a cross-sectional study was conducted in addis ababa on hcws who received training on transmission, vaccination, and diagnosis of hbv to assess their knowledge of risk factors for hbv. the result showed that more than 80% of the respondents had the knowledge of the modes of transmission and prevention of hbv; 83.3% had a positive attitude towards following infection control guidelines [18] . similarly, a study conducted in ethiopia at dire dawa university on the medical and health sciences students reported that almost all of the respondents had good knowledge of the transmission, treatment, and prevention of hbv. also, 94.2% of them had good attitudes towards the importance of standard precautions, but 58.4% had poor practices in applying the recommended standard precautions [19] . in another study, a hospitalbased cross-sectional investigation was conducted among 150 health workers who were taking training about infection and prevention of hospital-acquired infections at debre markos hospital in ethiopia. the results showed that 84.7% of them were found to be knowledgeable; however, only 57.3% of the respondents demonstrated good practice in infection prevention. moreover, respondents with older age, longer work experience, and higher educational status excelled in both knowledge and practice of infection prevention. inservice training, availability of infection prevention supplies, and adherence to infection prevention guidelines were also associated with the practice of infection prevention [19] . on the other hand, a study conducted at the university of gondar hospital about hand hygiene compliance on 405 study participants showed that only 8.97% had the knowledge about it and 5.62% had received training about hand hygiene compliance, respectively [20] . hand hygiene is an important means for the control and prevention of nosocomial infection. therefore, the current study intended to determine the impact of the medical education system on the knowledge, attitude, and practice of graduate health sciences students about the prevention and control of nosocomial infections at the university of gondar. an institution-based cross-sectional study was conducted with graduate students from the college of medicine and health sciences at the university of gondar. data collection was made between february and june of 2018. students who attended the regular academic program at the university of gondar, college of medicine and health sciences were the source population. however, only graduate classes of health science students were taken as the study population. before data collection, eleven departments were selected for sampling: health informatics, medical laboratory sciences, health officer, physiotherapy, environmental and occupational health and safety, psychiatry, optometry, midwifery, nursing, pharmacy, and anesthesia. there were 465 graduate students of health sciences in the academic year of 2018. the sample size was determined using a single random sampling method. since no similar study was found in the area, 95% were taken as a confidence interval. then, the calculated sample size was 384 and by adding a 10% non-response rate, the final calculated sample size became 422 (fig. 1) . the questionnaire was constructed from emergent themes reviewed in the literature and items were derived from the established guidelines set by a task force committee on infection control practices advisory committee [5, 21] . the questionnaire includes 40 questions subdivided into four categories: sociodemographic, knowledge, attitude, and practice towards nosocomial infections. knowledge was assessed using 11 questions containing three alternative choices each. the answers from the given alternatives were symbolized as '1' for poor, '2' for fair, and '3' for good. a higher score in the questions concerning nosocomial infections is considered to be good knowledge of nosocomial infections. attitude was measured using 8 questions in which answers for each question were assigned as 1 for 'disagree', 2 for 'neutral', and 3 for 'agree.' higher score achieved was considered as a positive attitude toward standard precaution. moreover, their practice towards the prevention and control of nosocomial infections was assessed using 14 questions with three alternative answers which were assigned as '1' for poor, '2' for fair, and '3' for good (table 1, table 2, and table 3 ). bloom's cut-off point was used to determine the level of knowledge, attitude, and practice because the conceptual framework of the present study was based on the taxonomy of educational objectives developed by bloom (1956) . according to bloom's taxonomy (1956), human behaviors are derived from the integration of the cognitive, affective, and psychomotor domains. knowledge, attitudes, and practices could be representatives of the cognitive, affective, and psychomotor domains, respectively. knowledge refers to the factual, conceptual, procedural, and met cognitive thought [22] . attitude is an internal or covert feeling and emotion; or selective nature of intended behavior which represents the affective domain. practice represents the psychomotor domain which refers to the physical movement, coordination and use of motor or neuromuscular activities [22] . accordingly, participants' overall knowledge and practice are considered as good if the score is 80% and above; moderate if the score is between 60 and 79%; and poor if the score is less than 60%. similarly, attitude towards nosocomial infection was assessed using 8 questions. responses to questions related to attitude were graded on a 3-point likert scale with an agreement scale ranging from '1' for disagree to '3' for agree [7] . the overall level of attitude was categorized using bloom's cut-off point: positive if the score was 80% and above; neutral if the score was 60-79%; and negative if the score was less than 60%. a simple random sampling method and lottery technique were used to select the respondents and a quantitative method of data collection was employed through a self-administered questionnaire. the quantitative method involves assessment of the impact of medical education on the knowledge, attitude, and practices of 422 graduate students on the prevention and control of nosocomial infection. the data collection instrument format was developed in english by different individuals for its accuracy and desired results. the data collectors used a self-administered questionnaire for graduate students of the health sciences, class of 2018. after receiving a complete response of the questionnaires, data were analyzed using descriptive statistics by ibm spss version 20.0. demographic characteristics are presented in tabular form using descriptive statistics and reported as mean, median, standard deviation, frequency, and percentage as presented in tables. the study was conducted after a written ethical clearance is obtained from the ethical research committee of the school of biomedical laboratory sciences and college of medicine and health sciences. moreover, the consent forms of the participants were completed voluntarily by the study participants themselves. pretest to evaluate the understandability and applicability of the instruments used, pretest data were collected and checked from 10 medical laboratory graduate students using a self-administered questionnaire. self-administered questionnaires were collected from 422 respondents. out of these respondents, 32% were female while only 2% of them were above 25 years of age. the proportions of respondents were: (6.9%) health informatics, (6.9%) medical laboratory sciences, (15.6%) health officers, (4.5%) physiotherapy, (4.3%) environmental and occupational health and safety, (6.6%) psychiatry, (5.5%) optometry, (14.7%) midwifery, (17.5%) nursing, (12.3%) pharmacy, and (5.2%) anesthesia graduated students. only 40% of the respondents had been trained in infection, prevention out of which 39% took the training at least a year ago ( table 4) . even though 40% of the respondents stated that they had taken training on nosocomial infections, only 35.5% had good knowledge of nosocomial infections (fig. 2) . from the 55 questions administered, the score of knowledge of the respondents ranged from 11 to 55, with a mean score of 40.64 at std. of + 12.73 (table 5) . as reflected in table 5 , out of the total respondents, 59.5% had good knowledge of the modes of transmission and risk factors for nosocomial infections; 57.3% of the respondents also stated that they were fully aware of hand-washing guidelines; 47.2% knew where and how the contents in biohazard bags or containers are being disposed. it is also shown that table 1 questioner for the assessment of knowledge of graduate health sciences students towards nosocomial infection instruction: to complete this section, please make a tick "✓" on the number corresponds to how you agree with the given alternatives 1 = poor, 2 = fair and 3 = good table 2 questioner for the assessment of practice of graduate health sciences students towards nosocomial infection instruction: to complete this section, please make a tick "✓" on the number corresponds to how you agree with the given statement 1 = poor, 2 = fair, 3 = good 65.2% of them knew that nosocomial infections could be transmitted via fomites, and 59.6% of the respondents understood that healthcare facilities harbor a variety of microorganisms that could be transmitted by healthcare workers. then, 62.1% of respondents were fully aware of safety precautions for the disposal of used medical equipment, and 56.2% of them believed that neutropenic patients like those with diseases of the respiratory system should be kept in private rooms. furthermore, 57.8% of the graduates were knowledgeable in the use of alcohol-based formulations, and 63.5% of them stated that some microorganisms were not totally removed by alcoholbased solutions. the overall practice scores showed that 32.5% have good practice in the prevention and control of nosocomial infection. the score of the practice of the respondents ranged from 14 to 70, with a mean of 45.61 atstd+ 15.35 (table 6) . respondents reflection to correctly following guidelines for the use of alcohol-based solutions before and after patient care activities were 42.2%; before opening vascular access equipment were 39.3%; between each patient contact were 48.6%; before and after direct contact with patients' intact skin were 41.2%; moving from a contaminated body site to a clean table 3 questioner for the assessment of attitude of graduate health sciences students towards nosocomial infection instruction: to complete this section, please make a tick "✓" on the number corresponds to how you agree with the given statement 1 = disagree, 2 = neutral, 3 = agree body site were 46.7%; before and after drawing or manipulating patient's body fluid samples were 52.1%; before inserting indwelling urinary catheters were 50.0%; and after touching inanimate objects and equipment in the patients' room were 45.7%. of all, 31.8% of the respondents used their computer keyboards with their glove during busy workload. finally, 43.1% of the respondents removed their rings, watches, or bracelets during hand hygiene (table 6) . the attitudes of students towards the prevention and control of nosocomial infection were 36% (fig. 3) . of the total respondents, 61.9% believed that nosocomial infections are posing serious negative outcomes but 44.8% responded the opposite while a colleague is noncompliant with the recommended guidelines for patient safety. moreover, 51.9% of the respondents regularly (table 7) . nosocomial infection is one of the most important challenges in health institutions. therefore, this study assessed the knowledge, attitude, practice, and associated factors of infection prevention among health science graduate students. the overall score of knowledge (35.5%) was lower than the study conducted in the usa (93%) and nepal (97%) [16, 23] . similarly, 59.6% had good knowledge of their etiology, modes of transmission, and risk factors of nosocomial infections which were also lower than the study conducted in new jersey, usa (95.7%), and nepal (82%). moreover, only 59.6% of the participants knew fomites as transmission factors, which is still lower than the study conducted in the usa (98.9%) [16, 23] . this might be due to a difference in study participants. in the usa, the study participants were registered nurses who were working in health care institutions and they might develop knowledge from their experiences and/or in-service training. however, in this study, the participants were graduate students of which 60% never took training on the prevention and control of nosocomial infections. this shows that not all health science students in this college are taking training before their clinical attachments. in the findings of this study, only 31.5% had good practice in the prevention and control of nosocomial infections of which only 50.4% followed the guidelines for the use of alcohol-containing hand sanitizer which is lower than the study conducted in the usa (78%), but higher than a study conducted in china (11%) [16, 17] . this may be due to the difference in study participants, the accessibility of alcohol-containing hand sanitizer, and the large difference in course curriculum where infection prevention might not be incorporated in all of the target population. possibly for similar reasons, the attitude of study participants towards the prevention and control of nosocomial infection (36%) was lower than a study carried out in the usa (79.66%) and nepal (66.0%) [16, 23] . a 52.2% positive attitude towards following the recommended guidelines for reducing the transmission of nosocomial infections was lower than a study conducted among health care workers at addis ababa in ethiopia (83.3%) [18] . this might be due to a difference in study participants in addis ababa, where they were registered health care workers who were working in the health institution and they might develop knowledge, attitude, and practice either through their experience or in-service training. however, in this study, the participants were graduate students who reported that 60% of them never took training on the prevention and control of nosocomial infections before their clinical attachment. generally, more than half of the respondents had poor knowledge, attitude, and practice on nosocomial infection and the application of infection and prevention procedures. the medical education system is the most important and effective tool to bring a better outcome for controlling and preventing nosocomial infections. incorporating the necessary knowledge into the regular course curricula, organizing training modules to medical students before starting clinical attachment, providing different guidelines and standard operating procedures are also helpful in understanding the nature of infections and how, when, and where to prevent and control nosocomial infection. therefore, this study showed that a smaller number of respondents had taken infection prevention training on their regular medical system. consequently, smaller proportions of them had good knowledge, attitude, and practice on the nature of the infection, prevention, and control strategies for nosocomial infections. therefore, to improve the level of knowledge, attitude, and practice of students towards nosocomial infections, strengthening the medical education system through relevant seminars including short and long-term training is essential. at the same time, the availability of infection prevention guidelines, standard operating procedures, and personal protective equipment like alcohol-based solutions in health institutions are important. departments, schools, and college administrative officers should work together to facilitate infection prevention training programs for all health science students before starting their clinical attachments. the ministry of health and ministry of education should work to enforce the universities to incorporate infection prevention knowledge into the course curricula for all health science students. all health care institutions must be prepared to give vaccination of common hospital-acquired diseases by making available infection prevention materials and standard operational author's contributions te: conception of a research idea, study design, data analysis, interpretation, and manuscript write up. the author(s) read and approved the final manuscript. no one was responsible for the funding of this research. all data generated or analyzed during this study are included in this article. ethics approval and consent to participate ethical clearance was obtained from the research and ethical review committee of the school of biomedical and laboratory sciences, college of medicine and health sciences, university of gondar. moreover, written consent was taken from each participant after they understood the purpose of the study. all the subjects' data were kept in full confidentiality and were not being disclosed to an unauthorized person. not applicable. extended hospital stay days, mortality and increased cost of healthcare) nurse or physician) is non-compliant with the recommended guidelines for patient safety in my opinion, healthcare workers should be sanctioned for non-compliance with protocols for reducing transmission of nosocomial infections (for example, yearly assessment, and denied promotion) in my opinion, healthcare workers should be rewarded (for example, given plaques, certificate) for compliance with protocols aimed at reducing transmission of nosocomial the number of graduate health science students; % indicates percentage engda prevention of hospital acquired infections: a practical guide is us health really the best in the world? decreasing urinary tract infections through staff development, outcomes, and nursing process statistical abstract of the united states guidelines for hand hygiene in health care settings an outbreak of hepatitis c virus infections among patients at a hematology/oncology clinic new jersey: pearson education cluster of cases of severe acute respiratory syndrome among toronto healthcare workers after implementation of infection control precautions: a case series use of influenza a (h1n1) monovalent vaccine: recommendations of the advisory committee on immunization practices (acip) the global burden of disease attributable to contaminated injections given in health care settings guidelines for preventing the transmission of mycobacterium tuberculosis in health-care settings training self-assessment and task-selection skills: a cognitive approach to improving self-regulated learning student self-assessment: the key to stronger student motivation and higher achievement. educ horizons impact of education on knowledge, attitudes and practices among various categories of health care workers on nosocomial infections knowledge and performance of the universal precautions by nursing and medical students in korea exploring knowledge, attitudes and practices of registered nurses regarding the spread of nosocomial infections. seton hall university dissertations and theses (etds): paper 1865 poor knowledge predictor of nonadherence to universal precautions for blood borne pathogens at first level care facilities in pakistan assessment of knowledge, attitudes and practices toward prevention of hepatitis b virus infection among students of medicine and health sciences in northwest ethiopia knowledge, practice and associated factors of infection prevention among healthcare workers in debre markos referral hospital, northwest ethiopia. bmc hand hygiene compliance and associated factors among health care providers in gondar university hospital attitude and practice of nursing students regarding hand hygiene in the western region of nepal taxonomy education attitude and practice of nursing students on hospital acquired infections in western region of nepal the author declares that there is no competing interest.received: 6 june 2019 accepted: 1 october 2020 key: cord-029672-y2ii6r3u authors: van assche, ari; lundan, sarianna title: from the editor: covid-19 and international business policy date: 2020-07-24 journal: j int bus policy doi: 10.1057/s42214-020-00065-7 sha: doc_id: 29672 cord_uid: y2ii6r3u we introduce a special collection of papers on covid-19 and international business policy that looks at the global policy challenge from different perspectives. the combination of confinement and protectionist measures including export bans, import tariffs and border closures that were introduced in response to the covid-19 pandemic affect mnes by increasing trade costs and amplifying the costs of communicating tacit knowledge. in this editorial, we draw specific attention to how these policies impact the heart of mnes’ activities – the knowledge-intensive intangibles. physical distancing limits the face-to-face meetings that undergird the production of intangibles in large urban hotspots. at the same time, travel restrictions constrain mnes’ abilities to connect to and source knowledge from their foreign partners. virtual conferencing has helped mnes to cope with these challenges but it remains an imperfect substitute. a protracted public health crisis that would require limiting social interactions in the medium term might push mnes to reconfigure their knowledge management strategies both locally and globally. in 2016, the academy of international business launched the ambitious project of adding a public policy journal to its line of publications -the journal of international business policy. the rationale was that international business scholars have a deep knowledge about how the economic landscape influences firm behavior and vice versa, and this knowledge can be highly valuable in policy discussions (lundan, 2018) . the goal of the journal was therefore to promote a policy turn in international business research by pushing scholars to not only reflect on the implications of their research for managers, but also to investigate the societal implications of their studies (van assche, 2018) . the timing of the launch was auspicious as it coincided with a period where international business policy mounted to the forefront of public policy discussions. populism had been rising on both sides of the atlantic in the wake of the great recession of [2008] [2009] , which fed a wave of antitrade sentiment (rodrik, 2018) . several government leaders jumped on the populist bandwagon by throwing their support behind export mercantilism (evenett, 2019) , renegotiating or leaving trade agreements, and initiating trade wars (bown & irwin, 2019) . international business (ib) scholars were quick to seize the moment and have in ten jibp journal issues put their own stamp on ib policy debates in an era of ''slowbalization''. the 2020 ''great lockdown'' related to the covid-19 pandemic is yet another policy challenge that can benefit from insights from ib scholarship. the instigator of the acute economic downturns across the globe is not the health crisis as such, but rather the policy response that has been remarkable in its suddenness, magnitude and synchronicity across countries. to halt the spread of the disease, governments have adopted extraordinary containment policies to flatten the epidemiological curve, that is, to slow the rate of new covid-19 infections in order to ease the burden on local health facilities. when the epidemic initially hit, governments put a hard stop on economic activity by suspending operations of non-essential businesses, by closing international borders to nonessential traffic, and by implementing stringent stay-at-home orders. once the initial outbreak was contained, countries have gradually loosened restrictions on those non-essential businesses where virus transmission can be more easily controlled. lockdown exit strategies have varied substantially across countries and regions, but a common element is that physical distancing measures will need to remain in place until a vaccine or treatment becomes available. policy is also central to countries' efforts to cope with the resulting economic crisis. governments have adopted a mix of fiscal, monetary and financial policies to avoid economic collapse related to the ''hard stop'' containment policies (baldwin, 2020) . as of april 2020, japan and the united states have committed to fiscal stimulus packages that already exceed 10 percent of their annual gross domestic products (elgin, basbug, & yalaman, 2020) . international business policies have been part of the mix. past experiences have shown that governments turn to protectionism during economic downturns (evenett, 2019) , and this time is no different. when the epidemic initially hit, several countries reacted to global shortfalls of personal protective equipment and testing kits by banning or limiting the export of medical equipment and medicines (evenett, 2020) . once the attention turned from infection containment to economic recovery, several governments have switched to more traditional mercantilist policies by imposing import tariffs, putting into place ''buy domestic'' measures and adopting export subsidies (bown, 2020) . most countries in the world have imposed partial or complete border closures to foreign nationals during the coronavirus outbreak (chugh, 2020) . businesses -including multinational enterprises (mnes) -have been left scrambling trying to respond to the new policy reality. much has been written about distilleries, textile companies and automotive firms that have ventured into the production of hand sanitizers, masks and ventilators (fleming, 2020) . other studies have pointed out that firms are reorganizing their global value chains by reshoring production, diversifying their supplier base and adopting industry 4.0 technologies (javorcik, 2020; kilic & marin, 2020; seric & winkler, 2020) . while these initial insights are valuable, it is fair to say that extant studies have only scratched the surface. our understanding of the complex twoway interaction between covid-19-related policies and ib remains murky and can benefit from fresh insights originating from ib scholarship. for this reason, we in early april contacted several leading scholars in ib and adjacent fields and asked them to provide their thoughts on the impact of the covid-19 pandemic on international business both in the short and medium run. our aim is to publish these articles in a jibp special collection on covid-19 and international business policy that will span the next two issues of the journal. in this editorial, we introduce the special collection by drawing the attention of jibp readers to the impact of covid-19-related policies that strike at the heart of mnes -the knowledge-intensive intangibles -as their managers rush to cope with local physical distancing measures, non-essential travel restrictions and stringent immigration policies. at the end of the article, we also provide an overview of the first four papers that are included in the special collection. of intangibles in a recent book, baldwin (2016) divided the history of globalization into three periods: the pre-globalization period (before 1820) when both trade and communication costs were high, the first unbundling period (1820-1990) when trade costs dropped but communication costs remained high, and the second unbundling period (1990-present) when both trade and communication costs became low (see figure 1) . 1 he argued that each period has organized production and consumption in a different way, affecting trade and foreign investment flows. the pre-globalized period was effectively a phase of autarky. high trade and communication costs isolated production and consumption to such a degree that the global economy was little more than a patchwork of local ecosystems that took care of themselves. while there was some international business (e.g., china's silk road trade), it had only a limited impact on a country's aggregate performance. the arrival of steamships and railroads in the nineteenth century substantially lowered the cost of long-distance transportation. with cheaper international shipping, more consumers started buying faraway goods, allowing for the gradual separation of production and consumption. this first unbundling period led to a spurt in international business, but the high communication costs across borders confined production processes within countries. globalization accelerated again at the end of the twentieth century, when the ict revolution slashed the cost of communicating codifiable information at a distance. this instigated globalization's second unbundling period, which involved the international fragmentation of production. improvements in communication technology now made it feasible and profitable to coordinate production activities across different countries, allowing for the emergence of global value chains. ib scholarship has brought to the table several refinements that have enriched this depiction of globalization. first, it has highlighted the central role that mnes have played in driving change over the two unbundling periods. as mne managers observed shifts in the costs and risks of doing business abroad, they reassessed how to compete, where to extend their geographical footprint and which activities to conduct within firm boundaries (buckley & strange, 2015; verbeke, coeurderoy, & matt, 2018) . it is mnes' adoption of new forms of intra-and inter-organizational arrangements on the micro-level that instigated shifts in macro trade and investment flows. second, ib scholarship has emphasized that mnes not only adapt the tangible parts of their activities -the production, trade and sale of physical goods -but also the intangibles which are at the heart of mne operations. intangibles are those claims to future benefits that do not have a physical or financial embodiment (lev, 2001) . highly knowledge-intensive in nature, they include the intellectual property, brand equity and other economic competencies that provide mnes the economic power to set product strategy, place orders, and take financial responsibility for the goods and services that their supply chains turn out (sturgeon, 2009; teece, 2018) . the drop in communication costs during the second unbundling period has pushed mnes to not only fragment production internationally but also to globalize the production of intangibles. a highly influential ib literature describes the complex global knowledge management strategies that mnes adopt to transfer and exploit existing repositories of knowledge around the world and to engage in the exploration of new knowledge (alcácer, cantwell & piscitello, 2016; cantwell, 1989; kogut & zander, 1993) . one of the reasons these knowledge management strategies are complex is that mnes need to consider the paradoxical geography of knowledge creation which is both concentrated in few local hotspots and increasingly global (lorenzen, mudambi & schotter, 2020) . the tacit nature of much of the knowledge that is embedded in intangibles implies that there are significant agglomeration economies in the production of intangibles (glaeser & gottlieb, 2009 ). one reason for this is that complex forms of knowledge are difficult to communicate over distance and require direct and repeated face-to-face contact for their exchange (storper & venables, 2004; d'este, guy, & iammarino, 2013) . for firms, co-locating with similar and related companies thus can boost collective learning processes through frequent opportunities for formal and informal knowledge exchanges. several recent studies provide strong evidence that the production of intangibles concentrates disproportionately in a few urban hotspots (crescenzi et al., 2019; moretti, 2019; balland et al., 2020) . a complementary feature is that lead firms have developed sophisticated strategies to allow teams of people in different global cities to collaborate in the production of intangibles. many lead firms nowadays deliberately establish linkages to other global knowledge hotspots to tap into pockets of complementary knowledge and resources that are unavailable or more expensive locally (bathelt, malmberg, & maskell, 2004) . they do so by setting up intrafirm linkages to competence-creating subsidiaries or by developing inter-firm partnerships (bathelt, cantwell, & mudambi, 2018) . once new knowledge combinations are created, these can then be transferred to the home country or other locations to enhance the parent firm's innovation performance (cano-kollmann et al., 2016) . regardless of the governance structure adopted, and the knowledge exchanges facilitated by ict, the mobility of key personnel is a key factor that keeps these ''pipelines'' open. several scholars have voiced their concern that the current covid-19 pandemic may have a lasting impact on globalization in general and more specifically on the ways in which mnes configure their cross-border activities (altman, 2020; irwin, 2020; kobrin, in this issue) . the combination of covid-19-related confinement and protectionist measures is effectively a policy mix that affects mnes by (1) increasing trade costs and (2) amplifying the costs of communicating tacit knowledge (see figure 1 ). the increase in trade costs reflects in part the struggle to adapt the international transportation system to the covid-19-related policy measures. the unprecedented grounding of passenger planes has disrupted air transport since a significant portion of air cargo is transported in the hold of passenger planes (van assche, 2020). many ports' restrictions on vessels and crew changes have led to similar disruptions in sea freight (heiland & ulltveit-moe, 2020) . protectionist measures have added to these trade cost increases. with governments trying to limit the health and economic impact on their people, some 75 countries have restricted their exports of hundreds of products, ranging from antibiotics to cotton face masks and medical ventilators (evenett, 2020) . the united states has reignited its trade war with china. japan has promised subsidies to japanese companies that are willing to reshore production from china to japan. the policy response to covid-19 has also amplified the costs of communicating tacit knowledge. locally, the physical distancing rules have severely limited both the planned and unplanned face-toface meetings that undergird the buzz of local innovation ecosystems. internationally, the closing of borders to non-essential travel and limits to immigration have limited firms' abilities to collaborate and exchange tacit knowledge with their foreign partners. the availability of digital infrastructure has allowed firms to limit the scope of disruptions related to the reduction in face-to-face interactions and may well have a permanent impact on global knowledge management practices in the post-pandemic period. for many office workers, telework and virtual conferencing have become the new norm. some companies such as twitter have even announced that staff can continue to work from home permanently. despite vast improvements in virtual conferencing capabilities, they nonetheless remain imperfect substitutes to face-to-face meetings. the effect of covid-19 on the development of intangibles will depend on both the duration of the health crisis and the extent to which the economy can return to normal once the pandemic itself is behind us. a relatively short crisis with a swift relaxing of physical distancing rules, an opening up of international borders and elimination of protectionist measures will likely lead to the return of established practices. a protracted public health pandemic, then again, that continues to limit social interactions in the medium term and potentially leads to a global descent into protectionism will require firms to reconfigure their business models on both the tangible and intangible side. in the special collection in this editorial we have highlighted just one topic area where international business scholarship can be instrumental in understanding the link between covid-19 and eventual economic recovery. the jibp special collection on covid-19 and international business policy includes several other original papers, including four that are published in this issue. we hope that these papers stimulate members of the ib community and beyond to engage in conversations about the ongoing policy challenges facing the global economy, and to place these issues on their research agenda alongside other ongoing sources of disruption such as climate change and rising inequality within and between countries. the collection opens with steve kobrin's thoughts on the impact of covid-19 on globalization. he argues that the pandemic has for the second time in little more than a decade laid bare the very real social, political and economic dangers of high specialization and connectedness between markets. after acting as a potent vehicle for the transmission of a global financial crisis during the great recession of 2008-2009, the globalized system has during the covid-19 pandemic of 2020 acted as a virulent medium for the spread of a health calamity that continues to wreak havoc across the globe. according to kobrin, these events further undermine the longstanding belief that globalization is an indicator of progress and they have reinforced fears of ''the other'' that are all too easy to exploit for political gain. he points out three major changes that need to occur to restore a positive view of globalization: restoring the balance between sovereignty and integration; making globalization more inclusive; and insuring some degree of resilience in global supply chains. however, kobrin is skeptical that the current geopolitical climate is conducive to these changes. in a second paper, gary gereffi digs deeper into one of the ''choke points'' that have been uncovered by the covid-19 pandemic, namely the global shortages of medical supplies in the early months of the health crisis. he questions the oftenheard assertion that these shortages were due to structural flaws in medical equipment supply chains and conducts a detailed case study of the face mask value chain in the united states to evaluate this claim. according to gereffi, the shortage of masks in this case was much more a policy failure than a market failure. while just-intime production techniques could make supply chains fragile in times of crisis, they did not stop the american company 3m from quadrupling its u.s. supply of n95 masks from just over 20 million in january to 95 million per month in may. rather, the critical bottleneck appeared to be the limited government stockpiling of n95 masks before the crisis and the regulatory constraints that in early 2020 prevented 3m and its competitors from scaling up local production. in a third paper, rajneesh narula turns to the other side of these global value chains and focuses on the impact of the covid-19 pandemic on the informal economy in developing countries. according to narula, informal workers and enterprises have been especially vulnerable to the pandemicinduced economic shock. for decades, global value chains have used the informal sector in developing countries as a valve that takes on extra tasks during periods of peak demand. unfortunately, the reverse side of the coin is that informal workers and firms have also been the first to be cut from the roster when the valve is turned off during economic downturns. often living from hand to mouth, and with limited government rescue packages to help them, covid-19 has thus condemned millions of workers in the informal economy to poverty. narula nonetheless sees a (faint) silver lining: the severity of the economic shock and the precariousness of the informal sector may finally convince governments to actively engage with informal actors without pushing for their formalization, for example by placing them on par with formal firms when implementing industrial policy. in a fourth paper, peter buckley reflects on the impact of covid-19 for china's highly ambitious infrastructure-building belt road initiative (bri) that spans a large number of developing economies across asia. even before the current pandemic, there were concerns in several bri countries that the high costs of the investment projects were leading to unprecedented levels of debt. buckley points out that these apprehensions have only increased in the wake of the covid-19 pandemic as governments face severe revenue shortfalls, as they embrace new budgetary priorities in light of the health crisis, and as china faces growing geopolitical tensions with the united states. he nonetheless suggests that china's need to address its own domestic overcapacity combined with the amorphous nature of bri provides the chinese government with the means to flexibly rethink its bri strategy in the face of the covid-19-related setbacks. this classification is similar to that presented by jones (2005) with the exception that the (policy-induced) collapse of international trade in the interwar years is not considered separately. internationalization in the information age: a new era for places, firms, and international business networks will covid-19 have a lasting impact on globalization the great convergence the supply side matters: guns versus butter, covid-style complex economic activities concentrate in large cities clusters and knowledge: local buzz, global pipelines and the process of knowledge creation overcoming frictions in transnational knowledge flows: challenges of connecting, sense-making and integrating covid-19 could bring down the trading system trump's assault on the global trading system: and why decoupling from china will change everything the governance of the global factory: location and control of world economic activity knowledge connectivity: an agenda for innovation research in international business technological innovation and multinational corporations will covid-19 change how we think about migration and migrant workers? the geography of innovation: local hotspots and global innovation networks shaping the formation of university-industry research collaborations: what type of proximity does really matter economic policy responses to a pandemic: developing the covid-19 economic stimulus index protectionism, state discrimination, and international business since the onset of the global financial crisis sicken thy neighbour: the initial trade policy response to covid-19 how big business is joining the fight against covid-19 the wealth of cities: agglomeration economies and spatial equilibrium in the united states an unintended crisis: covid-19 restrictions hit sea transportation the pandemic adds momentum to the deglobalization trend the greater trade collapse of 2020: learnings from the 2008-09 great trade collapse multinationals and global capitalism: from the nineteenth to the twenty first century how covid-19 is transforming the world economy how globalization became a thing that goes bump into the night knowledge of the firm and the evolutionary theory of the multinational corporation intangibles: management, measurement, and reporting international connectedness and local disconnectedness: mne strategy, city-regions and disruption from the editor: engaging international business scholars with public policy issues the effect of high-tech clusters on the productivity of top inventors populism and the economics of globalization covid-19 could spur automation and reverse globalisation -to some extent buzz: face-to-face contact and the urban economy from commodity chains to value chains: interdisciplinary theory building in an age of globalization business models and dynamic capabilities from the editor: steering a policy turn in international business-opportunities and challenges trade is among the casualties in the covid-19 pandemic the future of international business research on corporate globalization that never was… ari van assche is professor of international business at hec montréal and deputy editor of the journal of international business policy. his research focuses on the organization of global value chains and their implications for international trade sarianna lundan holds the chair in international management and governance at the university of bremen and she is the editor-in-chief of the journal of international business policy key: cord-319908-10b7de22 authors: naser, abdallah y; dahmash, eman zmaily; alwafi, hassan; alsairafi, zahra khalil; al rajeh, ahmed m.; alhartani, yosra j; turkistani, fawaz mohammad; alyami, hamad s. title: knowledge and practices towards covid-19 during its outbreak: a multinational cross-sectional study date: 2020-04-17 journal: nan doi: 10.1101/2020.04.13.20063560 sha: doc_id: 319908 cord_uid: 10b7de22 background: the emergence of covid-19 globally coupled with its unknown aetiology and its high transmission rate has created an unprecedented state of emergency worldwide. public knowledge and awareness about covid-19 are essential in suppressing its pandemic status. method: a cross-sectional study using an online survey was conducted between 19th of march and 6th of april 2020 in three middle eastern countries (jordan, saudi arabia and kuwait) to explore the knowledge and practices of middle eastern population towards covid-19. a previously developed questionnaire was used. multiple linear regression analysis was used to identify predictors of covid-19 knowledge. results: a total of 1,208 participants were involved in this study from the three countries (jordan = 389, saudi arabia = 433, and kuwait = 386). the majority of participants (n = 810, 67.2%) were females and aged 30 to 49 years (n = 501, 41.5%). participants had moderate overall covid-19 knowledge with a mean score of 7.93 (sd = 1.72) out of 12, 66.1%. participants had better knowledge about disease prevention and control with 83.0%, whereas the lowest sub-scale scores were for questions about disease transmission routes (43.3%). high education level was an important predictor of greater covid-19 knowledge scores (p<0.01). conclusion: middle eastern participants are of a relatively low level of knowledge about covid-19, particularly regarding its transmission routes. policymakers are recommended to develop informative covid-19 related campaigns targeted specifically towards university students, unemployed individuals and those with lower levels of education. coronavirus disease 2019 (abbreviated "covid-19") is an infectious disease with unknown aetiology characterised with acute pneumonia that has been recognized in wuhan, china in december 2019. predominantly, it is characterised by fever, fatigue, and dry cough. approximately, 20.0% of covid-19 patients developed to the severe stage, which is characterised by acute respiratory distress syndrome, bleeding, and coagulation dysfunction (1) . in the middle east region, this virus has appeared in some countries before others. for example, it has been firstly identified in iran then spread in kuwait, saudi arabia, and jordan. the spread of the virus in kuwait earlier than others could be attributed to the fact that there was a national holiday in kuwait in the period between the 23 rd to the 29 th of february, where most people used to spend it abroad. as people returned from vacations, confirmed cases with covid-19 started to raise. in march 12, 2020, the world health organization (who) declared that the covid-19 outbreak is a pandemic (2). the pandemic lingers to take a hefty toll on healthcare professionals, families, communities and the whole world. in response to that, jordan, saudi arabia, and neighbouring kuwait took drastic measures early on in a bid to contain the covid-19 pandemic, which includes halting air travels, imposing curfews, and quarantining and testing thousands of individuals (3, 4) . once a new pandemic starts, the focus of research and action within the medical and public health communities is chiefly and rightly directed towards the identification of the cause, clinical presentation, diagnosis as well as treatment (5) . few studies will address the epidemiology of the disease, the effectiveness of preventive measures, and the population psycho-behavioural directions. nevertheless, addressing public health preventive measures deserve equal attention. although there are several ongoing clinical trials assessing potential vaccines and treatments for covid -19, currently there is no specific treatment that can combat the virus within medical practice. therefore, public health measures are of crucial value. such measures mandate the public to be aware of transmission means and have safe preventive practices (6) . in this regard, people had concerns about the safety measures to be taken in order to protect themselves and their families from being infected. generally, the spread of any infectious disease is associated with a high level of fear among the population (7) . a particular concern in this regard is the spread of misinformation about covid-19 on social media sites (8) . universal data collection and analysis of pandemic control hardly contains information about the knowledge and practices of the population regarding the disease and their significance (9, 10) . to guarantee the best control over the transmission of the disease, community's adherence to important control measures are vital, which is affected to large extent by their knowledge and practices towards covid-19 (9, 10) . public practices that are built on knowledge ought to be evaluated during pandemics as it will enable policymakers to unravel generic issues from culture specific concerns and then communicate the best practices that appear to have an impact in successfully controlling the outbreak across different communities (11, 12) . thus there is an urgent need to understand what people knows about covid-19, and which misperceptions they hold about this condition, in order to help healthcare authorities and the media to design effective information campaigns, to know which group of people should be targeted, and to help controlling the spread of the virus among population. therefore, this study aims to assess the knowledge and practices of middle eastern population towards covid-19 during its rapid rise period. a cross-sectional study by means of online survey was conducted between 19 th of march and 06 th of april 2020 in three arab countries (jordan, saudi arabia and kuwait) to explore the knowledge and practices of middle eastern population towards covid-19. a convenience sample of eligible participants was invited to participate in the study from the three countries through social media (facebook and whatsapp). all participants voluntarily participated in the study and were thus considered exempt from written informed consent. study aim and objectives were clearly explained at the beginning of the survey. the inclusion criterion was participants aged 18 years who was living either in jordan, saudi arabia or kuwait. participants were excluded if they were: a) below 18 years of age; or b) unable to understand arabic language. . these questions were answered on a true/false basis with an additional "i don't know" option. a correct answer was given a score of one point, and any false or unknown answer was given a score of zero point. the total knowledge score could range from 0 to 12, with a higher score indicating a better knowledge about covid-19. cronbach's alpha coefficient of the knowledge questionnaire was 0.7 in our sample, showing acceptable internal consistency. respondents' practices were assessed using two behaviour questions, which asked the participants about whether they have gone to any crowded place, and wearing a mask when going out in recent . cc-by-nc-nd 4.0 international license it is made available under a author/funder, who has granted medrxiv a license to display the preprint in perpetuity. is the (which was not peer-reviewed) the copyright holder for this preprint . https://doi.org/10.1101/2020.04.13.20063560 doi: medrxiv preprint days. in addition, the following information was collected about participants' demographics: age, gender, marital status, education level, income, and employment status. using a confidence interval of 95%, a standard deviation of 0.5, a margin of error of 5%, the required sample size was 385 participants from each study population. this study was approved by the research ethics committee at faculty of pharmacy in isra university, amman, jordan. as participation in the study was voluntary, the research ethics committee approved consent waiver. descriptive statistics were used to describe participants' demographic characteristics. continuous data were reported as mean ± sd. categorical data were reported as percentages (frequencies). independent samples t-test/one-way analysis of variance (anova) was used to compare the mean knowledge scores between different demographic groups. participants' scores were interpreted as a continuous scale based on the scale midpoint, where scores above the midpoint identified stronger covid-19 knowledge. mean score was expressed in percentage out of 100% to facilitate the comparison between different sub-scales. multivariable linear regression analysis was used to identify predictors of knowledge score. a two-sided p<0.05 was considered as statistically significant. the statistical analyses were carried out using spss (version 25). a total of 1,208 participants took part in this study (jordan = 389, saudi arabia = 433, and kuwait = 386). table 1 details the baseline characteristics of the participants in the three countries. the majority of participants (n = 810, 67.2%) were females, aged 30 to 49 years (n = 501, 41.5%), married (n = 685, 56.8%), has bachelor degree (n = 723, 60.4%), employed (n = 568, 47.1%), and with an income of 2,113$ and above (n = 497, 44.6%). around 15.3% (n = 183) of the participants reported that they have gone to crowded places recently. besides, around 50.1% (n = 605) of the participants reported that they wore mask when they leave their home. . cc-by-nc-nd 4.0 international license it is made available under a author/funder, who has granted medrxiv a license to display the preprint in perpetuity. is the (which was not peer-reviewed) the copyright holder for this preprint . https://doi.org/10.1101/2020.04.13.20063560 doi: medrxiv preprint . cc-by-nc-nd 4.0 international license it is made available under a author/funder, who has granted medrxiv a license to display the preprint in perpetuity. is the (which was not peer-reviewed) the copyright holder for this preprint . https://doi.org/10.1101/2020.04.13.20063560 doi: medrxiv preprint table 3 below details the correct answer rates of the 12 questions on the covid-19 knowledge questionnaire. participants had the highest rates of correct answers for questions related to the necessity of isolation and some procedures related to prevention and control of the infection (k10 -k12). on the other hand, the lowest rates of correct answers were for questions related to susceptibility of children and young adults to covid-19, transferability of the infection from asymptomatic patients, and eating habits related to covid-19 transmission (k5, k6, and k9). . cc-by-nc-nd 4.0 international license it is made available under a author/funder, who has granted medrxiv a license to display the preprint in perpetuity. is the (which was not peer-reviewed) the copyright holder for this preprint . https://doi.org/10.1101/2020.04.13.20063560 doi: medrxiv preprint table 4 below presents participants' demographics data and their covid-19 knowledge scores. participants' knowledge scores significantly differed by country, age, marital status, education level, and whether they wear mask upon leaving home or not (p<0.05). multiple linear regression analysis showed that individuals who had a bachelor degree or higher education level had a greater knowledge scores (p<0.01). on the other hand, divorced individuals, university students, individuals with an income 1,408 $ and above had a lower knowledge score (p<0.05) table 5 . . cc-by-nc-nd 4.0 international license it is made available under a author/funder, who has granted medrxiv a license to display the preprint in perpetuity. is the (which was not peer-reviewed) the copyright holder for this preprint . https://doi.org/10.1101/2020.04.13.20063560 doi: medrxiv preprint to the best of our knowledge, this is the first study that assesses the knowledge and practices towards covid-19 among the arabic speaking middle eastern countries. the results revealed that participants are still embracing misconceptions about covid-19, resulting in an insufficient practice of protective measures against covid-19 infection. however, some of those individuals who did not adhere to protective measures could be the one who were in duties that necessitate being in a crowded places such as port employees and working in the healthcare sector. the findings of this study . cc-by-nc-nd 4.0 international license it is made available under a author/funder, who has granted medrxiv a license to display the preprint in perpetuity. is the (which was not peer-reviewed) the copyright holder for this preprint . revealed that the overall knowledge score among the three countries was around 66.1% (score 7.93 out of 12) with the highest was among jordanians with a mean score of (8.44 (±1.46); 70.3%) indicate an average level of knowledge about the pandemic. among the three countries, the participants achieved the lowest knowledge score in the "transmission routes" sub-scale with a score of 43.3%. whereas the highest knowledge score was recorded on the "preventive and control measures" subscale with a score of 83.0%. overall, the total knowledge score (66.1%) was moderate, which was unexpected as this epidemiological survey was conducted during the advanced stage of the pandemic. further analysis of the findings was directed towards determining demographic factors associated with knowledge gaps among participants. such findings will be valuable for public-health policymakers and healthcare professionals to recognize target populations for health education activities on the covid-19 outbreak. this is critical as the countries involved in the study are still at the beginning of the outbreak and governments took extreme measures to contain the pandemic. but such measures, which include imposing a curfew, are virtually useless unless were accompanied by responsible individuals' preventative practices (3) . in this study as well as in previous studies in china, us, and uk, an association has been noticed between particular demographic characteristics and knowledge, beliefs, and practices towards covid-19 (1, 15) . for example, in china, an association between non-adherence to preventive and control measures and male gender, occupation of "students", marital status of "others", residing in other parts of china, and poor covid-19 knowledge has been reported (1). those findings were supported by previous studies regarding age and gender patterns of risk-taking behaviours (16, 17) . in the us and uk, geldsetzer (2020) has . cc-by-nc-nd 4.0 international license it is made available under a author/funder, who has granted medrxiv a license to display the preprint in perpetuity. is the (which was not peer-reviewed) the copyright holder for this preprint . https://doi.org/10.1101/2020.04.13.20063560 doi: medrxiv preprint reported lower fatality rates due to covid-19 infection among individuals of east-asian ethnicity and children (15) . in the current study, there was a significant difference between knowledge scores in the three countries included in the analyses, and the highest knowledge score among the participants has been found in jordan. both males and females showed similar levels of knowledge about covidhigher covid-19 knowledge scores were found to be significantly associated with age and educational attainment, which is in line with the study conducted in china during the covid-19 pandemic (1). older individuals above the age of 29 years showed a significant increase in knowledge scores than younger ones. similarly, an appreciable increase in knowledge score with the increase in education level was noted. this could suggest that the contents and forms of health-related information about the pandemic could not be understandable and acceptable to young adults and the less educated individuals (12). our findings are consistent with several previously published studies that related age and the level of education with knowledge and awareness about outbreaks (18, 19) . furthermore, several studies reported that the lack of knowledge contributed to the emergence and spread of the outbreak. therefore, providing targeted health education programs to raise awareness is vital. such programs need to be tailored to young individuals and particularly those with lower education levels (19) (20) (21) . furthermore, a significant negative effect on the knowledge score was the employment status, specifically among students (p<0.05). the results are aligned to the finding that lack of knowledge is related to young people (18 -29 years) . such findings are alarming, as the lack of knowledge about covid-19 is correlated with a higher prevalence rate of the infection (1, 22) . therefore, there is a need for improving students' knowledge about covid-19 by means of health education, which may also result in improvements in their practices towards covid-19. such results could support the decision made by many countries including jordan, kuwait, and saudi arabia in taking a political commitment to temporally close all educational institutions to contain the spread of the covid-19 pandemic. globally, the action involved 188 countries, affecting around 1.5 billion learners. the mode of teaching moved into emergency remote learning (22, 23) . yet, although the closure of schools and educational facilities is a good mean to prevent young adults from mixing with others, it will not prevent exposure even with curfew laws. young adults will always go out and get exposed to the risk of getting infected as the level of knowledge is low. a recent study on covid-19 concluded that educational institutions closures alone would preclude only 2-4% of deaths, which is a lot less than other social distancing interventions. the integration of additional social distancing strategies along with educational institutions closures needs to be considered (24) . furthermore, the results of our study pertinent to knowledge scores stratification according to demographics are in line with studies on covid-19 in china and severe acute respiratory syndrome (sars) outbreaks among . cc-by-nc-nd 4.0 international license it is made available under a author/funder, who has granted medrxiv a license to display the preprint in perpetuity. is the (which was not peer-reviewed) the copyright holder for this preprint . https://doi.org/10.1101/2020.04.13.20063560 doi: medrxiv preprint others (1, 25) . these findings also proposed that health education strategies would be more effective if it was designed to targets certain demographic groups, such as university students. good knowledge is vital to enable individuals to have better practices in pandemics and outbreaks. a study on sars demonstrated that a higher knowledge score was found to be connected with improved adherence to precautionary practices (11, 26) . of interest, an income of ≥ 1,409 $ (p<0.05) had a significant negative association with the knowledge score in our cohort, showing that high income does not improve knowledge and practices. however, this is contrary to a previous study on the influenza pandemic, where there was no association between knowledge and income (26). a possible reason for such result is that the higher income of individuals is accompanied with low education level. the current study showed that knowledge scores significantly differed across marital status categories (p<0.002), divorced participants had a significant negative association with covid-19 knowledge score (p<0.01). our findings were in line with the results obtained from the study on covid-19 in china, which revealed a statistically significant difference in knowledge score according to marital status, however, they found that the category denotes as others (included divorced and widowed, separated, remarried) showing the highest score in knowledge (11.0 out of 12.0) (91.7%), p <0.001) (1). this difference could be justified by two elements. firstly, different countries might have different cultural dimension and hence results could not be compared. secondly, the group used in the chinese study included all divorced, separated, widowed and remarried individuals and hence we could not build a good correlation (1). overall, knowledge scores within our cohort were significantly associated with a higher likelihood of dangerous practices towards covid-19 pandemic. such findings indicate the presence of knowledge gap and the need to improve individuals' covid-19 knowledge via extensive health education strategies. such strategies are to be tailored and targeted towards specific demographic groups particularly, university students, individuals with low education, higher income and divorced. to the best of our knowledge this is the first study in the arabic speaking middle eastern countries that investigated the public knowledge and practices during covid-19 pandemic. a strength of the study is that a large sample of participants were recruited during this critical period -the actual covid-19 outbreak -and hence replies reflect the actual status. the participants were from three countries namely, jordan, saudi arabia and kuwait, which increases the generalisability of these findings. additionally, the use of previously used assessment tool that allowed comparison with other population was another strength of the study. however, there are some limitations. the study design itself, a cross-sectional survey design, limited our ability to identify causality between study variables. there are limited studies that assessed knowledge and practices of individuals during covid-19 pandemic worldwide and in the middle east specifically, which limited our ability to compare our . cc-by-nc-nd 4.0 international license it is made available under a author/funder, who has granted medrxiv a license to display the preprint in perpetuity. is the (which was not peer-reviewed) the copyright holder for this preprint . https://doi.org/10.1101/2020.04.13.20063560 doi: medrxiv preprint findings with arabic-speaking countries of a similar socioeconomic level and culture. in this study we employed a quantitative methodology with pre-set responses, which might not have allowed participants' views to provide varied but useful qualitative information. in addition, we used an online survey for data collection and therefore, some vulnerable populations within the three countries under the covid-19 pandemic could not be reached and we may have missed some of the targeted population. however, we tackled this by distributing the survey among three different countries and widely used the social media. finally, we were not able to estimate the response rate for our questionnaire study. findings of this study suggest that middle eastern participants are of a relatively low level of knowledge about covid-19, particularly regarding its transmission routes. the majority of the population showed appropriate practice by avoiding crowded places and staying home during the rapid rise period of the covid-19 outbreak. however, not wearing a mask when leaving home was predominant. as good covid-19 knowledge is associated with optimistic attitudes and appropriate practices towards covid-19, policymakers are in a position to develop targeted in information campaigns provided to targeted population specifically university students, unemployed individuals and those with lower levels of education. this information could be conveyed to people by clinicians to their patients, and news coverage supplied by the media and social media platforms. . cc-by-nc-nd 4.0 international license it is made available under a author/funder, who has granted medrxiv a license to display the preprint in perpetuity. is the (which was not peer-reviewed) the copyright holder for this preprint . https://doi.org/10.1101/2020.04.13.20063560 doi: medrxiv preprint knowledge, attitudes, and practices towards covid-19 among chinese residents during the rapid rise period of the covid-19 outbreak: a quick online cross-sectional survey corona crisis shocked the world and revealed surprises 2020 hard times shape speedy saudi and kuwaiti coronavirus response a fast-moving virus pits treating patients against finding a cure bloomberg2020 fear and stigma: the epidemic within the sars outbreak how to fight an infodemic communicable diseases network australia outbreaks in residential care facilities world health organization. covid 19 public health emergency of international concern (pheic) global research and innovation forum : towards a research roadmap a tale of two cities: community psychobehavioral surveillance in hong kong and singapore during the severe acute respiratory syndrome epidemic national health commission of the people's republic of china. a protocol for community prevention and control of the 2019 novel coronavirus (2019-ncov) infected pneumonia (trial version) knowledge and perceptions of coronavirus disease 2019 among the general public in the united states and the united kingdom age patterns in risk taking across the world sex differences in everyday risk-taking behavior in humans australian hajj pilgrims' knowledge, attitude and perception about ebola assessing the knowledge, attitudes, and practices of students regarding ebola virus disease outbreak global and domestic legal preparedness and response assessing knowledge and attitude about ebola in the us: a cross sectional survey. value heal covid-19 educational disruption and response school closure and management practices during coronavirus outbreaks including covid-19: a rapid systematic review. the lancet child & adolescent health survey of knowledge of villagers in prevention and control of sars in this study was supported by isra university (amman, jordan) and najran university and cancer society in najran, saudi arabia. the authors declare no conflict of interest.author contributions an conceived the study, wrote the methods, conducted the formal analysis, and coordinated the study. an, ezd, za, and aa drafted the manuscript with input from all authors. all authors have been involved in drafting the manuscript or revising it critically for important intellectual content. all authors read and approved the final manuscript. the data that support the findings of this study are available from the corresponding author upon reasonable request. key: cord-311264-zn7ydrvh authors: deurenberg-yap, m.; foo, l. l.; low, y. y.; chan, s. p.; vijaya, k.; lee, m. title: the singaporean response to the sars outbreak: knowledge sufficiency versus public trust date: 2005-06-17 journal: health promot int doi: 10.1093/heapro/dai010 sha: doc_id: 311264 cord_uid: zn7ydrvh during the outbreak of severe acute respiratory syndrome (sars) in singapore from 1 march to 11 may 2003, various national prevention and control measures were undertaken to control and eliminate the transmission of the infection. during the initial period of the epidemic, public communication was effected through press releases and media coverage of the epidemic. about a month into the epidemic, a public education campaign was mounted to educate singaporeans on sars and adoption of appropriate behaviours to prevent the spread of the disease. a survey was conducted in late april 2003 to assess singaporeans' knowledge about sars and infection control measures, and their concerns and anxiety in relation to the outbreak. the survey also sought to assess their confidence in the ability of various institutions to deal with sars and their opinion on the seemingly tough measures enforced. the study involved 853 adults selected from a telephone-sampling frame. stratified sampling was used to ensure adequate representation from major ethnic groups and age groups. the study showed that the overall knowledge about sars and control measures undertaken was low (mean per cent score of 24.5 ± 8.9%). while 82% of respondents expressed confidence in measures undertaken by tan tock seng hospital (the hospital designated to manage sars), only 36% had confidence in nursing homes. however, >80% of the public agreed that the preventive and control measures instituted were appropriate. despite the low knowledge score, the overall mean satisfaction score of the government's response to sars was 4.47 (out of possible highest score of 5.00), with >93% of adult singaporeans indicating that they were satisfied or very satisfied with the government's response to sars. generally, singaporeans had a high level of public trust (satisfaction with government, confidence in institutions, deeming government measures appropriate), scoring 11.4 out of possible maximum of 14. the disparity between low knowledge on the one hand and high confidence and trust in the actions of the government on the other suggests that singaporeans do not require high knowledge sufficiency to be confident in measures undertaken by the government to control the sars crisis. severe acute respiratory syndrome (sars) is the first severe and easily transmissible new disease to emerge in the 21st century (world health organization communicable disease surveillance and response, 2003). between 1 march and 11 may 2003, singapore experienced the harshness of this outbreak, and the country went into high alert once it was apparent that it was battling an unknown new infection capable of spreading rapidly in hospitals. the national prevention and control strategy for sars focused on: (i) eliminating nosocomial transmission through substantially enhanced infection-control practices in health care institutions; (ii) preventing additional importations of infection through temperature screening, health (singapore ministry of health, 2003) . during the initial stages of the epidemic, public communication was effected mainly through press conferences, media coverage and a website set up by the ministry of health. prominent press and radio advertisements were also placed in the early phase of the epidemic from mid april 2003. as more became known about the infection, and with control measures being put in place, an intensive public education campaign utilizing multiple channels was mounted between 30 april and 13 may 2003. this public education campaign aimed to educate singapore residents about sars, to encourage them to adopt appropriate personal hygiene and socially responsible habits to prevent the spread of sars, to encourage those with suspected sars infection to seek medical attention at ttsh, and to encourage cooperation with contact tracing and home quarantine orders (hqo) when necessary. in a heuristic-systematic model (griffin et al., 1999) , the assumption is that relevance of the message is likely to motivate subjects in 'risky and severe' condition to process risk information more systematically and to rely less on superficial cues in the structure and style of the message. in this paper, the informing seeking and processing mindset of singaporeans during a severe outbreak situation is assessed by testing the level of knowledge on sars and its preventive/ control measures following the earlier communication efforts and subsequent public education campaign. it seeks to understand the dynamics between knowledge sufficiency and public trust in infection containment measures instituted by the state, and how these in turn affect the general satisfaction with the handling of the sars outbreak by the state. a telephone survey was conducted from 30 april to 13 may 2003. the sampling frame consisted of telephone numbers and personal particulars (name, gender and ethnicity) in singapore. because of the urgent nature of the survey, the sampling frame was stratified by ethnicity and gender, and quota sampling carried out for each stratum. a total of 863 adults aged 19-69 years were interviewed. there was an over sampling of indians and malays to ensure that there were enough subjects from the three major ethnic groups (of which chinese form the majority) in singapore (distribution in the general population for chinese, malays and indians is about 76.8, 13.9 and 7.9%, respectively). the results were weighted to reflect the actual distribution in the population during analysis. the study conformed to the principles embodied in the declaration of helsinki (world medical association, 1996) and was approved by the ministry of health, singapore. knowledge on sars (symptoms, spread, protective measures, treatment) and infection control measures (measures at ttsh, other health care institutions, immigration check points, special ambulance arrangement for suspects), confidence in institutions to deal with the outbreak, and opinions about measures implemented by the government were assessed by a pre-tested questionnaire (table 1 ). in addition, respondents were asked to score their satisfaction level for the measures undertaken by the government. a per cent knowledge score was computed for the knowledge items. an index of 'public trust' was computed using a composite index comprising: (i) an index for overall confidence in ability of public institutions (ttsh, other government hospitals/institutions, private hospitals/institutions, nursing homes, schools, workplaces and the transportation system) to deal with the crisis; (ii) an index for assessing public perception of appropriateness of sars control measures (home quarantine order, punitive measures for quarantine breakers, infection control measures in hospitals and closing down of affected public places); and (iii) public satisfaction towards the response of the government to the outbreak. the data were analysed using spss version 11.5 for windows (spss inc., 2001) . demographic differences in knowledge of sars were assessed using the kruskal-wallis test (siegel, 2000; spss inc., 2001) . stepwise multiple linear regression (spss inc., 2001; gujarati, 2003) was carried out to assess the impact of demographic variables and overall knowledge about sars on , 2001; gujarati, 2003) , cook-weisberg test (spss inc., 2001; gujarati, 2003) and normality plot (spss inc., 2001; gujarati, 2003) were applied to check for multi-collinearity, heteroscedasticity and non-normality, respectively. all statistical tests were conducted at the 5% significance level. singaporean response to sars outbreak 323 of the 853 respondents covered in this paper, 458 were females and 395 were males ( table 2 ). the mean age of the respondents was 41 years (range: 19-81 years), with no differences between males and females. in terms of educational attainment, significantly more males had tertiary education (36 versus 26%), while more women had primary only or no education (38 versus 28%). the mean ± standard deviation per cent score for knowledge of sars was 26 ± 8%. of the four key domains under knowledge of sars, knowledge about treatment of sars was highest (mean score of 62 ± 24%) and knowledge about the protective measures against sars was lowest (mean score of 17 ± 11%). while 58% of the respondents were able to cite personal hygiene as a protective measure against sars, comparatively fewer of them mentioned temperature taking on a daily basis (21%) and wearing of a mask (23%). the mean per cent score for knowledge about the symptoms of sars was 40 ± 15%. most subjects (94%) were able to cite fever as a symptom, a key message of the campaign. generally, the indian respondents had a significantly higher mean per cent score for knowledge of sars (29 ± 10%) compared with the chinese (25 ± 8%) and malays (26 ± 8%). younger respondents fared significantly better compared with those aged у50 years (27 ± 8% for those below 49 years of age versus 24 ± 9% for those aged 50 years and above). likewise, respondents with tertiary education had significantly higher knowledge scores (29 ± 7%) compared with those with primary or no formal education (23 ± 9%). no gender difference was found. the mean per cent score for knowledge of sars infection control measures was 22 ± 13%. of the four key knowledge domains, knowledge about precautionary measures for air and sea passengers (mean score of 30 ± 21%) and types of people who needed special ambulance arrangement (mean score of 29 ± 16%) was higher. knowledge about the sars control measures implemented at ttsh and other health care institutions lagged behind (mean scores of 20 ± 16% and 17 ± 14%, respectively). the indians and those with higher educational attainment had significantly higher scores for knowledge about the control measures. age and gender did not have a significant impact in this area. the mean per cent score for overall knowledge was 25 ± 9%. knowledge of sars was slightly higher compared with knowledge about infection control measures. indian respondents performed significantly better than their chinese and malay counterparts. younger respondents had higher knowledge scores compared with those aged 50 years and above, but the difference was not significant. on the other hand, respondents with tertiary education fared significantly better than those with lower educational attainment. respondents were also asked to rate their confidence in the ability of various institutions in dealing with sars. the proportion of respondents citing confidence in the ability of ttsh to deal with sars was highest (82%) followed by confidence in schools (63%). confidence in the ability of nursing homes to handle sars was lowest (36%). the proportions of respondents citing confidence in restructured hospitals/institutions, private hospitals/institutions, workplaces and transport system were comparable, ranging from 48 to 56%. males were more likely to express confidence in the ability of these institutions to deal with sars. a similar pattern was seen among the older respondents. conversely, the proportion of respondents with tertiary education expressing confidence in these institutions was smaller compared with their counterparts with lower educational attainment, with confidence in ttsh being the only exception. the majority of the respondents, regardless of gender, race, age and educational attainment, found the sars control measures undertaken by the government appropriate. more than nine out of 10 respondents thought that infection control measures undertaken at hospitals were overall, the public trust index was high at 11.4 out of a maximum score of 14, with no significant difference between gender, age groups and educational levels. there was a slight but significant correlation with knowledge of sars (r = 0.23, p ͻ 0.001), knowledge of control measures (r = 0.18, p ͻ 0.001) and overall knowledge score (r = 0.23, p ͻ 0.001). a stepwise multiple linear regression model to predict public trust showed that knowledge scores for sars was significantly associated with a higher public trust index, controlling for other variables. among the demographic variables, only age turned out to be a significant predictor. in total, the covariates explained slightly ͼ5% of variations in the public trust index (table 3 ). the study shows that the overall knowledge about sars and control measures undertaken to control the virus was not high (mean per cent score of 25%). respondents' confidence in the ability of various institutions' ability to deal with sars were varied, with confidence in ttsh at one end of the spectrum (82% of the respondents expressed confidence in the hospital's ability to handle sars) and nursing homes at the other end (36% cited confidence in the ability of nursing homes to deal with the virus). respondents' perceptions about the appropriateness of sars control measures were high, with the majority of the respondents endorsing measures undertaken as being appropriate. while a regression analysis indicated the trend that higher knowledge of sars was significantly associated with higher level of public trust, it explained only ~5% of the variation for public trust. knowledge about sars control measures did not contribute significantly to the level of public trust. in addition, the study reveals some rather unexpected findings upon further analysis. first, while knowledge about infection control measures undertaken at ttsh was low (mean per cent score of 20 ± 16%), the level of confidence was high, with 82% of the respondents expressing confidence in the hospital's ability to deal with sars. secondly, despite the low knowledge level of the public about infection control measures undertaken at ttsh and other healthcare institutions (mean per cent score of 17 ± 14%), 91% of the respondents felt that these measures were appropriate. the study also found while knowledge about sars and infection control measures were lowest among those with primary education or below, they were more likely to express confidence in the ability of institutions' ability to handle sars. there is thus an apparent contradiction between low knowledge about sars and infection control measures yet high level of public trust in the government and their actions. several possible explanations can be given for this phenomenon. the model of risk information seeking and processing (griffin et al., 1999) seeks to understand how individuals respond to messages about health risks. the model adapted and synthesised theories from the heuristic systematic model of information processing (eagly and chaiken, 1993) and proposes that people who engage in more effortful information seeking and processing are more likely to develop risk-related cognitions, attitudes and behaviours that are more stable over time. some of the key concepts proposed by the model, information sufficiency, perceived hazard characteristics and social trust, provide useful theoretical grounding to facilitate understanding the responses of the survey respondents to the sars crisis. 'information sufficiency' is the amount of information people say they need in order to deal adequately with a given risk in their own lives (griffin et al., 1999) . information would be pursued and processed until perceived knowledge reaches the sufficiency threshold, which would represent the point at which the individual is confident that he or she can cope behaviourally with the risk (e.g. take effective preventive action). however, people might avoid risk content if, for example, it produces worries with which they cannot cope. some may not pay attention or simply avoid such information all together. this might explain why while basic knowledge about sars (e.g. awareness that fever is a symptom of sars) was high, the overall knowledge score remained low in the study as most singaporeans may feel that basic knowledge about sars allows them to cope adequately with the crisis at a personal behavioural level. 'perceived hazard characteristics' reflect the degree to which a risk is understood and the degree to which it evokes a 'feeling of dread' (griffin et al., 1999) . some studies have found that the more dreaded the hazard, the more people want strict regulations employed to reduce its risks (eagly and chaiken, 1993; geogory and mendelsohn, 1993) . the fear of sars was widespread, causing stress and anxiety wherever it struck (clark, 2003; cong et al., 2003; falsey and walsh, 2003; lee et al., 2003; wenzel and edmond, 2003) . this can explain why most singaporeans endorsed the sars control measures implemented in singapore as being appropriate even though there was low knowledge sufficiency. the concept of 'social trust' also offers useful insights. researchers have identified social trust as a key mediating factor in circumstances requiring collective action (slovic, 1992; earle and cvetkovich, 1994) . social trust serves as a cognitive heuristic tool that decreases the complexities of social life to workable levels. when people face threatening events or conditions, they often make risk judgements based on social trust, an expectation that assigns to others the responsibility for working on some necessary task (slovic, 1992; earle and cvetkovich, 1994) . it was clear that social trust was rampant during the sars crisis in singapore. singapore has sometimes been described as having a communitarian ideology (chua, 1995) . state interventions in social life are often viewed as pre-emptive measures for ensuring the collective well-being or as measures of good government. during the sars crisis, the government actively promoted public confidence in the ability of ttsh to cope with sars. several leaders of the country shared their personal views about sars and expressed confidence in the ability of the hospital to handle the crisis. the 'social norm' (fishbern and ajzen, 1975; ajzen and fishbein, 1980) of accepting ttsh as being capable of dealing with sars was further enforced when individuals who did not admit themselves into the hospital when suspected of sars were publicly criticised. as people placed increasing trust in the government, institutions and hospitals to manage sars, it could have led them to experience less personal worry and perceived vulnerability to sars. given the situation that little was known about sars and that there was constant appeal for people to live life as normally as possible, people tried to manage their lives by reducing complexities through increased social trust, which also implied an expectation on the government to take over the responsibility for fighting sars. consequently, they developed high confidence in the measures undertaken by the government, despite having a low level of knowledge about sars. in summary, the low level of knowledge about sars and infection control measures may indicate that the information sufficiency threshold is low, meaning singaporeans need very little information in order to feel confident to cope with sars or they do not see the need to know all the control measures before feeling confident with what the government is doing to handle the sars crisis. this is no doubt mediated by the high perceived hazard associated with sars and the high social trust ascribed to the government. taken together, these concepts help to explain the apparent disparity between the low knowledge about sars and the control measures on the one hand, and high confidence in key institutions dealing with sars, high endorsement about sars control measures being appropriate and high satisfaction with the government's response to sars on the other hand. in conclusion, some useful lessons can be gleaned from the study. the low knowledge level but high level of trust implies that public compliance is high. while public compliance may be high, it may not necessarily translate to private conviction and conversion about the importance of adopting appropriate preventive measures. however, even though private conversion may be the best societal control in the long-term, high-risk situations require quick short-term measures (asch, 1955; moscovici, 1985) , and those taken for sars were effective and appropriate for singapore. taking into consideration the cultural and societal background, instilling public compliance seemed to be an efficient tool of helping to cope with the immediate sars crisis. moving forward, it is important that future educational efforts promote private conversion and personal responsibility, utilising veracious and creative messages and taking into consideration different situations within the social and cultural context. understanding attitudes and predicting behaviour opinions and social pressures communitarian ideology and democracy mental stress and crisis intervention in the patients with sars and the people related fear of sars thwarts medical education in toronto the psychology of attitudes. harcourt brace risk communication: the social construction of meaning and trust. in future risks and risk management novel coronavirus and severe acute respiratory syndrome perceived risk, dread and benefits proposed model of the relationship of risk information seeking and processing to the development of preventive behaviours basic econometrics a major outbreak of severe acute respiratory syndrome in hong kong social influence and conformity nonparametric statistics for the behavioural sciences chronology of sars events in singapore. ministry of health perception of risk: reflections on the psychometric paradigm spss manuals, version 11.5 managing sars amidst uncertainty world health organization communicable disease surveillance and response (2003) severe acute respiratory syndrome (sars)-multi-country outbreak ethical principles of medical research involving human subjects we would like to express our appreciation to dr jeffrey lake for his comments on the questionnaire and to dr paul deurenberg for guiding us in drafting the manuscript. we would also like to acknowledge the ministry of health, singapore, for funding the study. key: cord-333015-xgudk2h0 authors: della polla, giorgia; pelullo, concetta paola; napolitano, francesco; lambiase, chiara; de simone, caterina; angelillo, italo francesco title: knowledge, attitudes, and practices towards infectious diseases related to travel of community pharmacists in italy date: 2020-03-24 journal: int j environ res public health doi: 10.3390/ijerph17062147 sha: doc_id: 333015 cord_uid: xgudk2h0 pharmacists should be educated about travel medicine, since they could influence their own choices and those of the individuals they encounter. this study aims to investigate the knowledge, attitudes, and behaviors towards infectious diseases related to travel among community pharmacists in italy. the data was collected from september 2018 to september 2019 using semi-structured telephone interviews. only 1.8% answered correctly to all seven questions regarding the infectious diseases related to travel. community pharmacists who had heard about travel medicine and those who had received information were more likely to have good knowledge. more than two-thirds of the respondents believed that it is important to provide information to the public about travel medicine. pharmacists who worked a higher number of hours per week, were more knowledgeable about the more frequent infectious diseases related to travel, believed that travel medicine was a pharmacist competency, believed that they could give advice to the public, and had received information from scientific journals and educational activities were more likely to have this positive attitude. more than two-thirds often/always informed the public about the importance of having travel health center counseling. pharmacists who had heard about travel medicine and those who believed that they could give advice to the public were more likely to inform. interventions are needed to improve knowledge in order that community pharmacists can play an active role in counseling the public. during the past decade, the number of international travelers to many destinations in various geographic regions has steadily risen leading to an increased risk in travel-related health problems with a subsequent impact on international public health and an increased potential demand for health services [1, 2] . although the destination can be one important risk factor for certain health problems, many travelers are unaware and often fail to seek appropriate preventative pre-travel care. healthcare professionals have a central role in providing recommendations on travel diseases. in particular, primary care providers should be familiar with destination-specific disease risks, be knowledgeable in pre-travel health advice in order to reduce the risks for travelers, be prepared to prescribe and recommend medications for treatment that can be taken during the trip, and be able to identify those who might need additional follow-up after the trip [3] . among healthcare professionals, pharmacists potentially have an important role in travel medicine within the community, and they are able to provide useful insights. in italy, community pharmacists have a primary role also in providing recommendations to patients on the use of the healthcare services, and they can sell over-the-counter drugs to the public without medical counseling. indeed, pharmacists can be one of the most common sources of healthcare information as they are closer to home and individuals often seek their services directly without consulting traveler medical centers. therefore, pharmacists should be highly educated about travel medicine, since they could influence their own choices and those of the individuals they encounter. several investigations on the knowledge, attitudes, and behavior of the general population and physicians regarding travel medicine have been conducted in different countries [4] [5] [6] , and, to date, there is also research on the involvement of community pharmacists in this field [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] . this is of utmost importance in order to provide important information for planning training interventions targeted towards these professionals to positively affect travelers' health. therefore, to enhance the existing information, the current study investigated the level of knowledge, attitudes, and behaviors towards infectious diseases related to travel and their influencing factors among community pharmacists in italy. the study was one of the key components of a larger national survey and a detailed description of the procedure has been published previously [17] and is described in brief below. this cross-sectional survey was conducted from september 2018 to september 2019, enrolling a random sample of 550 community pharmacists derived from the national list of those practicing in italy. for the sample size calculation, the prevalence of respondents who believe that it is important to provide information to the public about infectious diseases related to travel was assumed to be 70%, with a significance level of 5%, a margin of error of 5%, and to account for a response rate of 70%. a sample size of 461 participants was calculated. five trained and experienced interviewers initially contacted every selected community pharmacy by phone, and the pharmacist-in-charge or pharmacist manager were invited to participate in the survey. data were collected from the individuals who responded. at the beginning of each telephone call, participants were informed about the purposes and confidentiality of the survey and data protection, that participation was voluntary, that they could not skip questions and whole sections, but they were free to terminate the call whenever they wished without any consequences. the selected person was able to indicate whether he/she wished to participate in the survey. all participants expressed their verbal informed consent to their inclusion before initiating the interview. to improve the response rate, additional telephone reminder calls with each of the pharmacists that were not able to complete the survey the first time were made every week. at least three additional calls were made at different times and days to reach a pharmacist before he/she was considered as a non-responder. survey participation was not compensated. structured telephone interviews were designed to collect information divided into five sections. the first part consisted of the participants' socio-demographic and professional characteristics and included questions regarding their age, gender, marital status, year of graduation, number of years in practice, and practice size. the second part consisted of questions related to the pharmacists' knowledge about infectious diseases related to travel. fourteen main infectious diseases selected according to the most frequently visited countries by the italian population [5, [18] [19] [20] [21] were listed and pharmacists had to self-evaluate their knowledge giving their answers by choosing one of the three options: "yes", "no", or "don't know". the scale was then dichotomized ("yes" = 1 and "no"/"don't know" = 0) and the total knowledge score for each participant was computed by adding up the scores (maximum score of 14). the total score was then categorized as poor knowledge (score of ≤2), moderate knowledge (score of 3-4), and good knowledge (score ≥5). the third part assessed participants' attitudes with three statements on the importance of travel medicine for their work activity, asking whether they agreed that travel medicine is a pharmacist competency and whether they should give advice regarding travel medicine as part of their work activity, with response options on a 5-level likert scale ranging from "strongly disagree" to "strongly agree", and whether they believed that it is important to provide information to the public about travel medicine, with the response on a numerical 10-point likert scale with higher values corresponding to a stronger attitude. the fourth part collected information about their practice and communication with the public regarding travel medicine. three responses were collected on a 5-level likert scale ranging from "never" to "always" and the others through multiple-choice alternatives. the fifth part included two questions on whether respondents had received information about travel medicine and whether they had educational needs. a pilot study for clarity, length, and simplicity was conducted on a total of 20 interviews and some minor wording changes were made in the revised version. these 20 pilot study interviews were not included in the main study. then, the final protocol, including the informed consent form, and the questionnaire were approved by the ethics committee of the teaching hospital of the university of campania "luigi vanvitelli". the statistical analysis was conducted using stata statistical software, version 15 (statacorp., college station, tx, usa) [22] . first, descriptive analyses were performed to assess all characteristics of the participants. second, chi-square and student's t-test were conducted to examine the relationship between the independent variables and the outcomes of interest. third, multivariate ordered, linear, and logistic regression analysis was conducted by including in the models the variables with a p-value ≤ 0.25 at the bivariate analysis in order to estimate the independent association between potential predictors and the outcomes of interest. the following three models were developed: (1) pharmacists' level of knowledge about the more frequent infectious diseases related to travel (poor = 1; moderate = 2; good = 3) (model 1); (2) pharmacists who believed that it is important to provide information to the public about travel medicine (continuous) (model 2); (3) pharmacists who often or always inform the public about the importance of having travel health center counseling (no = 0; yes = 1) (model 3). initial candidate variables included in all models were age (continuous), gender (male = 0; female = 1), marital status (unmarried/separated/divorced/widowed = 0; married = 1), number of years since degree (continuous), number of years in practice (continuous), number of hours worked per week (continuous), employment type (owner = 1; employee = 2; director = 3), having heard about travel medicine (no = 0; yes = 1), sources of information on travel medicine (none = 1; scientific journals and educational activities = 2; internet and mass media = 3), and need of additional information on travel medicine (no = 0; yes = 1). moreover, the following variables were also included: pharmacists' level of knowledge about the more frequent infectious diseases related to travel (poor = 1; moderate = 2; good = 3), pharmacists who believed that travel medicine is a pharmacist competency (strongly disagree/disagree/uncertain = 0; agree/strongly agree = 1), pharmacists who believed that they could give advice to the public about travel medicine (strongly disagree/disagree/uncertain = 0; agree/strongly agree = 1) in models 2 and 3, and pharmacists who believed that it is important to provide information to the public about travel medicine (continuous) in model 3. a stepwise backward procedure was used to select which variables to include in the final models and in order to find the most parsimonious models. p-values of 0.2 and 0.4 were considered as a threshold to include and to eliminate variables. odds ratios (ors) and their respective 95% confidence intervals (cis) were calculated using ordered and logistic regression. standardized regression coefficients (β) were presented for the linear regression models. all tests were two-sided with p-values less than or equal to 0.05 considered statistically significant. among the 550 community pharmacists who were approached for the study, a total of 390 agreed to participate, resulting in a response rate of 70.9%. the socio-demographic and professional characteristics of the community pharmacists who participated in the survey are presented in table 1 . the majority of pharmacists were female (59.9%), the average age was 47.9 years, half of the sample were pharmacy owners (50.5%), the average number of years in practice was 18.1, and the mean number of hours worked per week was 41.2. number for each item may not add up to total number of study population due to missing values; * mean ± standard deviation (range). regarding the level of knowledge, an overwhelming majority of respondents were aware of travel medicine (85.4%). notably, the highest achieved score on the knowledge of infectious diseases related to travel was 7 out of a maximum score of 14, and only 1.8% provided correct answers to seven questions and were aware of none of these diseases. travel diarrhea (91.9%), hepatitis a (44.5%), malaria (44.3%), and cholera (36.5%) were the more recurrent diseases known by pharmacists. for all participants, the total score for knowledge of infectious diseases related to travel ranged between 0 and 7, with a mean of 2.9. associations between the different outcomes of interest and potential predictor variables using multivariable linear and logistic regression analyses are shown in table 2 . from the initial model and after the stepwise backward procedure, the final ordered logistic regression model with the outcome the score of the knowledge about the infectious diseases related to travel comprised two variables: level of knowledge and sources of information about travel medicine. pharmacists who had heard about travel medicine (or = 2.28; 95% ci = 1.26-4.11), having received information from scientific journals and educational activities (or = 5.57; 95% ci = 3.42-9.1) and from the internet and mass media (or = 3.96; 95% ci = 2.37-6.6), compared with those who did not receive information, were more likely to have a good knowledge about the more frequent infectious diseases related to travel (model 1). with regard to attitudes, 44.3% and 56.2% of the respondents agreed respectively with the statements that travel medicine is part of their professional responsibilities as healthcare workers and that they could properly give advice to the public on this topic. moreover, more than two-thirds of respondents (77.6%) believed that it is important to provide information to the public about travel medicine, with a mean value of 6.9, on a scale of 1 to 10. the results of the multivariate linear regression model, built to test the variables associated with this outcome of interest, showed that pharmacists who worked a higher number of hours for week, those who had a higher level of knowledge about the more frequent infectious diseases related to travel, those who believed that travel medicine is a pharmacists' competence, those who believed that they could give advice to the public about travel medicine, and those who had received information from scientific journals and educational activities compared with those who did not receive any information were more likely to believe that it is important to provide information to the public about travel medicine (model 2 in table 2 ). regarding the behaviors, more than one-third of the sample (37.2%) indicated that they sometimes receive requests for advice on travel medicine from the public, mainly regarding travel diarrhea (83.4%), safeness of food and water (77.6%), insect punctures (65.9%), and vaccinations (31.2%). moreover, more than half (52.1%) reported that the public often ask for medications for travel purposes without a prescription. the reasons for these requests included the inability to get a prescription in time (76.5%), advised information from the internet (47%), and unavailability of the physician (19.1%). more than two-thirds (69.5%) often or always informed the public about the importance of having travel health center counseling. multivariate logistic regression analysis showed that pharmacists who had heard about travel medicine (or = 6.44; 95% ci 3.2-12.97) and those who believed that they could appropriately give advice to the public about travel medicine (or = 3.61; 95% ci 1.9-6.84) were more likely to inform the public about the importance of having travel health center counseling (model 3 in table 2 ). among the survey participants, only two-thirds (65.6%) reported searching for information about travel medicine. when asked about the source of information, a higher number of respondents stated that the internet was the most frequently reported trusted source (52.9%), followed by scientific journals (27.9%), and educational activities (20.7%). the vast majority of responders (84.6%) felt they did not have sufficient information on travel medicine and expressed willingness to acquire more knowledge. the present study is to our knowledge the first and largest published national survey that has attempted to provide an overall picture about the knowledge, attitudes, and practices of community pharmacists in italy regarding infectious diseases related to travel and of the factors associated with these main outcomes of interest, adding to the existing literature by providing key findings. the study findings suggest that the participants had a knowledge gap in travel medicine and this situation is alarming as a basic level of knowledge about common infectious diseases is expected from healthcare staff. indeed, a striking finding depicted that less than a quarter of participants seemed knowledgeable about the seven more frequent infectious diseases related to travel, with a total mean score of knowledge of 2.9. the finding that respondents had a lack of knowledge is in accordance with previous reports, which have shown that community pharmacists have a knowledge gap and low confidence in providing this type of care [15, 23] . this fact emphasizes the importance that pharmacists should be informed since their level of awareness should primarily be improved. moreover, pharmacists who had a higher level of knowledge were far more likely to believe that it is important to provide information to the public about travel medicine and to inform them about the potential risks of travel diseases and the importance of having travel health center counseling before their departure. therefore, it is necessary that all pharmacists choose to improve their knowledge in order to offer travelers accurate and expert counselling. with respect to the attitudes towards travel medicine, this study indicated that 44.3% and 56.2% of the respondents agreed that travel medicine is part of their professional responsibilities as healthcare workers and that they could properly give advice to the public. a similar result was observed among malaysian community pharmacists who agreed that they are in a position to provide health information and recommendations regarding travel health [7] . this is a relevant finding because pharmacists should give advice to patients in order to guarantee the appropriate behaviors before and during their travels and they may be a key and trusted source of information. multivariate linear regression analysis showed that pharmacists who believed that knowledge about travel medicine is a part of their role were more likely to believe that it is important that they provide information to the public. this result can be explained by the fact that pharmacists who had this attitude were more likely to acquire information because they thought travel medicine was a key professional skill. therefore, policy makers and healthcare managers should implement programs that actively involve the pharmacists because pharmaceutical services are available throughout the territory and can easily be accessed, and this could lead to a greater adherence of the public to preventive measures and may contribute to patient safety and appropriate use of medications. concerning the practices for travelers, community pharmacists play a key role in the public's awareness of travel diseases and their recommendations may be an important determinant of travel health. in this study, 69.5% of the participants often or always informed the public about the importance of having a travel health center counselling session. this underlined the availability of the community pharmacists in providing information to the public regarding travel diseases. however, it is important to highlight that in italy pharmacists cannot dispense some medicines without a physician's prescription, and it should be considered that more than half reported that the public often asked for medications for imminent travel because their general practitioners were unavailable. this result is in line with those of a similar study conducted in australia that showed that pharmacists had a role in advising travelers, who would not normally visit a physician before travelling, on travel-related health issues before visiting certain destinations [24] . furthermore, the result of the logistic regression analysis showed that the positive attitude of the study participants about different health seeking behaviors predicted their real practices since those who believed that they could give advice to the public regarding travel medicine were more likely to often or always inform the public about the importance of having travel health center counselling than those who did not. the findings from this survey are consistent with several previous results from similar studies among different groups of individuals that have demonstrated the importance of scientific information for their knowledge and attitudes. indeed, as described in prior studies [5, 17, [25] [26] [27] [28] [29] , it is evident from the multivariate analysis that receiving information from scientific journals and educational activities facilitates the emergence of pharmacists who have better knowledge and have more positive attitudes. this is important because the pharmacists themselves, in order to effectively influence others, must be well equipped with an appropriate level of knowledge, and interventions should be aimed at including such sources as an important conduit of travel medicine-related information, which can support the pharmacists' confidence. however, an unsurprising finding was that a large majority of pharmacists identified the internet as being their main source of information, rather than the medical establishment. it should be underlined that several studies have expressed concern about the quality and accuracy of health information on the web [30] [31] [32] , and, therefore, it may not provide all of the details necessary to allow the pharmacists to make well-informed suggestions. consequently, it is possible that there are missed opportunities for pharmacists to provide high quality information. health information via the internet should not replace other healthcare professional experts in travel medicine who can give advice for specific situations. furthermore, the vast majority of responders felt they did not have sufficient information to adequately answer questions on this field. clear communication and knowledge about travel medicine would help to instill confidence in pharmacists and keep them adequately informed in order to meet the needs of their community, as well as enable them to discuss any concerns with their customers that may arise from what their customers have read. therefore, pharmacists must have the capacity needed to effectively educate and address public questions and concerns since the delivery of information from a trusted or known source with which the individual has already developed a relationship may be beneficial. this is crucial also in light of the measures taken for the coronavirus disease (covid-19) by the italian authorities [33] , as pharmacists represent an important element in order to provide appropriate recommendations to the public about preventive measures. when considering the study findings, it is important to take into account some potential methodological limitations such as those normally observed in similar survey-based studies. first, the cross-sectional design of this study can only demonstrate associations between the different outcomes of interest and the observed determinants, and it is not possible to say anything about causality. second, data collection through the telephone survey was based on self-reported information, and so may be subject to reporting bias. third, there is also the risk for social desirability bias, by which participants do not report attitudes and behaviors fully or accurately and may provide the responses they believe the researcher wants to know instead of the truth, as opposed to answering honestly, and it is possible that they may over-report socially desirable attitudes and behaviors or under-report socially undesirable attitudes and behaviors. efforts were made to minimize the risk of these biases by ensuring participants that the study was anonymous and confidential and that their data would be de-identified. fourth, it is possible that pharmacists with specific positive or negative opinions or interest in the topic were more or less likely to respond to the survey. if such bias exists, it may lead to an over-or underrepresentation of the rate of positive responses. however, the high response rate could offset this bias. moreover, no difference has been observed between respondents and non-respondents regarding the geographic area of activity. in conclusion, this survey provides some insights into the knowledge, attitudes, and behaviors regarding infectious diseases related to travel among community pharmacists in italy and identifies their associated characteristics. the findings will prove useful when designing and implementing targeted interventions to improve the level of knowledge of pharmacists so that they can play an active role in counseling the public and also in working more closely with the health services in travel medicine. crossref] 2. world health organization (who). international travel and health travel medicine: what's involved? when to refer? destination specific risks of acquisition of notifiable food-and waterborne infections or sexually transmitted infections among finnish international travellers travelers' knowledge, attitudes, and behavior related to infectious diseases in italy knowledge, attitude and practice of travel medicine among primary care physicians in oman: the need for intervention travel health-related activities and services provided by community pharmacies in selangor, malaysia: a cross-sectional analysis pharmacy-based travel health services in the united states the role of community pharmacists in travel health and vaccination in switzerland is travel health a new destination for pharmacy practice and business? an examination of revenue opportunities from pre-travel consultations australian pharmacists' in travel health perceptions and practices the role of pharmacists in travel medicine in south training pharmacists in travel health pharmacy travel health services: current perspectives and future prospects provision of travel medicine advice through community pharmacies: assessment of knowledge, attitudes and practices of pharmacists in malaysia needs assessment study for community pharmacy travel medicine services the knowledge, attitudes, and practices of community pharmacists in their approach to antibiotic use: a nationwide survey in italy principali malattie dei viaggiatori infectious diseases of potential risk for travellers international travelers' sociodemographic, health and travel characteristics: an italian study statacorp. stata statistical software: release 15; statacorp llc: college station are pharmacists ready for a greater role in travel health? an evaluation of the knowledge and confidence in providing travel health advice of pharmacists practicing in a community pharmacy chain in alberta australian pharmacists' perceptions and practices in travel health investigating knowledge, attitudes, and practices regarding vaccinations of community pharmacists in italy knowledge, attitudes, and behaviors of parents towards recommended adult vaccinations: an explanatory survey in the geographic area of rotavirus infection and vaccination: knowledge, beliefs, and behaviors among parents in italy seasonal influenza: knowledge, attitude and vaccine uptake among adults with chronic conditions in italy vaccinations among italian adolescents: knowledge, attitude and behavior parent resources for early childhood vaccination: an online environmental scan the growing vaccine hesitancy: exploring the influence of the internet internet exposure associated with canadian parents' perception of risk on childhood immunization: cross-sectional study decreto del presidente del consiglio dei ministri 11 marzo 2020 this article is an open access article distributed under the terms and conditions of the creative commons attribution (cc by) license the authors gratefully acknowledge all community pharmacists participating in this survey. the costs of the open access publication were supported by the "programma valere 2020" of the university of campania "luigi vanvitelli" (naples, italy). the authors declare no conflict of interest. key: cord-298778-wnbqqzot authors: hamza, marwa s.; badary, osama a.; elmazar, mohamed m. title: cross-sectional study on awareness and knowledge of covid-19 among senior pharmacy students date: 2020-06-15 journal: j community health doi: 10.1007/s10900-020-00859-z sha: doc_id: 298778 cord_uid: wnbqqzot extraordinary actions have been implemented in an effort to control the rapid spread of the ongoing covid-19 epidemic in egypt. people’s adherence to control measures is influenced by their knowledge, attitudes and practices towards the disease. therefore, in the present study we assessed pharmacy senior students’ knowledge, attitudes and practices towards the covid-19 pandemic. an online questionnaire was created and it consisted of 12 questions testing their knowledge about covid-19 clinical characteristics, transmission routes and prevention and control steps. among senior pharmacy students (n = 238), 70% were females and 63% were living in greater cairo. their main source of information included social media (70%), published articles (48%) and television (48%). the overall correct knowledge score was 83%. most of the students displayed a good covid-19 knowledge level (72.5% of the students). the students were least informed when trying to answer questions about hyper-coagulation, as a major cause for death in patients with severe covid-19, and about the timings on the necessity to wear masks. assessment of students’ attitudes and practices towards covid-19 reflected that 87% of them were confident that health care teams and scientists could win the fight against the virus. in addition, 72% of students agreed that covid-19 will be controlled successfully. the greater the students’ knowledge, the more confident they felt that covid-19 will be controlled successfully (or 2.2, 95% confidence interval [ci] 1.03–4.72). good behavioral practice towards covid-19 control was confirmed when 87% of students answered that they didn’t go out to any crowded place. females were 3.6 times (95% confidence interval [ci] 1.03–3.11) more likely to avoid going out than males. bad behavioral practice became evident when approximately 50% of students admitted that they did not wear masks when they left their house. therefore, more efforts should be taken to protect future pharmacists from this pandemic. an epidemic, triggered by the new coronavirus sars-cov-2, has recently become the focus of the scientific community attention [1] . covid-19 represents the illness generated by this virus. its clinical presentation ranges from being an asymptomatic infection to developing into a severe disease with high mortality rate [2] . at the time of writing (may 16, 2020), over 4,425,485 covid-19 cases and 302,059 death have been registered worldwide [3] . in particular, in egypt, there are 11,228 positive individuals and 592 death cases [4] . now, covid-19 has been classified as a pandemic by the world health organization. it is both a highly contagious and a life-threatening disease [3] . its reproduction ratio (rr) is defined as the expected number of cases generated by a single infected individual within a susceptible population [5] . it varies between 2.4 and 2.7 [6] . these values are high compared to influenza (rr = 1.3-1.8) [7] . currently, a fight is being undertaken against the covid-19 pandemic. adherence to control and prevention steps is paramount for ensuring a complete success over covid-19. lessons were learned from the 2003 sars outbreak. these indicate that fear exists within the population awareness and attitude to infectious diseases. this factor may hinder further efforts to prevent the virus spread. therefore, it is crucial to understand covid-19 public's awareness at this critical stage to facilitate pandemic management [8, 9] . as members of the health care team, pharmacists play an essential role in patient care. this includes contributing to the decision-making process within the multidisciplinary health care team, taking responsibility for medication administration and assessment of patient care [10] . moreover, together with physicians, pharmacists' knowledge, attitude and practices are crucial to prevent and control the disease [11] . health care teams include pharmacists are responsible for providing knowledge, delivering good quality management and protecting individuals from illness during epidemic prevalence period. therefore, pharmacy students' education should include training in precautionary measures, effective treatment and follow-up. this is critical together with their behavior in these fields [12] . normally, training in disaster medicine occupy a very small place in regular medical curricula worldwide [13] . therefore, the present study was carried out to evaluate knowledge about covid-19 as an attempt to identify senior students' degree of awareness and its effects. furthermore, this study objective aims to contribute in developing a fit for purpose education program that creates awareness among future practitioners. to the best of the authors' knowledge, this is the first study directly exploring senior pharmacy students' knowledge, attitudes and practice regarding this topic. this is an institutional based cross sectional study conducted in faculty of pharmacy at the british university in egypt during the period from april 28, 2020 to april 30, 2020. the targeted population included students who had completed ≥ 85% of the credit-hours required for graduation. the sample size was calculated by using the online sample size calculator raosoft®. based on an estimated population of 470 students in the final years (year 4 and 5) and on the anticipated response of 50%, the minimum required sample size was 204 participants with a confidence level of 95% and a 5% margin of error. a pre-validated questionnaire [14] was modified after a comprehensive literature search and based on the most recent available information from the world health organization, the center for disease control and prevention (usa) and the egypt ministry of health. the initial draft was sent to a group of experts, chosen according to their experience and expertise in related fields, to appraise the questions in terms of relativity, simplicity and importance. then, the questionnaire was sent by email, through survey monkey software®. a pilot study was conducted on 10 subjects, to test the questionnaire validity. following a group discussion, the questionnaire was completed. the data from the pilot study was removed from the final analysis. the questionnaire included two sections: demographics, and knowledge, attitudes and practices questions (kap). demographic variables included age, gender and place of current residence (cairo, other governorates in egypt). the first part of the kap represented the knowledge investigation, consisting of 12 questions ( fig. 1 ): 4 questions were about covid-19 clinical presentations (k1-k4), 3 asked about transmission routes (k5-k7), and the following 5 questions quizzed on prevention and control (k8-k12). these questions were answered on a true/false basis, including an additional "i don't know" option. one point was assigned to a correct response, while no points were given for an incorrect/unknown answer. the total score ranged from 0 to 12, with a higher score suggesting better covid-19 awareness. in the pilot sample, the knowledge questionnaire cronbach's alpha coefficient was 0.73, indicating reasonable internal consistency [15] . attitudes towards covid-19 were appraised with 2 questions (a1-a2, fig. 2 ) that asked about the level of agreement on the covid-19 control implementations and the confidence perception for winning the war against this coronavirus. the assessment of students practices was performed through 4 questions (p1-p4, fig. 2 ), that investigated their actions towards going to a crowded place, wearing masks when going out in recent days, maintaining social distance between others and if they agree that patients should disclose their exposure to covid-19. finally, one last question was added to inquire about the students' source of covid-19 information. before study start-up, permission was taken from the ethics committee of the faculty of pharmacy at the british university in egypt. the participants were briefed about the study rationale and were reassured about the confidentiality safeguards for their personal information and responses. the answered survey submission was considered as consent for study participation. respondents' participation was completely consensual, anonymous and voluntary. the completed questionnaires data were analyzed using the statistical package for social science, version 21.0 (spss 20.0, chicago, il). the significance level was set at a value of p < 0.05 for all analyses. initially, all information gathered through the questionnaire was coded into variables. in the pilot study, the cronbach's alpha coefficient test was used to measure internal consistency. data normality was tested using the kolmogorov-smirnov test. descriptive and inferential statistics, involving chi-square test, mann-whitney u test, correlation and logistic regression, were used to analyze the results. fever, tiredness, and dry cough, k2 some patients infected with the covid-19 virus may suffer from aches and pains, nasal congestion, runny nose, sore throat or diarrhea, k3. coagulation dysfunction is one of the major causes for death in patients with severe covid-19, k4. older persons and persons with pre-existing medical conditions (such as high blood pressure, heart disease, lung disease, cancer or diabetes) appear to develop serious illness with covid-19 more often than others, k.5 people with covid-19 will not transmit the virus to others when fever is not present, k6. covid-19 can spread from person to person through small droplets from the nose or mouth which are spread when a person with covid-19 coughs or exhales, k7. only wear a mask if you are ill with covid-19 symptoms or looking after someone who may have covid-19, k8. it is not necessary for children and young adults to take measures to prevent the infection by the covid-19 virus, k9. to prevent the infection by covid-19, individuals should avoid going to crowded places such as public transportation, k10. covid-19 is caused by a virus, so antibiotics do not work. antibiotics should not be used as a means of prevention or treatment of covid-19. they should only be used as directed by a physician to treat a bacterial infection, k11. isolating and treating people with covid-19 infections are effective means of minimizing viral spread, k12. people who have contact with someone who has covid-19 infection should be isolated in a proper place immediately. the observation period is usually 14 days in the present study, the scoring system and categorization were utilized to assess knowledge levels towards the covid-19 pandemic. the questions determination was based on one point for every correct answer and zero points for false answers. by adding the results of all the knowledge questions in the survey, the participants mean knowledge scores (kss) were determined. ks ranged from 0 to 12, with the higher scores demonstrating an increased level of pandemic covid-19 knowledge. knowledge level categories were defined by an 80% cut-off point as any score above 80% is good and any score below 80% is poor [16] . during the course of this study, a total of 248 participants completed the devised online survey questions. after excluding 10 respondents who participated in the pilot study, the final sample consisted of 238 participant students. among these, the average age (measured in years) was 22 ± 1, 166 students (70%) were females. when identifying the place of residence, 149 students (63%) were living in greater cairo while 89 (37%) resided in other governorates (table 1) . with regards to the students' source of information, results disclosed that the main source of information was social media (70%) followed by published articles (48%) and television (48%). in the present study, the main objective was to measure covid-19 related knowledge of pharmacy senior students. the mean score of covid-19 related knowledge was 83% (10 ± 1.2, ranging from 4 to 12) (table 1 ). in particular, 72.5% of the students possessed a good knowledge level, while 23.5% displayed poor knowledge based on an 80% cut-off point [16] . the first 4 survey questions (k1-k4) measured students' knowledge towards covid-19 clinical presentations. the highest correct answer rate (98%) related to the covid-19 related knowledge items identified by 2 questions; 'the main clinical symptoms of covid-19 are fever, tiredness, and dry cough (true)' and 'older persons and persons with pre-existing medical conditions (such as high blood pressure, heart disease, lung disease, cancer or diabetes) appear to develop serious illness with covid-19 more often than others (true)' (fig. 1) . however, the item with the lowest correct-answer rates was: 'coagulation dysfunction is one of the major causes for death in patients with severe covid-19 (true)' (37%) (fig. 1 ). the answer 'i don't know' was selected by 12% of students for the question 'covid-19 may cause aches and pains, nasal congestion, runny nose, sore throat or diarrhea in some patients' (fig. 1) . then the following 3 questions (k5-k7) measured students' knowledge towards the transmission routes. in this case, the percentage of knowledgeable students is high. k.5: no transmission if fever is not present (88%); k6: covid-19 can spread from person to person through small droplets (98%); and k7: wear a mask only if you are ill or looking after someone who may have covid-19 (96%). a similar result was achieved with the next 5 questions (k8-k12), which concerned assessing knowledge about covid-19 prevention and control (fig. 1) . this study showed a significant relation between the knowledge score and published articles as source of information (p < 0.05). the students' attitude was measured with 2 questions. the highest positive result was attained by the following covid-19-related attitude item: 'are you confident that the health care team and scientists can win the fight against the covid-19 virus?' (87% answered yes). however, the lowest positive attitude result corresponded to the following: 'do you agree that covid-19 will be successfully controlled?' (only 72% answered yes) (fig. 2) . the highest performance rate was attained by the following covid-19-related practice item: 'in recent days, have you gone to any crowded place?' (87% answered no). however, the item with the lowest performance rate was: 'in recent days, if you left your house, have you worn a mask when leaving home?' (50.5% gave a negative answer) (fig. 2) . as expected, it was found that the covid-19 practice for going out in crowded place was significantly different between males and females: females displayed a lower tendency to go out than males (p < 0.05). in addition, the practice for keeping social distance between students was significantly different between males and females: females maintained social distance more successfully than males (p < 0.05). it is worth mentioning that significant relations existed between the two practices and the students' source of information. in particular, the practice of wearing a mask when leaving home was directly correlated with using television as source of information (p < 0.05). furthermore, the practice of keeping social distance of at least 2 m from other people was directly associated with utilizing physicians' advice as source of information (p < 0.05). moreover, the correlation coefficient, existing between the major knowledge, attitude and practice variables was analyzed. a weak positive correlation was observed between attitude and practice and this result appeared to be highly significant (r = 0.2; p < 0.01). however, no correlation appeared to exist between knowledge and attitudes or knowledge and practice (table 2) . logistic regression analyses showed that students with a good knowledge score are 2.2 times more optimistic and agreed that covid-19 will be successfully controlled (95% confidence interval [ci] 1.03-4.72). students who live in cairo were 1.7 times more likely to maintain social distance of at least 2 m from other people (95% confidence interval [ci] 0.9-3.2)., and 1.7 times more likely to agree that patients should disclose potential covid-19 exposure, than students who live in other governorates in egypt (95% confidence interval [ci] 0.9-3.3). moreover, female students were 3.6 times more likely to avoid going to any crowded place than male students (95% confidence interval [ci] 1.5-8.6), and 2.3 times more likely to keep their social distance of at least 2 m from other people than male students (95% confidence interval [ci] 1.2-4.4) ( table 3 ). the coronavirus disease 2019 (covid-19) is an emerging contagious respiratory disease caused by a novel coronavirus. it was first detected in december 2019 in wuhan, china [17] . emerging infectious diseases like covid-19 can occur anywhere in the world. by being part of the health care team, pharmacists can be exposed to the risk of infection [18] . here, we present the results of a questionnaire study about the knowledge, attitudes and practices of egyptian senior pharmacy students towards the covd-19 disease. in the present study, one of the most significant findings was that the highest main source of covid-19 information among senior students was social media, television and published articles. the results strongly support similar findings in which the main source of middle east respiratory syndrome (mers) information was reported to be the internet and social media [19] . this clearly indicates that senior pharmacy students are relying consistently on the internet, social media and online information as the principal ways to obtain information, compared with friends and other resources. the egyptian ministry of health and population and the world health organization (who) have provided clusters with covid-19 information through their websites and are recommending everyone to be aware of updates relating to knowledge about covid-19 [3, 4] . similarly, a previous study reported that internet was the main source of information about mers-cov [20] . therefore, it is feasible to suggest that researches should urgently find ways to utilize the internet to promote health, especially in emerging infectious diseases and pandemics. at present, the quality of the information available on the internet is insufficient and needs to be revised. furthermore, a very small percentage of participants reported that healthcare professionals, either physicians or pharmacists, were their primary source of information. this scenario illustrates the lack of healthcare team engagement in increasing the public awareness regarding different aspects of covid-19. the present study showed a significant relation existing between the knowledge score and published articles as source of information (p < 0.05). that may be due to the role played by the who and the egyptian ministry of health and population in providing cluster with covid-19 information through their websites. therefore, these organizations recommend everyone to be aware of updates about covid-19. as mentioned earlier, 238 senior pharmacy students participated in the present study. the overall mean knowledge scored among students was 83% (10 ± 1.2). most of the students possessed a good covid-19 knowledge level (72.5% of the students), possibly owing to the fact that the egyptian ministry of health and population (mohp) and the world health organization (who) are providing a significant amount of covid-19 information through their websites [3, 4] . the present study results showed that 98% of participants were knowledgeable about the main clinical symptoms of covid-19. in addition, the same percentage of students knew that older persons and persons with preexisting medical conditions were more liable to develop serious illness with covid-19 than others. these findings confirm observations from previous studies, which reported that the majority of participants believed that the disease is more dangerous for the elderly and for those suffering from chronic diseases [21, 22] . the students were least knowledgeable towards the question of coagulation dysfunction being one of the major death causes in patients with severe covid-19. suitable remedial actions should be taken since vascular, venous thromboembolism and arterial thrombosis risks have been documented in severe and critically ill covid-19 patients [23, 24] . a perceived lack of information regarding some of the covid-19 clinical presentation or preventive measures highlights the requirement for the inclusion of more training, seminars in their courses. this raises concerns regarding the ability of newly-graduated pharmacists to deal with the covid-19 pandemic disease, or any other potential similar epidemic events. also, the students were least knowledgeable about when to wear masks. masks and sanitizing products use has evolved since the beginning of the coronavirus pandemic, resulting in their rapid depletion. lack of appropriate security measures represents a major concern for health workers. this is a cause of concern in a highly populated country such as egypt, where the public may experience anxiety and even panic feelings, because basic protection measures are unavailable [25] . knowledge is a prerequisite for establishing positive attitudes and promoting positive behaviors. therefore, it makes sense to invest in individuals' cognition towards the infection. regarding the senior students' attitude towards covid-19, in the present study 72% of them agreed that covid-19 will be successfully controlled and 87% were confident that health care teams and scientists can win the fight against the coronavirus. interestingly, it was found that a greater students' knowledge relates to more confidence in covid-19 becoming successfully controlled (or 2.2, 95% confidence interval [ci] 1.03-4.72). correlation between the students' attitude and practice indicates that their positive attitude is affecting their action towards covid-19. this present study results showed that 87% of participants did not go to any crowded place, 76% of the students agreed that patients should disclose their exposure to covid-19 and 73% of the students kept the social distance of at least 2 m from other people. however, the students showed bad practice when using masks as a protective measure (approximately 50%). that was directly related to their lower level of knowledge towards the importance of wearing masks (p < 0.05). that is supported by findings from a previous study, which reported that only about 35% of the participants were willing to wear face masks [21] . interestingly, it was reported that females preferred not to go to any crowded place more than male students (or 3.6, 95% confidence interval [ci] 1. 5-8.6 ). in addition, females preferred to keep the social distance more than male students (or 2.3, 95% confidence interval [ci] 1.2-4.4). moreover, students who live in other governorates kept social distance and agreed that patients should disclose their exposure to covid-19 rather than the students who live in the capital (or 1.7, 95% confidence interval [ci] 0.9-3.2). it is noteworthy that a significant association existed between the practice of wearing masks when leaving home and using television as source of information (p < 0.05). likewise, an association existed between the practice of keeping the social distance at least 2 m from other people and utilizing physicians as a source of information (p < 0.05). this encourages the public to avoid public meetings, including sports, ceremonies, meetings and school classes, in order to prevent a global transmission of coronavirus infection [26] . however, many people ignore the importance of maintaining social distance because of attitudinal issues. it can be concluded that senior pharmacy students are generally knowledgeable about covid-19. they lack knowledge about some of the clinical presentation of severe covid-19 infection such as hyper-coagulation. they also lack knowledge about the indication of wearing face masks. the present study constitutes the first attempt to examine knowledge and perceptions of future pharmacists in egypt on pandemic diseases. finally, the analysis of senior pharmacy students' knowledge and the factors affecting their attitudes and practices towards covid-19 could provide a reference for preventing further spread of this disease. it is recommended that strategies and education programs should be implemented to reduce the risk of infection among future pharmacists. senior pharmacy students should acquire increased awareness regarding the seriousness and consequences of a covid-19 infection. some of the results of this study were compared with research on other new emerging infectious diseases similar to coronavirus, because no analyses have been conducted concerning pharmacy students or students in general. covid-19-navigating the uncharted the loss of igm memory b cells correlates with clinical disease in common variable immunodeficiency covid-19) situation report-117. situation-reports/20200311-sitrep-51-covid-19.pdf?sfvrsn=1ba62e57_10 2020 covid-19 overview pandemic potential of a strain of influenza a (h1n1): early findings the reproductive number of covid-19 is higher compared to sars coronavirus estimates of the reproduction number for seasonal, pandemic, and zoonotic influenza: a systematic review of the literature college students' knowledge, attitudes and adherence to public service announcements on ebola in nigeria: suggestions for improving future ebola prevention education programmes the impact of knowledge and attitudes on adherence to tuberculosis treatment: a case-control study in a moroccan region health care professional development: working as a team to improve patient care knowledge, attitudes, and practices regarding pandemic h1n1 influenza among medical and dental residents and fellowships in shiraz iran awareness of the pandemic h1n1 influenza global outbreak 2009 among medical students in karachi pakistan disaster 101: a novel approach to disaster medicine training for health professionals knowledge, attitudes, and practices towards covid-19 among chinese residents during the rapid rise period of the covid-19 outbreak: a quick online cross-sectional survey the use of cronbach's alpha when developing and reporting research instruments in science education knowledge, attitude and practice regarding dengue fever among the healthy population of highland and lowland communities in central nepal epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in wuhan, china: a descriptive study community pharmacists: on the frontline of health service against covid-19 in lmics knowledge and attitude of healthcare workers about middle east respiratory syndrome in multispecialty hospitals of qassim saudi arabia identification of information types and sources by the public for promoting awareness of middle east respiratory syndrome coronavirus in saudi arabia knowledge, perceptions, and attitude of egyptians towards the novel coronavirus disease (covid-19) covid-19 patients' clinical characteristics, discharge rate, and fatality rate of meta-analysis anticoagulant treatment is associated with decreased mortality in severe coronavirus disease 2019 patients with coagulopathy clinical characteristics of coronavirus disease 2019 in china mass gathering events and reducing further global spread of covid-19: a political and public health dilemma. the lancet the researchers acknowledge our senior pharmacy students who took part in this study. the authors declare that they do not have conflict of interests. key: cord-016556-tdwwu43v authors: kawtrakul, asanee; yingsaeree, chaiyakorn; andres, frederic title: semantic tracking in peer-to-peer topic maps management date: 2007 journal: databases in networked information systems doi: 10.1007/978-3-540-75512-8_5 sha: doc_id: 16556 cord_uid: tdwwu43v this paper presents a collaborative semantic tracking framework based on topic maps which aims to integrate and organize the data/information resources that spread throughout the internet in the manner that makes them useful for tracking events such as natural disaster, and disease dispersion. we present the architecture we defined in order to support highly relevant semantic management and to provide adaptive services such as statistical information extraction technique for document summarization. in addition, this paper also carries out a case study on disease dispersion domain using the proposed framework. this paper gives an overview of a generic architecture we are currently building as part of the semantic tracking project in cooperation with the fao aos project [32] . initiated by fao and ku, the semantic tracking project aims at providing a wide area collaborative semantic tracking portal for monitoring important events related to agriculture and environment, such as disease dispersion, flooding, or dryness. this implies to deal with any kind of multilingual internet news and other online articles (e.g. wiki-like knowledge and web logs); it describes the world around us rapidly by talking about the update events, states of affairs, knowledge, people and experts who participate in. therefore, the semantic tracking project targets to provide adaptive services to large group of users (e.g. operator, decision makers), depending on all the knowledge we have about the environment (users themselves, communities they are involved in, and device he's using). this vision requires defining an advanced model for the classification, the evaluation, and the distribution of multilingual semantic resources. our approach fully relies on state of the art knowledge management strategies. we define a global collaborative architecture that allows us to handle resources from the gathering to the dissemination. however, sources of these data are scattered across several locations and web sites with heterogeneous formats that offer a large volume of unstructured information. moreover, the needed knowledge was too difficult to find since the traditional search engines return ranked retrieval lists that offer little or no information on semantic relationships among those scattered information, and, even if it was found, the located information often overload since there was no content digestion. accordingly, the automatic extraction of information expressions, especially the spatial and temporal information of the events, in natural language text with question answering system has become more obvious as a system that strives for moving beyond information retrieval and simple database query. however, one major problem that needs to be solved is the recognition of events which attempts to capture the richness of event-related information with their temporal and spatial information from unstructured text. various advanced technologies including name entities recognition and related information extraction, which need natural language processing techniques, and other information technologies, such as geomedia processing, are utilized part of emerging methodologies for information extraction and aggregation with problem-solving solutions (e.g. "know-how" from livestock experts from countries with experiences in handling bird flu situation). furthermore, ontological topic maps are used for organizing related knowledge. in this paper, we present our proposal aiming to integrate and organize the data/information resources dispersed across web resources in a manner that makes them useful for tracking events such as natural disaster, and disease dispersion. the remainder of this paper is structured as follows: section 2 describes the key issues in information tracking as nontrivial problems; in section 3 we introduce the framework architecture and its related many-sorted algebra. section 4 gives more details of the system process regarding the information extraction module. section 5 discusses the personalized services (e.g. knowledge retrieval service and visualization service) provided for collaborative environments. finally, in section 6, we conclude and give some forthcoming issues. collecting and extracting data from the internet have two main nontrivial problems: overload and scattered information, and salient information and semantic extraction from unstructured text. many experiences [20, 21, and 35] have been done regarding event tracking or special areas or areas related to events monitoring (e.g. the best practice for governments to handle bird flu situation), the collection of important events and their related information (e.g. virus transmission from one area to other locations and from livestock to humans). firstly, target data used for semantic extraction are organized and processed to convey understanding, experience, accumulated learning, and expertise. however, sources of these data are scattered across several locations and websites with heterogeneous formats. for example, the information about bird flu consisting of policy for controlling the events, disease infection management, and outbreak situation may appear in different websites as shown in fig. 1 . egypt -update 5 16 february 2007 the egyptian ministry of health and population has confirmed the country's 13th death from h5n1 avian influenza. the 37-year-old female whose infection was announced on 15 february, died today. consequently, collecting required information from scattered resources is very difficult since the semantic relations among those resources are not directly stated. although it is possible to gather those information, the collected information often overload since there is no content digestion. accordingly, solving those problems manually is impossible. it will consume a lot of time and cpu power. the system that can collect, extract and organize those information according to contextual dimensions automatically, is our research goal for knowledge construction and organization. secondly, only salient information must be extracted to reduce time consumption for users to consume the information. in many case, most of salient information (e.g. time of the event, location that event occurred, the detail of the event) are left implicitly in the texts. for example: in the text in fig. 1 , the time expression "15 february" mentioned only "date and month" of the bird flu event but did not mention the 'year'. the patient and her condition (i.e. '37-year-old female', and 'died') was caused by bird flu which is written in the text as 'avian influenza' and 'h5n1 avian influenza'. accordingly, the essential component of computational model for event information capturing is the recognition of interested entities including time expression, such as 'yesterday', 'last monday', and 'two days before', which becomes an important part in the development of more robust intelligent information system for event tracking. information extraction in traditional way processes a set of related entities in the format of slot and filler, but the description of information in thai text such as locations, patient's condition, and time expressions can not be limited to a set of related entities because of the problems of using zero anaphora [17] . moreover, to activate the frame for filling the information, name entity classification must be robust as it has been shown in [5] . in this section, we give an overview of the modeling we are providing. preliminary parts of our framework have been previously introduced to the natural language processing and database community [15] . in the following, we present our p2p framework and related many-sorted algebra modeling. let us introduce our design approach of an ontological topic map for event semantic tracking. the ontological topic map [22] helps to establish a standardized, formally and coherently defined classification regarding event tracking. one of our current focus and challenges has to develop a comprehensive ontology, which defines the terminology set, data structure and operations regarding semantic tracking and monitoring in the field of agriculture and environment. the semantic tracking algebra is a formal and executable instantiation of the resulting event tracking ontology. our algebra has to achieve two tasks: (1) first, it serves as a knowledge layer between the users (e.g. agriculture experts) and the system administration (e.g. it scientists and researchers). let us remind the notion of many sorted algebra [13] . such algebra consists of several sets of values and a set of operations (functions) between these sets. our semantic tracking algebra is a domain-specific many-sorted algebra incorporating a type system for agriculture and environment data. it consists of two sets of symbols called sorts (e.g. topic, rss postings) and operators (e.g. tm_transcribe, semantic_similarity); the function sections constitute the signature of the algebra. its sorts, operators, sets, and functions are derived from our agriculture ontology. second order signature [14] is based on two coupled many-sorted signatures where the toplevel signature provides kinds (set of types) as sorts (e.g. data, resource, semantic_data) and type constructors as operators (e.g. set). to illustrate the approach, we assume the following simplified many-sorted algebra: kinds data, resource, semantic_data, topic_maps, set type constructor -> data topic -> resource rss, htm // resource document type -> semantic_data lsi_sm, rss_sm, htm_sm // semantic and metadata vectors -> tm tm(topic maps) unary operations ∀ resource in resource, resource → sm: semantic_data,tm tm_transcribe ∀ sm in semantic_data sm → set(tm) semantic_similarity the notion sm:semantic_data is to be read as "some type sm in semantic_data," and means there is a typing mapping associated with the tm_transcribe operator. each operator determines the result type within the kind of semantic_data, depending on the given operand resource types. the semantic merging operation takes two or more operands that are all topic maps values. the select takes an operand type set (tm) and a predicate of type topic and returns a subset of the operand set fulfilling the predicate. from the implementation of view, the resource algebra is an extensible library package providing a collection of resource data types and operations for agriculture and environment resource computation. the major research challenge will be the formalization and the standardization of cultural resource data types and semantic operations through iso standardization. as shown in fig. 2 , the proposed framework consists of six main services. the detail of each service is outlined as followed: to generate useful knowledge from collected documents, two important modules, information extraction and knowledge extraction, are utilized. ontological topic maps and domain-related ontologies defined in owl [9] are used as a knowledge base to facilitate the knowledge construction and storage process as it has been shown in garsho's review [11] . the standard iso / iec topic maps (iso 13250) facilitates the knowledge interoperability and composition. the information extraction and integration module is responsible for summarizing the document into a predefined frame-like/structured database, such as <disease name, dispersion location and time, status of patient's condition>. the knowledge extraction and generalization is responsible for extracting useful knowledge (e.g. general symptom of disease) from collected document. latent semantic analysis will be applied to find new knowledge or relationships that are not explicitly stored in the knowledge repository. language engineering and knowledge engineering techniques are key methods to build the target platform. for language engineering, word segmentation [31] , named entity recognition [6] , shallow parsing [28] , shallow anaphora resolution and discourse processing [6, 7, and 12] have been used. for knowledge engineering, ontological engineering, task-oriented ontology, ontology maintenance [16] and topic maps [5] model have been applied. the information, both unstructured and semi-structured documents are gathered from many sources. periodic web crawler and html parser [33] are used to collect and organize related information. the domain specific parser [17] is used to extract and generate meta-data (e.g. title, author, and date) for interoperability between disparate and distributed information. the output of this stage is stored in the document warehouse. to organize the information scattered at several locations and websites, textual semantics extraction [27] is used to create a semantic metadata for each document stored in the document warehouse. guided by domain-based ontologies associated to reasoning processes [23] and ontological topic map, the extraction process can be taught of as a process for assigning a topic to considered documents or extracting contextual metadata from documents following xiao's approach [36] . knowledge retrieval service: this module is responsible for creating response to users' query. the query processing based on tmql-like requests is used to interact with the knowledge management layer. knowledge visualization: after obtaining all required information from the previous module, the last step is to provide the means to help users consume that information in an efficient way. to do this, many visualization functions is provided. for example, spatial visualization can be used to visualize the information extracted from the information extraction module and graph-based visualization can be used to display hierarchal categorization in the topic maps in an interactive way [27] . due to page limitation, this paper will focus in only information extraction module, knowledge retrieval service module and knowledge visualization service module. the proposed model for extracting information from unstructured documents consists of three main components, namely entity recognition, relation extraction, and output generation, as illustrate in fig. 3 . the entity recognition module is responsible for locating and classifying atomic elements in the text into predefined categories such as the names of diseases, locations, and expressions of times. the relation extraction module is responsible for recognizing the relations between entities recognized by the entity recognition module. the output of this step is a graph representing relations among entities where a node in the graph represents an entity and the link between nodes represents the relationship of two entities. the output generation module is responsible for generating the n-tuple representing extracted information from the relation graph. the details of each module are described as followed. to recognize an entity in the text, the proposed system utilizes the work of h. chanlekha and a. kawtrakul [6] that extracts entity using maximum entropy [2] , heuristic information and dictionary. the extraction process consists of three steps. firstly, the candidates of entity boundary are generated by using heuristic rules, dictionary, and statistic of word co-occurrence. secondly, each generated candidate is then tested against the probability distribution modeled by using maximum entropy. the features used to model the probability distribution can be classified into four categories: word features, lexical features, dictionary features, and blank features as described in [7] . finally, the undiscovered entity is extracted by matching the extracted entity against the rest of the document. the experiment with 135,000 words corpus, 110,000 words for training and 25,000 words for testing, shown that the precision, recall and f-score of the proposed method are 87.60%, 87.80%, 87.70% respectively. to extract the relation amongst the extracted entities, the proposed system formulates the relation extraction problem as a classification problem. each pair of extracted entity is tested against the probability distribution modeled by using maximum entropy to determine whether they are related or not. if they are related, the system will create an edge between the nodes representing those entities. the features used to model the probability distribution are solely based on the surface form of the word surrounding the considered entities; specifically, we use the word n-gram and the location relative to considered entities as features. the surrounding context is classified into three disjointed zone: prefix, infix, and suffix. the infix is further segmented into smaller chunks by limiting the number of words in each chunk. for example, to recognize the relation between victim and condition in the sentence "the [victim] whose [condition] was announced on ....", the prefix, infix and suffix in this context is 'the', 'whose', and 'was announced on ....' respectively. to determine and to assess the "best" n-gram parameter and number of words in each chunk of the system, we conduct the experiment with 257 documents, 232 documents for training and 25 documents for testing. we vary the n-gram parameter from 1 to 7 and set the number of words in each chunk as 3, 5, 7, 10, 13, and 15. the result is illustrated in fig. 4 . the evident shows that f-score is maximum when n-gram is 4 and number of words in each chunk is 7. the precision, recall and f-score at the maximum f-score are 58.59%, 32.68% and 41.96% respectively. after obtaining a graph representing relations between extracted entities, the final step of information extraction is to transform the relation graph into the n-tuple representing extracted information. heuristic information is employed to guide the transformation process. for example, to extract the information about disease outbreak (i.e. disease name, time, location, condition, and victim), the transformation process will starts by analyzing the entity of the type condition, since each n-tuple can contain only one piece of information about the condition. it then travels the graph to obtain all entities that are related to considered condition entity. after obtaining all related entities, the output n-tuple is generated by filtering all related entities using constrain imposed by the property of each slot. if the slot can contains only one entity, the entity that has the maximum probability will be chosen to fill the slot. in general, if the slot can contain up to n entities, the top-n entities will be selected. in addition, if there is no entity to fill the required slot, the mode (most frequent) of the entity of that slot will be used to fill instead. the time expression normalization using rule-based system and synonym resolution using ontology are also performed in this step to generalize the output n-tuple. the example of the input and output of the system are illustrated in fig. 3 . distributed adaptive and automated services require exploiting all the environmental knowledge stored in ontological topic maps that is available about the elements involved in the processes [24] . an important category of this knowledge is related to devices' states; indeed, knowing if a device is on, in sleep mode, off, if its battery still has autonomy of five minutes or four days, or if it has a wired or wireless connection, etc. helps adapting services that can be delivered to this device. for each device, we consider a state control that is part of the device's profile. then, of course, we use the information contained in communities' and users' profiles. personalized services rely on user-related contexts such as localization, birth date, languages abilities, professional activities, hobbies, communities' involvement, etc. that give clues to the system about users' expectations and abilities. in the remainder of this section, we present the two main adaptive services: the knowledge service and the knowledge visualization service based on our model. the knowledge retrieval service module is responsible for interacting with the topic maps repository to generate answers to user's tmql-like queries [33] . the framework currently supports three types of query. the detail of each query type is summarized in table 1 . the knowledge visualization service is responsible for representing the extracted information and knowledge in an efficient way. users require to access to concise organization of the knowledge. schneiderman in [12] pointed that "the visual information-seeking mantra is overview first, zoom and filter, then details ondemand". in order to locate relevant information quickly and explore the semanticrelated structure, our flexible approach regarding two ways of visualizations (spatialbased or graph-based visualization) is described in the following. the spatial-based visualization functions help users to visualize the extracted information (e.g. the bird flu outbreak situation extracted in fig. 3 .) using web-based geographical information system, such as google earth. this kind of visualization allows the users to click on the map to get the outbreak situation of the area according to their requests. in addition, by viewing the information in the map users can see the spatial relations amongst the outbreak situations easier than without the map. one usage example of google earth integrated system for visualizing the extracted information about bird flu situation is shown in fig. 5 . 6 related works we agree that distributed knowledge management has to assume two principles [4] related to the classification: (1) autonomy of classification for each knowledge management unit (such as community), and (2) coordination of these units in order to ensure a global consistency. having a decentralized peer-to-peer knowledge management, the swap platform [25] is designed to enable knowledge sharing in a distributed environment. pinto et al. provide interesting updates and changes support between peers. however, vocabularies in swap have to be harmonized; which implies to have some loss of knowledge consistency. but even if we share the approach of core knowledge structure that is expendable, the vocabulary, in our case, is common and fully shared by the community, so the knowledge evaluation and comparison can be more effective. moreover, swap provides some kind of personalization (user interface mainly) but does not go as far as the semantic tracking does. from our point of view, swap definitely lacks environmental knowledge management that is required to perform advanced services; on the other hand, dbglobe [26] is a service-oriented peer-to-peer system where mobile peers carrying data provide the base for services to be performed. its knowledge structure is quite similar to our project as it is using metadata about devices, users and data within profiles; moreover, communities are also focused on one semantic concept. dbglobe relies on axml [3] in order to perform embedded calls to web services within xml. thus, it provides a very good support for performing services but does not focus on users and environments knowledge in order to offer optimized authoritarian adaptive services. described as a p2p dbms, ambientdb [10] relies on the concept of ambient intelligence, which is very similar to our vision of adaptive services with automatic cooperation between devices and personalization. however, although ambientdb is using the effective chord distributed hash table to index the metadata related to resources, it lacks the environmental knowledge management provided inside our project that is necessary to achieve adaptive collaborative distribution and personalized query optimization. the extraction framework described in this paper is closely related to promed-plus [37] , a system for the automatic "fact" extraction from plain-text reports about outbreaks of infectious epidemics around the world to database, and mitap [8] , a prototype sars detecting, monitoring and analyzing system. the difference between our framework and those systems is that we also emphasize on generating the semantic relations among the collected resources and organizing those information by using topic map model. the proposed information extraction model that formulates the relation extraction problem as a classification problem is motivated by the work of j.suzuki et. al. [30] . this innovated work has proposed a hdag kernel solving many problems in natural language processing. the use of classification methods in information extraction is not new. intuitively, one can view the information extraction problem as a problem of classifying a fragment of text into a predefined category which results in a simple information extraction system such as a system for extracting information from job advertisements [38] and business cards [19] . however, those techniques require the assumption that there should be only one set of information in each document, while our model could support more than one set of information. as communities generate increasing amounts of transactions and deal with fast growing data, it is very important to provide new strategies for their collaborative management of knowledge. in this paper, we presented and described our proposal regarding information modeling for adaptive semantic management which aims at extracting information and knowledge from unstructured documents that spread throughout the internet by emphasizing on information extraction technique, event tracking and knowledge organizing. we first motivated the need for such modeling in order to provide personalized services to users who are involved in semantic tracking communities. the motivation for this work is definitely to improve user's access to semantic information and to reach high satisfaction levels for decision making. then, we gave an overview of our approach's algebra with its operators, focusing on update and consistency policies. we finally proposed and defined adaptive services that enable collaborative project to automatically dispatch semantic and to make the query results more relevant. this challenging work needs more complicate natural language processing with deeply semantic relations interpretation. know-what: a development of object property extraction from thai texts and query system a maximum entropy approach to natural language processing atomicity for p2p based xml repositories the role of classification(s) in distributed knowledge management thai named entity extraction by incorporating maximum entropy model with simple heuristic information elementary discourse unit segmentation for thai using discourse cue and syntactic information mitap for sars detection owl web ontology language reference. w3c recommendation ambientdb: p2p data management middleware for ambient intelligence living with topic maps and rdf centering: a framework for modeling the local coherence of discourse proceedings of the 15th international conference on very large data bases. very large data bases second-order signature: a tool for specifying data models, query processing, and optimization a framework of nlp based information tracking and related knowledge organizing with topic maps automatic thai ontology construction and maintenance system a unified framework for automatic metadata extraction from electronic document know-what: a development of object property extraction from thai texts and query system information extraction by text classification profile-based event tracking event recognition with fragmented object tracks, icpr application framework based on topic maps a flexible ontology reasoning architecture for the semantic web on data management in pervasive computing environments ontoedit empowering swap: a case study in supporting distributed, loosely-controlled and evolving engineering of ontologies (diligent) dbglobe: a serviceoriented p2p system for global computing topic management in spatial-temporal multimedia blog bootstrap cleaning and quality control for thai tree bank construction the eyes have it: a task by data type taxonomy for information visualizations kernels for structured natural language data thai word segmentation based on global and local unsupervised learning know-who: person information from web mining topic map query language (tmql) event recognition on news stories and semi-automatic population of an ontology, wi using categorial context-shoiq(d) dl to integrate context-aware web ontology metadata information extraction from epidemiological reports information extraction by text classification: corpus mining for features the work described in this paper has been supported by the grant of national electronics and computer technology center (nectec) no. nt-b-22-14-12-46-06, under the project "a development of information and knowledge extraction from unstructured thai document". key: cord-020180-fpx27v7z authors: smuts, hanlie; maramba, george title: a knowledge asset management implementation framework for information systems outsourcing projects date: 2020-03-10 journal: responsible design, implementation and use of information and communication technology doi: 10.1007/978-3-030-45002-1_23 sha: doc_id: 20180 cord_uid: fpx27v7z organisations are increasingly outsourcing information systems (is) to external service providers. these is outsourcing decisions are driven by multiple organisational factors such as outsource vendor expertise and knowledge, process performance improvement due to better is, and enabling the organisation’s ability to focus on its core capabilities. in order to harnass such vendor knowledge to achieve business outcomes, the importance of a shared knowledge asset base, between the client organisation and outsource vendor, is emphasised. however, outcomes from is outsourcing remain poor despite consideration of experience and research. therefore, the aim of this study is to design and propose a knowledge asset management implementation framework that may be applied in is outsourcing projects. the proposed framework was evaluated by an experienced programme director and its applicability was tested against a large scale is outsourcing project. the purpose of such a framework is to enable organisations to manage and institutionalise knowledge assets that are created during the is outsourcing project and to ensure that the organisation may gain the benefit from such knowledge assets as an outcome of the is outsourcing arrangement. information systems (is) outsourcing, where an organisation contracts external service providers to effectively deliver is-enabled business processes, application services and/or infrastructure solutions for business outcomes, is regarded as an important business strategy [1] . recently, as experience and knowledge have deepened, outsourcing as a business phenomenon has been positioned as an opportunity to be applied to is activities in line with an organisation's overall sourcing strategy [2] . several reasons are cited by organisations for adopting is outsourcing, such as; ensuring a high level of productivity, and to offer maximum quality to their customers [3] . the importance of knowledge management (km) and a shared knowledge base between the client organisation and outsource vendor, are highlighted as a basis for organisational performance gains [4] . in order to realise these performance gains, an environment for client organisation and outsource vendor knowledge integration must be created through common language and frequent interaction, consequently fostering knowledge transfer and ultimately, knowledge asset management [5, 6] . knowledge asset management is a commonly stated objective for organisations, although difficult to achieve, as knowledge creation is rewarded rather than knowledge reuse [7] . however, outcomes from is outsourcing continue to remain poor despite consideration of experience and research [8, 9] . recent studies suggest that one of the reasons for this lack of improvement is due to the complex nature of outsourcing agreements as they shift from being a mere cost cutting exercise to one which holds significant strategic and social importance to the organisation [9, 10] . furthermore, knowledge asset management is considered most valuable in the knowledge-driven economy, although the focus on tasks of knowledge asset understanding and management, have not been prioritised compared to their physical counterparts [11] . while scholars highlight the significance of knowledge asset management in addressing is outsourcing complexity, little research has been conducted on how organisations manage knowledge assets and knowledge reuse in is outsourcing circumstances [7, 12, 13] . therefore, the research question that this paper aims to address is: "what are the components of a knowledge asset management implementation framework for managing knowledge assets in is outsourcing projects?". by addressing this question, organisations are able to reference an approach to the management of knowledge assets during is outsourcing projects, ensuring that institutional knowledge is not lost during outsourcing and that new knowledge that is created through the partnership, is captured and managed. in sect. 2 we present the background to the study followed by the research approach in sect. 3. section 4 details the data analysis and findings, while sect. 5 concludes the paper. the management of knowledge in the context of is outsourcing, is a comprehensive course of action that requires focus and commitment throughout an organisation in order to achieve the desired results [13, 14] . the outcome of this organisational focus and commitment, is an important factor for is outsourcing arrangements, seeing as the aim is to increase the collective knowledge of each other's knowledge domain [5, 15] . technology-specific knowledge, such as the is services provided, flows from the organisation to the outsource supplier, and business-specific knowledge about processes and procedures flows from the supplier to the organisation. the purpose of this knowledge transfer is to increase the knowledge shared by the organisation and the outsource vendor [5, 16] . knowledge sharing and management in the context of is outsourcing is not a stand-alone practice; it should be integrated into all aspects of the outsourcing arrangement [5, 15, 17] . the inseparability of is from the internal production service in the client organisation implies that even in situations of absolute outsourcing, a minimum set of capabilities are retained in-house by the client organisation [18] . in the next sections we consider is outsourcing, km, and highlight the role of knowledge asset management in order to manage knowledge in an is outsourcing arrangement. is outsourcing refers to the contractual agreement between the client organisation and an outsource vendor for the transfer of assets and/or the development and implementation of is within an agreed time period and a specified cost [9, 18] . the transfer of assets view of is outsourcing is also applicable to subcontracting in is [19] . once the outsource vendor has been selected and prior to embarking on transition, several elements should be considered to further guide the outsourcing arrangement process such as a transition planning, communication strategy planning, transfer conditions identification and resource mobilisation [20] . is outsourcing initiatives consist of many tasks that have to be executed in an interrelated manner [9] . these interdependent steps must enable an organisation to obtain the correct information in order to be able to select the right services for the right reasons, consequently maximising business leverage in the outsourcing arrangement [20, 21] . several outsourcing lifecycle models exist each consisting of numerous steps. examples of such interrelated project steps include; the is outsourcing programme lifecycle consisting of the request for proposal and vendor selection, contracts and negotiations, setup and logistics, programme execution, implementation and testing, and programme completion stages [22] . the three phase is outsourcing building block approach consists of the architect, engage and govern stages [20] . alborz et al. [23] defined 3 stages: (1) pre-contract stage (scoping and evaluation), (2) contract stage (negotiation) and (3) post-contract stage. outsourcing strategy and due diligence operationalise the precontract stage and contract development the contract stage. the post-contract stage is operationalised through governance, performance management, contract management, working relationship management and knowledge management. such an is outsourcing project lifecycle guides an organisation in realising the full value that outsourcing can provide to become an informed purchaser, to plan and design the commercial arrangement, to carefully select the best value for money supplier and to put in the appropriate management skills and effort [20, 24] . outsourcing should give an organisation a strategic advantage and involves judgements about quantitative and qualitative factors [19] . if it fails to deliver these advantages then an outsourcing arrangement should not be considered [22] . in the next section, we present an overview of the km and knowledge asset management. the importance of a shared knowledge base between the client organisation and outsource vendor is highlighted as a basis for performance gains, as it creates sensitivity to the organisational environment of the other party and encompasses goals, constraints, interpretations and behaviour [4, 13] . such an environment for knowledge integration is created through a common language and frequent interaction, consequently fostering knowledge transfer and adding value [5, 6] . such knowledge asset value add constitutes what is known by the organisation and employees resulting in a potential long lasting, open-ended value. the organisational knowledge asset value may be derived in two ways: firstly, the degree to which knowledge assets may be abstracted and generalised, and secondly, the extent to which a knowledge asset may be codified [25] . as the same knowledge is used to solve multiple problems, it is recognised that the same is capability can transform different organisations in different ways [13] . outsource vendors benefit fully from their knowledge resources as they reuse the same knowledge in different contexts for different customers. similarly, is organisations have to design adequate knowledge transfer strategies to build expertise so that new problems can be addressed by reusing the same knowledge [26] . however, the fact that an organisation relies on an outsource vendor does not mean that it should ignore the importance of an ongoing knowledge management programme specifically related to knowledge transfer [6, 27] . knowledge transfer has come to the fore in response to the increasing size, complexity and scope of organisations, as well as the increasing capabilities of modern is to support knowledge-orientated activities [13, 28] . organisations discover the effective deployment of is management and derive business value from it through experiential learning and hands-on experience. challenges are not always appreciated unless they are experienced, as the understanding of the value of an is innovation tends to materialise in an evolutionary manner. organisations choosing to outsource may unintentionally fragment this knowledge by missing critical learning opportunities, with a resulting loss of ensuing business gains. this necessitates constant assessment of the impact of is outsourcing decisions on the protection and enhancement of an organisation's knowledge base [6, 27] . before we present the proposed knowledge asset management implementation (kami) framework, we present an overview of the research methodology followed for this research paper. the overall objective of this paper was to define a kami framework for managing knowledge in is outsourcing projects. the purpose of such a framework is to assist organisations with the management of knowledge in is outsourcing projects. in order to achieve this outcome, we followed a design-based approach [29] . design based research is a "systematic but flexible methodology aimed to improve educational practices through iterative analysis, design, development, and implementation, based on collaboration among researchers and practitioners in real-world settings, and leading to contextually-sensitive design principles and theories" [31:6] . design based research produces both theories and practical interventions as its outcomes [31] and encompasses five basic characteristics [30] . the first characteristic is pragmatic, referring to the research focus on solving current real-world problems through the design of interventions. the grounded characteristic points to the fact that the research is grounded in both theory and the real-world context, while interactive, iterative and flexible refer to the nature of the research process. integrative highlights that researchers integrate a variety of research methods and approaches from both qualitative and quantitative research paradigms, depending on the needs of the research. the final characteristic, contextual, emphasises that research outcomes are connected with both the design process through which results are generated and the setting where the research is conducted. with these characteristics guiding our research, we built upon prior literature about is outsourcing and knowledge asset management in order to create a kami framework (pragmatic nature of our research). our research approach was of a qualitative nature and the context of our research was is and organisations. we approached our research on both theory and a real-world context (grounded) by considering existing knowledge management implementation frameworks and gaps identified. thereafter, we designed our proposed kami framework through an iterative process by considering and finally, we evaluated the proposed kami framework with a real-world case in order to establish its practical application. multiple knowledge management system implementation frameworks focusing on operations exist, however, none of the frameworks specifically encompass knowledge asset management across all phases of an is outsourcing project. table 1 summarises the frameworks; in each instance listing the focus of the framework, the steps included in the framework and the reference. by mapping the kami activities in table 1 to the stages of is outsourcing presented in sect. 2.2, a proposed kami framework for is outsourcing projects is derived and depicted in fig. 1 . cognisance is taken of the unique attributes of the is outsourcing lifecycle stages and arrows indicate the flow of the diagram. the pre-contract stage [23] of [42] an is outsourcing project is operationalised through considering the outsourcing strategy where an organisation reflects on the reason and scope for outsourcing, followed by a due diligence evaluating the commercial potential of the outsourcing project. as the organisation is still considering the viability of the is outsourcing project, the knowledge asset management activities only centre around high level strategic activities such as agreeing on the associated principles and governance for managing knowledge assets during the is outsourcing project and the identification of the accountable organisational structure, as well as the organisational sponsor. the scope of which knowledge assets will be impacted, which knowledge assets will be required for the project and what new knowledge assets will be created during the is outsourcing project, must also be considered during this strategising activity. once the organisation approves that the is outsourcing project should continue, the project proceeds to the contract stage [23] . during this stage the organisation appoints the outsource vendor, normally through a procurement process, and negotiate all aspects of the is outsourcing contract with the chosen outsource vendor. as it is fairly certain that the is outsourcing project will proceed, but the contract is just not signed yet, the knowledge asset management activities now move to the scope phase. during the scope phase a more detailed knowledge asset requirements analysis must be concluded now that the scope of the is outsourcing agreement is clear. the identification of knowledge assets impacted will also advise prioritisation as the is outsourcing project proceeds and with this scope in mind, a high level design of the knowledge asset management system can also be concluded. the scoping of knowledge asset management activities may kick off towards the end of the pre-contract stage, especially once the organisation has completed the due diligence as this activity informs the scope of the knowledge assets impacted. towards the end of the contract stage, the initial activities of the knowledge asset management implementation development phase may be initiated to ensure that the kami is prepared to kick off shortly after the outsourcing contract is signed. if the time between the signing of the contract and the means to capture knowledge assets is too long, knowledge assets may be jeopardised and even lost. these initial activities include the completion of the kami planning, as well as the initiation of a knowledge asset audit relevant to the scope of the is outsourcing agreement. once the organisation and the chosen outsource vendor signed the is outsourcing agreement, the post-contract stage is initiated, [23] focusing on the active management of the outsourcing agreement, the working relationships management and the performance management of the vendor. these activities take place within the agreed contract governance framework. from a kami perspective, focus is now on the organisation, building and storing knowledge assets. knowledge elicitation during this phase is key as the elicitation process may include outsource vendor resources. the definition and implementation of knowledge asset management processes and the training of all stakeholders are important activities during the kami development phase. the activities in the development phase are cyclical in nature as it has to be managed throughout the entire is outsourcing contract management stage and last for the duration of the contract term. once the is outsourcing contract reached its termination date, the organisation may choose to extend the contract, procure and appoint a new outsource vendor or ensure that the outsource vendor hands operations over to the organisations. for the first two options, the kami process will kick off again as the new extended scope has to be considered. in the instance where an outsource vendor must hand over to the organisation, the kami must also be institutionalised and handed over to the identified sponsors and organisational structures. this will ensure that the knowledge assets created during the is outsourcing project, are maintained and sustained. we applied the proposed kami framework in a proof of concept evaluation in an actual is outsourcing project. the application and subsequent findings of the kami framework are discussed in the next section. in order to do a proof of concept evaluation of the kami framework, we applied the kami framework to a real world is outsourcing project from an organisation operating in the information communication technology (ict) sector in south africa (sa). the organisation utilises an is multi-sourcing approach as a strategy with a significant number of is projects and is operational functions that are outsourced. the specific is outsourcing project used for the proof of concept evaluation was for the replacement of a large scale legacy system that was outsourced to two vendors, necessitating the appointment of a systems integrator. the scope of work outsourced included business process modelling, high level solution design and architecture, an implementation programme work plan and a data migration strategy. a programme director, an experienced, independent consultant, appointed by the organisation to work with both outsource vendors, systems integrator, as well as the organisation's is outsourcing project team members, applied the proposed kami framework to the project in order to establish its applicability to a real-world situation. we believed that such an experienced programme director overseeing a complex is outsourcing agreement will be in a position to provide valuable insight into the practical application of the proposed kami framework. as the is outsourcing project was in progress already i.e. in the post-contract phase, the programme director mapped the proposed kami framework to the project and could, in addition to providing feedback, also share lessons learnt in terms of the early stages of the is outsourcing project. the programme director provided comments based on each stage of the is outsourcing project and related kami framework activities. for the strategise phase as part of the pre-contract stage, the programme director indicated that by considering an overview of the knowledge asset management principles as well as high level scope, presents great advantages. market knowledge, knowledge of the strategic objectives informing the is outsourcing and knowledge about the operational benefits that will be derived from the outsourcing, are important. a key enabler identified by the programme director pointed to the outsource vendor knowledge of the client organisation. such knowledge ensures that engagement takes place at the correct organisational level, that the external skills from the outsource vendor complement existing skill sets and that emphasis is placed on the up-skilling of the client organisation in terms of managing an outsource partner. as knowledge exchange in the pre-contract phase of is outsourcing deals with the structure of the agreement, negotiation position knowledge and negotiation skills are key. in terms of requirements analysis at this stage of the project, identification and documentation of organizational intellectual property, knowledge of different outsource models, quantification of the business case for outsourcing and multiple benchmarks and performance measure identification present major knowledge exchanges. these particular knowledge exchanges are considered as part of the high level knowledge asset requirements analysis. in terms of the scope phase of the framework, the programme director highlighted knowledge exchange in terms of how the knowledge audit methodology procedure step attracted significant input. these included a knowledge repository of the client organisation business processes, knowledge of service level agreements, project planning, support constructs, scope of outsourcing, resource mobilization time and the definition of a model to retain key knowledge and skill in the organisation. due diligence, live demonstrations and presentations, research and intelligence re outsource vendors and the knowledge embedded in a lessons learnt repository, informed the initiative scoping procedure. the knowledge audit and initiative scoping steps were relevant in the handover between the post-contract and contract phases. prioritisation focuses on the understanding of the key objectives of knowledge transfer into measurable services and products in order to inform vendor selection. technology solution assessment in the contract phase pointed to the assessment of a solution benchmarking repository and an outcomes based assessment of the usability of the perceived end product. the programme director highlighted that the post-contract stage included multiple knowledge exchanges in terms of planning. these include knowledge of joint execution key drivers and processes, contract management, escalation paths and the planning of the work contribution of the outsourcing arrangement with appropriate stakeholders. a key step in the development phase is the knowledge elicitation procedure where knowledge is exchanged through integration sessions, knowledge sharing sessions and a culture of knowledge sharing is fostered. during the building procedure, traceability of knowledge elements, agreed way of work, organisational monitoring capabilities and joint delivery presented opportunities for knowledge exchange. the planning and knowledge elicitation procedures are positioned within the hand over between the contract and post-contract stages. the implement phase includes key knowledge exchange identified and points to conducting formal knowledge transfer workshops of learning that has taken place during the project in order to create a lessons learnt knowledge base. as this procedure is in the post-contract phase, formal knowledge transfer mechanisms from outsource vendor to client organisation must be facilitated and sufficient planning for hand over time must be allocated. with a strong focus on maintenance and support, knowledge transfer regarding the monitoring of the service level agreement, housekeeping and general operational tasks, and decommissioning of vendor resources must be enabled. a knowledge repository with all relevant solution information, decisions made during the project and risks identified and mitigated, including proper security controls, must be created. here well-defined quantifiable business benefits linked to well defined and measureable business processes, facilitation of proof of concept and demonstrations of the solution, as well as the creation of a common knowledge repository between the client organisation and outsource vendor as part of the outsourcing arrangement, are important knowledge exchanges. the programme director highlighted the utilisation of joint documentation responsibilities between the client organisation and outsource vendor. the kami activities in the implement phase of the kami framework are cyclical in nature and has to be managed throughout the entire period of is outsourcing contract execution. finally, in the contract ended/terminated stage, the programme director emphasised knowledge maintenance procedures in two important areas. firstly, the institutionalisation of appropriately transparent knowledge transfer channels, where client organisation stakeholders are kept up to date on the processes used, including the rationale for vendor selection and subsequent contract terms. secondly, the facilitation of focus group meetings where all key stakeholders are mandated to attend and participate with access to a central document repository to ensure transparency to all participants in the process. he also identified the unique contribution of communication and change management and indicated that it should not be limited to the final phase only. he stressed the value of having communication and change management at initiation of the outsourcing project through to the close out of the project. in addition, the senior programme director commented on the advantages of the kami framework based on his experience with large scale is outsourcing programmes and also made kami framework improvement suggestions. firstly, he reflected on the cyclical nature of the kami framework as he experienced that knowledge assets are often lost when an is outsourcing contract is terminated or ends. the cycles indicated in the kami framework ensures that the organisation considers to capture these knowledge assets before the outsource vendor exists. the programme director reflected on the management of 3 knowledge flows namely, new knowledge created by the joint is outsourcing team, organisation-to-vendor knowledge exchange, and vendor-toorganisation knowledge exchange. he believed that although it was implied in the kami framework, it will add value to make it explicit, especially as different mechanisms may be applied in order to manage these different knowledge flows. the programme director highlighted that he believes two distinct sets of knowledge assets must be managed: firstly, the knowledge assets created by the is outsourcing project related to the business rationale for using outsourcing, e.g. the implementation of a new software solution. secondly, he indicated that knowledge assets are also utilised in the management of the is outsourcing agreement, e.g. the project documentation, outsourcing process knowledge, negotiation guidelines, contract management knowledge etc. he mentioned that he believes it will add great value of the kami framework guided both of these knowledge asset sets. by including the second knowledge asset set, namely the is outsourcing project assets, it will enable the commercial team to negotiate and include key knowledge asset management principles in the commercial agreement with the outsource vendor. the programme director stressed that the fact that a contract which includes knowledge asset management clauses, will assist in enforcing the vendor to share knowledge as some contract payments may be linked to the delivery of knowledge sharing. lastly, he suggested that the inclusion of an agreement on a knowledge taxonomy related to the project within the context of the bigger organisation should form part of the strategise phase of kami. this will be important for the institutionalisation of the knowledge assets created during the is outsourcing project and when the assets have to be integrated into the organisational knowledge context. the implementation of knowledge management at an organisational level is a key enabler for managing the knowledge assets in the organisation during the implementation of an is outsourcing arrangement. in order to assist organisations embarking on is outsourcing to pro-actively manage their knowledge assets during the is outsourcing project, the aim of this study was to develop a kami framework. the kami framework was designed based on the analysis of multiple km frameworks and then evaluated by an experienced senior programme director. the objective with this proof of concept evaluation was to establish whether the kami framework was applicable to is outsourcing projects where multiple knowledge exchanges take place, namely; organisation-to-vendor knowledge exchange, vendor-to-organisation knowledge exchange and new knowledge that was created by the is outsourcing implementation project. the proof of concept evaluation was done through an interview with the senior programme director and the data regarding knowledge exchange in an is outsourcing project that he managed, was analysed and reported in the context of the kami framework. the data included comments regarding each of the stages of the is outsourcing lifecycle. it was established that due to the iterative nature of the kami framework, the activities were well suited to be applied in the context of an is outsourcing project. the kami framework captured the knowledge asset management activities well and the proposed activities aligned well to the execution of the is outsourcing stages. the programme director made suggestions for the improvement of the proposed kami framework and this may be included in future research. furthermore, the revised kami framework may be applied in an actual is outsourcing project to test the applicability of the framework in a real-world is outsourcing project scenario. in addition, a comparative study may be conducted between the kami framework and the iso standard for knowledge management systems (iso 30401:2018). outsourcing of information technology and business processes in poland: motivations and environmental factors enhancing the application and measurement of relationship quality in future it outsourcing studies outsourcing management: outsourcing services worldwide and in romania the influence of employee motivation on knowledge transfer knowledge transfer processes in it outsourcing relationships and their impact on shared knowledge and outsourcing performance knowledge management as a framework for understanding public sector outsourcing process management for knowledge work the client-vendor offshore relationship: success factors factors influencing knowledge transfer in onshore information systems outsourcing in ethiopia global outsourcing relationships and innovation: a conceptual framework and research propositions knowledge asset management pertinent to information systems outsourcing knowledge management in client-vendor partnerships an architectural design for knowledge asset management system strategic knowledge management and enterprise social media analyzing it maintenance outsourcing decision from a knowledge management perspective knowledge sharing and cooperation in outsourcing projects: a game theory analysis critical team-level success factors of offshore outsourced projects: a knowledge integration perspective modularity and innovation in knowledge-intensive business services: it outsourcing in germany and the constructing customised models and providing information for it outsourcing decisions intelligent it outsourcing: eight building blocks to success gartner: marketplace realities in strategic outsourcing it, software and services: outsourcing & offshoring a model for studying it outsourcing relationships outsourcing relationships: the contract as it governance tool a holistic view of the knowledge life cycle: the knowledge management cycle (kmc) model. electron information systems outsourcing drivers and service delivery of commercial banks in kenya framework for managing shared knowledge in an information systems outsourcing context international business and mis approaches to multinational organisational research: the cases of knowledge transfer and it workforce outsourcing design-based research methods for studying learning in context: introduction design-based research and technology-enhanced learning environments design research: what we learn when we engage in design towards a practical guide for developing knowledge management systems in small organizations improving knowledge management practices in the south african healthcare system the emergence of knowledge management bridging knowledge management life cycle theory and practice knowledge management frameworks: a review of conceptual foundations and a kmf for it-based organizations. strengthening the role of ict in development study of knowledge management frameworks analysis of the south african surrogate for a tmrp-6 anti-tank mine an agile methodology for implementing knowledge management systems: a case study in component-based software engineering methodology for the implementation of knowledge management systems 2.0. a case study in an oil and gas company evolutionary concepts in end user productivity and performance: applications for organizational progress a framework and methodology for knowledge management system implementation key: cord-302384-gljfslhs authors: al-hanawi, mohammed k.; angawi, khadijah; alshareef, noor; qattan, ameerah m. n.; helmy, hoda z.; abudawood, yasmin; alqurashi, mohammed; kattan, waleed m.; kadasah, nasser akeil; chirwa, gowokani chijere; alsharqi, omar title: knowledge, attitude and practice toward covid-19 among the public in the kingdom of saudi arabia: a cross-sectional study date: 2020-05-27 journal: front public health doi: 10.3389/fpubh.2020.00217 sha: doc_id: 302384 cord_uid: gljfslhs background: saudi arabia has taken unprecedented and stringent preventive and precautionary measures against covid-19 to control its spread, safeguard citizens and ensure their well-being. public adherence to preventive measures is influenced by their knowledge and attitude toward covid-19. this study investigated the knowledge, attitudes, and practices of the saudi public, toward covid-19, during the pandemic. methods: this is a cross-sectional study, using data collected via an online self-reported questionnaire, from 3,388 participants. to assess the differences in mean scores, and identify factors associated with knowledge, attitudes, and practices toward covid-19, the data were run through univariate and multivariable regression analyses, respectively. results: the majority of the study participants were knowledgeable about covid-19. the mean covid-19 knowledge score was 17.96 (sd = 2.24, range: 3–22), indicating a high level of knowledge. the mean score for attitude was 28.23 (sd = 2.76, range: 6–30), indicating optimistic attitudes. the mean score for practices was 4.34 (sd = 0.87, range: 0–5), indicating good practices. however, the results showed that men have less knowledge, less optimistic attitudes, and less good practice toward covid-19, than women. we also found that older adults are likely to have better knowledge and practices, than younger people. conclusions: our finding suggests that targeted health education interventions should be directed to this particular vulnerable population, who may be at increased risk of contracting covid-19. for example, covid-19 knowledge may increase significantly if health education programs are specifically targeted at men. coronavirus disease 2019 is defined as an illness caused by a novel coronavirus, now called severe acute respiratory syndrome coronavirus 2 (sars-cov-2; formerly called 2019-ncov). covid-19 is an emerging respiratory infection that was first discovered in december 2019, in wuhan city, hubei province, china (1) . sars-cov-2 belongs to the larger family of ribonucleic acid (rna) viruses, leading to infections, from the common cold, to more serious diseases, such as middle east respiratory syndrome (mers-cov) and severe acute respiratory syndrome (sars-cov) (2) . the main symptoms of covid-19 have been identified as fever, dry cough, fatigue, myalgia, shortness of breath, and dyspnoea (3, 4) . covid-19 is characterized by rapid transmission, and can occur by close contact with an infected person (5) (6) (7) (8) (9) . the details on the disease are evolving. as such, this may not be the only way the transmission is occurring. covid-19 has spread widely and rapidly, from wuhan city, to other parts of the world, threatening the lives of many people (10) . by the end of january 2020, the world health organization (who) announced a public health emergency of international concern and called for the collaborative effort of all countries, to prevent its rapid spread. later, the who declared covid-19 a "global pandemic" (11) . following the who declaration, countries around the globe, including the kingdom of saudi arabia (ksa), have been leaning on response plans to respond to the pandemic and contain the virus. following the confirmation of its first case of covid-19, on monday 2 march 2020, the saudi government has been vigilantly monitoring the situation and developing countryspecific measures that are in line with the who guidelines in dealing with the outbreak (12) . these includes suspending all inbounds and outbounds flights, closing all malls and shops in the country, except pharmacies and grocery stores, and closing down schools and universities. umrah visas have been suspended, as have prayers at mosques, including the two holy mosques in mekkah and almadina. on 24 march 2020, the government imposed a nationwide curfew to restrict people movements for most of the day hours. despite the unprecedented national measures in combating the outbreak, the success or failure of these efforts is largely dependent on public behavior. specifically, public adherence to preventive measures established by the government is of prime importance to prevent the spread of the disease. adherence is likely to be influenced by the public's knowledge and attitudes toward covid-19. evidence shows that public knowledge is important in tackling pandemics (13, 14) . by assessing public awareness and knowledge about the coronavirus, deeper insights into existing public perception and practices can be gained, thereby helping to identify attributes that influence the public in adopting healthy practices and responsive behavior (15) . assessing public knowledge is also important in identifying gaps and strengthening ongoing prevention efforts. thus, this study aims to investigate the knowledge, attitudes and practices (kap) of ksa residents, toward covid-19 during the pandemic spike. to the researchers' knowledge, this is the first study to investigate covid-19 kap, and associated sociodemographic characteristics among the general population of the ksa. the findings of this study are expected to provide useful information to policymakers, about kap among the saudi population, at this critical time. the findings may also inform public health officials on further public health interventions, awareness, and policy improvements pertaining to the covid-19 outbreak. this cross-sectional study was conducted among the general population of saudi arabia, from 20 march 2020, to 24 march 2020. given the social distancing (physical distancing) measures and restricted movement and lockdowns, data were collected online, via a self-reported questionnaire, using surveymonkey. given the high internet usage among people in the ksa, a link to the survey was distributed to respondents, via twitter and whatsapp groups. the link was also posted on the king abdulaziz university website. the larger the target sample size, the higher the external validity and the greater the generalizability of the study (16) . this study aimed to maximize reach and gather data from as many respondents as possible. according to the latest ksa census, saudi arabia has a population of 34,218,169 (17) . the representative target sample size needed, to achieve the study objectives and sufficient statistical power, was calculated with a sample size calculator (18) . the sample size calculator arrived at 1,037 participants, using a margin of error of ±4%, a confidence level of 99%, a 50% response distribution, and 34,218,169 people. the self-reported questionnaire was developed by the authors, according to guidelines for the community of covid-19, by the centers for disease control and prevention (cdc) (19) . the questionnaire was conducted in arabic language. it was initially drafted in english by h.z.h., and y.a., and was translated from english to arabic by m.k.a and m.a. the questionnaire was translated then back to english by n.a and w.k to ensure the meaning of the content. on the first page of the online questionnaire, respondents were clearly informed about the background and objectives of the study. respondents were informed that they were free to withdraw at any time, without giving a reason, and that all information and opinions provided would be anonymous and confidential. respondents living in saudi arabia, aged 18 years or older, understand the content of the questionnaire, and agree to participate in the study were instructed to complete the questionnaire. online informed consent were obtained before proceeding with the questionnaire. the questionnaire consisted of four primary sections. the first section gathered information on respondents' sociodemographic characteristics, including age, gender, marital status, education level, work status, region of residence, and income level. the second section assessed participants' knowledge of covid-19. this section included 22 items on modes of transmission, clinical symptoms, treatment, risk groups, isolation, prevention and control. the third section assessed participants' attitudes toward covid-19, using a five-point likert scale. for each of six statements, respondents were asked to state their level of agreement, from "strongly disagree, " "disagree, " "undecided, " "agree, " or "strongly agree." the final section of the questionnaire assessed the respondents' practices. this section consisted of five questions related to practices and behavior, including (a) going to social events with large numbers of people, (b) going to crowded places, (c) avoiding cultural behaviors, such as shaking hands (d) practicing social distancing, (e) washing hands after sneezing, coughing, nose-blowing, and, recently, being in a public place. for sociodemographic variables, gender was coded as one for men, and zero for women. the age variable was divided into categories: 18-29 (reference category), 30-39, 40-49, 50-59, and ≥60. marital status was captured as binary, and a value of one was used for marriage and zero for otherwise. education was categorized into high school or below (reference category), college/university degree, and postgraduate degree. work status was broken down into government employee (reference category), non-government employee, retiree, selfemployed, and unemployed. monthly income (saudi riyal, sr 1 = usd 0.27) was divided into eight categories: <sr 3,000 (reference category); sr 3,000 to <5,000, sr 5,000 to <7,000, sr 7,000 to <10,000, sr 10,000 to <15,000, sr 15,000 to <20,000, sr 20,000 to <30,000, and sr 30,000 or more. we also controlled for the 13 administrative regions: almadina almonawra, albaha, aljouf/quriat, aseer/bisha, eastern region, haiel, jazan, najran, northern borders, qaseem, riyadh, tabouk, and the western region. respondents were asked to respond to knowledge items as either true or false, with an additional "don't know" option. incorrect or uncertain (don't know) responses were given a score of zero, and correct answers were assigned a score of one. the total score for knowledge ranged from zero to 22, with high scores indicating better knowledge of covid-19. items were evaluated for internal reliability, using cronbach's α. cronbach's alpha coefficient was 0.70, indicating internal reliability (20) . in the section on attitudes, scores were calculated based on the respondents' answers to each attitudinal statement, 1 = strongly disagree, 2 = disagree, 3 = undecided, 4 = agree, and 5 = strongly agree. scores were calculated by averaging respondents' answers to the six statements. total scores ranged from six to 30, with high scores indicating positive attitudes. the likert scales were assessed for internal reliability, using cronbach's α. cronbach's alpha coefficient was 0.81, indicating internal reliability. in the section on practices, respondents were asked to respond "yes" or "no" to the items. a score of one was given to answers that reflected good practice, and a score of zero was given for answers that reflected bad practice. the total score ranged from zero to five, with high scores indicating better practices. this study employed primarily univariate and multivariable regression data analyses. univariate analysis was used to tabulate the frequency of social and demographic statistics. one-way analysis of variance (anova) was used to assess differences in mean values for kap scores. because the scores were continuous, the overall mean differences were estimated using a bartlett test (21, 22) . a multivariable linear regression analysis was performed, to identify factors related to knowledge, attitudes, and practice. all analyses were conducted using stata software (statacorp lp, texas, usa). all procedures performed in this study, involving human participants, complied with the institutional and/or national research committee ethical standards, and the 1964 helsinki declaration and subsequent amendments or equivalent ethical standards. the study was designed and conducted in accordance with the ethical principles established by king abdulaziz university. therefore, ethical approval was obtained from the biomedical ethics research committee, faculty of medicine, king abdulaziz university . a total of 3,427 participants completed the questionnaire. after excluding 39 respondents who reported living outside the ksa, the final sample consisted of 3,388 participants. table 1 shows the social and demographic characteristics of the study participants. as shown in table 1 , the mean covid-19 knowledge score was 17.96 (sd = 2.24, range: 3-22), and the overall accuracy rate for the knowledge test was 81.64% (17.96/22 * 100). the mean attitude score for covid-19 was 28.23 (sd = 2.76, range: 6-30), indicating positive attitudes. the mean score for practices for covid-19 was 4.34 (sd = 0.87, range: 0-5), indicating good practices. of the total sample, 1966 (58.03%) were women, and 1422 (41.97%) were men. the majority of the sample (57.73%) were between the ages of 18 and 39. of the participants, 2,149 were married (63.43%) and 1,239 were unmarried (36.57%). more than half of the sample (56.20%) had a college or university degree. respondents were grouped according to monthly income, with 846 (24.97%) in the < sr 3000 group, and 246 (7.26%) in the ≥ sr 30,000 group. in terms of work status, 1,073 (31.76%) were unemployed, and 314 (9.27%) were retired. tables 2-4 show the responses to items related to kap towards covid-19. we also assessed the level of kap, across the various income groups. figures 1-3 show the results. figure 1 shows that the covid-19 knowledge score increases with income. the lowest score was for respondents in the lowincome category, < sr 3000, and the highest score was for respondents with an income of sr 20,000 to < 30,000. for attitudes, figure 2 shows that there were no discernible patterns across income groups. furthermore, with regard to practices, there was little variation between income groups, as shown in figure 3 . looking at the univariate statistics for each variable of interest, we took another step to assess the difference in the scores for kap. table 5 shows the results. as shown in table 5 , all scores for kap, were statistically different at the 1% significance level, for all age and income groups. although, across genders, there was a statistically significant difference in attitudes and practices scores, there was no difference in knowledge scores. although assessments showed that knowledge scores were not statistically different between regions, attitudes and practices scores were significantly different at the 1% level. apart from the univariate and non-parametric analyses performed in previous sections, we also focused on regression analysis. scores were logged for all variables, and interpreted using ordinary least squares (ols). increased scores imply increased knowledge, practices, and attitudes. the results are shown in table 6 . table 6 shows that, for knowledge of covid-19, age groups 30-39, 40-49, 50-59 and ≥60, are more knowledgeable about covid-19 than the reference group (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) . all variables for age groups 30-39 (β = 0.047; p < 0.001), 40-49 (β = 0.041; p < 0.001), 50-59 (β = 0.057; p < 0.001) and ≥60 (β = 0.051; p < 0.001), are statistically significant at the 1% level. however, attitudes follow a different trend. only the age group 50-59 (β = −0.021; p < 0.001) is significantly different from baseline. in practices for covid-19, age groups 30-39 (β = 0.039; p < 0.001), 40-49 (β = 0.033; p < 0.05), and 50-59 (β = 0.051; p < 0.001), are associated with good practices. regarding gender, the results indicate that, compared to women, men have lower knowledge (β = −0.018; p < 0.001), lower positive attitudes (β = −0.018; p < 0.001), and few good practices for covid-19 (β = −0.064; p < 0.001). no difference in kap toward covid-19 was observed by marital status. however, the relationships between income, region, education, and variables of interest, are heterogeneous, suggesting substantial differences. we also examined the association between kap scores. because all were logs, the interpretation would be akin to the elasticity. therefore, table 6 shows that every increase of 1% in knowledge score is associated an increase in attitude and practices scores, of 0.095 and 0.16, respectively. covid-19 is an emerging infectious disease that poses a significant threat to public health. given the serious threats imposed by covid-19 and the absence of a covid-19 vaccine, preventive measures play an essential role in reducing infection rates and controlling the spread of the disease. this indicates the necessity of public adherence to preventive and control measures, which is affected by their knowledge, attitudes, and practices (23, 24) . in our study, the high rate of correct answers to knowledge-related questions among participants, was not surprising. this may be due to the characteristics of the sample, as 84% had a college or university degree, or above, and 70% were over 30 years old. it may also be due to the distribution of the questionnaire, amid the covid-19 outbreak. in that time, people may have gained awareness and knowledge about the disease and its transmission, via television, news and media platforms, to protect themselves and their families. the positive association found between knowledge, and educational background and age, supports our claim. most of the participants in our study (98%) were aware of the clinical symptoms, and 96% knew that there is no clinically approved treatment for covid-19 as of the date of this manuscript. viral infections have been documented to be highly contagious among people in close proximity (19) . however, approximately half of the respondents were unaware that sars-cov-2 could spread from person-to-person in close proximity. it was also evident that the current general population (44%) had little knowledge of when and whom wearing masks to prevent infection. according to the who and the cdc, faces mask should only be worn by those who are sick or caring for people suspected of having covid-19 (9, 19) . these findings highlight the need to continue to encourage and emphasize maintaining social distancing, as a means of preventing the spread of the virus. it is important to note that there has been a great deal of efforts at all levels by the government, including public awareness campaigns. the saudi arabian ministry of health (moh) has conducted an intensive awareness campaign, communicated via its website, television and various social media. the moh has produced a guide to covid-19, to provide residents with facts and precautionary messages in more than 10 languages. the moh also works with the public and the media, especially via social media platforms. these early actions on engaging the public in prevention and control measures, as well as efforts to combat rumors and misinformation, have been greatly expanded (25). it is worth noting that the ksa is in the unique position of having dealt successfully with two outbreaks of viral origin, of related viruses (26) (27) (28) (29) (30) . this unique experience has helped the government in taking prompt response and precautionary measures against covid-19 to control its spread. significant predictors of participant knowledge in this study were age, gender, educational level, and income level. this finding is supported by other studies that have found that older, female, and more educated respondents are more knowledgeable about emerging communicable diseases (23, 31) . we also found that high income earners are more knowledgeable about covid-19. education, age, and income have been documented to be highly relevant to knowledge (32) . our findings suggest that greater emphasis should be placed on mass media, to target low-income, low-educated, young people, and men to improve public knowledge on the covid pandemic, through awarenessraising interventions. concerning attitudes, participants showed a positive and optimistic attitude toward covid-19. approximately 94% concur that the virus can be successfully controlled, and 97% are convinced that the saudi government will control the pandemic. positive attitudes and high confidence in the control of covid-19 can be explained by the government's unprecedented actions and prompt response in taking stringent control and precautionary measures against covid-19, to safeguard citizens and ensure their well-being. these measures include the lockdown, and the suspension of all domestic and international flights, prayer at mosques, schools and universities, and the national curfew imposed on citizens. this finding is consistent with a recent study conducted in china, where the majority of participants were convinced that the disease is curable and that their country will combat the disease (33) . however, these results contrast with other findings that suggest people tend to express negative emotions, such as anxiety and panic, during a pandemic that could affect their attitude (34) . nevertheless, our results show that the participants' high knowledge of covid-19 translates into good and safe practices, during the covid-19 pandemic, which suggests that the practices of saudi residents are very cautious. almost 95% of respondents refrained from attending social events, 94% avoided crowded places, and 88% avoided shaking hands. respondents adopted good and safe practices, as a result of saudi arabia's health authorities providing education and outreach materials, to increase public understating of the disease, and influence behavioral change. finally, the study findings may be useful to inform policymakers and healthcare professionals, on further public health interventions, awareness-raising, policies, and health education programs. men were significantly less likely to have knowledge, optimistic attitudes, and appropriate or safe practices toward covid-19. these findings are consistent with other studies showing that, in response to sars and mers, men were significantly less likely take preventive and protective measures than women (24, 35, 36) . our finding suggests that targeted health education interventions should be directed to this particular vulnerable population at high risk of contracting covid-19. for example, covid-19 knowledge may increase significantly, if health education programs are specifically targeted at men. health information can be sent to women (wives, sisters, mothers) who live with men, which may influence their practices, as suggested by a study in hong kong (35) . to the researchers' knowledge, this is the first study to investigate kap toward covid-19, in the general population of the ksa. data collection took place 2 weeks after the ksa confirmed its first covid-19 case. therefore, the prompt results may help health authorities to plan preventive strategies for future events. however, in interpreting the results of this study, some limitations should be considered. data used in the analysis of this study were self-reported, which might suffer from reporting bias. future research might employ administrative data to address this issue. furthermore, there may be some endogenous variables: general attitude and expectation from the government, personal hygiene, for example. however, even if we were to undertake causality analysis, we could not be in a position to do proper econometric identification because the data we used could not have a valid instrument to eliminate the endogeneity. additionally, community-based national sampling surveys were not feasible during this particular period. as such, data were collected online, through self-reported questionnaires, depending on the authors' networks. therefore, the majority of the respondents were in the western region, where most of the authors come from. further research should cover the perceptions of all regions of the country. finally, this study did not address causation. therefore, the regression results should be interpreted as relevant, as some variables may be endogenous. that said, the implication is for the objective of future research to assess whether there is a relationship between covid-19 knowledge and mortality, household consumption patterns, and the demand for unprescribed flue medication. this is the first study to investigate kap for the covid-19 outbreak, among the general population of saudi arabia. our findings suggest that saudi residents, especially women, have good knowledge, positive attitudes, and good practices toward covid-19. knowledge of the disease is considered the first stepping stone to any health education activity that is implemented. knowing the causes and transmission sources of a disease, increases the likelihood that people will become more aware of the spread of communicable diseases, and of the preventive measures to slow transmission. the results of this study suggest that more emphasis should be placed on less educated, lower income, and men. the findings may help policymakers identify the target populations, for covid-19 prevention and health education. the datasets generated and/or analyzed during the current study are not publicly available due to privacy and confidentiality agreements as well as other restrictions, but are available from the corresponding author (mohammed k. al-hanawi) on reasonable request. all authors made substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data, took part in drafting the article or revising it critically for important intellectual content, gave final approval of the version to be published, and agreed to be accountable for all aspects of the work. a new coronavirus associated with human respiratory disease in china a pneumonia outbreak associated with a new coronavirus of probable bat origin a familial cluster of pneumonia associated with the 2019 novel coronavirus indicating person-to-person transmission: a study of a family cluster covid-19 infection: origin, transmission, and characteristics of human coronaviruses importation and human-to-human transmission of a novel coronavirus in vietnam china coronavirus: cases surge as official admits human to human transmission early transmission dynamics in wuhan, china, of novel coronavirus-infected pneumonia who. coronaviruses (covid-19) 2020. available online at clinical characteristics of 138 hospitalized patients with 2019 novel coronavirus-infected pneumonia in wuhan prevention and control strategies for emergency, limited-term, and elective operations in pediatric surgery during the epidemic period of covid-19 saudi arabia announces first case of coronavirus explaining socioeconomicrelated inequality in knowledge about socio-economic inequality in comprehensive knowledge about hiv in malawi community perception and risk reduction practices toward malaria and dengue: a mixedmethod study in slums of chetla applied business research: qualitative and quantitative methods authority for statistics: the kingdom of saudi arabia 2020 size calculator 2020 cdc. centers for disease control and prevention: coronavirus (covid-19) 2020. available online at evidence based physical therapy a general anova method for robust tests of additive models for variances a default bayesian hypothesis test for anova designs is the saudi public aware of middle east respiratory syndrome? attitude and practice (kap) study about middle east respiratory syndrome coronavirus (mers-cov) among population in saudi arabia saudi arabia join forces to fight covid-19 nationally, regionally and globally 2020 preventive strategies to keep saudi arabia sars-free awareness, attitudes, and practices related to coronavirus pandemic among public in saudi arabia a lesson learned from middle east respiratory syndrome (mers) in saudi arabia public awareness of coronavirus in al-jouf region, saudi arabia knowledge of middle east respiratory syndrome coronavirus (mers-cov) and its management: a survey among saudi people in taif; kingdom of saudi arabia mers-cov infection: mind the public knowledge gap determinants of health knowledge: an investigation of age, gender, abilities, personality, and interests knowledge, attitudes, and practices towards covid-19 among chinese residents during the rapid rise period of the covid-19 outbreak: a quick online crosssectional survey the public's response to severe acute respiratory syndrome in toronto and the united states the impact of community psychological responses on outbreak control for severe acute respiratory syndrome in hong kong a meta-analysis of the association between gender and protective behaviors in response to respiratory epidemics and pandemics the authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.copyright © 2020 al-hanawi, angawi, alshareef, qattan, helmy, abudawood, alqurashi, kattan, kadasah, chirwa and alsharqi. this is an open-access article distributed under the terms of the creative commons attribution license (cc by). the use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. no use, distribution or reproduction is permitted which does not comply with these terms. key: cord-258125-2kkqqgop authors: wagner, peter title: knowing how to act well in time date: 2020-08-25 journal: j bioeth inq doi: 10.1007/s11673-020-10018-7 sha: doc_id: 258125 cord_uid: 2kkqqgop numerous scholars in the social sciences and humanities have speedily analysed and interpreted the covid-19-induced social and political crisis. while the commitment to address an urgent topic is to be appreciated, this article suggests that the combination of confidence in the applicability of one’s tools and belief in the certainty of the available knowledge can be counter-productive in the face of a phenomenon that in significant respects is unprecedented. starting out from the plurality of forms of knowledge that are mobilized to analyse covid-19 and its consequences as well as the lack of any clearly hegemonic knowledge, the article tries to understand how a limited convergence in the politico-medical responses to the crisis emerged, and speculates on what would have happened if this had not been the case. in conclusion, it is argued that this pandemic demands a greater awareness of the uncertainty of our knowledge and of the consequences of our actions, both in terms of being situated in time and of aiming at timeliness. abstract numerous scholars in the social sciences and humanities have speedily analysed and interpreted the covid-19-induced social and political crisis. while the commitment to address an urgent topic is to be appreciated, this article suggests that the combination of confidence in the applicability of one's tools and belief in the certainty of the available knowledge can be counterproductive in the face of a phenomenon that in significant respects is unprecedented. starting out from the plurality of forms of knowledge that are mobilized to analyse covid-19 and its consequences as well as the lack of any clearly hegemonic knowledge, the article tries to understand how a limited convergence in the politico-medical responses to the crisis emerged, and speculates on what would have happened if this had not been the case. in conclusion, it is argued that this pandemic demands a greater awareness of the uncertainty of our knowledge and of the consequences of our actions, both in terms of being situated in time and of aiming at timeliness. keywords action . covid-19 . democracy . knowledge . pandemic . technocracy . time . uncertainty there is no shortage of analyses and interpretations of the crisis that the covid-19 has inflicted on humankind and of its consequences. as it seems, many of my colleagues in the social sciences and humanities only had to take out their toolkit and put it to use on this new phenomenon. at times, one could even have the impression that they were waiting for something like this to happen for them to rush to go public with their analyses and proposals. characteristically, the interpretations of the crisis come in two forms. the optimists hold that humankind will grow through this experience. the world after covid-19 will show more solidarity, moral commitment, and concern for the common good. the era of unrestrained profit-making, boundless pleasure-seeking, and selfish pursuit of one's goals without concern for others will come to an end. these observers call for action to grasp the unique occasion for creating a better world, unique because, for once, the world is already moving in that direction. in turn, the critics observe a crisis-driven move to authoritarianism, to new nationalism and restrictions to movement, to technocracy with political decisions based on expertise rather than deliberation, in short, the return to disciplinary society combined with isolation of individuals and loss of sociality. the short era of liberal democracy and convivial civil society during which the power of dominating groups could effectively be limited will come to an end. these observers call, if they do, for resistance, and their view is marked by suspicion and denunciation of the politicians and their advisors whom we see and hear every day now. faced with this avalanche of striking analyses, i have asked myself: why do i not have anything to contribute to this debate? it's not that i lack the impetus. like many others, i have the impression to experience a sociopolitical event of possibly greater significance than any other during my lifetime. and as someone who studies society, history, and politics, i feel i should have something to say on this event. and if so, i should say so loudly, to be heard. so, why don't i? helped by the postponement of other commitments due to the lockdown, at the time of writing (mid-april 2020) already in its fifth week in spain where i live, i collect as widely as possible information on the ongoing events but also on their background and their conditions. thus, despite not having specific medical expertise, i do not feel that i know too little-too little in the sense of knowing so much less than others that i should keep my mouth shut. rather, i feel that we all know too littletoo little to be certain to act well, as far as the politicians and medical advisors are concerned, and too little to understand well what the appropriate action is, as concerns all others, for most of us. this lack may be exactly that which is most characteristic of the current situation. covid-19 emerged at a moment when, despite evident and accumulating signs to the contrary, the view was still widespread that we have all the knowledge we need and all the required understanding for necessary action; that things are basically under control or can at least be brought under control; and that this control emerges as the aggregate outcome of uncoordinated action rather than through communication and coordination among those affected and involved. this is a view that our current "modernity" is largely built on. covid-19 shakes the building and imposes a reflection on our assumptions about knowledge, action, and time. what follows is a small step in that direction. epidemiology and virology, so i learnt quickly, were expecting something like this to happen. they have accumulated knowledge of considerable quality and coherence, dating back at least to the influenza pandemic of 1918-1920 (still known as the spanish flu; a century ago, such an inadequate naming could still be made to stick) and building up in concise terms with hiv, sars, h1n1, ebola, and mers across the past few decades. the problem, though, is that every virus is specific and the behavioural properties of this one were unknown and are still not sufficiently known. they can be understood only through observation, after it already emerged. while comparing covid-19 with its predecessors could usefully guide research, drawing conclusions from quickly made comparisons could easily mislead-or delay-remedial action. in the next step, the gradually increasing knowledge about the virus is confronted with two other kinds of knowledge. on the one side, techniques of mathematical modelling have considerably developed in recent time. they are at the forefront of attempts at trying to understand uncertain futures. they have a prominent place in the deliberations of the international panel on climate change (ipcc), and for good reason, since without them little would be understood at all. but public debate-and already preceding it, public science communication-tends to forget about the degree of uncertainty. with regard to climate change, a reasonable maxim for action can rather easily be devised: it is better to err on the side of prudence. with regard to covid-19, conclusions are less straightforward. among other institutions, the institute for health metrics and evaluation (ihme) at the university of washington has become widely known for its predictions of covid-19 infections and deaths in the united states and elsewhere. reasonably, the researchers steadily update their findings-and their modelling-by introducing new data. as a consequence, though, their predictions are corrected day by day, and they change by considerable margins. rather than orienting public debate and action, their effect is disorientation. it might be advisable to refrain from publishing these figures-even though this would less be a call on the researchers, who will wish to enhance timely scholarly collaboration, rather than on the media, which at the very least could underline the extremely fragile nature of this knowledge. on the other side, neither the virus nor mathematics know anything about institutions. but the number of medical doctors, nurses, intensive care units, personal protection equipment, and their distribution and availability across space and time is crucial for the development of the pandemic and for the chances of slowing it down or stopping it. this is knowledge that public health research can provide, but the recommendations that follow from it may be quite different from those that would be arrived at when relying only on epidemiology, virology, and mathematical modelling. there is evidence that such a knowledge clash occurred over covid-19, more pronounced in the early weeks of the pandemic, and in some countries, in particular in the united kingdom and sweden, with more consequences than in others. governments all across the planet claim that their covid-19 policies are "science-based." heads of government and ministers of health appear always accompanied by leading medical experts in their press conferences. some observers have seen this as a sign of an emerging new technocracy, with expertise overriding democratic deliberation. given the multiple kinds of knowledge, however, this seems far from the case. let us look at two earlier political uses of scientific expertise in comparison. after the second world war, the slogan about the "civilian use of nuclear energy" promised a radiant future based on an endless supply of cheap energy and, consequently, infinite growth of the wealth of all nations with guaranteed satisfaction of every material need. science seemed truly to have become the "endless frontier." it took considerable time and effort to change public debate by underlining not only the danger of operating nuclear power stations but also the hubris of claiming to control nuclear waste sites over millennia to come. this was gradually achieved by the late 1970s, and the transformation of the debate was not least due to the elaboration of counter-expertise of such a scholarly quality that it could no longer be easily rejected or ignored. at about the same time, the late 1970s, a shift in hegemony occurred within the economic sciences. as keynesian demand management seemed to have failed to effectively respond to the economic problems of the early 1970s, a return to more conventional versions of neo-classical economics happened, which in policy terms demanded the reduction of public debt, of labour cost, and of taxation of profits, supposedly to unleash again the beneficial workings of markets. the new policies clearly benefitted certain business interests and worked against the less well-off in society. in this light, it remains a bit of an enigma why also centre-left (social democrat) politicians complied with this new doctrine, but what we may call epistemic domination certainly played a role: the economic sciences claimed to have understood why earlier policies failed and, based on these insights, arrived at an apparently clear view of the unalterable laws of market economies. in this case, there was a preceding "counter-expertise," but it was discredited by having failed to avert crises. in both of these cases, therefore, there was a monopoly or hegemony of expertise, and the recommendations following from such expertise served powerful interests in society, served (parts of) the dominating groups, as one used to say. the critical theories of technocracy, as they were elaborated during the 1960s and 1970s, were based on observations of such a situation. but none of this is characteristic of the present moment, for at least two reasons. first, whatever there is in terms of hegemonic expertise is of a different nature. one may say that, after all, most governments follow the guidelines for dealing with pandemics that had been elaborated by the world health organisation (who) and the respective national institutes of health. but in contrast to the physics of nuclear energy and the claims of neo-classical economics, the knowledge of a pandemic requires detailed and nuanced empirical observation of the specificities of the situation. as said above, this kind of knowledge only becameand still becomes-gradually available. furthermore, this knowledge is then liable to somewhat different interpretations by epidemiology, virology, modelling, and public health. at the moment of writing, one can say that there has been a considerable, even though in some settings slow, process of convergence of knowledge-based policy recommendations over the first few months of the pandemic. this convergence may even be understood, optimistically, as an indicator of solidarity across borders, or some understanding of a common problem and the need to act in common. but such convergence is not yet assured for the further course of the pandemic and the measures for "deescalation." this is so because, second, the already plural set of bio-medical expertise, broadly understood, meets countervailing expertise from the social sciences-economics, sociology, political science, psychology-as soon as policy decisions need to be taken. this is not to say that truly scholarly knowledge from these disciplines was present at the beginning, not at all. but politicians have a kind of "practical expertise" in these matters. they have a sense of their dependence on the mood in circles of society: business circles, the media, social movements, and others, varying with the political parties. this "knowledge" tells them what they can do and not do. from this observation, two questions arise, one asking about the decisions regarding the past two months and one asking about the decisions to be taken during the next, say, half a year. the striking answer to the first question is that politicians have done what they knew could not be done, namely ordering what is now called a "lockdown" of economy and society. the two societies that have long (though in recent years much less) been seen as the standard-bearers of liberalism, the united kingdom and the united states, toyed for some time with taking another route, led by governments that incarnate the doubt about institutional collective action that has become more widespread across the past three decades in general (more surprisingly, and probably for somewhat different reasons, sweden did so as well). this initial divergence permits us to speculate about what would have happened if this laissez aller had become the dominant attitude. clearly, many more people would have died, and at some point "herd immunity" would possibly have been reached-from all we know now: very little and after many victims. the pandemic would have been described as a natural catastrophe, like a giant earthquake or a tsunami. the governments would have regretted the "human losses" but argued that there was nothing that could be done about them-or at least nothing that could reasonably be done, given that different concerns had to be balanced. such an attitude was not as widely off the mark as it seems in the current moment, since concerns about restrictions of civil liberties and human rights have been and keep being voiced in many countries. we will know better only much later, but it is possible that only the emerging concise view of the who at the core of a well-networked global epistemic community in health studies prevented this option from being pursued and forced initially deviating governments to fall in line. carrying the speculation one step further, it is likely that we would talk very differently about covid-19 if that road had been taken. the pandemic would have been placed in the line, sketched above, that leads from the "spanish flu" to more recent spreads of viruses and just be compared in quantitative terms: number of infections and deaths, spatial diffusion, time until development of a vaccine or a treatment. there would have been some debate about social consequences, such as the sudden loss of the older generation, comparable with the debate about hiv/aids and changes in sexual behaviour. this would have been a "crisis," to be sure, but one that would have been cognitively domesticated by analysing it as something that had already happened before, only in describably different form. if so, then it is the extended "lockdown" that marks the difference, because of the sudden and radical halt to most economic, social, and cultural activities. it is the reason why we talk about covid-19 as potentially leading to a radical social transformation in a way in which we did not do about any of the other diseases mentioned above, except in some limited respects for hiv/aids. this halt, though, is nothing like an earthquake or a tsunami. it has been decided by governments and implemented by public administration, even supported by the deployment of the states' monopoly of legitimate violence. at times, the current situation is compared to a war. but recent experience of extended warfare, in particular the two world wars of the twentieth century, meant the total mobilization of society for the purpose of winning the war. currently, some segments of society are extremely highly mobilized, most importantly health workers, but most of society has been forced into de-mobilization. this is what is unprecedented and appears to us as a harbinger of radical change. since the financial crisis of 2008, critical political sociology has re-focused its attention on the relation between capitalism and democracy. one influential interpretation holds that governments today have to address two constituencies: the people in the traditional sense of the citizenry, expressing itself through elections and opinion polls, on the one hand, and on the other hand, the global markets, granting or withdrawing credibility and, as an immediate consequence, resources to indebted states (e.g., streeck 2013) . when there is tension between the interests of the constituencies, so the argument goes, the market constituency tends to prevail. currently, though, governments appear to act against the interests of both of these constituencies: they have wiped out the profit-making opportunities of big business and the earnings of small business and selfemployed workers; they have forced workers into unemployment, most drastically those in informal work situations; they have closed down the consumption and leisure activities that were considered sacrosanct because they kept people content (for an early attempt of analysing the dynamics of financial-crisis management, see tooze 2020). pre-covid-19 political sociology is at a loss to explain such government action. at first sight, one plausible way to close the explanatory gap is to underline that these measures are temporary, that they are taken in an emergency, that everything will return to "normal" once the emergency is over. but this leads us to the second question, the one about the decisions that will be taken in the near future. in parallel to the mathematical modelling of the course of the pandemic, economic modellers shoot figures about the expected decline in economic growth and its duration. however, in this situation even more than before, rather than being based on sound knowledge and technique, these predictions are best seen as attempts to imagine futures in the hope that the imagination will guide action in such a way as to make the future less uncertain than it actually is (beckert 2016) . it cannot be ruled out that, some time from now, something like a return to "normality" will occur. the adjustments of institutions and practices may turn out to be small compared to some of the expectation and imagination that the current exceptionality has generated. at the current moment, however, uncertainty and lack of knowledge prevail. and it is under this angle that the entire crisis is most fruitfully analysed. in many countries, governments have recently been criticized for having reacted late and insufficiently to the emergence of covid-19. while such criticism is more valid in some cases than others (and what follows should not be read as exempting everyone from responsibility), it is very often based on a problematic use of temporality. it is argued that some piece of information had been available early but that the action that was the adequate response to the information was taken with a considerable delay. it is overlooked that it was only with hindsight that the information was seen as requiring a certain action, not at the moment it first emerged. the criticism assumes, on the one hand, a timeless certainty of knowledge. by implication, it also assumes a fundamental stability of the world, in which an occurrence in the present can only be a repetition, with at best minor and insignificant alteration, of an occurrence in the past. and on the other hand, it assumes an unquestioned hierarchy in the application of knowledge, namely to be effected by the state. in other words, the criticism operates with a kind of vulgar platonism, a timehonoured desire of human beings to have everything known and under control. the first assumption holds that the knowledge required for action is in principle available; and that actors who do not avail themselves of the required knowledge are at fault. but this is in denial of exactly that which many of us experience, namely the occurrence of something unprecedented, and thus the lack of actionadequate knowledge. in the light of uncertain knowledge, and furthermore of a knowledge base that was only evolving over time, politicians have hesitatedproblematically so but also understandably. for purposes of action, the novelty does not have to be absolute to be problematic. virologists are unlikely to see covid-19 as radically different from all other viruses, and even sociologists do not need to abandon all of the concepts they developed for social transformations. we only need to be ready to accept that the novelty of the current situation is such that the knowledge-at-hand is insufficient to adequately understand it; that this situation is not one that can be easily subsumed under the concepts we are used to employ. therefore, it is more important at the moment to observe closely, to try to describe, to look for possible comparisons, probably not any single one that explains everything, but many partial comparisons. and this requires time. but if our scholarly knowledge is not sufficient for fully and immediately understanding the covid-19induced situation we are in, we also have to accept that the situation is similar for those who have responsibility to act. it is, furthermore, made much more difficult by the fact that the adequacy of an action often depends on whether it was made in time. the question of adequacy in time leads to the second assumption, the one of state capacity. in this regard, the virus encountered our societies on the wrong foot. the idea that the aggregate of uncoordinated individual actions would on its own bring about desirable outcomes has been ever more widely accepted for decades, in as different areas as unregulated market exchange, the "non-governmental" workings of "civil society," and the clip-form of communicative exchange in global social networks. in some such cases, like a stock exchange movement, a crowd-funding initiative, or a social-network clip "going viral," this kind of "collective" action can today be extremely fast, can take the form of a "social avalanche" (borch 2019) . with more such phenomena, though, the doubts about the presupposed desirability of the outcomes also increase. in contrast to those phenomena, co-ordinated collective action, based on problem-oriented communication and sedimented in purpose-guided institutions, has been systematically discredited for over three decades. many of its sites have been weakened and dismantled, in some cases as concretely as in the authority and resource endowment of public-health institutions. such institutions were supposed to be slow both in receiving new information and knowledge and in acting efficiently on the basis of such information. ironically, such recent criticism of bureaucratic states mirrors the timehonoured objections to pluralistic democratic deliberation, supposed to be incapable to confront any urgencies. however, we have no reason to believe that action based on multiple advice and plural deliberation is necessarily slower than either so-called "self-regulation" or hierarchic technocratic action (flaig 2013) . there is no reason at all to glorify currently leading politicians. most often, their reasons for acting tend to be considerably less than noble: getting re-elected; not offending their main allies in society; having something plausible to communicate in the short term. they are not necessarily oriented towards, as their "vocation" (weber 1994 (weber [1919 ) should be, the common good or the benefit of society at large. but watching them these days, we see that they do not even know how to pursue their lessthan-noble objectives: their re-election prospects are highly uncertain; they have acted against the immediate interests of their clientele; they are at a loss of communicating anything very convincing. as said above, they all claim that what they do is "science-based," but they say so merely to provide a justification that they would otherwise lack. there is hardly an indication of a rising technocracy and/or of a new form of authoritarian control. suddenly, critics expect this systematically disabled state to get everything right, and immediately. what we observe, though, is rather the evidence of lack of control and even of the illusion of control that politicians are used to advertise because it comes with their office. today, politicians are weak in the face of both, the pandemic and the criticism of how they are handling it. clearly, they do not have the tools at hand that they would have had before austerity politics and the discrediting of collective institutional action. and having lost those (or rather: thrown them away), they also now face human and civil rights-based criticism for wanting to increase the monitoring of behaviour; business criticism for destroying a functioning economy; social criticism for disregarding the fate of those who are less well off; and more. rather than orienting our social-science analyses towards sorting out which criticism is most justified, and to determine "whose side we're on," it seems to me we should grant to those responsible for action that they encounter the same uncertainty as everyone else in the face of a situation that in crucial respects is unprecedented. the difference between "them," the politicians, and "us," the scholars, but also between politicians in office and those in opposition, is that that the former need to act in time-and with regard to covid-19 indeed with considerable urgency. social scientists, most of the time, are exempted from that obligation. as social scientists, though, their obligation would be to elaborate a knowledge of society, economy, and politics that takes uncertainty and temporality better into account to understand the conditions for action in time. such knowledge will always be difficult to achieve but we can at least briefly spell out what, for want of better terms, can be called an ontological attitude and a methodological maxim. with regard to the former, one should abandon the notion that social phenomena repeat themselves. this is not even true of viruses, as covid-19 in relation to other coronaviruses shows, and much less so for social phenomena. rather than looking for regularity and equilibria, as much social science does, it is often more useful to focus on moments of imbalance, asymmetry, dispute, which is when more of the social world and its dynamics reveals itself (see, e.g., boltanski 1990) . this attitude has consequences for the pursuit of two of the most common activities of the social sciences, namely trying to derive the present from the past and predicting the future. as to the latter, as mentioned above, a key technique is mathematical modelling, developing with great speed and considerable variety. its practitioners are mostly well aware of the uncertainty they are dealing with. but they rarely convey this in public and, at least as importantly, they try to reduce uncertainty mostly by feeding in more data and improving technique. thus, remedial action with regard to past errors is again done by forward-looking. more effort may need to be devoted to review past modelling exercises and the way their-social, not just mathematicalstatistical-assumptions may have created or exacerbated erroneous predictions. 1 the future is not merely an extrapolation of the present, as the present is not an extrapolation of the past. but in contrast to the future, the past has for us the advantage that something about it is known with some degree of certainty, notwithstanding all ongoing dispute in the theory and philosophy of history. we have to learn to relate to the past in such a way as to grasp the possibly common conditions for action without limiting our current possibilities to the existing range of past outcomes. with a specific issue in mind, such as covid-19 today, the first question concerns the identification of a comparable situation. this is neither the plague, because the social configuration was very different, nor sars, because the behaviour of the virus was very different. my intuition is that a covid-19oriented historical-comparative sociology could fruitfully look at the first world war and the 1918-1920 pandemic. a century ago, societies were not so different from present ones that an attempt at comparison would find too little to hold on to. this is the minimum condition for a fruitful comparison. furthermore, those societies had undergone a major social transformation in the decades before 1914, one that today is sometimes called a first globalization. the dynamics of this transformation led towards catastrophe; and when the virus emerged, the catastrophe of the first world war had already happened. today, covid-19 may have emerged just in time for us to avert catastrophe-a future catastrophe that does not yet have a name but several candidates of which climate change is arguably the front-runner. my apologies for having turned a conclusion into a new opening. my excuse is that maybe this is just what we need to do. imagined futures: fictional expectations and capitalist dynamics social avalanche: crowds, cities, and financial markets l'amour et la justice comme compétence to act with good advice: greek tragedy and the democratic public sphere. in the greek polis and the invention of democracy: a politico-cultural transformation and its interpretations buying time: the delayed crisis of democratic capitalism how coronavirus almost brought down the global financial system. the guardian publisher's note springer nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. key: cord-320928-flsaa1wx authors: aldohyan, meshal; al-rawashdeh, nedal; sakr, farouk m.; rahman, saeed; alfarhan, ali i.; salam, mahmoud title: the perceived effectiveness of mers-cov educational programs and knowledge transfer among primary healthcare workers: a cross-sectional survey date: 2019-03-21 journal: bmc infect dis doi: 10.1186/s12879-019-3898-2 sha: doc_id: 320928 cord_uid: flsaa1wx background: knowledge transfer of middle east respiratory syndrome coronavirus (mers-cov) involves the dissemination of created/acquired information on mers-cov in hospitals, making this information accessible to all healthcare workers (hcws). this study evaluated the perceived effectiveness of mers-cov educational programs and knowledge transfer among primary care hcws at a hospital in saudi arabia that witnessed the largest outbreak of confirmed mers-cov cases in this country. methods: a survey was distributed among primary care hcws at five clinics in saudi arabia in 2016. those with non-direct patient care responsibilities were excluded. their knowledge was evaluated against facts published by mayo clinic foundation, and its percentage mean score (pms) ± standard deviation was calculated. hcws’ perceived effectiveness of educational programs and knowledge transfer was classified as negative or positive. results: sample comprised of 404 hcws, of which 64% were females and 36% were males. almost 26% were ≤ 30 years old, and 42% had > 10 years of work experience. almost 46.5% were nurses, 23.0% physicians, 18.1% were pharmacists, and 12.4% were technical staff. pms for knowledge was 71.1 ± 19.4. the prevalence of negative perceptions towards educational programs was 22.5% and of knowledge transfer was 20.8%. older(> 40 years of age) and more experienced(> 10 years) hcws had the highest pms for knowledge(73.4 ± 18.9,p = 0.005 and 76.9 ± 15.7,p < 0.001 respectively). negative perceptions of educational programs (49.4 ± 20.7; p < 0.001) and knowledge transfer (46.0 ± 19.7; p = 0.001) were associated with a lower knowledge pms. males were 2.4[95% confidence interval 1.4–4.2] times and 2.0[1.1–3.5] times more likely to have negative perceptions of educational programs and knowledge transfer (adjusted (adj.)p = 0.001 and adj. p = 0.023, respectively). physicians/pharmacists were 1.8[1.03–3.11] and 2.8[1.6–5.0] times more likely to have negative perceptions of both outcomes (adj. p = 0.038 and adj. p = 0.001, respectively). less experienced hcws were 2.1[1.3–3.5] times and 4.9[2.6–9.2] times more likely to exhibit negative perceptions of the two outcomes (adj. p < 0.001 each). conclusions: a negative perception of the effectiveness of mers-cov knowledge transfer was associated with poorer knowledge and was more prevalent among male hcws, physicians/pharmacists and less experienced hcws. hospitals should always refer to efficient knowledge sharing and educational strategies that render beneficial outcomes to patients, hcws, and the public community. electronic supplementary material: the online version of this article (10.1186/s12879-019-3898-2) contains supplementary material, which is available to authorized users. middle east respiratory syndrome coronavirus (mers-cov) has created an epidemiological and clinical crisis within 27 countries in north africa, europe, asia and the usa but mainly in the middle east (kingdom of saudi arabia) [1] . it is a viral respiratory illness, initially discovered in 2012 and speculated to have originated from camels or bats in saudi arabia, with subsequent spread to humans and across borders [2] . since 2012, a minimum of 1415 laboratory-confirmed cases have been reported in saudi arabia, of which 610 patients have died, 833 have recovered and two were under treatment [3] . high-risk groups were those in close contact with camels, geriatric persons, pregnant women, healthcare workers (hcws) and those with pre-existing comorbidities [1] . mers-cov infection ranged from asymptomatic or mild respiratory symptoms to severe acute respiratory disease and even death, which was reported in three to four out of every 10 reported mers-cov cases [4] . biologic samples of subjects with a suspected mers-cov infection (based on clinical symptoms) and of those exposed to reported mers-cov cases are tested using real-time reverse transcription polymerase chain reaction (rrt-pcr) assays. serology, such as an enzyme-linked immunosorbent assay and immunofluorescence assay, is also used to confirm mers-cov by the presence of antibodies [5] . in saudi arabia, a series of modifications was applied to the patient pathways while visiting the emergency departments or admitted as in-patients. this included segregating patients during triage based on prioritizing the inflow of patients by their chief complaints, bed availability and screening of flu symptoms/history of exposure. the infrastructure of the medical facility, particularly the ventilation system and isolation capacity of rooms, was subject to changes. some hospital wards and staff (especially nurses) were dedicated specifically to confirmed mers-cov cases to limit the chance of cross-contamination across wards and hcws. the infection prevention and control department (ipcd) at smng-ha, in particular, was on high alert for such mers-cov outbreaks, especially with the evident transmission of viral infections between patients and hcws at smng-ha. crisis management required a rapid implementation of adequate infection prevention, control procedures and case isolation, in addition to collaboration and coordination with local and international consultations. exceptional efforts have been made by the ipcd to apply the latest and most effective means of universal standard precautions throughout the mers-cov crisis. rules and regulations pertinent to infection control and prevention have been revisited and environmental surveillance has been carried out regularly to ensure that all wards are equipped with suitable protection and precautionary gear. numerous seminars, workshops and awareness campaigns have been launched for hcws of all disciplines to boost their knowledge on mers-cov, as well as their morale, to maintain a high-quality, safe and dedicated service for the patients. the latest updates issued on mers-cov from the world health organization (who), the centers for disease control and prevention (cdc), collaborative task forces (local and regional) and researchers have been circulated regularly among all hcws and across all managerial levels. numerous research studies have been conducted and published on the perception, knowledge and attitude of hcws towards mers-cov. it is rare to find a hcw who has not attended an educational program on mers-cov in saudi arabia. dissemination of mers-cov information/updates or knowledge transfer within a healthcare organization is a process in which this information is created, generated or acquired, and then organized and distributed within the system to ensure it is accessible to all hcws. one of the mechanisms of knowledge transfer is personalization whereby knowledge is transferred from one individual to another, while the other is codification where knowledge is converted into products such as documents, images and videos [6] . the need to transfer efficiently the precautionary regulations and updates about mers-cov to large numbers of hcws necessitates the mechanism of codification [7] . in addition, knowledge transfer or information sharing was found to be positively associated with job satisfaction [8] . authors hypothesized that although the dissemination of knowledge and updates on mers-cov among hcws has been given full consideration, these hcws might have reservations on the effectiveness and quality of mers-cov related educational offerings. therefore, there was an emerging need to evaluate the perceived effectiveness of mers-cov educational programs and knowledge transfer from the hcw's perspective, in a setting that witnessed the largest outbreak of confirmed mers-cov cases in saudi arabia. this was a cross-sectional study, based on an anonymous survey in english distributed among hcws at the primary healthcare centers in smng-ha medical centers, in riyadh, saudi arabia, between october and december 2016. the smng-ha is the second-largest healthcare sector in the country, second only to the saudi ministry of health, and provides healthcare services to the community of national guards, their dependents and employees [9] . the targeted primary healthcare centers were five randomly selected clinics out of 11 clinics that employ physicians, pharmacists, technicians and nurses. these clinics serve a population of 60,000 registrants, with a rate of four visits per registrant annually. eligible primary care hcws were targeted as being in frontline contact with potential confirmed cases of mers-cov. those occupying positions of management, education or non-direct patient care were excluded. medical and nursing students were also excluded. this study was ethically approved by the institutional review board at king abdullah international medical research center, king saud bin abdul-aziz university for health sciences, sp16/191 . the provisioned sample size in this study was calculated based on a reported level of knowledge between 43.3 and 57.4% by alkot et al. among hcws in the western region of saudi arabia. assuming an expected level of knowledge of 57%, with a 95% confidence limit (z = 1.96), and a margin of error 5%, the estimated sample size for this study was 376. for convenience, all eligible hcws at the targeted setting were invited to participate in this survey, to overcome a 25% nonresponse rate. the survey was provided in a sealed envelope with a cover letter that described the objectives of the study. the survey was in english language, as the targeted study participants were english literate and the educational offerings provided at the targeted setting were also in english. participants who agreed to enroll in this study hand-signed an agreement statement at the end of an informed consent, with no need for any personal identifier. the data collection tool comprised the characteristics of the hcw, principally gender, age category (years), job position and experience (years) [10] . the knowledge of hcws was measured using 16 statements based on undisputed facts published in the literature and issued by the mayo foundation for medical education and research in 2018 [11] . correct answers were scored "1", while wrong/don't know answers were scored "0". the percentage mean score (pms) of knowledge was calculated by adding the correct responses of the 16 statements, dividing the score by 16 and multiplying it by 100 (range of score 0 to 100). the perceived effectiveness of the mers-cov educational programs was measured using one statement: "prevalence of mers can be reduced by active participation of healthcare workers in the hospital infection control program", while the perception of knowledge transfer was measured by one statement: "any related information about mers should be disseminated among healthcare workers". both statements were rated on a four-point likert scale (strongly disagree, disagree, agree and strongly agree). those who responded by disagree or strongly disagree were classified as having a negative perception, while those who responded agree or strongly agree were classified as having a positive perception. the negative perception rate was calculated by dividing the number of participants who had negative responses over the total number of respondents. in addition, participants were asked about the source of mers-cov information. the survey was piloted on a group of five hcws, and their subjective comments were considered. the internal consistency or reliability (cronbach's alpha) of the knowledge domain measured 0.67 (additional file 1). data were analyzed using the statistical package for social studies (spss 25; ibm corp., new york, ny, usa). hcw characteristics, perceptions (negative vs positive) and incorrect knowledge response statements were presented in terms of frequencies and percentages, while the pms of knowledge was presented as the mean ± standard deviation. missing data were replaced by the average of the total, and outliers were dropped out. pearson's chi-square test was used to test categorical outcomes across hcw characteristics, while a mann-whitney test and a kruskal-wallis test were used to test the non-normally distributed pms of knowledge scores. two binary logistic regression models were constructed to determine the factors significantly associated with negatively perceived effectiveness of mers-cov educational programs and knowledge transfer. due to the small subgroup size of job positions, nurses were grouped with technicians, while pharmacists were grouped with physicians. these two subgroups had job positions comparable in terms of the educational levels, scope of practice and nature of patient care. the adjusted odds ratios [95% confidence interval] were calculated, and statistical significance was set at a value of p < 0.05. initially, 600 surveys were distributed among hcws; 404 participants agreed to enroll and completed the survey (response rate 67.3%). those who did not participate were mainly either off duty or busy with their workload. females constituted 64.1% of the sample, while males comprised 35.9%. almost 26% were ≤ 30 years old, 29% were 30-40 years old and 45% were > 40 years old. the majority (46.5%) were nurses, followed by physicians (23.0%), pharmacists (18.1%) and technical staff (12.4%). most hcws (42.1%) had accumulated > 10 years of work experience, with 29.2% having < 5 years of experience and 28.7% having 5-10 years (table 1) . overall, 36.1% of respondents claimed that their main source of information was the internet, while 47.8% reported more than one source, including research studies, books, media and others. the pms of knowledge score was 71.1 ± 19.4. the most common incorrect response to the statements (83.4%) was that for "incubation time for virus", followed by 43.6% with an incorrect response to the statement that "antibiotics are the first-line treatment for the management of mers-cov". other incorrect responses to statements are listed in chronological order in table 2 . overall, 22.5% of participants reported a negative perceived effectiveness of mers-cov educational programs, while 20.8% had a negative perception of knowledge transfer. with regard to the perceived effectiveness of mers-cov educational programs, male hcws had significantly a more negative perception than female hcws (n = 52, 35.9%, vs n = 39, 15.1%, respectively; p < 0.001). pharmacists (n = 24, 32.9%) and physicians (n = 30, 32.3%) reported more negative perceptions than technical staff (n = 9, 18.0%) and nurses (n = 28, 14.9%) (p = 0.001). hcws with work experience of < 5 years had the most negative perceptions in comparison with the other groups (p = 0.001). a number of factors were associated with a negative perception of knowledge transfer of mers-cov information. male hcws had a greater negative perception than females (n = 47, 32.4%, vs n = 37, 14.3%, respectively; p < 0.001). physicians (n = 31, 33.3%) and pharmacists (n = 23, 31.5%) had more negative perceptions of knowledge transfer in comparison with technical staff (n = 9, 18.0%) and nurses (n = 21, 11.2%) (p < 0.001). junior hcws with work experience of < 5 years (35.4%) had the highest rate of negative perception of knowledge transfer (p < 0.001) ( table 3) . hcws > 40 years old (pms 73.4 ± 18.9) had the highest knowledge scores in comparison with the other age groups (p = 0.005). more experienced hcws (> 10 years) also had the highest knowledge scores (pms 76.9 ± 15.7; p < 0.001). those who had a negative perception of the effectiveness of mers-cov educational programs (pms 49.4 ± 20.7) and of knowledge transfer of mers-cov updates (pms 46.0 ± 19.7) both had lower knowledge scores in comparison with the positive-perception group (p < 0.001 and p = 0.001, respectively), table 4 . logistic regression analyses showed that male hcws table 5 . mers-cov educational programs at healthcare institutions are a formal and reliable channel to deliver essential knowledge to hcws. for the sake of personal safety, job satisfaction and work morale, hcws should not pass up any opportunity to increase their theoretical knowledge and practical skills. hospital administrators do not necessarily face the challenge of producing new information, as an immense amount of valuable information already exists in the literature. the problem arises from the fact that current knowledge is either poorly structured or inaccessible to hcws [12] . for example, advanced practice nurses are observed to be "knowledge brokers" in a sense that they act as disseminators of knowledge among the nursing body. furthermore, health educators retrieve different types of evidence, synthesize it in different forms, translate it by evaluation, interpret it and then distribute it among nurses [13] . health education can improve levels of awareness and perception among hcws towards mers-cov infections [14] , and these higher levels of knowledge can aid in the control n frequency, % percentage of disease outbreaks [15] . however, published evidence in saudi arabia has shown that there is limited knowledge on mers-cov (both microbiological and virological aspects) among hcws in southern saudi arabia [16] . another study also claimed that knowledge about emerging infectious diseases was poor, and that infection control practices were suboptimal and also seemed to be overestimated [17] . the association between younger age and less experience on one hand and lower knowledge scores on the other was a reasonable finding. similar to literature findings, the knowledge of hcws in this setting was suboptimal and gaps remain that should direct the focus towards the mechanisms and quality of knowledge transfer. dissemination of mers-cov updates using e-mail, the internet, institutional announcements, employee meetings, the media and even personal communications are all methods of knowledge transfer. hcws can experience knowledge transfer both passively, absorbing information unconsciously, and actively. investigators in this study were curious to know how hcws perceived the transfer of knowledge about mers-cov and why this would be of concern to hospital administrators. for instance, knowledge transfer has been adopted with regard to smoking as a health hazard, hiv transmission as a sexual risk and seat belts in motor vehicles as a safety measure. people are exposed almost daily to precautionary advice by a variety of methods but unfortunately still undertake high-risk activities and are exposed to these hazards. this occurs regardless of the duration, frequency and quality of awareness campaigns. therefore, it is an aggravating concern that the repetitive exposure of hcws to mers-cov campaigns might have created some sort of "tolerance". hcws might disremember or take lightly the acquisition of current or new updates about mers-cov precautions due to routine attendance of educational programs or repetitive circulation of e-mails. knowledge and skills must be passed on in a systematic way from expert to novice employees in a way that makes sense [18] . managers who support work-empowering environments are actually boosting the engagement of participations in terms of knowledge transfer [19] . in fact, one of the key elements in seeking accreditation or managing crises such as mers-cov is knowledge communication, in the sense that effective communication ensures a purposeful exchange of information, thus allowing a more thorough understanding of the outbreak [20, 21] . interactive workshops remain highly recommended for the sharing and transferring of knowledge among hcws. however, one study noted that, although those who attended such workshops valued the expert input and discussions, after few months their sustainability of attendance was lost [22] . some barriers to mers-cov knowledge transfer could be the inability of hcws to recognize and articulate the instructions, personal opposition or resistance to change, the quality of the communication technologies, the absence of visual representations, language and cultural differences, deficiency in expertise, the work environment, a lack of job incentive/motivation, the organizational culture and others [23, 24] . current efforts to manage the mers-cov crisis are directed towards developing educational programs that target both the community and hcws [25] . a negative perception of mers-cov educational programs in this setting might result in outdated knowledge among hcws, which jeopardizes their compliance with disease precautionary and control measures. a mers-cov task force committee pointed out that the saudi arabian ministry of health has posted updates on mers-cov through videos, posters, handouts, posters and an official website. resilience against mers-cov increases with enriched education and awareness [25] . a saudi arabian study reported that hcws were unaware of the availability of mers-cov information at their work areas; they did not feel they had sufficient training and were not confident about infection control guidelines. these factors may also contribute to having a negative perception of mers-cov-related educational programs [26] . one study reported that the interest in following disease updates among hcws improved significantly after the implementation of a mers-cov educational program [14] . these programs improved the attitude of the hcws towards governmental measures taken regarding the crisis [14] . hcws often grasp their mers-cov educational information primarily from watching tv reports, or from the internet. a negative perception of knowledge transfer might be due to a pre-existing lack of trust in the media or in websites that might, to some degree, lack scientific credibility in comparison with educational programs provided in healthcare centers [27] . knowledge itself is complex, and its transfer process within healthcare institutions carries many challenges [28] . one way to overcome these challenges is to determine the characteristics of hcws who might be more likely to exhibit negative perceptions of knowledge transfer for significant mers-cov updates. in the literature, knowledge transfer has been investigated more frequently in manufacturing industries and firms, or among the public community. it has been seldom evaluated among hcws [29] , and never in a middle eastern setting or related to a mers-cov outbreak. a crossnational study suggested that organizational culture was a significant influence on the capacity of hcws to engage in knowledge transfer [30] . a systematic review paper study stated that knowledge transfer could streamline productivity and coordinate the use of resources more efficiently [29] . this review paper claimed to be the first to review published research focused on the perceptions of hcws about knowledge management [29] . knowledge management was defined as having an efficient idea or new practice accepted and adopted by an individual or a group through communication channels (successful diffusion of ideas) [31] . this definition also applies to the dissemination of updated regulations on the outbreak of mers-cov. this information, once absorbed by people, should be sustained for as long as it is useful, and not decay over time [32] . accordingly, a negative perception of the importance of knowledge transfer could be a warning sign of an interruption in this sustainability of retained information about mers-cov. signs of information decay were evident among hcws in this study, as those who had negative perceptions had lower knowledge scores about mers-cov in comparison with those who had positive perceptions. one of the key goals of knowledge transfer is to educate and train the less experienced and/or the newly hired hcws [33] . this phase of staff development is crucial yet stressful for novice hcws, who are expected to acquire skills and competencies rapidly to ensure that a safe and quality service flow is maintained at the institution. this explains why hcws with less work experience (< 5 years) had significantly more negative perceptions of knowledge transfer and the perceived effectiveness of mers-cov educational programs. as they gain more work experience, this perception improves as they realize the importance of education not only for their patients but also for their career advancement. the level of knowledge on mers-cov among hcws in primary healthcare clinics in this setting was found to be less than optimal. as the frontline in the battle of disease prevention and control, hcws are expected to be equipped with the relevant theoretical updates about mers-cov. special consideration should be paid to younger and less experienced hcws whose knowledge on mers-cov was moderately low. a negative perception of the knowledge transfer of mers-cov information and educational programs was associated with poorer knowledge. this negative perception was more prevalent among male hcws, physicians/pharmacists and less experienced hcws. this study has been conducted at one setting, yet the struggle against mers-cov has not ended and will continue against future emerging strains of viruses and bacteria causing communicable diseases in other settings too. knowledge is a valuable asset, and its holders within any healthcare institution should be retained and motivated so that they continue to spread their benefit among other hcws. all healthcare institutions should always identify and refer to reliable sources of knowledge. for instance, the center for disease control and prevention is a leading national public health institute and accountable for disseminating up-to-dates on various infectious topics. in saudi arabia, the ministry for public health has designated communication channels to release updates on mers-cov on their websites, through scientific arenas, memorandums and helpdesks. knowledge sharing and management strategies in the healthcare sector can render beneficial outcomes to patients, hcws, the organization and the public community [29] . in addition to the attendance of seminars or workshops, other methods of knowledge dissemination might involve launching of journal clubs among hcws to discuss updates on mers cov. audiovisuals at hospitals, such as educational videos on tv screens in lobbies or corridors, constantly enlighten hcws. deeper understanding of the negativity in the perception towards the quality or method of knowledge transfer necessitates a qualitative methodological approach, as face to face interviews with hcws aid in determining the underlying reasons and at a more personal level. furthermore, the execution of these strategies needs to be routinely monitored and evaluated so that the transfer of knowledge is time efficient and effective. optimal theoretical knowledge and practical competence are two main indicators of successful knowledge transfer among hcws. last but not least, a number of key points can be noted: as well as their morale, to maintain a high-quality, safe and dedicated service for patients. -the perceived effectiveness of mers-cov educational programs and knowledge transfer among health workers in this high risk setting was evaluated. -primary health workers were expected to be aware of the most recent updates on mers-cov, yet younger and less experienced hcws had moderate knowledge. -a negative perception of the effectiveness of mers-cov knowledge transfer was associated with poorer knowledge, and was more prevalent among male hcws, physicians/pharmacists and less experienced hcws. global summary and risk assessment middle east respiratory syndrome effects of educational program on mers-coronavirus among nurses students at jazan university 1439-2017 an observational, laboratory-based study of outbreaks of middle east respiratory syndrome coronavirus in jeddah and riyadh, kingdom of saudi arabia laboratory testing for middle east respiratory syndrome coronavirus (mers-cov). cdc what's your strategy for managing knowledge? knowledge management research & practice. journal article does perception of knowledge sharing ,transfer and recognition have an impact on job satisfaction? an empirical study in saudi arabia smoking cessation advice: the selfreported attitudes and practice of primary health care physicians in a military community, central saudi arabia predictors of attitude and intention to use knowledge management system among korean nurses what is mers-cov, and what should i do? : mayo clinic implementing knowledge management practices in hospital-in-the-home units the role of advanced practice nurses in knowledge brokering as a means of promoting evidence-based practice among clinical nurses knowledge, attitude, and practice toward mers-cov among primary health-care workers in makkah al-mukarramah: an intervention study knowledge and perception of health practitioners towards mers-cov in hail region, kingdom of saudi arabia knowledge and attitude towards the middle east respiratory syndrome coronavirus among healthcare personnel in the southern region of saudi arabia knowledge, attitudes and behaviours of healthcare workers in the kingdom of saudi arabia to mers coronavirus and other emerging infectious diseases sense making and knowledge transfer: capturing the knowledge and wisdom of nursing leaders nurses' participation in personal knowledge transfer: the role of leader-member exchange (lmx) and structural empowerment knowledge communication: a key to successful crisis management l'accréditation, source de connaissance et d'enrichissement using interactive workshops to prompt knowledge exchange: a realist evaluation of a knowledge to action initiative culture as an issue in knowledge sharing: a means of competitive advantage academic conferences limited understanding change and change management processes: a case study an educational programme for nursing college staff and students during a mers-coronavirus outbreak in saudi arabia questionnaire-based analysis of infection prevention and control in healthcare facilities in saudi arabia in regards to middle east respiratory syndrome middle east respiratory syndrome-related knowledge, preventive behaviours and risk perception among nursing students during outbreak intra-firm knowledge transfer-a qualitative case study of knowledge transfer and its implications in a soft service firm knowledge management practices in healthcare settings: a systematic review the importance of knowledge transfer between specialist and generic services in improving health care: a cross-national study of dementia care in england and the netherlands diffusion of innovations impaired memory retrieval correlates with individual differences in cortisol response but not autonomic response expatriate knowledge transfer, subsidiary absorptive capacity, and subsidiary performance this study was approved and monitored by king abdullah international medical research center, king saud bin abdulaziz university for health sciences, riyadh, saudi arabia. the authors would like to thank the research office and the institutional review board for their tremendous support. none to declare. the datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. all authors conceptualized and designed the study. md, fs, smr and aaf supervised the conduct of the study and data collection. md, smr and aaf undertook the recruitment of subjects and managed the data. fs, smr and aaf were accounted for the quality control. nar and ms provided statistical advice on study design, data analysis and responded to reviewers' comments. all authors drafted the manuscript, and contributed substantially to its revision as submitted and agree to be accountable for all aspects of the work. ethics approval and consent to participate a self-explanatory letter of invitation to participate was presented to each of the participants. all participants had given written informed consents for their participation in the research presented in this manuscript with full knowledge of the possible risks and benefits of participation. participants consented by ticking "agree", indicating their agreement to provide their feedback for this research study. study was approved by the institutional review board of the saudi ministry of national guard health affairs, riyadh, saudi arabia (protocol # sp16/191). this study followed the recommendations of the international conference on harmonization for good clinical practice (ich-gcp). not applicable. the authors declare that they have no competing interests. springer nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.author details key: cord-312678-81gnmxbk authors: elayeh, eman; aleidi, shereen m.; ya’acoub, rawan; haddadin, randa n. title: before and after case reporting: a comparison of the knowledge, attitude and practices of the jordanian population towards covid-19 date: 2020-10-15 journal: plos one doi: 10.1371/journal.pone.0240780 sha: doc_id: 312678 cord_uid: 81gnmxbk coronavirus disease2019 (covid-19) is an emerging contagious infectious disease. it is pandemic and has affected more than 21 million people and resulted in more than 750,000 deaths worldwide (https://www.worldometers.info/coronavirus/#countries; 14/08/20). our research group initiated a study to ascertain the knowledge, attitude and practices (kap) of jordanians toward covid-19 prior to any initial case report in jordan. this project was underway when the first jordanian case was reported. we extended our study to identify how case reporting would alter public kap towards covid-19. this cross-sectional study randomly selected and recruited 2104 jordanian adults. a four-section questionnaire was devised to address the sociodemographic characteristics of the subjects and their kap toward covid-19. the mean knowledge score for the study population was 15.9 ± 2.2 (out of the 20 knowledge questions), with 60.9% of the participants having good knowledge about covid-19. participants’ practices to prevent transmission of covid-19 were adequate in more than 60% of participants. most participants had positive attitudes regarding their role in preventing covid-19 and many of the participants’ attitudes and practices changed to more appropriate ones after reporting the first case of covid-19 in jordan. the percentage of participants who trust the government in confronting covid-19 increased significantly (p value < 0.001). however, one alarming and unexpected finding was that the prevention practice score of participants working in the medical field was similar to those from the general population. this may necessitate stricter training and guidelines for this group who will be in the frontline in combating the disease. impact of this study: the data generated from this study shows that when cases of disease were reported, the public’s attitudes and practices improved in many aspects, and that confidence in the government to contain the disease was boosted. we believe that this study is important in allowing other, international governments to develop an understanding of public kap during pandemic disease outbreaks. a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 after reporting 70 cases which were all isolated in dedicated hospitals, the jordanian government applied a nationwide lockdown and activated the defense law, which allows the army and police to ensure the adherence of public to the curfew [16] . under these circumstances, where public health is threatened by the pandemic and since the people's attitudes and practices play a crucial role in limiting or spreading the disease in their community, evaluating the knowledge, attitude, and practice (kap) of the general public towards the covid-19 is crucial. therefore, the initial aim of this study was to evaluate the overall knowledge, attitude and practices (kap) of jordanian citizens to the ongoing international situation regarding the emergence and pandemic nature of covid-19. however, we were lucky that our study was almost complete when the first case of the disease was reported in jordan. hence, we rapidly adapted our study in order to engage more participants and to evaluate how this reported case altered the kap of jordanian citizens. the aim of this report is to aid decision makers in their understanding of the public knowledge and awareness of the disease, the attitudes of the public and their practices to take necessary measures to prevent disease spread. this was a cross-sectional study conducted during the early days of covid-19 pandemic of 2019 and 2020 and over the period from 12 th february to 19 th march 2020. when the study started there were no reported cases in jordan. during the study, the first covid-19 case was reported in jordan, which encouraged the authors to extend the study and include more participants in order to identify how case reporting would alter the public kap towards covid-19. the study targeted adult resident jordanian nationals. it was a questionnaire based and self-selection study. the institutional review board (irb) at the deanship of scientific research, the university of jordan, approved the study protocol and the questionnaire (irb 28/2020). the questionnaire was developed based on extensive literature review [3, [17] [18] [19] [20] . it consisted of four main sections; (a) demographic and general characteristics of the participants, (b) participants knowledge regarding covid-19, its mode of transmission, and its preventative measures (20 questions), (c) participants' attitudes towards covid-19 and its preventative measures (13 questions) , and (d) participants' practices towards prevention of covid-19 transmission (6 questions). the questionnaire content was translated from english into arabic and then critically revised and face-validated by several academic colleagues. since the jordanian general public knows covid-19 as "coronavirus disease", this term was used in the questionnaire. amendments were made according to the notes and comments received after piloting to a sample of 50 adults chosen from the friends, neighbors and coworkers of the authors. the questionnaire was then developed as a google form and disseminated to participants via mobile applications, eg: whatsapp [21, 22] . snow ball sampling technique was used to distribute the google form and enroll participants [22, 23] . the link to the google form was sent to the contact lists of the authors, in addition to neighbors, friends, relatives, coworkers of all levels and departments, friends of friends, different whatsapp groups, who were also asked to complete the form and to disseminate it to all the people they know in jordan. this method of disseminating the questionnaire was selected in view of the need to reduce direct contact with participants under the pandemic situation. participants were asked not to identify themselves in any fashion. also they were informed that completing the questionnaire and returning it were considered as a formal consent to participate in the study since the participant can, at any moment, stop answering questions or choose not to "submit" the final form. a minimum sample of 1050 adults aged 18 and older was estimated based on the following equation [24] : n ¼ where n is the sample size, zα: type one error = 1.96 when α = 5%; zβ: type two error = 1.28 when β = 10%; q = 1-p: expected non-prevalence; p = proportion in the population possessing the characteristic of interest (based on the estimate that 50% of the respondents knew general information about covid-19, its routes of transmission and the main preventative measures), d = one-half of the desired interval of confidence, in this study d = 5%. accordingly, by filling in the equation, n = 0.5x0.5 (1.96+1.28) 2 / 0.05 2 = 1049.76. statistical analysis was performed using spss version 20.0 (spss inc., chicago, il). descriptive statistics were used to describe demographic characteristics of participants. categorical variables were presented as valid percentages to account for missing data with their frequencies, while continuous variables were presented as mean with standard deviation (sd). two scores were calculated for the participants: knowledge about covid-19 and its preventative measures, and participants' practices towards preventing transmission of covid-19. participants' knowledge scores were evaluated using the number of correct questions they answered out of 20 covid-19 knowledge questions. in order to calculate the knowledge score, correct answers were assigned a score of one, while wrong or "i don't know" answers were assigned a score of zero. participants were considered adequately knowledgeable if their knowledge score, for all knowledge questions, was higher than or equal to the sample median of the knowledge score. internal consistency of the knowledge questions was tested using cronbach alpha coefficient as a reliability test. the results showed that the cronbach alpha for knowledge (20 items) was 0.7. cronbach alpha within the range 0.6 to 0.7 is considered adequate and reliable [25] . similarly, participants practice towards preventing transmission of covid-19 score was calculated by assigning a score of one for appropriate practices, while inappropriate practices were assigned a score of zero. the maximum value of practice score is six based on the total number of correct practice items. participants' knowledge, attitudes and practices were compared before and after the reporting of the first confirmed case of coronavirus in jordan which was documented on 2 nd march 2020. by default, 'google forms' identifies the date and time of each participant's response which enabled us to sort the responses submitted into those made before and after case reporting. chi-square test was used for these comparisons. parametric tests including independent sample t-test and one way anova test were used to test the differences among the variables that affect both knowledge and practice scores (bivariate analysis) as specified in the results section (data were first tested for normality using kolmogorov-smirnov (k-s) test and for homogeneity of variances using levene's test). in addition, independent sample t-test was used to test the differences in the scores between the two groups (before and after case reporting). all hypothesis testing was two-sided. a p-value of < 0.05 was considered significant. the total number of participants recruited in the study was 2104. 832 participants were recruited before case reporting and 1272 participants were recruited after case reporting. more than 50% of participants were in the age group of 18-35 (62.7%, n = 1321). females accounted for 75.4% (n = 1586) of the study population. more than 75% of participants had a university or postgraduate degree (81.6%, n = 1717), and 43�6%, (n = 917) had education in the medical field. since the cohort participating in the study after case reporting differed from those participating before case reporting, there were variations in the sociodemographic data of the two groups (p< 0.05, chi square) except for gender and working in the medical field which were similar (p>0.05). participants' sociodemographic characteristics are presented in table 1 . the mean knowledge score for the whole study population was 15.9 ± 2.2 (out of the 20 knowledge questions), with 60.9% (n = 1281) of the participants having a score equal to or higher than the median (16.0) and consequently were considered to have adequate knowledge about covid-19. only 2.4% (n = 51) of the participants were able to correctly answer all of the 20 knowledge questions and 8% (n = 168) of participants were able to correctly answer 19 questions. in addition, 17% (n = 357) of participants knew the general information related to the covid-19, its symptoms and the highest risk group. however, more than 50% of the study population incorrectly thought that death is a common complication of covid-19 (44.2%, n = 929) ( table 2) . regarding covid-19 transmission, a major gap of knowledge regarding routes of covid-19 transmission was identified where only 2.3% (n = 49) of the population correctly recognized all the routes of covid-19 transmission. considering covid-19 preventative measures, more than half of the participants were able to identify the main preventative measures. however, only 18.3% (n = 385) of them knew plos one that a facemask should not be used daily as a preventative measure for covid-19 transmission (which was a recommendation by who and moh at the early days of the pandemic). in general, the mean knowledge scores of participants improved after case reporting (when compared with those exhibited before case reporting) in the three areas studied: definition, signs and symptoms, risk groups and complications (mean score 5.5 ±1.2 vs 5.1 ±1.2, p<0.005); mode of transmission (mean score 3.3 ± 0.8 vs 3.2 ± 0.7, p = 0.037) and prevention of transmission and treatment (mean score 7.6 ± 1.1 vs 6.9 ± 1.1, p<0.005). accordingly, the total knowledge score was improved from 15.2 ± 2.2 before case reporting to 16.3 ± 2.1 after case reporting (p value <0.001, independent sample t-test) ( table 2 ). the detailed frequencies for individual questions are presented in table 2 . the most common source of participants' information about covid-19 was social media, such as facebook, twitter or others (64.3%, n = 1353), followed by internet searching such as google (59.4%, n = 1249), and television (49�7%, n = 1046). a much lower proportion of respondents relied on newspapers (37.3%, n = 784), friends (29.3%, n = 616), brochures (20.2%, n = 426) or physicians' offices (6.6%, n = 138) to get information. a significantly higher knowledge score (p-value < 0.001) was associated with using internet search (score = 16.4) as sources of information (p-value < 0.001; independent sample t-test), while other sources were not associated with significantly higher knowledge score. most participants had positive attitudes regarding their role in preventing covid-19. in particular, 67.2% of participants thought that they could protect themselves against covid-19 and 88.7% of them thought that following advised preventative measures would be effective. however, more than 50% of participants didn't trust the jordanian ministry of health's (moh) approach to confronting covid-19 (57.6%, n = 1211) or the information provided by governmental authorities about the exact number of cases (66.1%, n = 2091). similarly, more than half of participants believed that the occurrence of coronavirus was related to international tension and trade wars (57.9%, n = 1219) and 49.7% (n = 1046) of participants believed that the recent coronavirus was created in a laboratory and was not naturally occurring. on the other hand, most participants were either unsure or didn't believe that herbal remedies (71.7%, n = 1508) or antibiotics (85.9%, n = 1808) were effective in treating or preventing covid-19. interestingly, many of the participants' attitudes had changed after reporting the first jordanian case of covid-19 (2 nd march 2020). most importantly, the percentage of participants who believe that covid-19 was a serious and life threating infection dropped significantly from 81.6% to 52.6% (p value < 0.001; table 3 ). on the other hand, the proportion of participants who trusted moh in confronting covid-19 increased significantly from 35.9% to 46.7% (p value<0.001). similarly, the percentage of participants who thought that they could protect themselves against infection with covid-19 (59.3% vs.72.3%, p value <0.005) and those who thought that treatment approaches were effective also increased significantly (from 31.0% to 42.7%, p value = 0.009). the mean practice score of all the participants was 3.95 ± 1.7 (out of the 6 practice questions) with 66.7% (n = 1403) of the participants having a score equal to or higher than 4 and consequently were considered as having adequate practices to prevent spread of covid-19. respondents' practices are presented in table 4 . participants' practices to prevent transmission of covid-19 in terms of avoiding hand shaking, hugging, kissing and crowded areas, and using disinfectants were appropriate in more than 60% of participants. interestingly, all these practices have increased significantly after reporting the first case of covid-19. on the other hand, the proportion of participants wearing facemasks and avoiding the purchase of chinese table 4 ). overall, the observed practice score improved significantly when compared between respondents before case reporting and after case reporting (3.65 ± 1.9 vs. 4.2 ± 1.4 respectively, p value < 0.001, independent sample t-test; table 4 ). factors affecting participants' knowledge and practice towards preventing covid-19 were determined by bivariate analysis. overall participants' knowledge and practice scores were associated with age, educational level and education in health or medical field. in addition, participants' knowledge score was associated with gender and work in the medical field. participants' practice score was also associated with marital status. for each sociodemographic characteristic, the calculated knowledge score (table 5 ) or practice score ( table 6 ) was improved after case reporting when compared to the value before case reporting. detailed results are shown in tables 5 and 6 . since the who declaration of covid-19 as a public health emergency on 30 th january 2020, health authorities around the world, led by who, have initiated huge campaigns to increase the awareness of the people toward the disease and to disseminate the appropriate practices to prevent its transmission [4] . as a member of these authorities, the moh in jordan started similar campaigns relying on different forms of media. the first case in jordan was reported on 2 nd march 2020 for a citizen returning from italy [28] . our study commenced by distributing the questionnaires on 17 th february before the reporting of the first case of covid-19 in jordan, and the questionnaire process ended on 19 th march. by that time, the measures taken by the country included closure of schools and universities, quarantining thousands of incoming air passengers in hotels, and prohibiting any kind of social gathering (including the closure of all mosques and churches). during that period, the number of cases reported increased to 52 without fatality. this gave our team an opportunity to compare the change in knowledge, attitudes and practices of jordanians towards covid-19 before and after the reporting cases of covid-19 illness. the sample population of this study (2104 participants) was largely well educated with females predominating. the observed skew in the sample toward females and well educated participants has also been seen in previous studies in jordan [29] [30] [31] [32] . in addition, the vast majority of participants were below the age of 55 years, which is representative of the generally younger population of jordan, where at the end of 2019 the estimated proportion of population below 55 years was around 92% [33] . the overall knowledge of the respondents was generally adequate (total knowledge score 15.9 out of 20). they exhibited excellent knowledge of the organs targeted by the virus, its nature as a zoonotic disease, and its signs and symptoms. however, the respondents had an exaggerated idea about the expected death rate of the disease where more than 55% considered death to be a very common outcome. this idea could have been drawn in their minds due to the effect of media focus on death rates and novel cases, rather than recovered cases [34, 35] , thus, giving the audience or readers the impression that this disease is highly fatal. the respondents were well aware of the most common routes of the disease transmission, such as close contact, respiratory droplets or touching contaminated surfaces. the vast majority of participants (> 90%) had an excellent knowledge about the required measures to prevent the disease. however, the more scientific questions related to the role of influenza vaccine in protecting from covid-19 or the one related to use of a specific drug to treat the illness were correctly answered by only around two thirds the participants. the participants' knowledge was gained through multiple sources of information. the electronic sources such as social media (64%) and internet (59%) were the top sources followed by television (50%). these results suggest that in this era, where electronic media sources are predominate at the expense of more traditional media sources, health authorities should focus their healthrelated campaigns on electronic media and they should consider adding social media platforms to their public communication tools, in order to reach the vast majority of population [36] [37] [38] [39] . it is noteworthy that people who obtained their information from internet searches had significantly higher knowledge scores than those using other sources. this is probably due to the fact that people who were keen to get more information about the disease actively sought this information from the internet. however, those relying on social media, such as facebook, twitter and whatsapp may have received inaccurate information since social media platforms often lack fact-checking and editorial control [40, 41] . when investigating the knowledge of the participants the major factors were; the educational level, education background, and the field of work (p value within group <0.005 for each) ( table 5 ). this was anticipated and is in line with previous studies that compare the knowledge of the general population about a specific health-related issue with those of higher levels of education or with a health-related backgrounds [42, 43] . the latter would have much better knowledge. comparing the mean knowledge scores of the respondents before and after case reporting (table 2) , showed an improvement in the respondents' knowledge after case reporting (p value between groups <0.05, independent sample t-test). we are aware that differences in the population characteristics before and after case reporting might contribute to observed changes in the knowledge score. however, statistical analysis of the factors affecting participants' knowledge (table 5 ) exhibits that the observed improvement in the knowledge was significant in all the classes of each characteristic (age, gender, marital status, etc, p value between groups <0.05), indicating a palpable effect of case reporting on the participants' knowledge. apparently, people become more interested in learning about a disease when it is seen to be proximal to their vicinity as has been seen in other similar situations [18] . this suggests the importance of pursuing health campaigns and associated recommendations when case reporting of any outbreak or epidemic commences in a country, as the public seems to become more receptive to that additional information. a significant change in the attitudes of participants in many aspects was noticed after reporting cases of covid-19 illnesses in jordan. unexpectedly, the participants seemed to become more optimistic and comfortable after the reporting of cases of covid-19. this optimism appeared in their perception to the severity of disease as a life threatening one, the effectiveness of preventative measures, their ability to protect themselves, the effectiveness of the treatment approaches and the ability of moh to confront the disease. although this is initially puzzling, it can be explained by the unique situation in jordan. shortly after first case reporting in jordan, there was a sense of panic among the population. in order to calm the public and fight the rumors, the first covid-19 patient was interviewed in hospital where he was isolated, by different national and regional tv and radio channels [44] . the patient tried to comfort the people about his situation, since he only suffered from mild respiratory illness. being a traveler returning back from italy, which was hard hit by the pandemic at that time, he compared the looser measures taken by italian authorities to contain the disease with the tougher measures taken in jordan, and he praised the latter. the same thing happened with the second case who only suffered from mild illness. the situation of both patients was of great public interest until they were discharged from hospital. apparently, the message perceived by the public from both patients was that the disease is mild and can be prevented, and that the measures taken by the authorities are effective. our study suggests that the jordanian government's open approach with the public about the disease, its spread, and the necessity for stringent control measures enhanced the optimistic attitudes of the participants towards the disease and its management after case reporting in the country. on the other hand, the attitudes of the participants in other areas remained the same before and after reporting cases in jordan. more than half the respondents indicated their willingness to buy the vaccine (most pharmaceuticals are paid for by the patient in jordan) and the number increased a little more if the vaccine was offered free of charge. these figures are relatively high compared to the general attitudes of jordanians towards vaccinations other than the mandatory ones at childhood [42, 45] . this improved attitude could be related to the worldwide worries about this disease, which have made the jordanians more willing to protect themselves against coronavirus. as for the emergence of this virus, almost half of the participants believed that the virus was created in a laboratory and more than half believed that its emergence was the result of political reason linked to international tension or trade war. this conspiracy theory is not limited to jordanians, and has been reported worldwide. some americans accused china of bioengineering the virus as a bioweapon and conversely, some chinese have accused the american military of introducing it into china. nevertheless, these theories have been refuted in the media and by scientific studies [46] [47] [48] [49] . the practices of the participants in order to prevent the transmission of the virus were found to be adequate for more than two thirds of the respondents. however, there was a remarkable improvement (p < 0.001) in the participants' practices after the reporting of cases of the disease in jordan. the practices evaluated included; avoidance of shaking hands, avoidance of kissing, avoidance of crowded places, and an increase in the use of disinfectants (table 4 ). this improvement in practices' scores was anticipated because the public was observing the disease having increased proximity to their areas, thus they were more willing to engage with the recommendations of the moh regarding covid-19. the improvement in practices' scores seen among the population after case reporting compared to those before case reporting cannot be attributed solely to differences in sample population, because the increase in practice score was obvious in all the population characteristics studied (table 6 ). after reporting cases in jordan, there was a significant decrease in the number of participants using facemasks when going outside. this contradicts the behavior of chinese people during the epidemic in their region, where 98% of a study population indicated wearing facemasks when leaving homes [50] . nevertheless in this respect, the practices of jordanian respondents exhibited better adherence to the instructions of moh at the time. in the early days of the pandemic who and moh instructions to the general public were not to use facemasks, as it was feared this would lead to shortage of masks for essential health and care workers. many other factors affected the overall practices of the respondents. participants with ages of more than 55 years had statistically better practices than those in the lower age groups (18-25yr and 26-35yr; p value within group <0.005). this implies that older respondents are more cautious about their health than younger participants due to who and moh warnings that people above the age of 60 years were at higher risk of developing severe complications than those of younger ages [51] . these findings suggest that the public is willing to implement the recommendations of the health authorities when continuously and firmly directed to do so. surprisingly, the negative factors affecting the overall practices of the participants towards the prevention of transmission are the education level and having a health-related education. these participants scored lower than their counterparts in terms of the best practices (table 6 ). this finding, albeit odd is not uncommon. in a study among health care workers in saudi arabia to evaluate their kap about middle east respiratory syndrome (mers), physicians, pharmacists, nurses and technicians showed low to average scores on practices, but better knowledge and attitude scores [52] . we speculate that having a better education, particularly in a health-related field, provides the participants with confidence that they are aware of the risks and know better how to protect themselves. another disappointing finding is that the overall participants practice score of those working in the health field is the same as those working outside this field (3.9 for each, table 6 ). this finding should be alarming to the health authorities and health care sector, since these workers will be in the front line in combating the epidemic. being lenient in applying the strictest measures and practices in protecting themselves and preventing the spread of infection could lead to a disaster. therefore, the health sector should immediately intensify education and extensive training and setting guidelines for the proper practices among healthcare workers of different specialties. the findings of the study should be interpreted with the following limitations in mind: • sampling of the study was made through social media, what's app in particular, which could pose some bias to the study where underprivileged people or those having problems in using electronic devices may not be able to participate, thus, will be under-represented in the sample. • females and people with education level of more than 12 years were over-represented in the sample studied when compared to the general jordanian population. • like any other self-reported study, responses (mainly of attitudes and practices) could have been reported based on social desirability, not the actual situation of the participants. severe acute respiratory syndrome coronavirus 2 (sars-cov-2) and coronavirus disease-2019 (covid-19): the epidemic and the challenges who [internet]. 2020 centers for disease control and prevention who. events as they happen-rolling updates on coronavirus disease early transmission dynamics in wuhan, china, of novel coronavirus-infected pneumonia novel coronavirus: where we are and what we know coronavirus disease 2019 treatment: a review of early and emerging options draft landscape of covid-19 candidate vaccines modes of transmission of virus causing covid-19: implications for ipc precaution recommendations q&as on covid-19 for older people refugee population by country or territory of asylum-| data health care and pharmacy practice in jordan higher health council & world health organization government suspends educational system in a bouquet of coronavirus preventive measures government announces 10 decisions to fight against coronavirus jordanians now ponder economic cost of coronavirus. in: al-monitor, the pulse of the middle east covid-19) advice for the public bin saeed a. knowledge, attitude and practice of secondary schools and university students toward middle east respiratory syndrome epidemic in saudi arabia: a cross-sectional study covid-19)-how coronavirus spreads covid-19)-how to protect yourself & others. in: centers for disease control and prevention public knowledge, attitudes and practices towards covid-19: a cross-sectional study in attitudes and practices towards covid-19: an epidemiological survey in north-central nigeria sage research methods foundations parental knowledg attitudes and behaviours towards influenza a/h1n1 in italy global patterns in students' views of science and interest in science who says people receiving packages from china not at risk of contracting coronavirus-society & culture-tass los angeles daily news who. coronavirus disease 2019 (covid-19) situation report-43 promotion of appropriate knowledge and attitude towards medicines among schoolchildren in jordan: the role of teachers knowledge and information sources about covid-19 attitudes and behavior regarding antibiotics use and misuse among adults in the community of jordan. a pilot study albsoul-younes am. knowledge, awareness and practices towards seasonal influenza and its vaccine: implications for future vaccination campaigns in jordan. fam pract table 2: estimated population of the kingdom by sex why coronavirus is killing far more men than women-the washington post. the washington post more covid-19 cases, deaths reported in rest of world than in china-who the use of social media among saudi residents for medicines related information internet and social media use for antibiotic-related information seeking: findings from a survey among adult population in italy taking the paper out of news mers-cov infection: mind the public knowledge gap fake news: spread of misinformation about urological conditions on social media the trump effect: with no peer review, how do we know what to really believe on social media? clin colon rectal surg influenza vaccination coverage rates, knowledge, attitudes, and beliefs in jordan: a comprehensive study hepatitis b knowledge, perceptions and practices in the french general population: the room for improvement jordan free from corona cases as infected citizen left quarantine seasonal influenza vaccination among older adults in jordan: prevalence, knowledge, and attitudes. hum vaccines immunother covid-19 coronavirus was not bioengineered. here's the research that debunks that idea covid-19 coronavirus epidemic has a natural origin-sciencedaily coronavirus did not come from a lab: experts debunk myths that china or usa bioengineered covid-19. mail online evolutionary origins of the sars-cov-2 sarbecovirus lineage responsible for the covid-19 pandemic. nat microbiol knowledge, attitudes, and practices towards covid-19 among chinese residents during the rapid rise period of the covid-19 outbreak: a quick online cross-sectional survey who. coronavirus disease 2019 (covid-19) situation report-51 middle east respiratory syndrome (mers): comparing the knowledge, attitude and practices of different health care workers the authors would like to thank professor phillip collier (visiting professor, faculty of pharmacy and medical technology, university of petra) for his critical review of the paper and proof reading. haddadin. the design and timing of this study could not have been more fortuitous, in that it allowed the authors to rapidly adapt and expand a preliminary study into a fully operational before-andafter study of kap towards covid-19. the findings of this study are necessarily complex, but generally show that the population reacts well to open and honest governmental advice about pandemics and can rapidly adopt safe practices in response to appropriate advice. the international relevance of this work is obvious, in that jordan achieved one of the lowest rates (per million population) of covid-19 infection ranking 186 and 177 out of 213 in the number of cases and related fatality respectively, in the world (https://www.worldometers.info/ coronavirus/#countries; 14/08/20). key: cord-315886-f4ofmdcm authors: karim, a.; akter, m.; mazid, a. t.; pulock, o. s.; aziz, t. t.; hayee, s.; tamanna, n.; chuwdhury, g.; haque, a.; yeasmin, f.; mitu, m. a.; rashid, h.; kuri, a. k.; das, a.; majumder, k.; barua, d.; rahaman, m. m.; akter, s.; munia, n. n.; sultana, j.; usaila, f.; sifat, s.; nourin, n. a.; uddin, m. f.; bhowmik, m.; ahammed, t.; sharik, n.; mehnaz, q.; bhuiyan, m. n. h.; banu, t. title: knowledge and attitude towards covid-19 in bangladesh: population-level estimation and a comparison of data obtained by phone and online survey methods date: 2020-05-26 journal: nan doi: 10.1101/2020.05.26.20104497 sha: doc_id: 315886 cord_uid: f4ofmdcm adherence of people to the guidelines and measures suggested in fighting the ongoing covid-19 pandemic is partly determined by the knowledge, attitude, and practices (kap) of the population. in this cross-sectional study, we primarily addressed two key issues. first, we tried to determine whether there is a significant difference in the estimated covid-19 knowledge level from the online and phone survey methods. second, we tried to quantify the knowledge and attitude of covid-19 in bangladeshi adult population. data were collected through phone calls (april 14-23, 2020) and online survey (april 18-19, 2020) in bangladesh. the questionnaire had 20 knowledge questions with each correct response getting one point and incorrect/do not know response getting no point (maximum total knowledge score 20). participants scoring >17 were categorized as having good knowledge. the percentages of good knowledge holders were 57.6%, 75.1%, and 95.8% in the phone (n=1426), online non-medical (n=1097), and online medical participants (n=382), respectively. comparison between phone and online survey showed that, overall, online survey might overestimate knowledge level than that of phone survey, although there was no difference for elderly, poor, and rural people. male gender, higher education, living in town/urban areas, good financial condition, and use of internet were positively associated with good knowledge. however, higher knowledge was associated with having less confidence in the final control of covid-19. our adult population-level estimates showed that only 32.6% (95% ci 30.1-35.2%) had good knowledge. this study provides crucial information that could be useful for the researchers and policymakers to develop effective strategies. cc-by-nc-nd 4.0 international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may 26, 2020. . https://doi.org/10.1101/2020.05.26.20104497 doi: medrxiv preprint 1 coronaviruses (covs) are a large family of enveloped, positive-sense rna viruses that are important pathogens of humans and other mammals [1] . in the past two decades, two highly pathogenic human covs, namely sars-cov and mers-cov, have emerged in 2003 and 2012 respectively [2] . recently, a third new type of cov, which is even more infectious, is spreading across the world in an unprecedented manner. this novel coronavirus was discovered to be the cause of unexplained pneumonia-like cases in wuhan, hubei, china in december, 2019 with the majority of initial patients had been exposed to huanan seafood market [3] . this coronavirus, which was provisionally named as 2019-ncov, has been renamed as severe acute respiratory syndrome coronavirus 2 (sars-cov-2). the disease caused by sars-cov-2 has been named as coronavirus disease 2019 (covid-19) [4] . it was first reported to the world health organization (who) on december 31, 2019. as of april 30, globally confirmed cases rose to 3 090 445 (with 217 769 deaths) from 212 countries/territories/areas [5] . bangladesh announced the first three cases of covid-19 on march 8, 2020. just by april 30, the total number of confirmed cases raised to 7667 (with 168 deaths) [6] . to reduce the spread of covid-19 and chance of being infected, world health organization (who) advises public to adopt some simple precautions of maintaining hygiene and social distancing [7] . authorities from different countries around the world are providing additional advice and adopting further measures e.g., complete or partial shutdown of different areas, restricting transport facilities, travel bans, transitioning educational and business activities to online, restricting access to public places. an important determinant of peoples' adherence to these guidelines and measures could be the knowledge, attitudes, and practices (kap) towards covid-19 of the respective population [8] [9] [10] . furthermore, information about kap of the general population may guide the policymakers and researchers tremendously to develop effective strategies to face this kind of rapidly evolving crisis. some covid-19 kap survey reports from different countries [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] , including bangladesh [21, 22] , are available already. however, most of these surveys were conducted through online questionnaire as it was not so feasible to do it otherwise during this pandemic disruption. online surveys probably have limited value when estimating population-level kap in countries like bangladesh, where only 15% of the population has access to internet compared to the world average of 50% (world bank -2017) [23] . therefore, kap of the general bangladeshi population towards covid-19 is largely unknown and it is imperative to conduct further studies to understand more about it so that strategies can be developed or adapted accordingly. in this study, we surveyed adults from bangladesh through phone call interviews and an online questionnaire. this allowed us to investigate whether one method tends to overestimate knowledge level over another. the survey with phone calls enabled us to access people from diverse sociodemographic backgrounds, thus allowing us to adjust the phone survey data to the population level sociodemographic characteristics to estimate covid-19 knowledge status of the adult population. to the best of our knowledge, this study is the largest in bangladesh conducted so far to reveal the knowledge and attitudes of its adult population towards covid-19 during its rapid rise period and the only study to do a comparative evaluation of phone and online survey methods in estimating covid-19 knowledge. ethical clearance was obtained from the ethical review committee of chittagong medical college, chattogram, bangladesh. a cross-sectional survey was conducted through phone calls from april 14-23, 2020. in addition, an online survey was conducted from april 18-19, 2020. as it was difficult to predict the number of participants we would be able to recruit during this disruption, we had no predetermined target sample size. due to the rapid spread of covid-19 and significant disruption (termed as "lockdown" by many, but not officially) of different areas of bangladesh during our study period, it was not feasible to conduct a community-based national sampling survey. therefore, we had to rely on the authors' network to recruit participants. for better representation of the diverse population and to cover all the administrative divisions of bangladesh, we first formed a research group with members studied/studying/worked/working in various universities/institutes throughout the country. members were trained to conduct the phone survey with same instructional videos and materials so that everyone could follow a similar approach while conducting the survey. inclusion criteria were: a) persons giving consent to participate, and b) at least 18 years old. exclusion criteria were: a) persons not giving consent, b) persons who could not communicate well, c) researcher's family members, d) researcher's relatives/friends/colleagues with whom the researcher previously discussed about covid-19 to make them aware, e) other family members/friends/colleagues living together with already included participant, and f) doctors/nurses/medical students and family members living with them. the researchers used their acquaintances, acquaintances of family members or friends or colleagues, and acquaintances of acquaintances to recruit a wide range of participants of different age groups, geographic locations, financial conditions, and educational backgrounds. when we requested our acquaintances to suggest some of their acquaintances to be included in our study, they were requested not to disclose the survey questions asked from them during their participation in our survey. conversations with the participants were made through phone calls. at the beginning, we obtained informed verbal consent. after that, we asked whether they had heard about the recent outbreak of covid-19 or sars-cov-2. if they answered "yes", then we proceeded with the questionnaire which took on average 10-12 minutes to complete. if the participants appeared to be not communicating well enough or not understanding the questions reasonably, we excluded them from the survey. after the end of the interview, they were informed briefly about the important information regarding covid-19 and the correct answers to the questions asked. the online survey was done with the same questionnaire with additional attitude questions that were not covered in the phone survey (to minimize phone call duration). the questionnaire was created on google forms and distributed through social media. the online link was posted and reposted on the facebook timeline of the researchers, different facebook groups, sent through friends or acquaintances through facebook messenger or whatsapp. besides, the researchers requested their acquaintances to post/share the survey link on their facebook timeline or social media apps. the link was accompanied by a summary of the study to allow the participants to make an informed decision. we requested bangladeshi citizens who understood the nature of the study, was willing to participate, and was not surveyed by the phone call to attempt the questionnaire. at the beginning of the survey, they were asked whether they had heard about the recent outbreak of covid-19 or sars-cov-2. if they answered "yes", then the questions of the questionnaire appeared section by section and at the end of the survey, they were provided with important information regarding covid-19 and referred to the website of who and institute of epidemiology, disease control and research (iedcr) of bangladesh for further information. if they answered "no", they were directly forwarded to the last information page. the sociodemographic data collected in both phone and online surveys included the age in years (≥18 to <35, ≥35 to <55, and ≥55), gender (male, female, other), education level (≤ grade 5, grade 6 to 12, > grade 12 to bachelor, > bachelor to master/above), lived mostly in the last four months (rural, urban/town, other), financial condition (poor, middle-class, rich), and administrative division where the participant lived most of the life. information about occupation (doctor/nurse/medical student, others) was asked in online survey. this auxiliary information allowed us to stratify the online survey data to "online non-medical" and "online medical" participants. because participants of <18 years old could not be restricted from attempting the questionnaire, we added one more age category (<18 years) in the online survey. this enabled us to exclude them from data analysis as our targets were the participants of ≥18 years old. the questionnaire is given in table 1 . it was developed based on the information provided by the who, iedcr, and centers for disease control and prevention (cdc) of usa to the general public, latest scientific evidence from literature, and previously published similar study [10] . both english and bangla versions of the questionnaire were agreed by a panel comprising of one clinical professor, two epidemiologists, five physicians, two medical students, and four non-medical researchers. it was tested by a pilot survey from 20 participants through phone call to ensure that all the questions were clearly understood. it is worth mentioning that there are several dialects of bangla language spoken throughout the country. during the phone survey, sometimes researchers had to conduct survey using the local dialect. the questionnaire had three main sections: section-1 asked about the source of information regarding covid-19 or sars-cov-2. section-2 (knowledge section) had two subdivisions: section-2a (questions related to the knowledge of disease) and section-2b (questions related to the knowledge of preventive practices). section-3 (attitude section) was used only for the online survey. in section-2a, there were 14 questions (kd01 to kd14) related to the knowledge of disease that included signs/symptoms, mode of spread, outcome of the disease etc. in section-2b, there were six questions (kp01 to kp06) related to the knowledge of preventive practices. in section-3, there were two attitude questions (a1 and a2). options to the knowledge questions were a) true, b) false, and c) don't know. options to the attitude questions were a) yes, b) no, and c) don't know. a correct response to a knowledge question was assigned 1 point and an incorrect/don't know response was assigned 0 point. total score of knowledge of disease (kd) and knowledge of preventive practices (kp) ranged from 0 to14 and 0 to 6, respectively. therefore, the total knowledge score ranged from 0-20. a higher score indicated a higher knowledge level. the cronbach's alpha coefficient of the knowledge questionnaire was 0.74 in the current study, indicating acceptable internal consistency [10, 24] . data analysis was primarily done with ibm spss statistics 20. we divided the survey participants into three main groups as "phone", "online non-medical", and "online medical". frequencies and percentages were used to describe the responses to the questions and sociodemographic characteristics of the participants. chi-square test for independence and mann-whitney u test were used for comparison between different groups wherever appropriate. correlations among kd, kp, and total knowledge scores were investigated with pearson product-moment correlation. binary or multinomial logistic regression was used as appropriate to identify factors associated with total knowledge score and attitudes. the statistical significance level was set at p < 0.05 (two-sided). benjamini-hochberg correction of p values was done as appropriate in case of multiple testing and statistical significance was determined using a false discovery rate (fdr) of 0.05. the r package "anesrake" version 0.80 was used for raking (also known as iterative post-stratification) the phone survey data for different purposes [25] . the unadjusted phone survey data was weighted (adjusted) to match the socio-demographic distributions (age, gender, education, lived mostly in last four months, financial condition of the family) of the online non-medical survey data for comparison of the phone and online survey methods. for population-level estimation, the unadjusted phone survey data was adjusted to match the socio-demographic distributions of bangladeshi adult population (as of 2011 bangladesh national census data, given in the supplementary methods). table 1 questionnaire used in the present study there were 1427 phone participants. all of them heard about covid-19 except one. in the online survey, we had a total of 1521 participants. 98.6% (1500) of them heard about the disease. from these 1500 online participants, 19 were excluded because of being less than 18 years old. two more participants were excluded because they reported their education level as not more than grade 12 (higher secondary school -hsc) but reported themselves as doctor/nurse/medical student. the results presented hereafter are based on the final dataset of 1426 phone participants and 1479 online participants. among these 1479 online participants, 382 were medical and the remaining 1097 were non-medical. the administrative division-wise distribution of the participants is given in table s1 and fig. s1 . we had participants from all the divisions of bangladesh in both phone and online surveys although most of them were from the two major divisions chattogram and dhaka. the other socio-demographic characteristics are shown in table 2 . chi-square test for independence showed that all the characteristics differed significantly between participant groups. the participants of the phone survey were less educated and from poorer families compared to the online participants. a significantly higher portion of the online participants had been living in urban/town areas (vs. rural) for the last four months compared to the phone survey participants. also, the online survey had comparatively more young participants than phone survey. both phone and online survey participants mentioned television and internet as major sources of information of covid-19 ( fig. 1a) . table 2 socio-demographic characteristics of the participants in the phone (unadjusted), phone (adjusted to online non-medical), online non-medical, and online medical participants, respectively (table 3) . chi-square test for independence showed that the correct response rates differed significantly for 11 out of 14 kd questions between the unadjusted phone survey data and online non-medical survey data. in general, for questions having significantly different correct response rates, the rates were higher in online survey compared to that of the phone survey, except for two questions (kd02 and kd10). however, the number of kd questions having significantly different correct response rates dropped to six when the phone survey data was adjusted to the socio-demographic characteristics of the online non-medical participant group. in addition, for 11 out of 14 kd questions, correct response rates were significantly higher for the online medical group than online non-medical. this suggests that the knowledge of disease was the highest in the online medical group which was followed by the online non-medical group. the phone survey group was the least knowledgeable about the disease even after adjusting to the online non-medical group. despite this difference in kd, correct response rates in the kp questions showed less difference except for one (kp01) question. table 3 percentage of the correct response to each question of the knowledge questionnaire participants were categorized into "poor knowledge" or "good knowledge" holders based on the total knowledge score achieved (≤17 and >17-20 respectively) ( fig. 1b and 1c) . the percentages of good knowledge holders in the phone (unadjusted), phone (adjusted to online non-medical), online non-medical, and online medical groups were 57.6%, 69.6% 75.1%, and 95.8%, respectively (fig. 1c) . the total knowledge scores based on sociodemographic characteristics and sources of covid-19 information are shown in table 4 . online medical participants achieved the highest total knowledge score (median 20, iqr 19-20) (table 4 ). it was followed by the online non-medical group (median 19, iqr [18] [19] [20] . the phone participants scored the least (median 18 [iqr [16] [17] [18] [19] and median 18 [iqr [17] [18] [19] for unadjusted and adjusted data, respectively). the median scores for almost all categories of sociodemographic characteristics and sources of information were higher in online non-medical group compared to that of the phone participant group with most of the differences being statistically significant. table 4 total knowledge score according to the sociodemographic characteristics and source of information pearson product-moment correlation coefficient showed that there was a positive association of medium strength (coefficient 0.33) between kd score and kp score (table 4 ). although both kd and kp had strong correlation with total knowledge score, the kd had nearly a perfect correlation with the total knowledge score with a coefficient of 0.98 and thereby explaining 96.2% of the variance in the total knowledge score. taken together with the high correct response rates to five of the six kp questions (as shown in table 3 ), it suggests that most participants had good knowledge of the preventive practices irrespective of their knowledge level of the disease. table 5 pearson product-moment correlations between different scores binary logistic regression analysis of the combined dataset of phone (unadjusted) and online non-medical participants showed that, male gender (vs. female), higher education (vs. ≤ grade 12), living in town for the last four months (vs. living in rural areas), middle-class/rich financial condition (vs. poor), and internet as a source of information (vs. not as a source) were positively associated with having good knowledge (vs. poor knowledge) ( table 6 ). table 6 results of binary logistic regression on factors predicting the likelihood of having good knowledge responses to the attitude questions are shown in table 7 (responses by sociodemographic characteristics and source of information are shown in table s2 and table s3 ). 61.1% and 56% among the online non-medical and online medical participants, respectively, were optimistic that this new coronavirus will finally be controlled completely. multinomial logistic regression analysis showed that for every unit increase in the total knowledge score, the odds of the participants responding "no" (compared to responding "yes") increased by a factor of 1.19 (95% ci 1.04 to 1.36), all other factors being equal (table 8) . similarly, the odds of a participant responding "no" (compared to "yes") was 2.95 times higher (95% ci 1.28 to 6.78) for someone who reported the internet as a source of information than for a participant who did not, all other factors being equal. interestingly, females had higher odds of responding "don't know" (compared to "yes" or "no") to this attitude question than males. in case of attitude question a2 ("this virus is created by humans" -this kind of discourse is heard. do you believe that it is true?), only 19% of the participants mentioned that the statement is not true ( table 7) . multinomial logistic regression analysis showed that for every unit increase in the total knowledge score, the odds of the participants responding "no" increased by a factor of 1.13 or 1.14 (compared to responding "yes" or "don't know", respectively), all other factors being equal (table 8 ). table 7 responses to the attitude questions table 8 multinomial logistic regression analysis on factors predicting the likelihood of different responses of attitude questions finally, we adjusted the original phone survey data to the sociodemographic distributions (gender, education, rural vs. urban/town, and poor vs. middle-class/rich) of the bangladeshi adult population in an effort to estimate the correct response rates to the knowledge questions at population level (table 9 ). it estimated that less than 50% of the adult population know that some covid-19 patients might have no symptoms, diarrhea is a symptom of covid-19, and this disease cannot be completely prevented by any medications. the estimated average total knowledge score was 15.85 (sd 3.14) with a median score of 17 (iqr 14-18). only 32.6% (clopper-pearson 95% ci 30.1-35.2%) of the population was estimated to have good knowledge (>85% total knowledge score). table 9 population-level estimates of the correct response rates to the knowledge questions bangladesh is a lower-middle-income country and one of the most densely populated countries in the world. apparently, like many other countries, it is struggling to combat covid-19. nearly all countries are relying heavily on non-pharmacological interventions against covid-19. success in this battle is partly determined by the knowledge and behavioral changes of the general population as well as how effectively these non-pharmacological control measures are being implemented by the authorities and abided by the citizens [26] . therefore, it is imperative to know about the knowledge and attitudes of the population towards covid-19 to develop effective strategies. given the social distancing measures in place in most countries, the studies quantifying kap of the population were mostly conducted through internet with some exceptions [26] . survey through the internet may quantify the true kap of the population of countries where a high proportion of the population use internet (e.g., 95% and 87% of the population of the united kingdom and the united states use internet, respectively (world bank -2017) [23] . however, online survey may not be able to quantify the true kap of the population of countries where access to the internet is limited and may result in overestimation of the kap. in addition, because of inadequate samples from underrepresented population groups (e.g., poor, less uneducated, and rural people), it becomes difficult to adjust the online survey data to the general population of the respective countries with poor internet access [21, 22] . in the present study, we primarily addressed two key issues. first, we tried to determine whether there is a significant difference in the estimated covid-19 knowledge level determined by the online and phone survey methods given the same questionnaire. second, we tried to quantify the knowledge of covid-19 in bangladeshi adult population. in our study, the sociodemographic characteristics were significantly different between the phone and online participants. the phone survey method was able to include more participants from poor, rural, and less educational backgrounds compared to the online survey. in fact, there were only a few participants in the online survey in some socio-demographic groups (e.g., education up to grade five) making it difficult to adjust the online dataset to the population level. to compare the estimated knowledge level by the phone survey and that by the online survey, we adjusted the phone survey data to the sociodemographic characteristics of the online survey dataset. we found that, the total knowledge score was still significantly higher in the online non-medical participants than that of phone survey participants in most of the socio-demographic groups. although investigating the factors responsible for the higher estimated knowledge level in the online survey was out of the scope of our current study, online participants probably had more opportunity to know the correct answers from others or by searching elsewhere than the phone survey participants who had to respond immediately to the survey questions during the phone calls. it is noticeable from our study that the total knowledge score did not differ significantly between the adjusted phone survey dataset and online non-medical dataset for the elderly (≥55 years), poor, and rural people. it suggests that, although these people are less likely to be accessed by the online surveys in countries like bangladesh, the knowledge estimates will be similar by both phone and online survey methods for these subgroups of the population. however, as these people are likely to be underrepresented in online surveys and more educated, urban, younger, and financially solvent people are likely to be overrepresented, we therefore advise caution in generalizing the unadjusted online survey results to the general population particularly in countries with limited internet access for the population. in the present study, inclusion of participants from medical backgrounds in the online survey allowed us to choose an informed cut-off score for dividing the participants into poor or good knowledge holders rather than choosing an arbitrary score as in some other covid-19 kap studies [21, 10] . we assumed that most of the medical participants should have good knowledge of covid-19. then we examined the distribution of total knowledge scores (fig. 1b) and decided to use a score of >17 (out of 20) as a cut-off score to categorize participants as good knowledge holders as it covered 95.8% of the medical participants as good knowledge holders. based on this criterion, we found that 75.1% of the online non-medical participants had good knowledge which was significantly higher than that of phone survey participants. in fig. 2 , we tried to summarize the timeline of all the major covid-19 related events that occurred in bangladesh (adapted and updated from [27] ). we also highlighted the timeline of two other online surveys conducted previously in bangladesh so that we can meaningfully compare our online survey results with those studies [21, 22] . the first study by haque, t et al. [21] was done at the very beginning of the covid-19 spread in bangladesh. the study ended on march 30, when bangladesh had total confirmed cases of 49 (deaths 5). the second study by rahman, a et al. [22] , was done for four days that ended on april 10, at the time when the total number of confirmed cases started to rise rapidly (from 164 to 426 in four days). the online survey of the present study was done approximately and 91.5%, respectively. the percentages of good knowledge holders were 54.9%, 86.2%, and 75.1%, respectively (although this is prone to more variation than the average knowledge score as the defining scores for "good knowledge" were somewhat arbitrary). therefore, it is reasonable to claim that there has been a clear improvement in the knowledge level among the online participants. however, to what extent this improvement applies to the huge portion of the bangladeshi population who do not have access to the internet, remains unclear as we could not find any previous data to compare our phone survey data. unfortunately, the improvement might not be as dramatic as online participants. there could have been a lot of factors associated with the improvement of knowledge level of the online participants. fig. 2 shows the trends of "corona"-related news on two popular bangladeshi online news portals and the google trends for search term "coronavirus" from bangladesh. the news trends rapidly raised soon after the initial declaration of the covid-19 cases on march 8 indicating that people were getting increasing amount of coronavirus-related news from the news portals as well as sharing of the news on social media. some major unprecedented steps taken by the authorities also followed soon. all these factors could gradually have made people more interested to actively learn about the disease as evident in the google trends. this is also supported by the findings of a positive association of good covid-19 knowledge level with higher education and use of internet as a source of information in our study. however, it is extremely difficult to quantify the relative contribution of all these factors to the improvement of the knowledge level. an online survey conducted in china approximately one month after the initial case detection in wuhan, found a similar knowledge level among the participants (average total knowledge score of 90%) [10] . in the study from china mentioned above, 90.8% of the online respondents were optimistic that the covid-19 will finally be successfully controlled. furthermore, higher covid-19 knowledge score was positively associated with higher likelihood of being optimistic about this [10] . apparently, compared to many other countries, china has been successful in controlling the local epidemic approximately within two months of the outbreak despite being the initial epicenter of covid-19. contrasting to this, and possibly a warning finding in our study, only 61.1% of the online non-medical participants had optimistic attitudes towards final control of covid-19. the medical participants were even less optimistic (56%). rahman. a et al. reported even lower proportion of online respondents having optimistic attitudes (41.7%) [22] . furthermore, in our combined dataset of phone (unadjusted) and online non-medical participants, we found that higher total knowledge score was positively associated with having negative attitude towards the final control of covid-19. if we look back to the timeline presented in fig. 2 and see the potentially major adverse events that happened in bangladesh (see the paper by [27] for detailed discussion) that were in stark contrast to the known containment strategies, it is tempting to associate these events to the negative feelings towards the final success. one might argue that bangladesh could suffer devastating consequences if more effective and coordinated approaches are not taken and people continue to defy social distancing advice. however, as pointed out by [27] , it is extremely challenging in bangladesh. our study estimated that the average total knowledge score would be approximately 79.3% (15.85/20) in the general bangladeshi adult population with an estimated 32.6% of the population having good knowledge. this finding is alarming for the country and appropriate educational public health measures must be taken to improve the knowledge level so that the people can have better chance of protecting themselves. one limitation of our study was the sampling techniques. as we had to rely on the authors' network for recruiting participants for phone survey, it suffered selection bias and we could not obtain a nationally representative sample for adult population. however, as we obtained good number of participants from each sociodemographic category and adjusted the dataset to national census data, this issue should be less problematic, and our population-level estimates should be fairly representative. the second limitation worth mentioning is that we merely assessed the knowledge level of preventive practices. thus, the knowledge level found for the preventive practice section does not necessarily reflect the actual physical behavior adopting those preventive practices. it is reasonable to argue that the proportion of the population adopting those preventive practices could be less. for example, our population level estimates showed that, 98% of the population know the information that "washing hands frequently and thoroughly with soap or alcohol-based hand sanitizers reduces the risk of being infected by the virus". but in practice, proportion of the population washing hands more frequently could be less. a third limitation was that we could not quantify the financial condition of the participants with any formal scale to limit the phone call duration. therefore, we had to rely on the "self-reported" financial condition. our findings suggest that online survey might overestimate knowledge level compared to phone survey, although there appears to be no difference for elderly, poor, and rural people. we also showed evidence supporting a rapidly increasing covid-19 knowledge level among the online participants. male gender, higher education, living in town/urban areas, good financial condition, and use of internet were positively associated with higher knowledge. however, higher knowledge was associated with having less confidence in the final control of covid-19. our population-level estimates showed that only 32.6% had good knowledge. table 1 questionnaire used in the present study do you think that this new coronavirus will finally be under control completely? a2 "this virus is created by humans" -this kind of discourse is heard. do you believe that it is true? table 8 multinomial logistic regression analysis on factors predicting the likelihood of different responses of attitude questions a all the independent variables shown table 6 were also used here with one additional independent variable (total knowledge score). only the significantly associated variables are shown in this a2 "this virus is created by humans" -this kind of discourse is heard. do you believe that it is true? (no vs. don't know) total knowledge score 0.12 0.05 6.97 1 0.008 1.14 1.03 1.23 participants were categorized into "poor knowledge" holders if total knowledge score was ≤17 (≤85%) and into "good knowledge" holders if the score was >17. an overview of their replication and pathogenesis clinical characteristics of 113 deceased patients with coronavirus disease 2019: retrospective study clinical features of patients infected with 2019 novel coronavirus in wuhan the species severe acute respiratory syndrome-related coronavirus: classifying 2019-ncov and naming it sars-cov-2 who. coronavirus disease 2019 (covid-19) situation report -101. 2020 covid-19 status bangladesh covid-19) advice for the public college students' knowledge, attitudes and adherence to public service announcements on ebola in nigeria: suggestions for improving future ebola prevention education programmes the impact of knowledge and attitudes on adherence to tuberculosis treatment: a case-control study in a moroccan region knowledge, attitudes, and practices towards covid-19 among chinese residents during the rapid rise period of the covid-19 outbreak: a quick online cross-sectional survey knowledge and behaviors toward covid-19 among u.s. residents during the early days of the pandemic. medrxiv coronavirus disease-2019: knowledge, attitude, and practices of health care workers at makerere university teaching hospitals, uganda. frontiers in public health knowledge and behaviors toward covid-19 among u.s. residents during the early days of the pandemic this paper was submitted to the bulletin of the world health organization and was posted to the covid-19 open site, according to the protocol for public health emergencies for international concern as knowledge, perception and attitudes in regard to covid-19 pandemic in peruvian population readiness in response the epidemic of coronavirus disease-2019 (covid-19) among young adults in chiang rai province knowledge and practices towards covid-19 during its outbreak: a multinational cross-sectional study. medrxiv knowledge, perceptions, and attitude of egyptians towards the novel coronavirus disease (covid-19) knowledge, attitude, practice and perceived barriers among healthcare professionals regarding covid-19: a cross-sectional survey from pakistan. medrxiv covid-19 related knowledge, attitudes, practices and needs of households in informal settlements in knowledge, attitude and practices (kap) towards covid-19 and assessment of risks of infection by sars-cov-2 among the bangladeshi population: an online cross sectional survey attitude, and preventive practices toward covid-19 among individuals using the internet (% of population) -bangladesh, world, united kingdom, united states the use of cronbach's alpha when developing and reporting research instruments in science education impact assessment of nonpharmaceutical interventions against coronavirus disease 2019 and influenza in hong kong: an observational study. the lancet public health covid-19 and bangladesh: challenges and how to address them. frontiers in public health government decided to scale down the 100 th birthday anniversary celebration of nation's founder sheik mujibur rahman 2. all flights from europe (except united kingdom) were cancelled 3. 142 passengers returned from italy (the then epicentre), many of whom were allowed to go home with advice to self-isolate despite the travel ban from europe, a flight carrying 304 bangladeshi passengers from europe landed in bangladesh decision to ban all rallies (religious, political, social, or cultural) was taken armed forces was deployed to supervise two important quarantine facilities in dhaka (capital) lockdown/shutdown" started for 10 days with ban on major transport facilities and announcement of public holidays resulting in closure of most non-emergency public and private offices 11. thousands of workers started to move towards the major cities in anticipation that "lockdown" will end, and offices/factories will resume activities. however, they had to go back again as extension of public holidays announced approximately 60 areas underwent a special form of lockdown with special monitoring by law-enforcement personnel 15 the authors would like to express great appreciation to mahadi shaded (city university of key: cord-331771-fhy98qt4 authors: huang, he; chen, yahong; ma, yefeng title: modeling the competitive diffusions of rumor and knowledge and the impacts on epidemic spreading date: 2021-01-01 journal: appl math comput doi: 10.1016/j.amc.2020.125536 sha: doc_id: 331771 cord_uid: fhy98qt4 the interaction between epidemic spreading and information diffusion is an interdisciplinary research problem. during an epidemic, people tend to take self-protective measures to reduce the infection risk. however, with the diffusion of rumor, people may be difficult to make an appropriate choice. how to reduce the negative impact of rumor and to control epidemic has become a critical issue in the social network. elaborate mathematical model is instructive to understand such complex dynamics. in this paper, we develop a two-layer network to model the interaction between the spread of epidemic and the competitive diffusions of information. the results show that knowledge diffusion can eradicate both rumor and epidemic, where the penetration intensity of knowledge into rumor plays a vital role. specifically, the penetration intensity of knowledge significantly increases the thresholds for rumor and epidemic to break out, even when the self-protective measure is not perfectly effective. but eradicating rumor shouldn’t be equated with eradicating epidemic. the epidemic can be eradicated with rumor still diffusing, and the epidemic may keep spreading with rumor being eradicated. moreover, the communication-layer network structure greatly affects the spread of epidemic in the contact-layer network. when people have more connections in the communication-layer network, the knowledge is more likely to diffuse widely, and the rumor and epidemic can be eradicated more efficiently. when the communication-layer network is sparse, a larger penetration intensity of knowledge into rumor is required to promote the diffusion of knowledge. epidemics continue to trigger world alarms in recent decades [1, 2] , and have become a serious threat to human health [3] [4] [5] . currently, a new type of coronavirus (covid-19) is circulating worldwide [6] , and its plateaus hasn't been reached up to the time of writing this manuscript [7] . however, the world is still not fully prepared for the outbreak of an epidemic [8] . there are many reasons behind this, including the widespread presence of viral vectors (such as mosquitoes) [9] , virus mutation [10] , low effect of treatment [11, 12] , seasonal outbreak (such as influenza) [13, 14] , etc. as a consequence, the global protection is not enough, and people are usually encouraged to strengthen personal protection to reduce the risk of being infected, and they are also willing to do that [15, 16] . previous researches have discovered the important role of individuallevel self-protection in eliminating epidemics and saving human lives [17] . wang et al. reviewed the researches on the coupled dynamics of behavior and epidemic, and summarized that people's precautions obviously affect the prevalence of an epidemic [18, 19] . more importantly, the self-protective awareness will spread across the population [20] , yielding an interdisciplinary research area: interaction between information diffusion and epidemic spreading [21] . previous complex network models have found that information diffusion plays an important role in promoting the spread of self-protective measures during an epidemic [22, 23] . however, in the era of information explosion, the accuracy of information is difficult to guarantee. the spread of incorrect information may mislead people to choose the measures of low effect or no effect. tai et al. reviewed in detail the rumors on protective measures during the outbreak of sars, such as fumigating vinegar [24] . in addition, even if the correct information is spreading in social networks, rumors against it may discourage people from adopting it [25] . one of the reasons for the inaccurate information to prevail in social networks is that, most people only have common sense and lack relevant professional knowledge to make accurate judgments. the professional knowledge is usually mastered by a few people. for example, prelec et al. conducted a survey on the knowledge of us state capitals [26] . one question is "philadelphia is the capital of pennsylvania, yes or no?" the answer given by most respondents is "yes", which is incorrect. although the respondents are from world-class universities, they failed to give the right answer. to explore the reason behind, prelec et al. found that almost all respondents have the common sense that philadelphia is the largest city in pennsylvania, while the knowledge that "harrisburg is the capital of pennsylvania" is mastered by a few people. similarly, during sars, people may have the common sense that acetic acid has a certain bactericidal effect, but the knowledge is that the concentration of acetic acid in vinegar is not enough to kill the virus. in this research, we define inaccurate information as rumor and accurate information as knowledge. as what we learn from the study of prelec et al., rumor may spread more widely than knowledge, which is very harmful to controlling an epidemic. therefore, it is important to know how the diffusion of rumor affects the spread of an epidemic, and whether rumor and epidemic can be contained by the diffusion of knowledge. we adopt a two-layer network to model the processes of information diffusion (including rumor diffusion and knowledge diffusion) and epidemic spreading. in the communication-layer network, we adopt a ua 1 a 2 model to describe the competitive diffusions of rumor and knowledge. specifically, we propose a unidirectional transition probability from the rumor-believed state to the knowledge-believed state, which represents the penetration of knowledge into rumor. in the contact-layer network, we adopt an seis model to describe the spread of epidemic. the infected state is divided into two stages: infected but undetected (e), infected and detected (i). when infected nodes are detected, they will be quarantined immediately. previous epidemic models often assumed that the diffusion of information is driven by local risk information (e.g., whether the linked neighbors are infected [27] ). however, in our epidemic model, the direct risk sources (i.e., unobserved infected nodes) are not detected, which makes the local risk information less "protective" [28] . moreover, people prefer to making pre-protection, regardless of whether there is any infected neighbor [29] . for instance, in the previously mentioned vinegar case, people's pre-protection is mainly induced by global risk information. as another example, since the covid-19 virus was proved to be infectious, the knowledge of wearing appropriate masks is widely spread on social networks of 1.4 billion chinese users. therefore, unlike previous models, we will focus on pre-protection and assume that once the infected node is detected globally, information diffusion will begin. when a node receives a piece of rumor or knowledge, s/he is likely to take corresponding protective measures. we use three methods to analyze the model: numerical prediction, agent-based simulation and mean-field analysis. the results of the three methods are very close. based on the results, we summarize three main findings. first , the diffusion of knowledge can not only eradicate rumor, but also help to eradicate the epidemic, and the penetration of knowledge into rumor plays a key role. we analyzed the relationship of the outbreak thresholds (of rumor and of epidemic) with the penetration intensity of knowledge into rumor. in the communication-layer network, when the penetration intensity of knowledge is increased, the rumor outbreak threshold is increased linearly, so that the rumor becomes easier to control. in the contact-layer network, no matter whether the self-protective measure is perfectly effective or not, the threshold of epidemic outbreak increases nonlinearly with the penetration intensity of knowledge, but only if the penetration intensity of knowledge is large enough to make the knowledge widely spread. second , it is not necessary to eradicate rumor in order to eradicate epidemic. eradicating rumor is undoubtedly helpful to contain the epidemic, but it may cost too much. when there are enough knowledge-believed nodes, the epidemic outbreak threshold will be increased greatly. our results show that when rumor and knowledge are diffusing simultaneously in the communication-layer network, the epidemic can also be eradicated by the diffusion of knowledge. however, it should be noted that when rumor is eradicated, if the effectiveness of self-protective measures is very low, the epidemic may not be eradicated. third , the topology structure of the communication-layer network is very important for eradicating epidemic in the contact-layer network. when more links are added to the communication-layer network, which allows people to receive information from more sources, knowledge is more likely to break out, and rumor and epidemic are more likely to be eradicated. in addition, in a sparse communicationlayer network, a higher penetration intensity of knowledge into rumor is required to diffuse the knowledge. the coupled dynamics of information diffusion and epidemic spreading have been extensively studied [18, 21, 30, 31] . a typical method is to build a two-layer network, with one layer to model the diffusion of information, and the other layer to model the spread of epidemic [32] [33] [34] [35] [36] . for this research, such method is more necessary because there are two competitive dynamics for information diffusion. the structure of the two-layer network is presented in fig. 1 . the nodes in the two layers are the same, while the links are not exactly the same. wang et al. considered a case that the links in the two layers are uncorrelated [27] . some other researches proposed a coefficient to denote the overlap (or correlation) of links in the two layers [30, 37] . in this research, we propose a different overlapping method 1 . we assume that the communication-layer network is the extension of the contactlayer network. it is because that the spread of epidemic is based on offline contact, while the diffusion of information can be induced by both online and offline communication. people may never have any face-to-face contact with some online friends. to build the two-layer network, we first create a random network of node count n and average node degree k using the classical watts-strogatz (ws) network model [38] . the first created random network is the contact-layer network. next, the nodes and links in the contact network are mapped to the communication-layer network. moreover, the network is extended by letting each node randomly link with k other nodes, generating the second random network with the average node degree k = k + k . the second created random network is the communication-layer network. previous studies have successfully modeled the diffusion of rumor [39] , and found that the denying of rumor affects the outbreak size and threshold of rumor [40, 41] . but the denying of rumor in their studies is usually described by a constant probability, not driven by the diffusion of knowledge. during an epidemic, it is easy to discover the diffusion of knowledge against rumor. for instance, during the covid-19 epidemic, knowledge about the functions of different medical masks was widely spread on social media, which helps people to choose the appropriate masks [42] . moreover, cyberspace administration of china (cac) builds a specialized zone for sharing knowledge and refuting rumor during the covid-19 epidemic [43] . inspired by previous rumor models and combined with the findings of prelec et al., we propose a ua 1 a 2 model to describe the competitive diffusions between rumor and knowledge, extended from the sis model [44, 45] . each node has three possible states: unknown (u), rumor-believed (a 1 ), and knowledge-believed (a 2 ). the transition of the three states are well presented in fig. 2 . when an unknown node receives information from a rumor-believed (or knowledge-believed) neighbor, s/he may become rumor-believed (or knowledge-believed), as well. the "infectivity" of rumor (or knowledge) is α 1 (or α 2 ). after a considerable period of time, the rumor-believed (or knowledge-believed) node may forget the information and return to the unknown state [46] [47] [48] . the expected period length is 1/ f 1 (or 1/ f 2 ). namely, the probability for the rumor-believed (or knowledge-believed) node to return to the unknown state is f 1 (or f 2 ). moreover, there is an important unidirectional transition between the rumor-believed (a 1 ) state and the knowledge-believed (a 2 ) state. when the rumorbelieved node receives information from a knowledge-believed neighbor, s/he is likely to become knowledge-believed. the "infectivity" of knowledge on the rumor-believed node is defined as α 3 . apparently α 3 denotes the penetration intensity of knowledge into rumor. it should be noted that a node can transit from the rumor-believed state to the knowledgebelieved state because the knowledge-believed nodes have the extra accurate information (namely knowledge). thus, the "infectivity" of knowledge may be lower than that of rumor because people have to take more time/efforts to learn such extra knowledge. besides, previous studies also suggested that rumor might be more contagious. for example, in the new york times, awe-inspiring tales are likely to be more contagious than the regular news [49] . therefore, we set α 2 < α 1 . 1 previous overlapping methods are also applicable in our model. here "u" denotes the state that people are unknown about rumor and knowledge. "a 1 " denotes the state that people believe rumor, and "a 2 " denotes the state that people believe knowledge. α 1 and α 2 denote the infectivities of rumor and knowledge. α 3 denotes the penetration intensity of knowledge into rumor. f 1 and f 2 denote the forgetting probabilities of rumor and knowledge. we propose ϕ 1 ( t ) and ϕ 2 ( t ) to denote the probabilities of a randomly selected link pointing to a rumor-believed node and a knowledge-believed node [50] . they are obtained as where a 1 k ( t ) and a 2 k ( t ) denote the densities of rumor-believed and knowledge-believed nodes among the nodes with degree k at time t. p ( k ) denotes the distribution function of the node degree. for ease of analysis, we assume that the communication-layer network is homogeneous and the nodes have similar degrees. thus, it can be approximately derived here a 1 ( t ) and a 2 ( t ) denote the densities of rumor-believed and knowledge-believed nodes, respectively. correspondingly, the probability that a node has n 1 rumor-believed neighbors and n 2 knowledgebelieved neighbors can be given by a trinomial distribution where k is the degree of node. the discrete-time information diffusion process can be described as follows where the discrete-time probabilities of state transitions are derived following the research of liu et al [41] as blow when an unknown node receives information from a rumor-believed neighbor and a knowledge-believed neighbor at the same time, s/he is first affected by the knowledge-believed neighbor and then affected by the rumor-believed neighbor, as shown in eq. (4) . namely, we set the priority of knowledge higher than that of rumor. based on the discrete-time probabilities of state transitions, we derive the continuous-time probabilities following wu et al. [51] , as detailed in the appendix a-1. the mean-field dynamic equations of information diffusion are then derived as below, we calculate the equilibrium of information diffusion, based on which we are able to derive the thresholds for rumor and knowledge to break out, as detailed in the appendix a-2. the outbreak of rumor requires the following inequality to be satisfied here "s" denotes the state that people are susceptible to the epidemic, "e" denotes the state that people are infected but are not detected, and "i" denotes the state that people are infected and detected. β 1 denotes the infectivity of epidemic. β 2 denotes the infectivity of epidemic on the knowledge-believed nodes. γ denotes the detecting probability of the infected nodes. f denotes the recovery probability of the infected nodes. while the outbreak of knowledge requires the following inequality to be satisfied previous studies have extensively modeled the spread of epidemic in the contact-layer network [52] [53] [54] [55] . as shown in fig. 3 , we adopt a transformed seis model (or siis model [28] ), which is slightly different from the typical seis models [56, 57] , to describe the spread of epidemic. when a susceptible node contacts with an undetected infected peer, s/he may become infected but is not detected (e.g., with mild symptoms). the infectivity of the epidemic is β 1 . the undetected infected nodes are detected at rate γ . once detected (e.g., with severe symptoms), the infected nodes will be quarantined immediately and cannot infect other susceptible nodes. after a period of treatment, the detected infected nodes recover to the susceptible state at rate f . in addition, the diffusion of information has an important impact on the spread of epidemic. when the knowledgebelieved nodes take an effective protective measure, s/he will reduce her/his susceptibility to the epidemic. without loss of generality, we propose that the infectivity of epidemic on the knowledge-believed nodes changes to β 2 . apparently, β 2 < β 1 . however, the rumor-believed nodes adopt an ineffective measure and the epidemic infectivity on them remains β 1 . previous studies assumed that the spread of epidemic would continuously promote the diffusion of information, because the infected nodes were identified as risk sources for their neighbors [27] . in this research, the detected infected nodes are quarantined and "of no risk", while the direct risk sources are undetected. as a result, the effect of local risk information on containing the epidemic is greatly reduced [28] . moreover, we focus on pre-protection instead of local-risk protection (e.g., high-risk immunization [58] ). therefore, we don't consider such local risk information in this research. we use ϕ e ( t ) to denote the probability of a randomly selected link pointing to an undetected infected node. it can be calculated as [50] ϕ where e k ( t ) denotes the density of undetected infected nodes among the nodes with degree k at time t . the contact-layer network is a sub-network of the communication-layer network. we assume that it is also homogeneous. the node degree is set as k . it can be approximately obtained that ϕ e (t) = e(t ) . the probability that a node has m undetected infected neighbors can be given by a binomial distribution correspondingly, the discrete-time epidemic process can be described as follows where the discrete-time probabilities of state transitions are derived as similar to the analysis in the information diffusion model, we derive the continuous-time probabilities of state transitions based on the discrete-time probabilities, as detailed in the appendix a-1. the mean-field dynamic equations of epidemic spreading are derived as below, we calculate the equilibrium of epidemic spreading, based on which we are able to derive the outbreak threshold of epidemic, as detailed in the appendix a-3. specifically, the outbreak of epidemic requires the following inequality to be satisfied where the equilibrium density of the knowledge-believed nodes a 2 ( ∞ ) is expressed in the appendix a-1. when a 2 (∞ ) = 0 , the epidemic threshold degenerates to β 1 = γ /k . to evaluate the model, we adopt three different methods including two numerical methods and one theoretical method. the two numerical methods are: prediction based on the discrete-time state transition equations (abbreviated as predicted results), and agent-based simulation (abbreviated as simulated results). the theoretical method is the mean-field analysis based on the continuous-time state transition equations (abbreviated as mean-field results). the predicted results are first compared with the simulated results for verification. and the mean-field outbreak thresholds are then presented with the predicted results to reveal more interesting findings. to get the predicted results, we adopt matlab to iterate the discrete-time state transition equations of information diffusion and epidemic spreading. for the simulated results, we adopt repast to perform a series of agent-based simulations. initially, rumor and knowledge are believed by only the minority. but rumor has a much larger "infectivity" than knowledge and diffuses much faster. the topology of the communication layer and the contact layer are set to be random network, with the average degree defaulted as k = 8 and k = 6 , respectively. some other parameters are defaulted as: compared with a neutral unknown node, a rumor-believed node is often less susceptible to knowledge. the concept of a popular decision model, that is belief decision model [59] , can support this conjecture. according to the belief decision model, each node beliefs on three options: rumor-believed, knowledge-believed, and neutral (or hesitating). when a neutral node receives knowledge, her/his belief on knowledge will be quickly increased. while a rumor-believed node receives knowledge, her/his belief is first transferred from the rumor-believed option to the neutral option, and then transferred from the neutral option to the knowledge-believed option. apparently, the knowledge-believed belief of the rumor-believed node is less increased due to information fading during belief transfer. thus, the value of α 3 is lower than the value of α 2 . when α 3 = 0 , for example, the two information diffusion dynamics are completely separated. while α 3 = α 2 , the rumor-believed nodes have no reluctance against knowledge, and rumor is unable to affect the diffusion of knowledge. therefore, the value of α 3 is vital for the competition between rumor and knowledge, and we select it as a key parameter for analysis. as shown in fig. 4 , the equilibrium densities of the rumor-believed nodes a 1 ( ∞ ) and the knowledge-believed nodes a 2 ( ∞ ) over α 3 are well presented. increasing α 3 decreases the density of the rumor-believed nodes and increases the density of the knowledge-believed nodes. more importantly, the simulated results and the predicted results are consistent with each other. besides, from the results, we infer that both rumor and knowledge may have outbreak thresholds. specifically, the outbreak of rumor may require α 3 to be smaller than 0.1, while the outbreak of knowledge may require α 3 to be larger than 0.01. that is to say, when α 3 is larger than 0.01 and smaller than 0.1, rumor and knowledge will diffuse simultaneously. we select the infectivity of epidemic ( β 1 ) to compare the predicted results and the simulated results of epidemic prevalence. as shown in fig. 5 , the equilibrium density of the (undetected and detected) infected nodes over β 1 is well presented. increasing β 1 will undoubtedly increase the density of the infected nodes. moreover, the simulated results and the predicted results are very close to each other. besides, from the results, we can detect the epidemic outbreak threshold. it is important to study whether the diffusion of information (especially the diffusion of knowledge) could affect the epidemic outbreak threshold. to analyze the outbreak thresholds in the communication-layer network, especially to study whether and how rumor can be eradicated by the penetration of knowledge into rumor, we focus on the parameter space α 3 − α 1 to draw the heat maps of rumor-believed densities and knowledge-believed densities. the value of α 2 is set as 0.1, α 3 is confined in the interval [0, 0.1], and α 1 is confined in the interval [0.1, 1]. as shown in figs. 6 and 7 , the equilibrium densities of rumor-believed nodes and knowledge-believed nodes over α 3 − α 1 are almost complementary, reflecting the competition between them. increasing rumor infectivity ( α 1 ) will increase the density of the rumor-believed nodes and reduce that of the knowledge-believed nodes, while increasing the intensity of knowledge penetrating into rumor ( α 3 ) will reduce the density of the rumor-believed nodes and increases that of the knowledge-believed nodes. these two parameters could determine which wins in the competition, especially whether the rumor can be eradicated by knowledge. moreover, from the two figures, we observe the effects of α 1 and α 3 on the outbreak thresholds of rumor and of knowledge, and we compare the numerical predictions with the mean-field thresholds in eqs. (6) and (7) . it can be found that the numerical threshold of rumor is highly matched with the mean-field rumor threshold. as α 1 (i.e., infectivity of rumor) increases, a larger α 3 (i.e., penetration intensity of knowledge into rumor) is required to eradicate rumor. and the rumor outbreak threshold indicates that α 3 should increase linearly with α 1 , as shown below 2 the numerical threshold of knowledge is also close to the mean-field knowledge threshold 3 with the increase of α 1 (i.e., infectivity of rumor), a larger α 3 (i.e., penetration intensity of knowledge into rumor) is required for knowledge to diffuse. but it should be noted that the equilibrium density of the knowledge-believed nodes increases sharply with α 3 . even if the infectivity of rumor is extremely high, a small α 3 (e.g. α 3 = 0 . 2 ) will make knowledge ineradicable. the knowledge outbreak threshold indicates that α 3 increases nonlinearly with the increase of α 1 , as shown below therefore, from the above results, we observe a three-phase phenomenon of information diffusion over α 3 . when α 3 < , knowledge starts to diffuse. as a result, rumor is gradually confined. when α 3 > (α 1 f 2 − α 2 f 1 ) (kα 2 − f 2 ) , knowledge takes the dominant position and rumor is eradicated. thus, the value of α 3 (i.e., penetration intensity of knowledge into rumor) is crucial to information diffusion. it largely determines whether rumor can be eradicated by knowledge. 3 compared with rumor, the gap between the predicted thresholds and mean-field thresholds of knowledge is less negligible. it is largely due to two reasons. first, the mean-field analysis may lead to some bias. second, the bias becomes clear when the parameter α 3 is confined in a very small range [0, 0.1] and the step is set as a very tiny value: 0.005. to analyze the outbreak threshold of epidemic in the contact-layer network, especially to study whether the diffusion of knowledge helps to eradicate the epidemic, we consider two situations: perfect self-protection and imperfect self-protection. firstly, we focus on the parameter space α 3 − β 1 with the value of β 2 set to 0 (i.e., perfect self protection). α 3 is confined in the interval [0, 0.1], and β 1 is confined in the interval [0, 1]. secondly, we fix the infectivity of epidemic ( β 1 = 0 . 1) , and focus on the parameter space α 3 − β 2 , in order to explore how imperfect self-protection ( β 2 ∈ [0, 0.1]) affects the spread of epidemic by knowledge diffusion. as shown in fig. 8 , in the case of perfect self-protection, increasing the value of epidemic infectivity ( β 1 ) will increase the density of the (undetected and detected) infected nodes, while increasing the penetration of knowledge into rumor ( α 3 ) will decrease the density of the infected nodes. the finding verifies that the diffusion of knowledge helps to contain the epidemic. moreover, the numerical epidemic threshold is very close to the mean-field epidemic threshold in eq. (13) . as epidemic infectivity ( β 1 ) increases, a larger penetration intensity of knowledge into rumor ( α 3 ) is required to eradicate the epidemic. the epidemic outbreak threshold indicates that the epidemic infectivity ( β 1 ) can increase nonlinearly with the increase of α 3 , as shown below as shown in fig. 9 , in the case of imperfect self-protection, improving the effectiveness of self-protection (i.e., reducing β 2 ) can significantly decreases the density of the infected nodes, but highly depending on the penetration intensity of knowledge into rumor ( α 3 ). when the intensity of knowledge is insufficient to penetrate the rumor-believed nodes (e.g., α 3 < 0.02), reducing β 2 will not have much impact on the density of infected nodes. while α 3 is large enough (e.g., α 3 = 0.08), reducing β 2 can eradicate the epidemic. moreover, the numerical epidemic threshold is compared with the meanfield epidemic threshold in eq. (13) . with the increase of the penetration intensity of knowledge into rumor ( α 3 ), a less effective self-protection, namely a larger β 2 , can be enough to eradicate the epidemic. the epidemic outbreak threshold indicates that β 2 can increase nonlinearly with the increase of α 3 , as shown below it can be seen from the above results that information diffusion has significant influence on the epidemic spreading. as knowledge becomes more "penetrating" into rumor, rumor and epidemic are more likely to be eradicated. moreover, based on the outbreak thresholds in the communication-layer network, information diffusion can be divided into three phases. in this section, we try to explore the phase diagram of the whole two-layer system based on the outbreak thresholds in both layers. as shown in fig. 10 , the two-layer system can be divided into five phases by α 3 − β 2 . first, information diffusion can be divided into three phases: "no knowledge ", "rumor vs. knowledge ", and "no rumor ". second, epidemic spreading can be therefore, it is not necessary to equate rumor eradication with epidemic eradication. when rumor cannot be eradicated, the epidemic can also be eradicated by the spread of knowledge. when rumor can be eradicated, the epidemic may be uneradicated when the effectiveness of self-protection is very low. moreover, the adjacent phases are interconvertible. for example, the "rumor vs. knowledge & no epidemic " phase can convert to the "no rumor & no epidemic " phase with the increase of α 3 , and can convert to the "rumor vs. knowledge & epidemic outbreak " phase with the increase of β 2 . based on the phase diagram, it is possible to find an optimal strategy to contain rumor or epidemic. for example, if the two-layer system is in the "rumor vs. knowledge & epidemic outbreak " phase (phase ii), the optimal strategy of containing the epidemic is to develop an effective self-protective measure to reduce β 2 . the network structure highly affects the outbreak size and threshold of the spreading dynamics [60, 61] . in a two-layer network, the overlap of the two layers is also important for the spreading dynamics [37] . this research assumes that the communication-layer network is an extension of the contact-layer network. in the information age, the cost of communication is greatly reduced, and it becomes easier to extend the communication-layer network. it is interesting to know whether the extension of the communication-layer network helps to contain rumor and epidemic. therefore, in this section, we try to explore the evolution of the rumor-believed density and the infected density with the change of average node degree k in the communication-layer network. as shown in fig. 11 , when k increases from 0 to 12, the rumor-believed density will increase first and then decrease, while the knowledge-believed density will increase monotonically. obviously, rumor is more likely to break out in a sparse network than knowledge because rumor infectivity ( α 1 ) is much larger than knowledge infectivity ( α 2 ). however, when knowledge starts to break out (e.g., k = 3 ), rumor will be confined immediately and the rumor-believed density will begin to decrease. it should be noted that a higher penetration of knowledge into rumor helps knowledge to break out in a very sparse network. moreover, when the knowledge-believed density starts to increase, the infected density begins to decrease. more importantly, with the increase of k , the infected density gradually decreases to 0, which indicates that extending the communication-layer network helps to eradicate the epidemic in the contact-layer network, consistent with eq. (13) . besides, a larger α 3 in the communication-layer network is also helpful to eradicate the epidemic. in summary, extending the communication-layer network is helpful to eradicate the epidemic, but it largely depends on the condition that knowledge is able to penetrate the rumor. when people can get information from diverse sources, s/he is more likely to get access to the accurate information (i.e., knowledge) and get rid of rumor. this will make her/him be better protected from the epidemic. besides, compared with reducing rumor infectivity or epidemic infectivity, the cost of widening information sources is usually lower. during an epidemic, individuals are usually willing to take protective measures to avoid being infected. however, the diffusion of rumor sometimes prevents them from choosing the effective measures. to make things worse, such rumor is often contagious and hard to be eradicated, making the epidemic difficult to contain. therefore, how to contain rumor and epidemic has become a critical issue to the human society. in this research, inferred from the study of prelec et al. and the practice of chinese authorities during covid-19, we propose that the diffusion of knowledge is the key to controlling both rumor and epidemic. to model interaction between information diffusion and epidemic spreading, we adopt a two-layer network structure. the ua 1 a 2 model is adopted to describe the competitive diffusions of information and the seis model is adopted to describe the spread of epidemic. specifically, in the ua 1 a 2 model, we explicitly propose a unidirectional transition probability from a 1 (rumor-believed state) to a 2 (knowledge-believed state), which denotes the penetration of knowledge into rumor. three frequently used methods are used to evaluate the impact of knowledge diffusion on the spread of rumor and epidemic. the diffusion of knowledge is able to eradicate both rumor and epidemic, where the penetration intensity of knowledge into the rumor-believed nodes ( α 3 ) plays a vital role. the rumor outbreak threshold increases linearly with α 3 , while the epidemic outbreak threshold increases nonlinearly with α 3 . in particular, even if the self-protective measure is not perfectly effective, increasing α 3 is helpful to contain and even eradicate the epidemic. it is inappropriate to equate eradicating rumor with eradicating epidemic, although eradicating rumor is helpful to eradicate the epidemic. on one hand, when the density of knowledge-believed nodes increases, the outbreak threshold of epidemic will significantly increase, no matter whether rumor dies out or not. on the other hand, when rumor is eradicated, the epidemic may still prevail if the self-protective measure is not effective enough. the overlapping of the contact layer and the communication layer is a research focus in the two-layer network structure. we assume that the communication-layer network is an extension of the contact-layer network. when adding links to the communication-layer network, namely when people get information from more sources, knowledge is more likely to break out, while rumor and epidemic are more likely to be eradicated. moreover, a larger penetration intensity ( α 3 ) makes knowledge more likely to break out in a sparse network. from the above conclusions, it can be obtained that information diffusion has significant influence on the epidemic spreading. increasing penetration intensity of knowledge or improving effectiveness of self-protection is helpful to eradicate an epidemic. therefore, for policy-makers, it is necessary to know which is more suitable for a specific case. the phase diagram provides lots of implications. first, it should be evaluated at which phase the present two-layer system is located. second, an effective strategy is the one that can convert the system from a bad phase (e.g., "rumor vs. knowledge " & "epidemic breakout ") to a nearby better phase (e.g., "rumor vs. knowledge & "no epidemic "). our research has several limitations, which can be extended in future studies. first, the dynamic impact of epidemic spreading on information diffusion is not fully considered. in particular, the spread of rumor may be dynamically affected by the severity of epidemic. for example, people's susceptibility to rumor may not be constant, but inversely related to the outbreak size of epidemic. in other words, when the epidemic outbreak size increases, the infectivity of rumor may increase, as well. we set the rumor infectivity to a very large value ( α 1 = 0 . 5 ), but we didn't consider its dynamic feature. in a dynamic case, rumor and epidemic may be more easily contained at the early stage. second, the transition from the knowledge-believed state to the rumor-believed state is not considered, which is consistent with most of previous rumor models [41] . in extreme cases, such transition may occur. for example, a wicked rumor may make people to abandon the effective practice [25] . introducing such transition into our model may make the competition between rumor and knowledge more salient. third, the topology structures of the contact-layer network and the communication-layer network may be different, which may affect the spreading process. fortunately, although we only study the random network, this research can be easily extended to different network structures. world on alert for potential spread of new sars-like virus found in china a game theoretic approach to discuss the positive secondary effect of vaccination scheme in an infinite and well-mixed population epidemics and rumours in complex networks economic impact of dengue illness in the americas a systematic review of the social and economic burden of influenza in low-and middle-income countries clinical features of patients infected with 2019 novel coronavirus in wuhan forecasting covid-19 are we ready for pandemic influenza? evolutionary enhancement of zika virus infectivity in aedes aegypti mosquitoes optimal control of influenza epidemic model with virus mutations mathematical analysis of an sis model with imperfect vaccination and backward bifurcation modelling the effect of imperfect vaccines on disease epidemiology interplay between cost and effectiveness in influenza vaccine uptake: a vaccination game approach is subsidizing vaccination with hub agent priority policy really meaningful to suppress disease spreading braess's paradox in epidemic game: better condition results in less payoff modelling and analysing the coexistence of dual dilemmas in the proactive vaccination game and retroactive treatment game in epidemic viral dynamics saving human lives: what complexity science and information systems can contribute coupled disease-behavior dynamics on complex networks: a review statistical physics of vaccination effect of information spreading to suppress the disease contagion on the epidemic vaccination game coupling dynamics of epidemic spreading and information diffusion on complex networks analysis of sir epidemic model with information spreading of awareness coevolution spreading in complex networks the rumouring of sars during the 2003 epidemic in china vaccination strategies in a two-layer sir/v-ua epidemic model with costly information and buzz effect a solution to the single-question crowd wisdom problem asymmetrically interacting spreading dynamics on complex layered networks identifying the direct risk source to contain epidemics more effectively how the individuals' risk aversion affect the epidemic spreading dynamical interplay between awareness and epidemic spreading in multiplex networks suppressing disease spreading by using information diffusion on multiplex networks the impact of awareness diffusion on sir-like epidemics in multiplex networks interplay between sir-based disease spreading and awareness diffusion on multiplex networks epidemic spreading with activity-driven awareness diffusion on multiplex network analysis of epidemic outbreaks in two-layer networks with different structures for information spreading and disease diffusion the impact of information spreading on epidemic vaccination game dynamics in a heterogeneous complex networka theoretical approach a colored mean-field model for analyzing the effects of awareness on epidemic spreading in multiplex networks collective dynamics of small-world networks dynamical analysis of rumor spreading model in homogeneous complex networks stochastic rumours modeling cyber rumor spreading over mobile social networks: a compartment approach tips for choosing and using masks to prevent novel coronavirus specialized zone for rumor-refuting during the 2019-ncov epidemic dynamical and correlation properties of the internet immunization of complex networks rumor spreading model considering forgetting and remembering mechanisms in inhomogeneous networks rumor spreading model with consideration of forgetting mechanism: a case of online blogging livejournal a rumor spreading model with variable forgetting rate what makes online content viral? epidemic spreading in scale-free networks the impact of awareness on epidemic spreading in networks epidemic spreading on heterogeneous networks with identical infectivity epidemic spreading in correlated complex networks thresholds for epidemic spreading in networks epidemic spreading on weighted complex networks global dynamics of an seis epidemic model with saturation incidence and latent period an seis epidemic model with transport-related infection efficient immunization strategies on complex networks a belief-based model for characterizing the spread of awareness and its impacts on individuals' vaccination decisions the small world yields the most effective information spreading absence of epidemic threshold in scale-free networks with degree correlations this work is supported by social science foundation of beijing municipality (grant #: 18gla009 ) and specialized foundation of beijing wuzi university (grant #: 2019xjqn07 , 2020yqyj01 ). supplementary material associated with this article can be found, in the online version, at 10.1016/j.amc.2020.125536 key: cord-103475-90k21u1w authors: de giorgio, andrea; lundgren, magnus; wang, lihui title: procedural knowledge and function blocks for smart process planning date: 2020-12-31 journal: procedia manufacturing doi: 10.1016/j.promfg.2020.05.148 sha: doc_id: 103475 cord_uid: 90k21u1w abstract in the age of digital manufacturing there is a need to elicit and transfer procedural knowledge between humans and machines. having proper knowledge is essential in decision-making. the more the knowledge, the better decisions are made. to capture experiences and turn them into knowledge is fundamental in learning processes and knowledge development. knowledge engineering and knowledge management have been subject for research for decades and several concepts about knowledge and knowledge transfer have been introduced, but a functional approach to exploit knowledge efficiently in manufacturing is still missing. in the era of industry 4.0, humans and machines must be able to collaborate in such way that both can exploit the best abilities of each other in a manufacturing process. this paper introduces a procedural knowledge process (pkp) approach to capturing and defining unexpected events, while a process step is able to perform its required functions and transfer that information as machine-understandable knowledge about a failure mode. function blocks (fbs), as per the iec-61499 standard, have been proposed as a way to achieve distributed process planning in which the manufacturing process can adapt itself to runtime conditions, e.g. machine availability, etc. however, fbs are event-driven systems and the approach is limited to work under well-known runtime conditions, e.g. machine configurations and states, or deviations which are impossible to foresee in advance, for instance the outcome of a process failure mode effects analysis (pfmea). the pkp introduced in this paper, aims at bridging this gap by integrating at runtime an expert operator’s solution based on root cause analysis (rca) in an fb architecture, making the fb knowledge-driven systems, for further executions of the same without redesigning it. natural language representations of procedural knowledge blocks (pkbs) allow to transfer procedural knowledge to human operators, i.e. explain the process flow of a machine decision, while machine representations of pkbs allow to embed procedural knowledge that is elicited from expert operators upon unexpected events into the fbs process. the resulting pkp enhances the fbs for smart industrial applications, such as the process planning use case described in this paper. computer-aided process planning (capp) is the use of computer technology to support process planning in manufacturing. the first capp systems were conceived more than forty years ago. one of the earliest capp systems was the variant planning system cam-i automated process planning in 1976 [1] . a decade earlier, niebel [2] discussed the use of computers for mechanized selection of optimum processes for manufacturing at design stage. even though niebel has been recognized as one of the first who discussed the use of computers in process planning [3] , the impact of computers in manufacturing was addressed by diebold [4] in 1952, thirteen years before niebel, who in a "provocative study of the possibilities, limitations, and social and economic consequences of the revolutionary new machines of the electronic age" envisioned future computerized automatic factories. capp has been subject for research since the early 1970s and the first process planning system known for using artificial intelligence (ai) technology, gari, was presented by descotte et al. in 1981 [5] . the year after, weill et al. published the first large survey of capp systems [6] . several capp systems and capp research reviews, e.g. [2, 3, [6] [7] [8] [9] [10] [11] [12] [13] , have since been presented. xu et al. [14] concludes that research has faced many still unsolved difficulties in attempting to implement ai techniques in capp. in a recent computer-aided process planning (capp) is the use of computer technology to support process planning in manufacturing. the first capp systems were conceived more than forty years ago. one of the earliest capp systems was the variant planning system cam-i automated process planning in 1976 [1] . a decade earlier, niebel [2] discussed the use of computers for mechanized selection of optimum processes for manufacturing at design stage. even though niebel has been recognized as one of the first who discussed the use of computers in process planning [3] , the impact of computers in manufacturing was addressed by diebold [4] in 1952, thirteen years before niebel, who in a "provocative study of the possibilities, limitations, and social and economic consequences of the revolutionary new machines of the electronic age" envisioned future computerized automatic factories. capp has been subject for research since the early 1970s and the first process planning system known for using artificial intelligence (ai) technology, gari, was presented by descotte et al. in 1981 [5] . the year after, weill et al. published the first large survey of capp systems [6] . several capp systems and capp research reviews, e.g. [2, 3, [6] [7] [8] [9] [10] [11] [12] [13] , have since been presented. xu et al. [14] concludes that research has faced many still unsolved difficulties in attempting to implement ai techniques in capp. in a recent computer-aided process planning (capp) is the use of computer technology to support process planning in manufacturing. the first capp systems were conceived more than forty years ago. one of the earliest capp systems was the variant planning system cam-i automated process planning in 1976 [1] . a decade earlier, niebel [2] discussed the use of computers for mechanized selection of optimum processes for manufacturing at design stage. even though niebel has been recognized as one of the first who discussed the use of computers in process planning [3] , the impact of computers in manufacturing was addressed by diebold [4] in 1952, thirteen years before niebel, who in a "provocative study of the possibilities, limitations, and social and economic consequences of the revolutionary new machines of the electronic age" envisioned future computerized automatic factories. capp has been subject for research since the early 1970s and the first process planning system known for using artificial intelligence (ai) technology, gari, was presented by descotte et al. in 1981 [5] . the year after, weill et al. published the first large survey of capp systems [6] . several capp systems and capp research reviews, e.g. [2, 3, [6] [7] [8] [9] [10] [11] [12] [13] , have since been presented. xu et al. [14] concludes that research has faced many still unsolved difficulties in attempting to implement ai techniques in capp. in a recent 48th sme north american manufacturing research conference, namrc 48 (cancelled due to review of the past 35 years 1981-2016 research on ai system applications in process planning and manufacturing, leo kumar [15] stated that a limited number of works has focused on knowledge acquisition methodology and key issues involved during knowledge acquisition. in addition, lack of standardized definition of what process planning encompasses in terms of activities [16] puts up challenges in designing capp systems capable of supporting all kind of applications. current knowledge-based methods are based upon expert knowledge codified in advance, e.g. as explained by park [17] . this knowledge is "fixed" and is not extended or updated at runtime. current capp approaches are also limited by the lack of ability to respond to unpredictable changes in daily industrial applications. this paper addresses the question of capturing human expertise during common operations and not just at knowledge systems design stage, considering industrial operations as a dynamic procedural knowledge process (pkp). as emphasized by gorecky et al. [18] and thoben et al. [19] humans are the most flexible entity in the industry 4.0 paradigm. in contrast to the technology oriented computer-integrated manufacturing (cim) paradigm of the 1980s, humans are not to be simply replaced by artificial intelligence and automation on the shop floor but integrated into the cyber-physical structure in such a way that their individual skills and talents can be fully realized. human and machine collaboration is discussed by zhong et al. [20] who suggested a "human-in-the-loop" machine learning approach as one future research direction under industry 4.0. advanced learning models for machines enable humans to interact efficiently and effectively in decision-making, for example traditional machine learning systems which are interjected with human knowledge to improve human-machine interactions and communications. however, a detailed outline of such collaboration is out of the scope for this paper and will be subject for further research. many researchers and philosophers have defined knowledge in various ways. attempts to formalize knowledge started more than two thousand years ago. aristotle's categories and his system of syllogisms for reasoning about the categories were the most highly developed systems of logic and ontology. sowa among the latest ones, defined the knowledge soup, which explains how some knowledge may be represented in symbolic or propositional form, but much, if not most of it, is stored in image-like forms [21] . making sense from an engineering perspective of the knowledge soup and unifying all the knowledge definitions might look challenging, nonetheless. therefore, this paper proposes a functional redefinition of declarative and procedural knowledge, together with the concepts of procedural knowledge blocks and process as a framework to deal with knowledge processes in industry. function blocks (fbs), as per the iec-61499 standard [22] , can support intelligent and autonomous feature-based machining. fbs have been successfully used to manage process planning in a distributed manner [14] and allow for a better portability among resources, dynamism and adaptability [23] . this work points out a gap in dealing with unforeseen events at runtime. in fact, these systems are not specifically designed with an ability to define new states at runtime, a property that this paper points out is needed to be able to define and handle unforeseen events that might be happening during their execution, in accordance with the current state of the art. this paper introduces in section 2 a framework for capturing procedural knowledge in a functional way through procedural knowledge blocks (pkbs). in section 3, fbs are introduced and it is explained how they can support distributed process planning. in section 4, it is briefly introduced how process failure mode effects analysis (pfmea) and root cause analysis (rca) contributes to the procedural knowledge process (pkp). finally, the paper explains how pkbs can bridge a functional gap and enhance current fbs approaches by capturing human actions upon unexpected events and embed successful solutions as procedural knowledge into common fbs. a simple industrial manufacturing use case shows how an operator can stay in the loop and transfer their knowledge. engineering operates with models able to reproduce the functionality of real world objects or meta-objects. if knowledge is something that can be stored, computed, used and transferred, then there must be an engineering model that can do so. this paper aims at redefining declarative knowledge (dk) and procedural knowledge (pk) in order to introduce a framework called procedural knowledge process (pkp), which is based on the engineering use of the new definitions. a diagram called procedural knowledge block (pkb) is introduced to visualize dk and pk and their relationships. the concept of knowledge representation [24] with respect to knowledge is considered as part of the new definitions in order to be able to connect previous scientific work on the topic. a simple example, not pertaining the area of manufacturing, is provided in order to learn to reason intuitively in terms of dk and pk before applying them to the manufacturing use case presented later in this paper. among the most recent definitions, joachim funke who is professor in theoretical and cognitive psychology defines knowledge as "a piece of subjectively acquired information about the world" that "can be construed as embodied information" [25] . nico stehr who is a pioneer in the knowledge of societies defines knowledge as "the capacity to act, whereas information does not enable one to set anything in motion" [26] . haapasalo et al. have done an extensively review on the matter of knowledge types and have showed that there is agreement on two kinds of knowledge: declarative (also called descriptive or conceptual) and procedural (also called practical) [27] . the most common distinction, originally from skemp and papert, is that declarative knowledge is identified as "knowing-that" or knowledge of facts, embodying concepts, principles, ideas, schemas and theories, while procedural knowledge is identified as "knowing-how", especially intended as how to perform some tasks [28] [29] [30] [31] . the discovery that biology might have different structures to store declarative and procedural memories [32] might have encouraged the scientific community to separate the definitions of dk and pk, making it difficult for engineering to apply a unified type that can be used for practical purpose. hence, a better definition of dk and pk needs to try to reconcile the differences as much as possible. a more functional definition of declarative knowledge, starting from funke's definition [25] , needs to consider knowledge as organized information, subjectively acquired by a system, about another system, that is non-necessarily the world. in order to make the definition applicable to any engineering systems, funke's subjective or world knowledge has to be generalized to system knowledge. as by definition, a system is intended as a regularly interacting or interdependent group of items forming a unified whole [33] . in none of the studied literature there has been any reference to time within the knowledge definitions. hence, a more functional definition of dk also needs to introduce time and characterize the system at an instant of time in order to imply its stationary nature: a description of a system can only be valid if the system does not change or if it changes really slowly with respect to the time scale considered. due to the introduction of the time variable, there is a possibility to simplify the definition of pk. in fact, a better definition of pk can be considered as the variation of dk in a time interval. the advantage is that one definition is directly based on the other one and connected by a variation of time. in light of these considerations, the following definitions are proposed: definition 1. declarative knowledge is systemic information of a system a, related to information about a system b, at an instant of time . procedural knowledge is systemic information of a system a, related to the variation of information about a system b, between two instants of time 1 and 2 . an immediate advantage of the definitions introduced by this paper is that they remove the need to express knowledge and knowledge representations as two separate concepts. the reduction of ambiguity is vital in engineering applications. in fact, imagine that a system a has knowledge about a system b. then, there must be structural information instantiated in system a, i.e. a certain reorganization of its parts, in order to represent information about system b. thereby follows that there is no knowledge without a representation. this outcome helps to reconcile previous research presented in literature on knowledge and knowledge representations and can be postulated as the following definition: definition 3. a knowledge representation is the instantiation of information in a system through the reorganization of its parts. this definition goes in line with a well-cited journal paper from davis et al. [34] in which a knowledge representation is defined as:  a surrogate;  a set of ontological commitments;  a fragmentary theory of intelligent reasoning;  a medium for efficient computation;  a medium of human expression. in fact, if a knowledge representation is achieved in a system (surrogate) through the reorganization of its parts, i.e. a change of states, any systems can in theory represent knowledge. a system can be described by a set of ontologies. an intelligent system can act upon its own knowledge and become a medium for efficient computation. moreover, humans, seen as systems, interact with other systems by means of language, images and every knowledgeable product of their being. the definitions of dk and pk can be interpreted in the following intuitive way: declarative knowledge is like having a snapshot of a system at a given time while procedural knowledge is like having a video sequence of a system within a time interval. photos and videos are familiar systems to capture knowledge about systems, e.g. a kid playing in the sand, it is intuitive to grasp how they differ for the ability to capture knowledge. an image is static, as it does not express time, unless explicitly stated with tricks such as time-arrows or referring to actions taken by static figures. a video, on the other hand, is composed of successive frames (images). when they are seen in succession, they produce a dynamic experience that takes into account the flowing of time. something that can be recognized as an event or an action. as an intuitive example, imagine a man who is swimming in a calm lake. the man and the lake are two separate systems that can be described by means of language (a system). reading this sentence changes the reader's system (brain) state and creates a mental image at time that is declarative knowledge. bringing the example further, think about the man getting out of the lake at time + 1 . the effect of this second declarative knowledge is an immediate interpolation of the two images as frames of a continuous video stream. procedural knowledge is generated by assuming that these two images are descriptions of a variation of states of the same systems (the man and the lake), happening between time and time + 1. this holds true even when the two states are equal but there is time running in between the states. for example, knowledge about a mobile system that instead maintains its spatial position during a period of time is also procedural knowledge, e.g. the man floating for a while at the same spot in the lake. there is plenty of literature that ties actions with knowledge [35] [36] [37] in a superfluous way, however the practical application sought by this paper requires a formal connection between action, dk and pk. it is a common habit to use an action (intended) or an event (unintended) to explain how time has progressed if a system varies in between two instants of time 1 and 2 . since this variation can be expressed by a pk, action/events and pks are interchangeable ways to describe how a system evolves. going back to the example of the man swimming in the lake, the mental sequence of images (dks) provided by the pk is a placeholder for an action/event label such as "a man is swimming in the lake". the correspondence between pk and actions/events can be visualized as a procedural knowledge block (pkb), which is a diagram composed of two instances of dk at different times, separated by an action/event label that refers to a systemic transformation in the interval of time between them (see fig 1) . the very correspondence between actions/events and pks makes the pkb a conceptual module for knowledge-driven systems that can be integrated with or upgrade event-driven systems, such as the function blocks introduced in the next section. since both human language and machine language can quite precisely describe a system state, two corresponding pkbs can be used to refer to the same pk, under two different instantiations. one is used to communicate with humans, in natural language, the other is used to communicate among machines, through direct machine state representations. an interesting observation related to automatic translations is that the approach of storing and using overall meaningful sentences in different languages has proven to be more efficient than word-by-word translations [38] . this intuitively leads to use a similar approach to transfer knowledge between operators and machines by means of pkbs which is conceptually demonstrated in this paper and will be explored further in future work. the problem of transferring knowledge from humans to machines and the other way around can be thus reformulated into the problem of translating knowledge from one systemic instantiation to another. a procedural knowledge process is a framework to structure any processes, especially the industrial ones, using the procedural knowledge blocks introduced before. in particular, there are two possible ways of connecting the blocks: in series or parallel. when the blocks are connected in series, the declarative knowledge part that ends the first block is the same that starts the second block. when two blocks are executed in parallel, the declarative knowledge at the beginning and at the end of both blocks coincides. a pkp is a diagram of a process that can be directly implemented as a system, exactly in the same way as a system can be described by a finite state machine (fsm) and then implemented. explanation (left) and abstraction (right) of a process through procedural blocks. explanation produces several procedural knowledge blocks starting from a single block. the initial state of the first produced block and the final state of the last produced block are the same of the original block. abstraction merges several knowledge blocks into one, which is labelled after the most abstract action/event that leads the first block initial state into the last block final state. by taking advantage of the holistic aspect of the pkbs, each of them can be replaced by several blocks. such operation is called explanation, see fig 2 (left) . in reverse, several blocks can be aggregated into one. such operation is called abstraction, see fig 2 (right) . thus, knowledge can be expressed at different detail levels or depths. starting from the most abstract level of a given procedure, e.g. machining of a gear box housing, which can be represented by a single pkb, one can explain the process task by task in sequential or parallel operations represented each by one pkb. the resulting diagram is a pkp. function blocks are defined in the iec-61499 standard [22] and can be immediately classified into two types: basic fbs and composite fbs. a basic fb is composed of two main parts that are the execution control chart (ecc), i.e. a finite state machine controller of the fb operations, and the core of the fb containing the algorithms. each basic fb, see fig 3 (a) , can encapsulate the algorithms needed to perform an operation that can be, for example, part of an assembly plan or a process plan for a specific assembly or process feature as shown by wang et al. [39, 40] . a composite fb, see fig 3 (b) , which is made of several basic fbs connected by events and data, relies on the internal states and embedded algorithms of the basic fbs. this allows to create a whole process made of several basic fbs that can be operated by one or more composite fbs, especially to handle a distributed process plan. in fact, the use of fbs solves the problem of handling the computer numerical control (cnc) language as machine neutral language. it adds portability in the form of fbs compatible to various machine tools with similar capability, where machine-specific data, e.g. cutting parameters or tool paths, are generated ad hoc. in case of a machine failure, the same function blocks can be redirected to another machine for local optimization and execution [40] . however, it is important to note that a basic fb can produce different outputs even when the input is the same, depending on its internal state info. it is a characteristic dictated by the choice of defining an fb as an explicit event-driven model: when a trigger event is passed as an input, the fb controller starts the execution of its embedded algorithms, despite of the system conditions. it is the ecc that has the role to overview that the execution of the fb algorithm is completed in the best possible way given varying system conditions, i.e. external inputs and internal variables. another way of seeing it is as a resource-driven model: once one or several resources are specified, the embedded algorithms of an fb will find the optimal parameters of the fb task for the given resources, e.g. machine. this, of course, allows for a great flexibility at runtime, but everything must be embedded in the fbs by design before execution, in order for it to handle all the varying but well-known runtime events. in fact, fbs are not structurally meant to define new and unexpected events at runtime, e.g. a product breaks during machining, and adjust the process to continue if the event was not foreseen at design stage. that is the part where a procedural knowledge process, introduced in section 2, can step in and bridge the gap, exploiting the preexisting fb structure to instantiate the missing events. with this approach, in section 5, it is explained how fbs can be transformed from event-driven systems to knowledge-driven systems. failure mode and effect analysis (fmea), as defined in the iso/iec 31010 standard, is a technique to identify the ways in which components, systems or processes can fail to fulfill their design intent [41] . when talking about process planning, process failure mode and effect analysis (pfmea) is often mentioned. pfmea can be actively used to design and implement fbs architectures that are fault-resilient. however, there are no dynamic applications for it at runtime, when the fbs architecture is already designed and running a process. if and when a fault occurs at runtime, it is because of missing knowledge of such possibility at design stage, which was not spotted by the pfmea. major faults can trigger a redesign of the fbs architecture, which implies a stop of the operations that induces further losses. root cause analysis (rca), as defined in the iso/iec 31010 standard, attempts to identify the root or original causes of a major loss instead of dealing only with the immediately obvious symptoms, in order to prevent its reoccurrence [41] . in order to include an rca into an fbs architecture, the latter needs to be redesigned to include ad hoc fbs to respond to events generated by states that are discovered to be not desired, i.e. the root causes of successive losses. this is a complementary stage for the lean design of systems: rca can spot missing knowledge that can be integrated in the design stage, when pfmea is performed. this opens up a gap for methods and subsequent systems that can elicit rca knowledge from industrial actors and learn from it at runtime. ideally, the outcome is a dynamical instantiation of new knowledge-driven fbs to handle the newly learned event, without the need to redesign the whole architecture. the demonstrative use of the framework proposed in this paper is to exploit skilled operators' knowledge to avoid critical faults in process planning by spotting and absorbing minor faults in the process at runtime before they become critical. the aid of pkbs in redesigning the fb process at runtime makes the fb process itself more adaptable and resilient to changing or new conditions. the research gap that is covered is about learning from unforeseen events at runtime through the operators' experience. in this regards, the machining of a gear box housing is implemented as a use case to demonstrate how an unexpected event, as well as the successful actions of an experienced user upon the event, can be captured in one or several pkbs. the gear box housing has three bearing seats with a mutual functional relationship between the center-axes of the bearing seats. most likely a four, or a five-axis milling machine would be a process planner's primary choice. however, as the intention with the case is to demonstrate how the pkp can capture unexpected events and successful actions upon them, the process planning scenario has been executed with a threeaxis machine due to unavailability of a more versatile five-axis. the choice of using the three-axis machine is also useful because it forces the involvement of an operator between each setup, which is essential for a human-machine knowledge transfer in form of pkbs between the operator and the automated fbs process. the process plan for machining a gear box housing is made in seven setups. in the first setup the top face, and a step around the workpiece is machined. the step fulfills an important process function in serving as intermediate location surfaces in following setups. it is important to obtain an accurate surface when machining the top face (indicated in green in fig 4) in setup 1, as this face is put against the solid jaw of the vise to achieve accurate location in setup 2 to 5. in fact, if the top face for some reason (which can be explained) becomes too rough in setup 1, that face might not fulfil its intended function as location surface roughness. an experienced operator may notice that the operation has not ended in an intended way. secondly the operator may find the cause of the problem, eliminate and adjust the process. the fb architecture is introduced, as showed in fig 5, to control the process from setup 1 to setup 7. the first function block in the simplified schema is not connected, assuming that the start signal will be given by other systems or manually by the operator. every fb responsible for each setup operations is connected in cascade to the fb responsible for the next setup operations. these are to be considered composite fbs, as they coordinate the whole set of operations after a certain setup. between setups, the operator is involved as they have to manually interact with the machine and the product to realize the setup. the fb simply waits for a confirmation of the operator, e.g. pushing a button, to proceed with the machining operations. this simple process is repeatable, and the complexity of the fb structure can handle any foreseen event that can be added to it in design phase, but in case of unforeseen events there is no possibility to redesign the operations on the fly. in order to see this, a finite state machine of the process can be drafted, see fig 6. in fact, the ecc of an fb is nothing else than an fsm. the number of states is finite and consists of all the expected ones, by design. each setup corresponds to an operator involvement in the process. the smooth surface obtained after setup 1 is needed for good clamping in setups 2, 3, 4 and 5 and for the success of the relative fbs operations. fig 5. a simplified schema of composite fbs controlling the machining operation to produce a gear box with a 3-axis machine. seven setups are required and manually executed by an operator, before that each fb starts controlling the cnc machine operations. in read, the event-driven flow of the process. fig 6. finite state machine. the first states of the machining process of the gear box housing are described through an fsm that is the equivalent of an ecc for an fb. in light blue the unexpected state that is reached with unrecognized operations* performed after setup 1. an expert actor can identify the unexpected state and restore setup 1 by expanding the fsm at runtime. this is possible with the joint use of procedural knowledge blocks and function blocks, as showed in this paper. despite many good executions of the fb machining process, unwanted outcomes might still arise. for example, after the fb execution of the face milling operation after setup 1, the blank surface that is supposed to be smoothly machined as in fig 7 appears instead as the rough surface showed in fig 8. in regular conditions, fbs are able to complete the process plan without errors. the flow of operations is continuous from setup 1 to setup 7. an actor intervenes at each setup and triggers the starting signal of the next function block (be that through a confirmation button or automated sensors). in case of failure, the only operation that is allowed by design is to abort the execution of the fb process plan or to restart it, placing a new blank and discarding the corrupted one. unintended outcome of face milling in setup 1: due to that cleaning of machine table has failed when the fixture was mounted; remaining swarfs contributes to create vibrations which causes the top surface of the blank unable to fulfill its location function. however, if not spotted by an operator, this failure will not prevent the process to advance where the rough surface of setup 1 will be the cause to a more critical failure in the critical setups 3, 4 or 5. failure mode and effect analysis (fmea) can be fixed at those stages, when it will be too late for the product to be saved but useful feedback can be learned from the production fault. on the other hand, anticipating the effect of an unexpected setup 1 would allow to save the product by restoring good conditions before the following critical setups. such a stop requires an actor to run an rca and determine the cause of the error, then restore the process if possible or restart it. in either cases that the operator does or does not have such expertise, they can only restart the process and report the malfunction for someone else to analyze it and redesign the system. the risk is that the next process might fail again because the cause of fault has not yet been identified and might happen again. if this is the case, the production is stopped and the interruption might bring additional costs to those determined by the presence of a faulty product once and then, which in case of an aluminum gear box might be not expensive, but surely not affordable for a turbine component machined from a titanium block [42] . both conditions can eventually be avoided by exploiting human expertise at runtime to prevent failure and improve the production process for the next executions. structuring the fb process as a pkp, allows to explain each fb operation by an associated pkb, as in fig 9 (a) . when the unexpected event occurs, an operator is requested to produce a pkb documenting the unexpected state with their own declarative knowledge. furthermore, if an rca can be successfully performed and a recovery procedure is known to the operator, it can be documented in another pkb for future executions, as in fig 9 (b) . (c) descriptive knowledge of the procedural knowledge blocks. it can be represented by system states, determined by internal and input variables. a is the blank, a* is the blank and swarf in the machine, b is the correctly milled surface after setup 1, b* is the incorrectly milled surface after setup 1, c is the outcome of the machining operations after setup 2. a simple scenario is introduced to explain the creation of new pkbs by the operators. the presence of swarf, state a* in fig 9 (c) , in the machine has led to a faulty execution of the machining operations after setup 1, see fig 9 (b) and state b* in fig 9 (c) . the rough surface, showed in fig 8, can be recognized as effect of something that went wrong, the swarf identified as the problem and setup 1 executed again to correct the smoothness of the surface, after cleaning the machine from any residuals of the previous operation, see state a in fig 9 (c) . if setup 1 is not executed again in the correct conditions, then setups 3, 4 and 5 will create a product out of tolerance. for this, several pkbs are created, see also fig 9 , as follows:  the creation of a pkb for a correct setup 1 can be done by an operator simply memorizing a natural language message of the kind "please, operator, check that the machine is clean" in an available media support (text, video, audio, etc.).  the pkb is directly translated in the instantiation at runtime of an fb that represents the event-driven manifestation of that pkb. the fb needs to act as a placeholder for the presence of a pkb in the next executions of the fb architecture. for example, in case that the same event occurs during a process, the newly instantiated fb is called and this triggers the pkb known as "please, operator, check that the machine is clean". meanwhile, the fb architecture pauses on the instantiated fb while an operator needs to manually retrieve the pkb associated to the fb (it can be visualized on a screen, or a message on a speaker, etc.), interpret its instructions based on natural language, execute them and confirm the execution to the fb architecture so that the instantiated fb is considered executed. this allows to keep the operators in the loop, while executing automated processes through fbs. this structure is embedded in the fb architecture by adding at design phase an additional composite function block that runs the pkps and can instantiate additional fbs representing each new pkb instantiated at runtime (pkb-fbs). the functionality to instantiate function blocks at runtime is present in the iec-61499 standard. the pkb-fbs control the flow of the standard fbs through the additional events that they define. such events are procedural knowledge-based and transform the fb architecture from an event-driven technology, to a truly knowledge-driven one. an unexpected event is shaped as a pkb and this into a runtime-instantiated pkb-fb, that is set to be run in between regular fbs, thus redefining the flow of the fbs process, as described in the ecc or fsm (see fig 6, state in light blue). when similar conditions are assessed by the new pkb-fbs, the event takes place as expected, instead of unforeseen, and the process flow continues uninterrupted. learning or knowledge transfer from the operator to the machine has happened. this paper redefines declarative and procedural knowledge and introduces the procedural knowledge blocks as useful diagrams to generate a procedural knowledge process, a framework to manage knowledge in industrial engineering processes. the fbs are presented as a viable technology to implement smart capp systems in industry 4.0, provided that they are enhanced with procedural knowledge capabilities as showed in this paper. it has been presented a way to use the procedural knowledge process framework to transform event-driven systems implemented with fbs to knowledge-driven systems implemented with pkbs and fbs. the clear advantage is the ability to represent and handle unforeseen events at runtime. a simple process planning use case about the machining of a gear box housing is used as a proof of concept. the scenario requires unforeseen minor faults to manifest themselves before that critical faults can happen and that these are spotted by operators before irreparably damaging the product. the application allows to exploit the pkp framework to elicit knowledge from the operators and reshape the fb process at runtime, without interrupting the production or redesigning the fb architecture. the outcome is the preventive recovery from failure mode and knowledge transfer from humans to machines of the recovery operations performed by the operators on the shop floor. future work can be oriented in the following directions:  implementing a full demonstrator for the pkb instantiation at runtime that can be tested preferably on a real industrial case.  demonstrating the possibility to transfer knowledge between humans and machines by corresponding pkbs in different languages, i.e. respectively natural language and machine language, as proposed in section 2.5.  reshaping the whole process plan as a pkp in design phase, in order to make fb architectures with greater explainability and modularity. research shall be conducted in such direction because procedural knowledge might have a determinant role in the way industrial processes are shaped and implemented in automated systems.  exploring further human-machine interaction and collaboration between operators and industrial systems with interfaces built with the aid of the pkp framework. because efficient and explicit declarative and procedural knowledge transfer is needed to achieve meaningful collaborations. computer-integrated design and manufacturing mechanized process selection for planning new designs computer aided process planning: the stateof-the-art survey the advent of the automatic factory gari: a problem solver that plans how to machine mechanical parts survey of computer-aided process planning systems computer-aided process planning: the present and the future generative computer aided process planning for part manufacturing: an expert system approach methods to improve the response time of a capp system that author name / procedia manufacturing 00 evolution and future perspectives of capp a review of knowledge-based expert systems for process planning principles of process planning : a logical approach process and operation planning computer-aided process planning -a critical review of recent developments and future trends state of the art-intense review on artificial intelligence systems application in process planning and manufacturing process design as fundament in efficient process planning, procedia manuf knowledge capturing methodology in process planning human-machineinteraction in the industry 4.0 era conf. ind. informatics, indin industrie 4.0" and smart manufacturing -a review of research issues and application examples the challenge of knowledge soup iec61499 -international standard for distributed systems yiming rong, step-nc and function blocks for interoperable manufacturing what is a knowledge representation? how much knowledge is necessary for action knowing and not knowing two types of mathematical knowledge and their relation the cambridge handbook of thinking and reasoning intelligence, learning, and action : a foundation for theory and practice in education mindstorms: computers, children, and powerful ideas the cognitive psychology of school learning the declarative/procedural model: a neurobiological model of language learning, knowledge, and use what is a knowledge representation? knowledge preconditions for actions and plans, readings distrib combining declarative, procedural, and predictive knowledge to generate, execute, and optimize robot plans knowledge representation and reasoning improving translation via targeted paraphrasing conf. empir. methods nat. lang. process design of adaptive function blocks for dynamic assembly planning and control designing function blocks for distributed process planning and adaptive control -risk management -risk assessment techniques modern machining of advanced aerospace alloys-enabler for quality and performance key: cord-317900-05y9re12 authors: senanayake, nari; king, brian title: geographies of uncertainty date: 2020-08-14 journal: geoforum doi: 10.1016/j.geoforum.2020.07.016 sha: doc_id: 317900 cord_uid: 05y9re12 abstract the question of uncertainty has generated substantial critical engagements across the social sciences. while much of this literature falls within the domains of anthropology, science studies, and sociology, this short introductory paper highlights how geographical scholarship can also enrich emerging transdisciplinary debates on uncertainty. specifically, we discuss how geographers engage with uncertainties produced through and reconfigured by some of the most formidable issues of our contemporary moment, including neoliberal transformation, disease and illness, resource conflict, global climate change, and ongoing struggles around knowledge, power, and justice. in conversation with debates in cognate fields, this special issue brings together contributions that grapple with uncertainty through key geographic concepts such as scale, power, spatiality, place, and human-environment relations. this work extends scholarly understanding of howuncertaintyarises, is stabilized, and also how people navigate, experience, challenge, and rationalizeuncertaintyin everyday life. in doing so, we signal the immense potential offered by emerging intersections between human geography and broader critical social science interventions on the question of uncertainty. uncertaintyand its cognates, indeterminacy, liminality, obscurity, confusion, and misperceptionincreasingly pervade social and environmental problems. they are, for instance, endemic features of neoliberal transformation, health-environment interactions, food (in) security, natural disasters, resource conflict, and global climate change. while the question of uncertainty is not new, it increasingly captivates popular imagination in the context of these highly dynamic, contested and intractable problems, thereby catalyzing academic interest across the social sciences. yet despite the pervasiveness of uncertainty in daily life and forms of academic inquiry, few of the contributions associated with this transdisciplinary literature are geographical. as a consequence, this special issue demonstrates new ways that geographers can enrich work in this field, while also highlighting the potential of the concept to reshape approaches to topics, themes, and foci that the discipline actively pioneers. the articles in this special issue collectively respond to two trends that have historically sidelined the influence of critical human geographic scholarship in transdisciplinary debates about uncertainty. first, while the problem of uncertainty has received substantial attention in academic work across the fields of science and technology studies (sts), anthropology, sociology, and development studies, texts that explore uncertainty through a critical and explicit engagement with geographical concepts are relatively few and far between (we discuss some recent and promising exceptions below). second, within the discipline of geography itself, the question of uncertainty has largely been interrogated by quantitative and spatial perspectives that have given rise to various taxonomies of incertitude (kwan and schwanen, 2018; kwan, 2012; griffith, 2018) . fewer geographers, however, use qualitative and ethnographic methodologies to explore how the limits of knowledge become spatially generative, and/or operate as a direct force shaping social life. this is surprising given that the question of uncertainty is increasingly intertwined with contexts and problems that human geographers have long engaged, such as globalization, disease and illness, migration, and ongoing struggles around knowledge, power, and justice. furthermore, the multiple challenges and/or opportunities created by uncertainties in such contexts arise from and often rework central preoccupations in the subdiscipline of critical human geography; namely, socio-spatial and human-environmental relationships. approaches from human geographers could therefore help enrich existing work through their perspectives on scale, power, spatiality, place, and human-environment relations. with all of this in mind, this special issue offers a timely reflection on geographies of uncertainty and uncertain geographies. we bring together papers that were presented at the 2018 annual meeting of the american association of geographers as part of two sessions organized around the question of uncertainty. our contributors engage with a wide spectrum of qualitative methodologies and empirical case studies in order to examine how uncertainty shapes places and landscapes, social relations, human health, and/or human-environment interactions. while these articles grapple with diffuse subjects, including banana disease, wolf management, pest eradication programs, and human health (hiv, ckdu, and silicosis), taken together they clearly articulate several key themes for future geographic scholarship. first, we explore how uncertainty is often produced through, and temporarily stabilized by, key geographic axes such as time, space, scale, and regimes of environmental governance. attending to the ways that uncertainty is experienced as a spatiotemporal condition, and how it frequently compounds across scales of knowledge production, enables the special issue's contributors to demonstrate how forms of incertitude work through geographic relationships. crucially, this work takes us beyond established approaches that tend to engage with geography only in so far as it helps situate the production of uncertainty within broader social, cultural and political-economic contexts. as part of this discussion, we also contribute to recent moves both within and beyond the discipline that theorize experiences of uncertainty from the spatial margins. beginning from what claire herrick (2017) dubs as "non-archetypal spaces of expertise" the papers assembled here reveal other knowledges and practices that teach us how to live with, and in some instances, to make a living from, experiences of uncertainty. in particular, our contributors are in dialogue with wider debates which suggest that responding to uncertainties often requires combining different styles of knowledge. as scoones (2019a, 29) concludes, bringing together "formal and informal, accredited and lay knowledge, and experiential and conceptual understandings… [are central to] addressing uncertainties in context through culturally embedded, experiential [forms of] learning" (also see: shattuck, 2020a) . a second intervention of the special issue examines how the production and productiveness of uncertainty can generate geographical outcomes and socio-ecological possibilities. a key point here is that while uncertainty can restructure social and ecological relationships, it does so in highly differentiated ways, often with unanticipated effects. as a consequence, our contributors demonstrate how uncertainty can reinforce exiting fault lines of inequality at the same time that it might generate new forms of social, spatial, and ecological difference. third, this collection of articles focuses on how uncertainty is reconfigured through human-non-human dialectics. drawing on a rich tradition of scholarship within the discipline of geography, our contributors grapple with what robbins (2012, 234) describes as "the stubbornness and intractability of certain properties of non-human things" such as the fugitive chemistry of suspected toxin(s), unstable viral loads, materially ambiguous and uncooperative agricultural pests, and illegible wildlife. these complex human-non-human interactions pose important challenges to western scientific and regulatory traditions with "huge implications for the practice of science, management and policy" (scoones, 2019a, 5) . indeed, the frequent failure to classify, record, and regulate unruly human-non-human dynamics opens up possibilities to engage with more experiential forms of learning, diverse knowledges, and democratic responses to the question of uncertainty. in short, these papers speak to ian scoones' (2019b) call to revolutionize "methodologies for science and policy that take uncertainties seriously" by exploring engagements with "experimentalism, citizen participation, transformative action and 'post-normal' science." fourth, our contributors raise critical questions about the politics of uncertainty and in particular, how it indexes new forms of accountability, governance, and regulation. on the one hand, several articles demonstrate how uncertainty is strategically deployed in ways that individuate responsibility for environmental risk. in particular these studies demonstrate how uncertainty is often weaponized to reproduce both the imperceptibility of links between health and environment and the unaccountability of the structural forces that produce them. on the other hand, the papers in this special issue explore the potential of embracing uncertainty as a pathway to open "up a more rigorous, robust, transparentand democratically accountableenvironmental politics" (stirling 2018, 120) . in sum, the existence of uncertainty is generative of politics and possibilities that can be mobilized for very different political (and geographical) ends. more broadly, in the context of covid-19, our intervention comes at a moment of heightened, albeit highly differentiated, experiences of uncertainty and risk. enduring uncertainties about disease prevalence, testing, spread, and treatment have intensified uneven life opportunities and unjust exclusions, while also reconfiguring patterns of mobility, care, and social reproduction (neely and lopez, 2020; dattani, 2020; hannah et al., 2020) . this crisis has also heralded new forms of intervention, regulation, and governance that are unfolding as this special issue goes to press. many of the contributions included in this collection speak to similar themes and demonstrate the entanglements of uncertainty and geography in our daily lives. in reviewing the broader literature associated with this special issue, we were struck by the marginal position human geography occupies within scholarship on uncertainty in the critical social sciences. this is not to say that human geographers have not recently taken up many of the substantive themes that are of interest in transdisciplinary debates. indeed, engagements with experiences of liminality, protracted uncertainty and indeterminacy within the subdiscipline have become increasingly wide-ranging in their ambit, investigating questions on migration and refugee resettlement (loyd, ehrkamp, and secor, 2018; mountz et al., 2002) , development (chung, 2017 (chung, , 2020 , climate futures and the anthropocene (nightingale, 2018) , neoliberalism (anderson et al., 2020) , identity and selfhood (march, 2020) , and humanitarianism (newhouse, 2017) . "uncertainty also appears as a key term in various geographic handbooks and encyclopedias (brown and damery, 2009; stirling, 2018) . within these volumes, scholars argue for greater analytical precision in approaches to uncertainty, differentiating it conceptually from risk, ambiguity, and ignorance (stirling, 2003; . 1 other authors reflect more broadly on how uncertainty poses "[1] philosophical challenges [for the discipline] regarding the nature, origins, and value of knowledge, [2] ethical challenges regarding acceptable levels of risk, and [3] political challenges concerning how to act and who has the mandate to decide" (brown and damery, 2009, 90) . cutting across all of these contributions, is the promise of geographical approaches to shed light on how these multidimensional challenges vary across space, time, and scale, and how they mediate place-specific and human-environment relations and practices. however, only a handful of existing contributions within geography explicitly reference transdisciplinary debates on uncertainty. as has been argued elsewhere, and for other topics, "at the most practical of levels, this means that this rich corpus of highly prescient geographical writing rarely appears in keyword database searches" (herrick, 2016, 672) . more broadly, we argue that the relative invisibility of these promising geographic contributions relates to our failure "to develop the same kind of critical conceptual mass" that has allowed scholars in science and technology studies (sts), anthropology, sociology and development studies to direct conversations on the production, circulation, and lived experience of uncertainty. to date, scholars of sts and histories of science, medicine and environment occupy among the most well-cited and influential niches in the literature on uncertainty. for more than two decades, this work has taken pains to: (a) re-embed uncertainty in material as well as social contexts of knowledge practice (langston, 2010; murphy, 2004; roberts and langston, 2008; hecht, 2009; jasanoff, 2004) , (b) disrupt stable and binary distinctions between certainty and uncertainty as categories of knowledge (proctor and schiebinger, 2008; mcgoey, 2012) , and (c) examine how uncertainty is reconfigured by struggles over definition, legitimacy, and responsibility (brown, morello-frosch, and zavestoski, 2011; dumit, 2006; michaels, 2008; oreskes and conway, 2010; zavestoski et al., 2004) . a central and enduring line of investigation turns on how "unknowing, ignorance, and imperceptions [about social and environmental problems are] not just accidentally but purposefully generated in the history of knowledge practice" (murphy, 2006, 9, emphasis added) . this collection of work investigates how uncertainty is often manipulated or manufactured by industry or interest groups with the aim of displacing responsibility and fending off regulation (see langston, 2010; michaels, 2008; oreskes and conway, 2010; proctor and schiebinger, 2008) . as mcgoey (2012, 4) explains, this well-established body of literature "explores how different forms of strategic ignorance and social unknowing…both maintain and disrupt social and political orders, allowing both governors and the governed to deny awareness of things it is not in their interest to acknowledge." work in these disciplines also documents how scientific uncertainty is strategically marshalled and renegotiated by social movements that organize in response to toxic exposures and environmental crises. this line of analysis has been established in work on medically contested illnesses such as multiple chemical sensitivity, gulf war syndrome, and chronic fatigue syndrome, where patient activism and struggles for diagnosis are often grounded in experiences of being denied healthcare, insurance claims, and/or social legitimacy as sufferers. zavestoski et al.'s (2004; 161) research into guld war syndrome for instance, has argued that these experiences of exclusion often give rise to new collective subjectivities and illness groups which contest 'what they see as… unresponsive medical [legal, and bureaucratic] system[s]' (also see dumit, 2005; ; brown, morello-frosch, and zavestoski, 2011) . as senanayake and king (2019) argue, the intention of this work is to decentralize and potentially democratize knowledge production in response to uncertainty, challenging what counts as risk or harm and who gets to decide the terms and stakes involved. the allied focus on situated knowledge and regimes of imperceptibility in feminist science studies represents another compelling cluster of scholarship on the question of uncertainty. this includes work that foregrounds the indeterminacy and partiality of knowledge claims that "put[s] ambiguity to critical use… through demonstrating how social categories such as woman or man, black or white, straight or queer as well as concepts such as identity and selfhood need not be fixed in order to inform scholarly inquiry of structural and unstructured dynamics of power" (hayes-conroy and hayes-conroy 2008, 469; also see: sedgwick 2008) . for example, a recent review article by loren march (2020) provides a route into the affective and biopolitical implications of liminality, foregrounding experiences of disorientation, performativity and embodiment among trans and queer folks who do not conform to the dominant grids available for categorizing bodies. by destabilizing binary identity categories, this work documents how illegibility and in-betweenness are experienced and produced through space. in dialogue with this work, other scholars investigate how uncertainty generates new forms of social difference and subjectivity that reinforce patterns of vulnerability (auyero and swistun, 2009; chung, 2020) . together this work encourages us to think more deeply about uncertainty as a lived and spatiotemporal condition and prompts us to engage with the disorienting and sometimes potentially deadly consequences of being in-between. similar debates around the question of uncertainty animate scholars in critical development studies and anthropology, finding concrete expression in several anthologies and special issues (samimian-darash and rabinow, 2015; cooper and pratten, 2014) . much of this scholarship grapples with how uncertainty articulates with narrow, modernist views of progress, development, and technocratic risk management (scoones, 2019a; scott, 1998; ferguson, 1994) . this theme is taken up by paprocki (2019), who investigates how donors and development agencies translate uncertainty about climate change into, narrowly defined, normative visions of development in coastal bangladesh. these "anticipatory development futures" elevate commercial shrimp aquaculture and the expansion of an urban-based, industrial economy as preferred adaptation strategies in the face of uncertainty, despite significant evidence that these adaptation regimes result in environmental degradation, agrarian dispossession and 'development-induced migrations.' as paprocki (2019. 307) argues, these forms of anticipatory governance "entail the end of rural livelihoods in the delta, replacing them with a highly stylized (and age old) vision of development where the rural population transitions into an industrial labour force." this work thus brings debates about uncertainty, governance and accountability into sharp relief, demonstrating how uncertain ecological crises can be claimed and mobilized by donors and development agencies to justify interventions that continue to displace agrarian lives and livelihoods. this concern is also taken in a different direction by a recent working paper by scoones (2019a, 5) that draws on uncertainty to challenge "linear, hierarchical, modernist vision of progress… and technocratic managerialism." by theorizing uncertainty from the margins, he asks: "what more can we learn from alternative (including non-western) cultures of uncertainty that construct the world in different ways, through different histories, social imaginaries, traditions of thinking, and everyday practices?" (2019, 10; also see: leach et al., 1999; scoones, 1994) . in doing so, he suggests that pastoralists in sardinia, tibet, and northern kenya have situated knowledge and experiences of living in limbo that could help upend conventional modes of development, progress, and risk management, and in turn create space for imagining more socially just and ecological sound futures (also see: street, 2011; 2014; ; krätli and schareika, 2010; amin, 2013; samimian-darash and rabinow, 2015) . while this transdisciplinary literature is vibrant and compelling, it tends to treat geography as context, situating uncertainty within broader fields of social relations, cultural practices, and political-economic systems. the purpose of this special issue is not to deny the significance of these interventions, but instead to explore how key geographical concepts such as scale, spatiality, place, and human-environment relations can deepen and enrich transdisciplinary conversations on the question of uncertainty. we argue that the relative silence of geographers in these debates is a missed opportunity for the discipline, especially given our scholarship on related themes as well as a long history of critical approaches to many of the topics that are at the heart of this literature. together, this signals the potential for geographical contributions to transdisciplinary debates. we now turn to a discussion of cross-cutting themes in this special issue, to help make some of these contributions more explicit. in organizing this special issue, we collected a diverse set of papers that, while in active dialogue with innovations in other disciplines, remain intent on documenting how experiences of uncertainty, insecurity, risk, indeterminacy, and illegibility are inherently geographical. each of the articles in this special issue make original contributions on their own; however, when taken together they collectively address four core themes for a geography of uncertainty. first, the contributions illustrate how uncertainty is temporarily stabilized across key geographic axes such as time, space, scale, and regimes of environmental governance. this theme helps demonstrate how uncertainty can be spatially fixed at key moments, either due to necessity, strategy, or through institutional action. reflecting this concern, arielle hesse's paper examines attempts to regulate respirable crystalline silica, which is carcinogenic dust that is generated through hydraulic fracturing. the expansion of natural gas extraction in the united states has introduced new forms of risk for workers and competing understandings of how to regulate silica exposure. the article is an excellent case study of how the industry attempts to construct illegible geographies in efforts to fend of regulation. by documenting the spatial and temporal flexibility of the worksite, hesse demonstrates how geography is strategically enrolled to position the hydraulic fracking industry as unknowable and thus, ungovernable. in doing so, she demonstrates how the variable spatial and temporal configuration of the industry is used to stabilize uncertainty about particular health risks and outcomes for workers. while uncertainty can persist and accumulate through spatial relationships, several of the articles also wrestle with how uncertainty comes to be functional or rational at some scales yet is counterproductive at others. as one example, nari senanayake examines local attempts to neutralize biomedical uncertainty about chronic kidney disease of unknown etiology or ckdu, in sri lanka's dry zone. given resource constraints and generic treatment regimes, senanayake argues that the disease that is most frequently enacted through clinical encounters is actually an undifferentiated form of kidney disease that envelopes both regular and mysterious forms of illness (ckd/ckdu). while the absence of diagnostic closure in kidney disease clinics enables important improvisations in the practice of care, it also inadvertently stabilizes uncertainty about actual disease burden at higher scales of analysis. specifically, she demonstrates how uncertainties about disease states compound across scales of knowledge production and how this then frustrates attempts to map the prevalence of ckdu at regional or national levels. in a related vein, jeff martin details the challenges for wolf management in the u.s. west, documenting how government agencies strategically harness uncertainty to navigate funding cuts and longstanding anti-federal and anti-regulatory sentiment. in a fascinating extension of sts scholarship, martin explores how illegibility about wildlife populations can be understood as logical in certain domains of bureaucratic management even as it produces new uncertainties and socio-ecological dilemmas at other spatial and temporal scales. on the one hand, a hesitancy to share information on wolf populations can protect mid-level bureaucrats from external political pressures to delist the species from the endangered species act. on the other hand, martin asserts that illegibility "from above" undermines agents' ability to effectively intervene around their charges and guarantee wolf conservation over the long durée. similar to hesse's discussion of how the production of uncertainty shapes silica regulation, martin asserts the production of uncertain data can allow state agencies to displace responsibilities for environmental protection in space and time. a second theme reveals the productivity or productiveness of uncertainty and how it can be generative of geographical outcomes and socio-ecological possibilities. papers by cavallo (2020) , martin (2020) , and senanayake (2020) all document pragmatic, experimental, and improvised practices of survival in contexts of uncertainty, institutional instability, and resource scarcity. here authors illustrate how farmers in uganda, midlevel public administrators in the american west, and doctors in the dry zone of sri lanka frequently engage in pragmatic strategies to maximize pathways for action in response to multiple and compounding uncertainties. farmer experimentation in the face of uncertainty about a deadly plant disease (bxw) is a focus for sara cavallo, where improvised knowledge networks partially realign relations of expertise. specifically, she demonstrates how strategic experimental networks between farmers and scientists utilize a hybrid set of practices that remain rooted in local cultural understandings and social networks. indeed, cavallo shows how farmers negotiate differing claims to certainty regarding the disease through a range of tactics that do not fully align with theories of the environmental subject as an extension of state logic. instead, compounding uncertainties allow farmers to make claims on the state and re-envision agricultural systems in ways that accommodate greater experimentalism, particularly in the face of uncertain plant health risks. discussion of uncertainty as a strategic resource is also evident in martin's analysis of controversial wildlife management in the american west. this paper documents how implicit and pragmatic choices around day-to-day practices and different technologies allow agents to navigate ecological complexity and multiple use mandates. the production of contingent geographies of care in the wake of sri lanka's kidney disease epidemic is the basis of senanayake's paper, which documents how local attempts to manage biomedical uncertainty about disease are socially and spatially generative. rather than attempting to achieve diagnostic closure in cases of mystery kidney disease, many local clinicians pragmatically suspend diagnosis and instead routinely treat disease-complexes, including different combinations of kidney disease, hypertension, diabetes and cholesterol giving rise to more holistic, contextualized and responsive ecologies of care. more broadly, this work demonstrates how geographic, as opposed to diagnostic, distinctions increasingly structure access to care and state resources. in doing so, she reveals how the island's kidney disease epidemic has cemented new forms of socio-spatial difference in the dry zone while simultaneously reproducing previously established geographical patterns of neglect. in an excellent example of how uncertainty about chemical harms generates socio-ecological possibilities, annie shattuck explores how the expansion of pesticide use within northern laos reconfigures smallscale farming livelihoods within an increasingly globalized agricultural economy. safe use education dominates this setting, which entails information on safe handling practices, personal protective equipment, and familiarity with the toxicity of pesticides. yet in grounding this form of knowledge translation with local livelihood practices, the article effectively demonstrates how safe use education elides contextual political economies and hybrid practices that result in lived experiences of uncertainty. this leads shattuck to conclude that "smallholder farmers are not operating from a complete deficit of knowledge. rather their knowledge of toxicity and risk is partial, situated in a particular context, laced with uncertainty and inequality, and often gained from embodied experience." the article thereby offers us a conceptual route through which to consider the entanglements between situated knowledges of risk and socio-ecological possibilities. a final critical point that emerges from these papers is that while uncertainty can restructure social and ecological relationships, it often does so in highly uneven ways. indeed, to a large extent, all of the papers in this special issue are concerned with how uncertainty is differentially experienced and embedded in multiple, interlinked struggles that constitute daily life (such as poverty and political marginalization) as well as, individual experiences of resource scarcity, health, and/or disease. along those lines, brian king's article details the ongoing uncertainties with hiv in rural south africa even within a setting that has drastically scaled-up the provision of care. the widespread provision of anti-retroviral therapy in the global south has led public health institutions to identify hiv as a chronic condition; however, there remain a wider set of related uncertainties that include ongoing stigma, food insecurities, and unequal bodily responses to treatment. king concludes that given these dynamics, hiv uncertainty and certainty simultaneously co-exist and are co-produced, while remaining materially uneven for those managing their health. likewise, cavallo (2020) and shattuck (2020b) , illustrate how different forms of government intervention in response to uncertainties about plant disease management and pesticide use can reproduce patterns of socio-spatial inequality, normalizing risk and harm for some bodies and not others. as a consequence, our contributors demonstrate how uncertainty is frequently enrolled in the production of social, spatial, and ecological difference. third, the contributions demonstrate how uncertainty is produced, transformed, and negotiated through a key focus of geographic inquiry: the interactions between human and non-human agents. in collecting these articles, it was not our intention to select case studies that dealt specifically with the non-human. but in reading these contributions together, we are struck that all authors deal in some meaningful way with non-human actors. whether it is a pathogen or virus, banana, wolf, chemical toxin, or an invasive insect population, all of the articles engage with the ways that the non-human generates uncertainties, and in turn poses important challenges to western scientific and regulatory traditions. martin and sedell for instance, address how dynamic and moving species, either wolves or invasive insects respectively, confound attempts to measure, surveil, and manage. and hesse and shattuck emphasize how chemical exposure cannot easily be fixed in space and time, as the pressures from broader political and economic forces, including corporations that want to mitigate their exposure to lawsuit and damages, draw upon uncertainty and personal responsibility to deflect culpability in undermining human and environmental health. in considering how uncertainty is refracted by human and nonhuman interactions, there are a few points worth emphasizing. first, western traditions of bureaucratic rationality, scientific management, and state power all require practices of containment. whether it is the containment of invasive pests within a particular agricultural region, wolves in a conservation zone, or an infectious disease within an individual body, scientific and bureaucratic management requires that the object of concern to be contained in space and time. the fact that these entities resist attempts at containment and measurement mean they are active participants in the production of uncertainty. perhaps most evident of this point is jennifer sedell's paper, which convincingly demonstrates that in terms of invasive pests and agricultural production in california, uncertainty about insect presence/ absence cannot be eradicated. as one of her informants explains, the absence of insects in the traps does not prove that they have been eradicated since it is impossible for existing measurement devices to observe the status of pest populations at sub-detectable or low levels. as sedell notes, the implications for the agricultural industry are significant and political responses are required to restore 'pest-free' status for the trade of many agricultural goods. this theme also takes center stage in martin's contribution. focusing on how and why government agents routinely collect less and/or less precise data about the wildlife under their charge, martin complicates traditional assumptions of bureaucratic rationality and state power. he skillfully demonstrates how, in the context of ecological complexity, logistical limitations and socio-political conflict, state agents engage in practices of obfuscation to manage risks and exonerate themselves from blame. in dialogue with mcgoey, 228 (2007) ; (2012;)), martin demonstrates how this apparent dysfunction is not a breakdown of bureaucratic rationality but rather is "indicative of a rational strategy [for managing antithetical demands] in itself." instead of precipitating more experimental and adaptive management approaches, martin demonstrates how the frequent failure to record and manage unruly nonhuman dynamics can end up reproducing forms of bureaucratic rationalityeven when uncertainties are central to managing socioecologically complex systems. in doing so, he draws our attention to how "politics matter, and adaptive, experimental governance without democratic politics, can end up similar to earlier [technocratic] expert modeseven when uncertainties are recognized" (scoones 2019a, 28) . finally, these papers raise critical questions about the politics of uncertainty and, in particular, how it indexes new forms of accountability, governance, and regulation. together the special issue invites geographers to grapple with how uncertainty often generates politically opposing ways of problematizing and intervening in human-environment and socio-spatial relationships. for instance, in this collection, several articles demonstrate how materializing health-environment interactions as uncertain have become powerful strategies for affirming ambiguity about toxic and pathogenic exposures, and ultimately, of fending off stricter government regulation (also see: murphy, 2006) . shattuck (2020a) examines farmers' knowledge of pesticide risk in northern laos in the wake of individually focused safe use messaging and enduring uncertainties about chemical harms. by documenting how situated experiences of toxicity articulate with safe use models, she demonstrates that local typologies of safety and harm continue to blame poor, 'risky' people for pesticide-related health impacts "instead of the socio-economic situations that require them to take risks" in the first place. partial knowledge about the impacts ofand responsibility forchemical exposures in this case, is marshalled to reinforce and legitimate individually based risk management. in the context of rural poverty, uneven access to health care and local inequalities, this approach renders invisible the myriad ways that farmers' decision making is refracted by other risks, principally the risks of being poor and rural. the tendency to reinforce individual responsibilities for exposure in the face of uncertainties about occupational health hazards is also a focus for hesse (2020), where, as argued above, the fossil fuel industry strategically conceptualizes the geographies of hydraulic fracturing as flexible, variable, and unknowable in order to contest government regulation of toxic silica exposures. specifically, this article reveals how the processes of rulemaking as well as the variable spatial and temporal configuration of the fracking industry helps decentralize and individualize responsibility for exposure. similarly, king (2020) asserts that the "production of uncertainty is relational and dynamic, changing not only with new 'certainties' about the virus and treatment protocols, but also the ways in which its management intersects with social and environmental dynamics." this reveals how the experiences of living with hiv are not uniform and continue to be shaped by a range of factors, such as socio-economic resources, food insecurity, employment, gender, age, and sense of responsibility for others. as a consequence, the special issue makes a powerful case for focusing on how uncertainty generates styles of politics that reproduce precarity and individual responsibility for managing exposures to multiple social and environmental harms. a different, albeit related, perspective is explored in other papers where embracing uncertainty is viewed as a pathway to open "up a more rigorous, robust, transparentand democratically accountableenvironmental politics" (stirling 2018, 120) . this work combines what scoones (2019a, 27) describes as a "transformational vision, [where] notions of justice are central" with a "more patient, sometimes unruly, bottom up approach to defining future pathways," often focused on adaptive management, incremental learning, and polycentric governance. as one example, sedell (2020) demonstrates how competing models for understanding invasive pest populations reveal the profound instability of metrics that determine pest absence/presence in california's agricultural sector. as part of this analysis, she demonstrates how alternative methodologies for visualizing pest presence/absence as well as bottom up diversified farming practices reconceptualize pests as neither exceptional nor temporary, but instead, as part of a new normal. doing so, opens up possibilities for alternative agro-ecological and political arrangements by locating the region's pest problem not with the insects themselves but in production systems that lack resilience to biological invasions. as a consequence, sedell reveals how uncertainty about pest presence/absence can precipitate a fundamental re-evaluation of california's agro-environmental system, and simultaneously transform our relationships to pest populations and how the region produces food. how networks of cross-species relations are productive of 'new ecological accommodations' or 'ways [of] learning to live endemically with our viral [microbial, and insect] companions' (greenhough 2012, 295) thus constitute important avenues for future geographic research on the politics of uncertainty. to a large extent, all of the papers in this special issue push us to grapple with uncertainty as a norm and not an anomaly. in other words, this work opens up space for engagement with uncertainty as more than a temporary phenomenon but as a "permanent condition" where stability and certainty about socio-spatial and human-environment relationships are often elusive (brown and damery 2009, 82) . rather than attempting to ensure a strict boundary between knowledge and its opposite, our contributors encourage us to think more deeply about uncertainty as a lived and spatiotemporal condition. it is these very entanglements between uncertainty, environments, and everyday life that provide the narrative thread uniting this diverse collection of articles. while engaging with a broad range of topics and qualitative methodologies, this work coalesces around four central themes: (1) how uncertainty works through geographic relationships; (2) how uncertainty generates geographical outcomes; (3) how uncertainty is reconfigured by human-non-human dialectics; and, (4) how uncertainty raises questions about politics. in doing so, this special issue develops a critical human geography of uncertainty, which not only articulates how the concept is useful for geographers, but also, argues that geography can enrich existing transdisciplinary work on the subject with its perspectives on scale, spatiality, power, place, and human-environment relations. credit authorship contribution statement nari senanayake: conceptualization, writing -original draft, writing -review & editing. brain king: conceptualization, writingoriginal draft, writing -review & editing. surviving the turbulent future brexit: modes of uncertainty and futures in an impasse flammable: environmental suffering in an argentine shantytown a companion to environmental geography contested illnesses: citizens, science, and health social movements navigating compounding uncertainty: farmer strategies amid biosecurity crises in western uganda governing a liminal land deal: the biopolitics and necropolitics of gender engendering the new enclosures: development, involuntary resettlement and the struggles for social reproduction in coastal tanzania ethnographies of uncertainty in africa come on, people… we* are* the aliens. we seem to be suffering from host-planet rejection syndrome': liminal illnesses, structural damnation, and social creativity illnesses you have to fight to get: facts as forces in uncertain, emergent illnesses depoliticization, and bureaucratic power in lesotho where species meet and mingle: endemic human-virus relations, embodied communication and more-than-human agency at the common cold unit 1946-90 uncertainty and context in geography and giscience: reflections on spatial autocorrelation, spatial sampling, and health data thinking through covid-19 responses with foucault -an initial overview taking back taste: feminism, food and visceral politics. gender, place cult africa and the nuclear world: labor, occupational health, and the transnational production of uranium global health, geographical contingency, and contingent geographies when places come first: suffering, archetypal space and the problematic production of global health geographies of uncertainty and negotiated responsibilities of occupational health states of knowledge: the co-production of science and the social order rethinking social reproduction in the time of covid-19 hiv as uncertain life living off uncertainty: the intelligent animal production of dryland pastoralists the uncertain geographic context problem context and uncertainty in geography and giscience: advances in theory, method, and practice toxic bodies: hormone disruptors and the legacy of des exploring understandings of institutions and uncertainty: new directions in natural resource management a geopolitics of trauma: refugee administration and protracted uncertainty in turkey queer and trans* geographies of liminality: a literature review between scylla and charybdis: environmental governance and illegibility in the american west on the will to ignorance in bureaucracy strategic unknowns: towards a sociology of ignorance doubt is their product: how industry's assault on science threatens your health lives in limbo: temporary protected status and immigrant identities uncertain exposures and the privilege of imperception: activist scientists and race at the u.s. environmental protection agency sick building syndrome and the problem of uncertainty: environmental politics, technoscience, and women workers care in the time of covid-19 uncertain futures and everyday hedging in a humanitarian city the socioenvironmental state: political authority, subjects, and transformative socionatural change in an uncertain world merchants of doubt: how a handful of scientists obscured the truth on issues from tobacco smoke to global warming all that is solid melts into the bay: anticipatory ruination and climate change adaptation agnotology: the making and unmaking of ignorance political ecology: a critical introduction toxic bodies/toxic environments: an interdisciplinary forum modes of uncertainty: anthropological cases no fly zone? spatializing regimes of perceptibility, uncertainty, and the ontological fight over quarantine pests in california searching for ckdu: mystery kidney disease, differentiated (in) visibility, and contingent geographies of care in dry zone sri lanka living with uncertainty: new directions in pastoral development in africa what is uncertainty and why does it matter embracing uncertainty: what are the implications for sustainability n. senanayake and b. king geoforum xxx (xxxx seeing like a state: how certain schemes to improve the human condition have failed epistemology of the closet health-environment futures: complexity, uncertainty and bodies toxic uncertainties and epistemic emergence: understanding pesticides and health in lao pdr risky subjects: embodiment and partial knowledges in the safe use of pesticide uncertainty and precaution: some instrumental implications from the social sciences uncertainty artefacts of not-knowing: the medical record, the diagnosis and the production of uncertainty in papua new guinean biomedicine biomedicine in an unstable place: infrastructure and personhood in a papua new guinean hospital patient activism and the struggle for diagnosis: gulf war illnesses and other medically unexplained physical symptoms in the us we thank all of the authors who contributed their articles to this special issue and the reviewers for their helpful and timely comments. we are especially grateful to harvey neo for his insightful feedback and patience in helping steer this project towards completion. the special issue emerged from a series of aag sessions in 2018, where these papers received constructive comments and encouragement from our generous discussants, rebecca lave and david demeritt. particular thanks to arielle hesse and a. marie ranjbar for valuable comments on previous drafts of this introduction. key: cord-325445-80p6wthw authors: goh, ong sing; fung, chun che; wong, kok wai title: query based intelligent web interaction with real world knowledge date: 2008-03-14 journal: new gener comput doi: 10.1007/s00354-007-0031-7 sha: doc_id: 325445 cord_uid: 80p6wthw this paper describes an integrated system based on open-domain and domain-specific knowledge for the purpose of providing query-based intelligent web interaction. it is understood that general purpose conversational agents are not able to answer questions on specific domain subject. on the other hand, domain specific systems lack the flexibility to handle common sense questions. to overcome the above limitations, this paper proposed an integrated system comprises of an artificial intelligent conversation software robot or chatterbot, called artificial intelligence natural-language identity (hereafter, aini), and an automated knowledge extraction agent (akea) for the acquisition of real world knowledge from the internet. the objective of akea is to retrieve real world knowledge or information from trustworthy websites. aini is the mechanism used to manage the knowledge and to provide appropriate answer to the user. in this paper, we compare the performance of the proposed system against two popular search engines, two question answering systems and two other conversational systems. traditional media such as television, radio, newspapers and magazines play an important role in reporting and providing the latest information on current events. however, nearly all of these media are linear and unidirectional in nature. in other words, they do not provide interactive communication nor can they answer any query from the users. most of these traditional means of media also do not link the answers to the source or reference of the information. hence, viewers or readers may require additional effort in locating or verifying the information. on the other hand, the internet, in particular, the world-wide-web (www), coupled with multimedia, browser and web service technologies, presents a powerful way of communication. if the system is designed properly, it could provide high level of interactivity between the users and the computer. this is clearly a major step in the advancement over the traditional means of communication. one of the possible applications of such system is to provide answers to queries on specific topics or knowledge domains. consider situations such as outbreak of diseases, natural disasters and terrorist attacks, they have caused much miseries, fear and confusion around the world. examples of such crisis are the severe acute respiratory syndrome (sars), bird flu, september 11, earthquakes and tsunamis. in such times, a lot of people will be eager to know as much information as possible. this is where the traditional media may be found inefficient as they are limited with constraints such as page number and air-time. they may also be unable to source all the information from around the world. in addition, people such as managers and decision-makers, frontline specialists or emergency services personnel, and concerned citizens who are directly or indirectly involved in the situations, require to be better informed of the situation and the development. to this end, the internet could have a major role to play as an essential communication channel in providing an intelligent interactive interface for the users. a form of a "global crisis communication system" could be used to provide a natural language query-and-answer system. this system should be capable of providing relevant responses to queries from the users in a conversational manner. in this paper, we describe part of a project which is currently under development. the aim of the project is to develop an intelligent conversation agent called aini to answer domain specific questions as well as open-domain (or common sense) questions. in this report, we have used the subject -bird flu as the domain knowledge of interest to demonstrate the feasibility of the developed system. the key contribution described in this paper is the integration of the common sense knowledge and domain specific knowledge in the form of a "knowledge matrix." the system is based on a layered and modular design, and the answers for the queries are searched from these modules. the proposed system is based on an intelligent conversation agent called aini and an akea. the objective of akea is to retrieve real world knowledge or information from trustworthy websites whereas aini is the mechanism used to manage the knowledge and to provide appropriate answer to the user. descriptions of the system are given in the subsequent sections. the prototype system is compared to two popular search engines -google and yahoo!, two question answering systems -start and askjeeves and two conversational systems -eliza and alice. while it is understood that the purpose of those systems are different, the objective of the comparison is to demonstrate the need of the proposed system described in this paper. this also shows that the proposed system is capable to provide appropriate answers by integrating open domain and domain specific knowledge into one conversation agent architecture. §2 aini's conversational agent architecture the aini architecture is shown in fig.1 . the architecture has been reported in previous publications by the authors. 19, 20) basically, the aini engine comprises a number of knowledge modules in the data layer. it has the ability to communicate with three layers: the application layer, the data layer and the client layer. the client layer is capable to communicate with the user via different channels such as web browser, mobile browser, wap browser and gsm interface. it can carry on multiple independent conversations at the same time. aini's knowledge bases are located within the data layer. there are a number of modules supporting the application layer which governs the manipulation and searching of the answers. the modules use plug-in principles that can quickly be augmented with domain knowledge for specific purposes. originally, this research project involves the establishment of an embodied conversational agent (eca) based on an aini architecture. 20) the prototype system is designed specifically for the web and mobile technology as shown in the client layer in figure 1 . the complete software agent can be considered as a multi-domain knowledge system with multimodal human-computer communication interface. the query and answer between the user and the computer are communicated via the common protocol tcp/ip. aini is designed to engage the user with focus on the chosen subject topic. in this particular application, the topic is on the possible pandemic virus, h5n1. aini communicates with the user in natural language via typed messages. the system is also capable to reply in text-prompts or text-to-speech synthesis together with appropriate facial-expressions on the displayed object which can be an animated avatar or a human face. as illustrated in fig.1 , aini employs an internet three-tier, thin-client architecture that may be configured to work with any web application. it comprises a client layer, an application layer and a data server layer. the hybrid architecture provides features of multimodal interface (client layer), multilevel natural language query (application layer) and multiple knowledge bases (data layer). the process of communication and answering is as follows. given a question, aini first performs a question analysis by extracting pertinent information to be used in query formulation, such as the noun phrases (nps) and verb phrases (vps) using the minipar parser. 26) minipar is a broad-coverage parser for the english language. an evaluation with the susanne corpus shows that minipar achieves about 88% precision and 80% recall with respect to dependency relationships. in our experiment by using corpus extracted by automated knowledge extraction agent (akea), 18) minipar parser is capable to parses nearly 500 words per second on a dell precision pws380 server 3gh with 1gb memory. the user interface resides in the thin-client layer and it supports webbased and mobile service interface. for the web-based interface, the system is based on multimodal agent markup language (maml) interpreter in order to handle the user interface. maml is a prototype multimodal markup language based on xml that enables animated presentation agents or avatars. it involves a virtual lifelike talking 3d agent character designed to carry out a more natural conversation with the user. for mobile devices, due to their small screens, there are limitations on the amount of information that can be presented at one time. reading large amounts of information from such devices can require excessive amount of scrolling and concentration. to reduce distraction, interactions, and potential information overload, a better way of presenting information might be through multilevel or hierarchical mechanisms. 9) a chat mode interface will suit the multilevel mechanism nicely and this will be a better solution for mobile service. in addition, current wireless network service vendors are now providing a wide bandwidth telephone network, known as 3g communication. 23) this development has led to the increasing adoption of smartphone as a client in the traditional distributed systems. on such system, the proposed system will require the mobile flash player. 2) for wap services, the application was embedded wap browsers from vendors such as openwave * 1 and nokia. * 2 all communication with aini takes place through typed text messages and they are processed based on natural language understanding and reasoning. aini's engine implements its decision making network based on the information it encounters in the six levels of knowledge modules. the input and output of each module is an xml-encoded data structure that keeps track of the current computational state. the knowledge modules can be considered as transformations over this xml data structure. the system accepts questions or queries from the users and it processes the queries based on the information contained in aini's knowledge bases. the system is implemented by open-source architecture employing a kannel mobile gateway, php, perl scripting language, apache server and knowledge base stored in a mysql server. in this server, it handles the process of the queries. here, one or more application servers are configured to compute the dialogue logic through the multilevel natural language query algorithm. in this layer, it is based on a goal-driven or top-down natural language query (nl-query) approach which is similar to the way that humans process their language. as indicated by literature in the field of natural language processing (nlp), the top-down approach is by far the best approach. as shown in fig. 1 , the top-down nl-query approach consists of six levels of queries, namely spell checker (level 0), natural language understanding and reasoning (nlur) (level 1), faqchat (level 2), metadata index search (level 3), pattern matching and case-based reasoning (pmcbr) (level 4) and supervised learning (level 5). all these have been discussed in reference. 17) the data layer serves as storage for data and knowledge required by the system. this is where the aini bot's conversational knowledge bases are stored. in the proposed approach, the architecture can be considered as a domain matrix knowledge model 16) for the support of the conversational system. it is well understood that true intelligent action requires large quantities of knowledge. such reservoir of knowledge is harvested from the internet and deployed in the domain matrix knowledge bases architecture. this form the basis for the construction of large-scale knowledge bases to be used as the engine for the intelligent conversation systems. these databases established so far are dictionary, domain-specific knowledge bases, open domain knowledge bases and chatlog (conversation logs and unanswered question logs). as mentioned previously, the first step taken by aini is to perform a spell check to eliminate possible wrong spellings. the spell checker is located at the application layer and the dictionary contains the database of words. the spell checker used in this study is called ispell. ispell is an open-source program that is used to correct spelling and typographical errors in a file. ispell was first run on tops-20 systems at the mit-ai lab. many people have contributed dictionaries to allow ispell to be used with their native languages. at least 50 dictionaries have already been established. as far as this project is concerned, only english is used. however, it can be anticipated that aini has the potential to be used as a true intelligent web interaction with users who are using different languages. combining the dictionary and the spell checker in the application layer, this combination can be considered as the level 0 knowledge. in terms of the domain-specific knowledge databases, it can be observed in fig. 1 that the domain specific knowledge modules are made up of a natural language corpus and a faq (frequent asked questions) module. the natural language understanding & reasoning (nlur) module in the application layer is used to gather data to be stored in the natural language corpus. in addition, the akea is used to extract information from the web and in particular trusted web sites. the architecture of akea has been reported previously and the details can be found in reference. 18) a description of its operation will also be given in the subsequent section. this forms the level 1 knowledge. the faq (frequent asked questions) module is connected to the faqchat module in the application layer. it is used to provide information based on the faq sections from websites and from the actual chat logs. in other words, the conversations and answers given by the user to the system are also used to enhance the information in the faq module. this forms the level 2 knowledge. in this study, the bulk of the information was collected during the height of the sars epidemic in 2003 19) and the bird flu pandemic crisis in 2005. 21) however, in this paper, the domain specific knowledge is only focused on bird flu. the open-domain knowledge modules in the data layer are based on trec corpus, minipixel and the open-domain knowledge base (aaa). they are further supported by the conversation logs which store the previous conversation for extraction of answers from the unanswered questions. it is assumed that such answers could be used in future interaction with the users. details of these open-domain knowledge modules are described in the section 3.1 and they form the subsequent levels of knowledge in the application layer. §3 domain knowledge matrix model in this paper, bird flu pandemic is the domain-specific knowledge used as an example to illustrate the application of this proposal. research and information on h5n1 pandemic have become increasingly important as the pandemic may have dire global implications. wall street journal online 7) predicted that this pandemic could be worse than the one in 1918 which killed at least 20 million people. in addition, the world health organization estimates the h5n1 virus could infect up to 30 percent of the world's population. shigeru omi, a who official, also warned that an estimation of 2 to 7 million deaths are "conservative" and that the maximum figure could be as high as 50 million. 29) the aini's domain knowledge matrix model incorporates several knowledge subjects. this is analogous to the consultation of expertise knowledge from multiple experts. for example, a sales domain knowledge should contain expertise or knowledge on how to improve sales. however, as a sales person, he or she is expected to have a wide range of common sense knowledge and the ability to engage the potential customer. hence, the intelligent system should also incorporate open-domain knowledge to handle general or generic questions. in here, the open-domain knowledge is not necessary the common sense knowledge which is assumed to be possessed by everyone. the open-domain refers to the accessibility of the knowledge. hence, they are still categorized according to knowledge domain. by including multiple domain knowledge bases with aini's single domain knowledge, the proposed aini will be able to hold "meaningful" conversations with the users. in this proposed system, the open-domain and domain-specific knowledge are pre-defined in the data layer. these modules are used to support the various knowledge levels at the application layer. depending on the user's input, the agent will respond or switch from one level to another. while the system is capable to communicate with the user beyond the domain knowledge, there are cases that the system will exhaust its capability to answer the queries. in such case, the system will attempt to divert the focus back to the current topic of interest by responding with some predefined random statements. the purpose is to direct the users' back to the system's domain-specific state. hence aini will attempt to "cycle" between the 6 levels of information processing within the application layer supported by the various knowledge modules in the data layer. a way to view the proposed domain knowledge matrix model is given in fig. 2 . in this approach, the knowledge base of the aini can be considered as a collection of specific conversation domain units. each unit handles a specific body of knowledge used during the conversation between aini and the user. the knowledge can be seen as arranged in the vertical columns making up the open domain or domain-specific knowledge. in addition, specific subjects are shown in the horizontal rows. for example, in the open domain knowledge, the subject units will cover topics such as personality, business, biology, computer, etc. in this report, our focus is on the medical subject and in particular, the bird flu pandemic, therefore there are additional modules being incorporated. the domain knowledge model plays a major role in conversational systems. such systems normally are comprised with two subcategories: the traditional or narrow domain, and, the open domain. in the traditional domain, systems attempt conversational fluency based on limited domains of expertise. eliza 34) for example simulates a rogerian psychotherapist, and this implementation is commonly known as doctor. the rogerian psychotherapist knowledge base attempts to engage the patient by repeating the patient's statements, and by encouraging the patient to continue talking. terry winograd's shrdlu 37) is another program simulating a robot which is able to interact within a simple world which consists of colored building blocks and boxes on a flat surface. shrdlu has knowledge about its world and it can answer questions in natural language. three decades have passed since eliza was created. computers have it is assumed that the human brain contains a world of knowledge, but it has limitations on memory retrieval. hence, knowledge needs to be specified. in the proposed system, the domain knowledge matrix model is designed along this line. the system uses custom domain-oriented knowledge bases and existing knowledge bases from online documents and training corpora. it contains a spectrum of possible solutions from queries on specific domains to general conversation questions. it can also be considered as a xml-like metadata model. this approach does not attempt to predict possible inputs from the user. instead, the system aims to handle conversations within a specified domain or focus on 'domain-specific' conversations. in this project, the novel contribution is the development of the "domain knowledge plug-in components." with this arrangement, the domain-specific knowledge could become portable, scalable and incorporated easily with other applications. this approach will allow future improvements to encourage collaborative contribution to the specific knowledge domain. the proposed system will enable the development of a wide range of information sources. open-domain conversational systems need to deal with questions about nearly any topic. it is very difficult to rely on ontological information due to the absence of wide and yet detailed banks of world knowledge. on the other hand, these systems have much more information and data to be used in the process of answering the queries then any domain specific systems. in aini's conversation system, information from the large-scale mass collaboration mindpixel 30) and training data sets from the text retrieval conference's (trec) training corpus 32) are used. aini also uses alice annotated aiml (aaa), 5) the loebner prize winner 28) and the chatterbox challenge winner 10) hand crafted knowledge bases. these are illustrated in the data layer in figure 1 and under the open-domain columns in fig. 2 . mindpixel is a common sense knowledge component and it is similar to openmind * 3 and cyc. * 4 the system accepts public contributions. however, cyc model and openmind had bottlenecks which prevent truly large-scale collaboration 33) . the first is the fact that knowledge does not grow by itself. every new rule or axiom has to be entered manually and the process takes a lot of patience and time. furthermore, information has to be input with the cycl programming language and to follow the rules of the system. the second drawback is the complexity of cyc. it will need months to install and implement a system that is based on the knowledge base of cyc. on the other hand, mindpixel started collecting their propositions privately via email in 1994 and then evolved to online mass collaboration. to date, the project's user base of nearly fifty thousand people has contributed more than one million propositions and recorded almost ten million individual propositional response measurements. ainis use only 10% of the mindpixel propositions. in practice, 10% of the training corpus is held back from training to act as a generalization test to ensure the system does not simply memorize the corpus. passing this generalization test would be the basis for claiming that the system is able to replicate human-level intelligence in a machine. although a lot of knowledge has been collected, it is recognized that the system is still less than the uncountable "pieces" of common sense knowledge that are estimated to be involved with human intelligence. 31) a second common sense knowledge component deployed by aini is a training corpus from trec as shown in table 1 . trec, organized each year by the national institute of standards and technology (nist), has offered a specific track to evaluate large-scale open domain question-answering (qa) systems since 1999. finding textual answers to open domain questions in large text collections is a difficult problem. in our system, we only extracted factoid questions to be incorporated in the aini's engine. our concern is the types of questions that could be answered in more than one way. we have tried to avoid such questions. in conversational systems, factoid questions should have only single factual answers. 3, 12, 13, 36) these are considered as a good stimulus-response type of knowledge unit. examples of such questions are, "who is the author of the book, the iron lady: a biography's of margaret thatcher? " "what was the name of the first russian astronaut to do a spacewalk? " or "when was the telegraph invented? " trec's corpus has a considerably lower rate of answer redundancy than the web and thus, it is easier to answer a question by simply extracting the answers from the matching text. to gather this data, we automatically classified questions in the trec 8 through trec 10 test sets by their 'wh'-word and then manually distinguished factoid questions, which represented * 3 www.openmind.org * 4 www.cyc.com around half of the initial corpus as shown in table 1 . the third knowledge base in the aini's open domain knowledge model is obtained from hand-crafted annotated alice aiml (aaa), * 5 a loebner prize winning 28) conversation system knowledge base. aaa is a free and opensource software package based on xml specifications. it is a set of artificial intelligence markup language (aiml) scripts and this is the backbone of the award winning conversation system. aaa is specifically reorganized to facilitate conversational system developers to clone the 'brain' of the conversation system and to enable the creation of customized conversation agent personalities. the approach has reduced the need to invest huge efforts in editing the original aaa content. aaa's knowledge bases covered a wide range of subject domains based on the conversation agent's "personality." example subjects include ai, games, emotion, economics, film, books, sport, science, epistemology and metaphysics. these subjects are shown in fig.2 as part of the domain knowledge matrix model. in order to illustrate the ability of alice in handling common sense or open domain queries, alice has won the 2000, 2001 and 2004 loebner prize for being the most lifelike machine. the competition is based on the turing test 1) which aims to determine whether the responses from a computer can convince a human into thinking that the computer is a real person. in the competition, alice used a library of over 30,000 stimulus-response pairs written in aiml to answer the queries. the development of alice is based on the fact that the distribution of the sentences in conversations tends to follow zipf's law. 25) it is indicated that the number of "first words" is only limited to about two thousand. the frequency of any word is roughly inversely proportional to its rank in the frequency table. the most frequently used word will occur approximately twice as often as the second most frequent word. it in turn occurs twice as often as the third most frequent word, and so forth. questions starting with "what is" tend to have zipf-like distributions. this type of analysis can now be accomplished in a few milliseconds of computer time. while the possibilities * 5 www.alicebot.org/aiml/aaa/ of what can be said are infinite, the range of what is actually said in conversation in most cases is surprisingly small. specifically, 1800 words cover 95% of all the first words input. it is this principle that aini is operating on which enables it to be able to respond in an efficient and mostly accurate manner. at present, the world-wide web provides a distributed hypermedia interface to a vast amount of information available online. for instance, google 14) currently has a training corpus of more than one trillion words (1,024,908,267,229) from public web pages. this is valuable for many types of research. the web is a potentially unlimited source of knowledge repository; however, commercial search engines may not be the best way to gather answers from queries due to the overwhelming number of results from a search. before the rise of domain-oriented conversational systems using on natural language understanding and reasoning, evaluation was never a problem, as information retrieval-based metrics were readily available for use. however, when conversation systems begin to become more domain specific, evaluation becomes a real issue. 15) this is especially true when natural language processing (nlp) is required to cater for a wider variety of questions and, at the same time, required to achieve high quality responses. as shown in fig. 1 and 2 , aini's domain-specific knowledge bases consist of natural language corpus and frequently asked questions (faq). both components are extracted from the online documents using the akea as described in reference. 18) another significant aspect of this paper is the objective of aini to deliver essential information from trusted sources while capable of interacting with the users. a discussion on the selection of the trusted websites is given below. as the web that we know today becomes increasingly chaotic, overpowering and untrustworthy, selection of trusted web pages is becoming an important factor contributing to its long-term survival as a useful global information repository. in our experiment, the selection of the trusted websites is based on pagerank tm . 8) pagerank tm is a system for ranking web pages developed by larry page and sergey brin at stanford university. pagerank tm relies on the uniquely democratic nature of the web by using its vast link structure as an indicator of an individual page's value and high-quality sites always receive a higher pagerank tm . in our study, the selection of trustworthy websites started with the initial six seed words: bird, flu, avian, influenza, pandemic and h5n 1. these seeds are supposed to be representative of the domain under investigation. the seed terms are randomly combined and each combination is used in google api * 6 and bootcat tool 6) for bootstrapping corpora and terms from the web. we used the seeds to perform a set of google searches, asking google to return a maximum of 20 urls per query and then we collected the corpus. after visual inspection of the corpus, we used the top 40 seeds extracted from token frequencies for the second run. finally, we retrieved 1,428 urls out of 1,500 urls related to the domain being investigated. the reduction in number is due to the duplicated and broken link urls being removed. based on the 1,428 urls, we sent a query to google's pagerank tm directory using parameter tool 11) to determine their rankings. figure 3 shows the results of the top 10 site based on the pagerank tm scale. the pagerank tm scale goes from 1 to 10. a less important site is the one with a pagerank (pr) of 1. the most referenced and supposedly important sites are those with a pr of between 7 and 10. the final set of urls was further culled to include only selected sites which are attributed to a regulated authority (such as a governmental or educational institution) that controls the content of the sites. once the seed set is determined, each url's page is further examined and rated as either reliable or reputable. this selection is reviewed, rated and tested for connectivity with the trusted seed pages. from this exercise, whitehouse.gov, pandemicflu.org, cdc.gov and who.int were selected due to their pagerank tm scale scores being above 7. the most important factors in determining the "reliable authority" of a site is based on its history and the number of back-links to the governmental and international organization links. the more established and relevantly linked the more it will be considered as "stronger" or "more reliable." this effectively gives the linked site a measure of "trust" and "credential." the selected urls are then used as the source knowledge base for akea to extract the contents on bird flu so as to build aini's domain-specific knowledge base. a discussion of the extracted information is given in the following section. in aini's domain knowledge matrix model, the unit domains in the natural language corpus component consist of knowledge and information harvested from or expressed in ontologies, gazetteers, named entities and wordnet. these have been implemented as domain-dependent modular components. the named entity module identifies named locations, named persons, named organization, dates, times and key measures in text formats. the information is obtained by akea. for example, information for diseases is based on symptoms, causes, diagnoses, vaccination locations, persons and organizations. in order to identify these entities, our system uses rules to specify the named entities' structure in terms of text tokens and information from the source such as tagger, morphosyntactic analyzer and knowledge bases of names, clue words and abbreviations. the web knowledge base is then continuously updated with facts extracted from online pandemic news using information extraction (ie) by akea. ie is the task of extracting relevant fragments of text from larger documents and to allow the fragments to be processed further in an automated manner. an example of an application of akea is to prepare an answer for a user's query. the ontology and gazetteer have been implemented as domain-dependent modular components which will allow future improvements in the domain knowledge. in aini's faq component, the unit domain consists of information concerning diseases, symptoms, causes, diagnoses, vaccinations, etc. the selection of faq trusted web pages has been carried out using pagerank tm as discussed above. but at this stage, each of the selected websites was evaluated in order to find the more suitable and reliable faq pages. from this experiment, the answers.pandemicflu.gov and who.int/csr/disease/avian influenza/avian faq pages have been selected as the source of information for akea to build aini's faq knowledge base. based on the proposed approach, the quality of the results returned from aini's engine using the faq knowledge base are either similar to or better than those generated by search engines such as google. aini's sql engine uses the most significant words as keywords or phrase. it attempts to find the longest pattern to match without using any linguistic tools or nlp analysis. in this component, aini does not need a linguistic knowledge unit and relies on just an sql query. all questions and answers can be extracted from the database which was built by akea after applying a filtering process to remove unnecessary tags. example results from the system are illustrated in the next section. table 2 , aini's open domain knowledge base currently has more than 150,000 entries in the common sense stimulus-response categories. out of these, 100,000 came from mindpixel, 997 factoid questions from the trec training corpus and 45,318 categories from aaa knowledge bases. on the domain-specific knowledge base, aini has more then 1,000 online documents extracted by akea. this makes up 10,000 stimulus-response items in total. aini also has 158 faq pairs of questions and answers which have been updated using akea. in addition, aini has also collected more than 382,623 utterances in conversations with online users since 2005. these utterances will be integrated into aini's knowledge bases through supervised learning by domain experts. at present, aini has learnt 50,000 categories from conversations with online users. all of this combined knowledge has made up the total of 206,473 stimulus response categories in aini's knowledge bases. in comparison to other system, the original conversational programs such as eliza, 35) written by professor joseph weizenbaum of mit, have only 200 stimulus response categories. alice silver edition was ranked the "most human" computer, and has about 120,000 categories which include 80,000 taken from mindpixel. in this experiment, three types of systems are compared. they are search engine, question answering system and conversational system. for each system, we compared two different engines against aini. the two search engines compared are google and yahoo. for the question answering engines, askjeeves and start are used. they are supposed to use natural language processing for their queries. for the conversational engines, eliza and alice are selected. in particular, alice was ranked as the "most human computer" in the turing test competition. 28) google is a well known search engine which determines relevancy of information primarily on their pagerank algorithm. 24) in our experiment, we developed a search engine interface using google soap search api service 22) and yahoo!. 38) for the question answering system, the idea behind ask jeeves and start is to allow users to get answers for questions posed in natural language. ask jeeves is the first commercially question answering system available on the internet. start 27) is the world's first web-based question answering system which commenced operation since december, 1993. eliza is a well known program in the discipline of artificial intelligence and it is also the oldest system in a way, aini can also be considered as an enhancement of alice with the inclusion of the pandemic domain-specific knowledge base which was extracted by akea, and it also has parsing capabilities based on a full natural language understanding engine for multilevel natural language query. 17) in the present study, appropriate or accurate domain responses are expected to be important. the control question set was submitted to the seven urls where the seven systems were located. the responses of the queries are then collected and displayed as shown in fig. 4 in the course of the present study, a question was raised on how to compare the performance of the proposed system. it was decided that the performance of the system would be based on the keywords instead of complex phrases. for example, if the question "what is bird flu?" is asked, responses from the search engines, question-answering engines and conversational engines are summarized in table 3 , 4 and 5 respectively. the responses are noted as follows: • search engines -the systems identify the key words and return documents that are relevant to the query. the results could be tens of thousand or even millions of hits as shown in table 3 . • natural language question-answering systems -they generate responses using nlp and return fewer possible answers. the answer is table 3 responses from popular search engines -google and yahoo table 4 responses from popular question answering engines table 5 respohnses from conversational engines possibly within the context of the document. the sample response is shown in table 4 . • conversation engines -provide quick response from the stimulus. simple pattern matching and substitution are used based on a surprisingly small number of pre-defined rules. the example is shown in table 5 . §6 discussion in this example, eliza responded with "does that question interest you? " it is observed that eliza tries to ask another question, instead of giving an answer. the objective is to encourage the user to continue with the conversation. on the other hand, alice attempts to convince the user by generating randomly answer from aiml knowledge base. alice's response does not need a grammatical parser as her knowledge base contains the pattern "what is bird flu?" and the witty reply is "xfind *" with an aiml categories. by using "xfind *" pattern, alice will randomly generated responses such as "is there only one," "let me think about it," "have you tried a web search," "i haven't heard of bird flu," etc. the pattern matching language used in alice permits only one wild-card ('*') match character per pattern. therefore, al-ice responds with a variety of inputs from the users. alice does not concern whether it really "understands" the input. it aims to provide a coherent response to the client in order to convey the impression that the system understands the client's intention. for the eliza and alice systems, they are not able to handle questions that demand specific answers. they are simply not designed for such purpose. the three possible ways to handle these types of questions are: (a) analyze the problems with nlp and then provide an appropriate answer, (b) rely on human to review the conversation logs and continually improves the knowledge base, or, (c) treat the query as impossible and then choose a pre-defined random answer. for the aini chatterbot, the response was "avian influenza, or "bird flu," is a contagious disease of animals caused by viruses. . . ". the answer was generated from the domain-specific knowledge base using natural language understanding parsing from level 1. in this query, the answers were discovered by aini from the trusted sites such as who. in addition, the response is based on the natural language understanding and reasoning. the reasoning mechanism of the aini is based on answer discovery in layer-oriented knowledge base. currently ongoing work includes quantitative measurement and assessment of the results and performance of these different systems. although the systems used in this study were built with different objectives in mind, the purpose of this study is to show that there is a need of the proposed system to handle domain specific applications. at the same time, this study also shown that the proposed conversation agent architecture can achieve the expected objectives. §7 conclusion in this paper, we have reported the use of an aini conversation agent architecture to develop a domain specific intelligent web interaction. aini consists of a knowledge acquisition tool called akea for the gathering of conversation and domain-related knowledge. the proposed system has the potential to be used in a domain specific application area. the study has demonstrated with one particular area of domain expertise -bird flu. in this paper, we only worked on selected pandemic crisis websites where we performed knowledge extraction through akea for the domain-specific knowledge databases on the server. it is believed that the approach would be useful and applicable for other domains. we have also compared and tested the flexibility of aini against other popular search engines, question answering systems and conversation systems. furthermore, we have found that domain-specific knowledge base has higher response than the corresponding conversational-style responses. further work will be done on expanding the sources of knowledge and to provide quantitative measurements of the quality of responses from the aini. computing machinery and intelligence mobile and devices analysis of factoid questions for effective relation extraction artificial linguistic internet computer entity the annotated a.l.i.c.e. aiml bootcat: bootstrapping corpora and terms from the web just how deadly is bird flu? it depends on whom you ask inside pagerank overcoming the lack of screen spaces on mobile computers alice winner of chatterbox challenge the effect of document retrieval quality on factoid question-answering performance factoid question answering over unstructured and structured web content all our n-gram are belong to you a black-box approach for response quality evaluation conversational agent system domain matrix knowledge model for embodied conversation agents top-down natural language query approach for embodied conversational agent automated knowledge extraction from internet for a crisis communication portal domain knowledge model for embodied conversation agent intelligent agents for an internet-based global crisis communication system embodied conversational agents for h5n1 pandemic crisis google soap search api (beta) 3gsm statistics google and yahoo search engine technology comparison zipf's law dependency-based evaluation of minipar building a reusable test collection for question answering loebner prize is asian bird flu the next pandemic? common sense in humans text retrieval conference (trec) building large knowledge bases by mass collaboration computer power and human reason eliza -a computer program for the study of natural language communication between man and machine monolingual web-based factoid question answering in chinese, swedish, english and japanese understanding natural language he was associate professor of faculty of information technology and communication, university technical malaysia melaka. his research interest is in the development of intelligent agent and conversational agents to facilitate graceful human-computer interactions. he is a member of ieee we would like to thank the anonymous referees for the constructive and helpful suggestions. this research project is funded by murdoch university research excellence grant scheme (regs), 2006. key: cord-178783-894gkrsk authors: zhang, rui; hristovski, dimitar; schutte, dalton; kastrin, andrej; fiszman, marcelo; kilicoglu, halil title: drug repurposing for covid-19 via knowledge graph completion date: 2020-10-19 journal: nan doi: nan sha: doc_id: 178783 cord_uid: 894gkrsk objective: to discover candidate drugs to repurpose for covid-19 using literature-derived knowledge and knowledge graph completion methods. methods: we propose a novel, integrative, and neural network-based literature-based discovery (lbd) approach to identify drug candidates from both pubmed and covid-19-focused research literature. our approach relies on semantic triples extracted using semrep (via semmeddb). we identified an informative subset of semantic triples using filtering rules and an accuracy classifier developed on a bert variant, and used this subset to construct a knowledge graph. five sota, neural knowledge graph completion algorithms were used to predict drug repurposing candidates. the models were trained and assessed using a time slicing approach and the predicted drugs were compared with a list of drugs reported in the literature and evaluated in clinical trials. these models were complemented by a discovery pattern-based approach. results: accuracy classifier based on pubmedbert achieved the best performance (f1= 0.854) in classifying semantic predications. among five knowledge graph completion models, transe outperformed others (mr = 0.923, hits@1=0.417). some known drugs linked to covid-19 in the literature were identified, as well as some candidate drugs that have not yet been studied. discovery patterns enabled generation of plausible hypotheses regarding the relationships between the candidate drugs and covid-19. among them, five highly ranked and novel drugs (paclitaxel, sb 203580, alpha 2-antiplasmin, pyrrolidine dithiocarbamate, and butylated hydroxytoluene) with their mechanistic explanations were further discussed. conclusion: we show that an lbd approach can be feasible for discovering drug candidates for covid-19, and for generating mechanistic explanations. our approach can be generalized to other diseases as well as to other clinical questions. anteed. on the other hand, de novo development and approval of an effective antiviral therapy can take more than a decade. in the absence of an effective vaccine or other therapies, there have been significant efforts in repurposing drugs approved for other diseases for covid-19 treatment, some of which have been tested in clinical trials (e.g., dexamethasone [9] , hydroxychloroquine and lopinavir/ritonavir [10] ). computational approaches to drug repurposing have also garnered much attention to accelerate discovery of therapies for covid-19 [11, 12] . common computational drug repurposing methods include drug signature matching, molecular docking, genome-wide association studies, and network analysis [13] . these data-driven approaches involve systematic analysis of various types of biological and clinical data (e.g., gene expression, chemical structure, genome and protein sequences, and electronic health records) to generate hypotheses regarding repurposed use of approved or investigational drugs [13] . the potential of recent advances in artificial intelligence (ai) and machine learning for covid-19 drug repurposing has also been highlighted [14] and several studies using these techniques have reported promising results [15] [16] [17] [18] . in particular, approaches leveraging network medicine [19] principles and biological knowledge graphs have been emphasized [14] . most of these computational approaches have focused on biological data, such as gene expression, protein-protein and drug-target interactions, and used sars-cov-2-related data. however, covid-19-specific data is meaningful in the context of the larger body of diverse knowledge underpinning medicine and life sciences, a primary source of which is the biomedical literature. while some covid-19 drug repurposing studies incorporated some literature-based knowledge [15, 18] , their focus has remained largely covid-19-specific. we argue that efficiently and safely repurposing drugs to treat covid-19 requires more effective integration of literature-based knowledge with biological data collected via high-throughput methods. in this paper, we propose a novel literature-based discovery [20, 21] ap-proach for covid-19 drug repurposing. similar to related work [18] , we cast drug repurposing as a task of knowledge graph completion (or link prediction). we use a large, literature-derived biomedical knowledge graph constructed from semmeddb [22] as well as covid-19 research literature [23] , as our data source. we use several state-of-the-art, neural network-based algorithms [24] [25] [26] for the task, and also complement these approaches with an approach based on discovery patterns [27] . furthermore, we highlight the role of discovery patterns in search of mechanistic explanations for the proposed drugs. unlike most approaches that focus on covid-19-specific knowledge [15, 18] , we consider a larger body of biomedical knowledge, as captured in the pubmed bibliographic database as well as in the covid-19 research literature. our results show that our approach can identify known drugs that have been used for covid-19 and discover other novel drugs that can potentially be repurposed for covid-19. significant computational work has already been done to prioritize fdaapproved drugs for repurposing to treat covid-19 [11, 12] . for the most part, these studies can be categorized as molecular docking-based drug screening studies and network-based studies, the majority of them belonging to the former category. in molecular docking studies, small molecules in compound libraries are screened for effectiveness against the host proteins in the sars-cov-2 host interactome. many studies of this kind have been reported, and some of the proposed drugs such as ritonavir, ribavirin, remdesivir, oseltamivir, have been used in practice and many are being evaluated in clinical trials [28] [29] [30] [31] [32] [33] [34] [35] . while not as common as docking studies, network-based approaches to drug repurposing have also been explored. in one early study, a virus-related knowledge graph which consists of drug-target and protein-protein interactions and similarity networks from publicly available databases (e.g., drugbank [36] , chembl [37] , biogrid [38] ) was constructed and network-based machine learning and statistical analysis were used to predict an initial list of covid-19 drug candidates. this list was narrowed down based on text mining from the literature and gene expression profiles from covid-19 patients, and a poly-adpribose polymerase 1 (parp1) inhibitor cvl218, was proposed for therapeutic use against covid-19 [15] . cava et al. [39] used gene expression profiles from public datasets to construct a protein-protein interaction network in conjunction with pathway enrichment analysis to identify 36 potential drugs, including nimesulide, thiabendazole, and fluticasone propionate. in another study, network proximity analyses of drug targets and hcov-host interactions in the human interactome were used to prioritize 16 potential repurposed drugs, including melatonin, mercaptopurine, and sirolimus, which were validated by enrichment analyses of drug-gene signatures and transcriptome data in human cell lines. potentially useful drug combinations (e.g., melatonin plus mercaptopurine) were also suggested [16] . a follow-up study combined network medicine approaches based on human interactome with clinical patient data from a covid-19 registry to show that melatonin was associated with reduced likelihood of a positive sars-cov-2 laboratory test [17] . the approach was further extended to explore deep learning [18] . a comprehensive knowledge graph of drugs, diseases, and proteins/genes (named cov-kge) was constructed by combining molecular interaction information from the literature with knowledge from drugbank. a knowledge graph embedding model, named rotate [25] was used to represent the entities and the relationships in the knowledge-based in low-dimensional vector space. using the ongoing covid-19 trial data as a validation set, 41 high-confidence repurposed drug candidates (including dexamethasone, indomethacin, niclosamine, and toremifene) were identified, and further validated via an enrichment analysis of gene expression and proteomics data in sars-cov-2-infected human cells. another study used node2vec graph embeddings and variational graph autoencoders for the same purpose [40] . [41] evaluated three algorithms (graph neural network, network proximity, and network diffusion) on a network of drug protein targets and disease-associated proteins for covid-19 drug repurposing. while they obtained low correlations across the three algorithms, an ensembling approach that combined the predictions of all algorithms was shown to outperform the individual methods, ranking ritonavir, chloroquine, and dexamethasone among the most promising candidates. some limited literature knowledge relevant to covid-19 has been incorporated to network-based approaches; however, their focus remains largely on structured molecular interaction information encoded in databases. literature-based discovery (lbd) [20, 21] is a method of automatic hypothesis generation pioneered by swanson [42] . based on the concept of "undiscovered public knowledge", lbd seeks to uncover valuable hidden connections between disparate research literatures, and has been proposed as a potential solution for the problem of "research silos" (the view that scientific research areas are largely isolated from one another). the primary lbd paradigm is the so-called abc model. in the open discovery form of this model, a relationship between two concepts a and b is known in one research area and another relationship between concepts b and c is known in another, and a potential relationship between concepts a and c is proposed. conversely, in closed discovery, relationship ac is known, and a concept b is proposed as an explanation for the relationship ac. extensions to abc model have also been proposed, such as discovery browsing model that aims to elucidate more complex relationship paths between biomedical concepts [43, 44] . most applications of lbd have been in the biomedical domain, beginning with swanson's discovery of fish oil as a treatment for raynaud disease [42] , a hypothesis supported subsequently by clinical studies. while early lbd systems focused primarily on term cooccurrence [45, 46] , semantic relations have been widely used in later years for representing scientific content of biomedical publications [27, [47] [48] [49] . more recently, distributed vector representations based on term or semantic relation co-occurrence have been gaining popularity [50] [51] [52] . drug repurposing has been one of the prominent applications of lbd [27, [53] [54] [55] [56] [57] [58] . for example, hristovski et al. [27] used semantic discovery patterns following the abc model to identify potential therapeutic uses for drugs. zhang et al. [56] used discovery patterns and semmeddb relations to identify potential prostate cancer drugs. cohen et al. [55] used a vector representation approach based on semantic relations to predict a small number of active agents within a large library screened for activity against prostate cancer cells. knowledge graphs are represented as a collection of head entity-relation-tail entity triples (h,r,t), where entities correspond to nodes and relations to edges between them. knowledge graph completion (or link prediction) is the task of predicting unseen relations between two existing entities or to predict the tail entity given the head entity and the relation (or head entity given the tail entity and the relation). recent approaches to knowledge graph completion rely on knowledge graph embedding methods [59] , which learn a mapping from nodes and edges to continuous vector space that preserve the proximity structure of the knowledge graph and are amenable to application of machine learning methods. such methods include translational models, which use distance-based scoring functions (e.g., transe [24] , transh [60] , rotate [25] ), and semantic matching models, which use similarity-based scoring functions (e.g., rescal [61] , distmult [62] , complex [63] , and holographic embeddings (hole) [64] ). graph convolutional networks [65, 66] as well as methods that use context-based encoding approach (kg-bert [67] , stelp [26] ) have also been recently proposed. knowledge graph embedding techniques based on a network of drug, disease, and gene/protein entities, have been used to support drug repurposing for rare diseases [68] . graph convolutional networks were used to model drug side effects resulting from drug-drug interactions [69] . a multimodal graph of proteinprotein interactions, drug-protein target interactions and drug-drug interactions was constructed from publicly available datasets. sang et al. (2018) [70] constructed low-dimensional knowledge graph embeddings from semmeddb relations and trained a long short-term memory (lstm) model using known drug therapies from therapeutic target database [71] , proposing potential drugs us-ing the trained model. in this section, we first describe our data sources and the preprocessing steps that were taken to construct a literature knowledge graph from these data sources. next, we discuss the knowledge graph completion methods that we used to predict candidate drugs for covid-19 as well as the discovery patterns used for providing mechanistic explanations. lastly, we detail various evaluation schemes that we used to automatically validate our predictions. a workflow diagram illustrating our approach is provided in fig. 1 . we constructed our biomedical knowledge graph primarily from semmeddb [22] , a repository of semantic relations automatically extracted from biomedi-cal literature using semrep natural language processing (nlp) tool [72, 73] . semrep-extracted relations are in the form of subject-predicate-object triples (also called semantic predications) and are derived from unstructured text in concepts are enriched with semantic type information (disease or syndrome, pharmacologic substance, etc.) and the relations are linked to the supporting article and sentence. semmeddb has supported a wide range of computational applications, ranging from gene regulatory network inference [76] to in silico screening for drug repurposing [55] and medical reasoning [77] , and has also found widespread use for literature-based knowledge discovery and hypothesis generation [44, 48, [78] [79] [80] . in its most recent release (version 43, dated 8/28/2020) 1 , semmeddb contains more than 107m relations from more than in this work, we focused on a subset of semantic relations derived from the combination of pubmed and cord-19 datasets, predicted to be accurate and informative for drug repurposing. first, we eliminated relations involving generic biomedical concepts (i.e., relations in which both subject and object were present in a generic concept table of semmeddb such as pharmaceutical preparations) and relations with identical subject and object arguments. next, we excluded a subset of predicate types that were not expected to be useful for drug repurposing, such as part of and process of. the predicate types used are affects, associated with, augments, causes, coexists with, complicates, disrupts, inhibits, interacts with, manifestation of, predisposes, prevents, produces, stimulates, and treats. lastly, we also excluded the relations in which the subin the second step, we eliminated uninformative semantic relations using loglikelihood ratio (g 2 ) and network degree centrality for the concepts (in-degree and out-degree). we assigned each semantic relation a g 2 score indicating how strongly the terms within a triple are associated with each other [82] . a high g 2 score means that the observed and expected frequencies are significantly different, indicating that the triple is less likely to occur by chance. for computational purposes, we created two three-dimensional contingency tables with indices i, j, and k. the first table (ot) holds observed frequencies of a triple from the knowledge graph and the second table (et) contains the expected values assuming independence of terms in each triple. g 2 was then calculated using the equation where n ijk is the cell i, j, k in ot, m ijk is the cell i, j, k in et, and t = n ijk . next, we normalized all three measures (g 2 , in-degree, and out-degree) to the range [0, 1] and summed them up into a final score. the lower the score, the more specific and informative the relation is. for example, the relation operative surgical procedures-treats-woman with high score is more general than relation interleukin-6-affects-autoimmune diseases. we kept all relations for which the score value was less than a threshold value î±. we manually tuned the î± value to achieve a balance between specificity of relations and their variability. we kept 30 % of all relations with the lowest score in the data set. we also kept all biomedical concepts that refer to covid-19 terms (cuis: c5203670, c5203671, c5203672, c5203673, c5203674, c5203675, c5203676). at the end of the preprocessing stage, the knowledge graph consists of 131 355 nodes and 2 558 935 relations. the precision of semantic predications generated by semrep vary by domain (e.g., molecular interactions are less precise than clinical relationships). to improve the precision of the relations used for drug repurposing, we extended the semrep accuracy classifiers previously proposed [83, 84] . we fine-tuned a collection of transformer-based pretrained language models to classify semantic predications as correct vs. incorrect. we used the following models: vanilla bert (base size, cased and uncased) [85] , biobert [86] , bioclinicalbert [87] , bluebert [88] , and pubmedbert [89] . to extend the coverage of our existing classifiers, we used 6492 predications annotated as correct vs. incorrect with respect to their source sentences. we leveraged 6000 annotations from a previous study [84] (cohen's îº of 0.80) and annotated 492 new predications. annotation guidelines generated in the previous study was used. two of the authors (hk and mf) and two health informat-ics graduate students annotated predications containing predicates of interest absent in the prior study (fleiss' îº = 0.410, indicating moderate agreement). the resulting annotated set was split into 80/10/10 training/validation/test sets. hyperparameters were determined empirically and the learning rate was set to 1 ã� 10 â��5 , the batch size was 16, the maximum number of epochs was set to 10 but early stopping was employed. optimization was done using the adam optimizer [90] with decoupled weight decay regularization using betas (0.9, 0.999) and decay 0.01. the pooled output from the bert model was fed through a linear layer to produce logits that then underwent a softmax transformation to return class probabilities. a single tesla v100 gpu was used to train the models. we compared the performance of various above-mentioned transformers. the best classifier was then used to filter incorrect semantic predications. consider a knowledge graph g = (e, r, e), where e refers to a set of entities, r denotes a set of possible relations, and t stands for a set of triples in the form (h)ead-(r)elation-(t)ail, formally denoted as {(h, r, t)} â�� e ã� r ã� e. the aim of knowledge graph completion is to infer new triples (h , r , t ) such that h , t â�� e and r â�� r. in this setting, the knowledge graph completion problem could be represented as a ranking task in which we learn a prediction function ï�(h, r, t) : e ã� r ã� e â�� r which generates higher scores for true triples and lower scores for false triples. we explored three classes of knowledge graph completion methods: transe [24] and rotate [25] for translational models, distmult [62] and complex [63] for semantic matching models, and stelp [26] for context-based encoding. these methods differ in the way that they encode entities and relations in a knowledge graph into a low-dimensional vector space (i.e., kg embedding). such distributed vector representation can be used for downstream reasoning and machine learning tasks. transe [24] describes a triplet (h, r, t) as a translation between head entity h and tail entity t through relation r in a continuous vector space, i.e., h + r â�� t, where h, r, t â�� r d is the embedding of h, r, and t, respectively. to measure plausibility of relations transe employs a distance-based score function we choose transe because of its simplicity and good prediction performance. however, transe is able to model only one-to-one relations and fails to embed one-to-many, many-to-one, and many-to-many relations. to solve this problem, many other solutions have been proposed including rotate [25] . rotate treats each relation in a complex vector space as a rotation from the head entity to the tail entity, i.e., s(h, r, t) = |h â�¢ r â�� t| l1 , where â�¢ is a hadamard product. we selected rotate as a counterpart to transe, as transe reportedly does not perform well on some data sets (e.g., b15k family of data sets), which require symmetric pattern modeling. distmult [62] is the simplest approach among semantic matching models. semantic triple encoder for link prediction (stelp) [26] , is a contextbased encoding approach to knowledge graph completion. at its core is a siamese bert model that leverages sharing one set of weights across two models to produce encoded, contextual representations of the predications that are then where d is the set of correct triples, n (tp) is the set of corrupted triples for where î³ is a scaling factor for the contribution of the contrastive loss. at inference, stelp considers every entity-context combination for a given partial predication, (h, r) to find (t) or (r, t) to find (h), and ranks every pair using the sum of the positive class probability and the scaled negative euclidean distance. we replaced the vanilla base bert model proposed in the stelp paper with biobert, trained on biomedical literature corpora. the 1 016 124 unique predications remaining after preprocessing were each corrupted to produce five negative predications for a total of 5 080 620 negative predications and a grand total of 6 096 744 predications. the hyperparameters were set to the same values as in the original stelp paper and the learning rate was set to 1e-5, the batch size was 16, the contrastive loss scaling factor was 1.0. optimization was done using adam with decoupled weight decay with betas (0.9, 0.999) and decay 0.01. training was run for 29 000 training iterations. ranking was done by adding the scaled contrast score to the positive class probability and entities ordered in descending rank order. all preprocessing was done using custom bash and python scripts. transe, rotate, distmult, and complex link prediction models were implemented in pytorch using the dgl-ke package [91] for learning large-scale kg embeddings. the bert models were based on huggingface bert implementations using pytorch. pre-trained weights for biobert (biobert-base v1.1 (+ pubmed 1m)) 2 , bioclinicalbert 3 , pubmedbert 4 and bluebert (bluebert-base, uncased, pubmed+mimic-iii) 5 came from various sources associated with each paper. stelp was also implemented using a combination of a hug-gingface bert model and pytorch. our source codes are also publicly available 6 . discovery patterns are defined as a set of constraints that need to be satisfied for the discovery of new relations between concepts [27] . herein, we used discovery patterns for two purposes. first, we explored open discovery patterns to identify drugs that can be repurposed for covid-19. second, we used closed discovery patterns to propose plausible mechanisms for drugs identified via knowledge graph completion methods described above. discovery patterns are expressed in terms of predication pairs (or predication chains). in particular, we focused on the following discovery pattern: druga -inhibits|interacts with -conceptb and 10/15/2020). we focus on the latter category in our qualitative evaluation below. we semi-automatically generated a ground truth drug list, similar to the approach in other computational drug repurposing studies for covid-19 [18] . we downloaded the interventions used in covid-19 drug trials from clinicaltrials.gov using the following search: https://clinicaltrials.gov/ct2/res this yielded a set of 1167 clinical trials. we extracted the interventions from these studies and mapped the intervention terms to umls cuis using metamap (v2016) [92] and filtered the resulting concepts by their semantic groups [93] , keeping only those concepts with the semantic group chemicals & drugs. we also considered the additional semantic types therapeutic procedure and gene or genome, which also appeared for some concepts in intervention lists. we removed the duplicates and some general concepts (e.g., therapeutic procedure, placebo) as well as incorrect mappings, which resulted in a final list of 285 concepts. the automatic evaluation (below) was performed against this set. time slicing is an evaluation technique often used in literature-based discovery and link prediction tasks [20] . the idea is to train models on data prior to a specific date and test them on data after that date and evaluate whether links that formed only after the cutoff date can be predicted from the trained model. in this study, we trained our models on semantic relations extracted from publications dated 03/11/2020 or earlier and tested whether they can predict the drugs that have been proposed for covid-19 since then or have been evaluated in clinical trials. this date was selected as cutoff, as it is the date on which who declared covid-19 a pandemic. it is also a date by which enough biological knowledge about sars-cov-2 had accumulated, although covid-19 therapies were still in their infancy, making it a suitable cutoff for time slicing experiments. all five link prediction models were automatically assessed using a link prediction evaluation protocol proposed by bordes et al. [24] . suppose that x is a set of triples, î� e be the embeddings of entities e, and î� r be the embeddings of relations r. in the first, corruption step, we go through a set of triples and for each triple x = (h, r, t) â�� x replace its head and tail with all other entities in e. each triple is corrupted exactly 2|e| â�� 1 times. formally, the corrupted triple is defined as: where h = h and t = t. we employ the filtered setting protocol not taking into account any corrupted triple that already appears in the kg. in the second, scoring phase, original and corrupted triples are tested using the constructed classifier ï�. intuition behind this is that the model will assign a higher score to the original triple and a lower score to the corrupted triple. in the third, evaluation phase, the proposed link prediction models are assessed using three measures: mean rank (mr), mean reciprocal rank (mrr), and hits@k measure. mr is an average rank assigned to the true predication, over all predications in a test set: where rank h i and rank t i denote the rank position: where the indicator function i[p ] is 1 iff p is true, and 0 otherwise. mrr is the average inverse rank for all test triples and is formally computed as: hits@k measures the percentage of predications in which the true triple appears in the top k ranked triples, where k â�� {1, 3, 10}; formally: our aim was to achieve low mr and high mrr and hits@k. in addition, we also performed a qualitative evaluation. one of the authors (mf) used neo4j browser to assess the plausibility of some of the drugs highly ranked by the knowledge completion models, guided by literature search and review, and following the closed and open discovery paradigms. we report the performance of the semantic relation accuracy classifier as well as the knowledge graph completion methods in this section. the full table of results for the comparison of various bert models for the accuracy classifier is included below ( table 1) the link prediction results for all employed models are presented in table 3 . for mr a lower score is considered better, for all others a higher score is considered better. the score for each method is the mean value over all triplets in the testing set. sion, l 1 norm, learning rate î· = 0.01 and regularization coefficient î» = 2ã�10 â��8 . model training was limited to 20 000 epochs. relatively small number of relations (15) ensure that all entities and relations can be smoothly embedded into the same vector space. next, we use t-sne (t-distributed stochastic neighbor embedding) [94] algorithm to graphically represent embeddings of computed concepts in a twodimensional space (figure 2 ). t-sne algorithm enables reduction of highdimensional data into a low-dimensional space such that similar concepts are presented by nearby points. the plot demonstrates relatively good co-localization of selected concepts, especially for suspected covid-19 and paclitaxel. our results indicate that more complex knowledge graph completion models might be less efficient in drug repurposing tasks. theoretical considerations suppose that transe is outperformed by its successors [25, 62, 63] . however, differences in performances among distmult, complex, and rotate are relatively small. all three models achieved low performance on mrr, hits@1, inherently model only one-to-one relations and fails to represent one-to-many, many-to-one or even many-to-many relations, it shows its efficiency in embedding a large-scale complex biomedical knowledge graph, such as the extended semmeddb used here. empirical evidence shows that distmult and complex usually perform well for high-degree entities, but fails with low-degree entities [95] . because we eliminated highly frequent concepts due to their lack of informativeness, it is possible that this is reflected in lower performance scores of both models. the context-encoding model, stelp, showed rather poor performance in evaluation. one possibility is that the model was only able to learn high-level groupings for the predicates. this is likely the case as it was observed the model versus affects covid-19 etc., it did not learn more granular features that allow it to differentiate between subjects within the context of treats covid-19. however, analysis of the t-sne embedding and the qualitative evaluation show that the model mostly clustered the ground truth drugs into a couple of large clusters. to further compare the drug rankings between transe and stelp, we performed the wilcoxon signed-rank test (p = 0.851), which indicates that no correlation was found between how the two models were ranking novel predications. spearman's rank correlation between the novel predication rankings for both models was found to be â��0.003 and this further supports the results of the wilcoxon test. table 4 and table 5 show that there is very little agreement between transe and stelp, particularly in the top 1000 rankings for each model. it is worth noting that there were 47 items in common in the top 1000 rankings for both models. ranked triples for the specified model, calculating the absolute difference between the rankings from the two models for each of those triples, and calculating the statistics. for example, the triples that transe ranked as the top 1000 triples we gathered, the absolute differences of rankings between transe and stelp for those 1000 triples were calculated, and the statistics were calculated from those differences. the can be possible to explore larger graphs than that explored in this work. on the other hand, with adequately large computational resources, it may be possible to optimize stelp hyperparameters and train over multiple random seeds to generate a model that obtains better results than transe or rotate, which are limited by their smaller representational capacity. discovery patterns based on semantic relations provide an intuitive way of exploring potential mechanistic links between biological phenomena. neo4j and cypher, its query language, are powerful tools that complement semantic relations nicely in quickly pinpointing promising research directions, although massive graphs present some challenges for effective query and retrieval. in addition, a domain expert is needed to sort out some of the noise in semantic relations (some of it obvious) due to text mining errors. however, given that predictions made by the knowledge completion models above are largely opaque, a human-in-the-loop discovery browsing approach based patterns [43, 44] remains an effective alternative to these more complex approaches, and also complements them by providing potential explanations. the following classes of drugs have been used for the management of covid-19 so far: antivirals (e.g., remdesivir), antibodies (e.g., convalescent plasma), anti-inflammatory agents (e.g., dexamethasone), immunomodulators (e.g., interleukin inhibitors), anticoagulants (e.g., heparin), antifibrotics (e.g., tyrosine kinase inhibitors), and adjuvants (e.g., vitamin d) [96, 97] . in addition, several trials have studied antimalarials (e.g., hydroxychloroquine) and antiparasites (e.g., ivermectin), but evidence from trials do not support their use. the knowledge graph completion models did not predict antivirals, antimalarials, and antiparasites, except for antivirals from the class neuraminidase inhibitors and the antimalarial artemisone. all the other drug classes and most of their members were predicted by the models. dexamethasone, currently considered the most effective drug for reducing mortality in patients receiving oxygen, was the highest ranking drug from the rotate model. it is possible that the models missed specific antivirals and antiparasites due to their mechanism of action, which usually involves binding to specific receptors, a relation type on which semrep does relatively poorly. despite this issue, qualitative assessment of the drugs predicted by the models was overall positive. using the open discovery pattern approach, we identified five promising drugs that were ranked highly and were not, to our knowledge, discussed in the literature, which we discuss below (paclitaxel, sb 203580, alpha 2-antiplasmin, pyrrolidine dithiocarbamate, and butylated hydroxytoluene). the same approach also yielded other highly ranked substances, which are currently evaluaetd in clinical trials, such as quercetin, melatoninm losartan, estradiol, and simvastatin. note that the knowledge graph completion models predicted 7 of these drugs (excluding sb 203580, alpha 2-antiplasmin, and pyrrolidine dithiocarbamate). figure 3 shows the resulting network from this discovery pattern generated by neo4j browser. paclitaxel is used to treat several cancer types, including ovarian cancer, breast cancer, lung cancer, cervical cancer, and pancreatic cancer. it stabilizes the microtubule polymer and protects it from disassembly. chromosomes are thus unable to achieve a metaphase spindle configuration. this blocks the progression of mitosis and prolonged activation of the mitotic checkpoint triggers apoptosis or reversion to the g0-phase of the cell cycle without cell division [98] . the following patterns support the paclitaxel discovery: 1. paclitaxel-inhibits-interleukin-6-causes-covid-19 2. paclitaxel-inhibits-nf-kappa b-associated with-covid-19 3. paclitaxel-inhibits-interleukin-1, beta-associated with-covid-19 4. paclitaxel-inhibits-granulocyte colony-stimulating factorassociated with-covid-19 5. paclitaxel-inhibits-interleukin-10-predisposes-covid-19 6. paclitaxel-inhibits-interleukin-8-predisposes-covid-19 7. paclitaxel-inhibits-thromboplastin-associated with-covid-19 the first six patterns support a role for paclitaxel in alleviating the cytokine storm of covid-19, triggered by dysfunctional immune response and mediating widespread lung inflammation. paclitaxel may plausibly help as an immunosuppressive therapy to immunomediated damage in covid-19 [99] . thromboplastin (pattern 7) is a complex enzyme found in brain, lung, and other tissues and especially in blood platelets and functions in the conversion of prothrombin to thrombin in the clotting of blood and may be elevated in patients with covid-19. as pulmonary microvascular thrombosis plays an important role in progressive lung failure in covid-19 patients, paclitaxel may reduce the state of hypercoagulability by acting as an inhibitor of thromboplastin [100] . the final pattern involves the interaction of paclitaxel with tlr4. paclitaxel is known to have high affinity for tlr4 receptors. sars-cov-2 spike protein binds with human innate immune receptors, mainly tlr4, increasing secretion of il-6 and tnf-î± and neuroimmune response. this suggests that paclitaxel may dislocate sars-cov-2 spike proteins [101, 102] . sb 203580 is a specific inhibitor of p38î±, which suppresses downstream activation of mapkap kinase-2, involved in many cellular processes including stress and inflammatory responses and cell proliferation. the following patterns support the sb 203580 discovery: ity, especially in patients with comorbidities such as hypertension, diabetes, and coronary heart disease [104] . the following patterns support the alpha 2-antiplasmin discovery: 1. alpha 2-antiplasmin-inhibits-plasmin-predisposes-covid-19 2. alpha 2-antiplasmin-inhibits-fibrinogen-associated with-covid-19 3. alpha 2-antiplasmin-interacts with-igy-associated with-covid-19 more specifically, plasmin may cleave a newly inserted furin site in the s protein of sars-cov-2, which increases its infectivity and virulence in covid-19. in addition, fibrinogen levels are higher in covid-19 patients and may contribute to hypercoagulability [104] . by inhibiting plasmin and fibrinogen (first two patterns), alpha 2-antiplasmin may confer protection to covid-19. in addition, pattern 3 suggests a mechanism of protection via immunoglobulin y (igy). in the immunology field, igy against acute respiratory tract infection has been developed for more than 20 years. several igy applications have been effectively confirmed in both human and animal health. igy antibodies extracted from chicken eggs have been used in bacterial and viral infection therapy. igy production has been proposed as immunization as an adjuvant therapy in viral respiratory infection caused by covid-19 infection [105] . chicken immunized with alpha 2-antiplasmin and the peptide-specific antibody (igy) was isolated from the egg yolks of hens that could be used as potential protections for covid-19 patients [106] . pyrrolidine dithiocarbamate is a family of drugs used for metal chelation, induction of g1 phase and cell cycle arrest. it binds to zinc and the resulting complex can enter the cell and inhibit viral rna-dependent rna polymerase [107] . it is supported by the following patterns: 1. pyrrolidine dithiocarbamate-inhibits-nf-kappa b-associated with2. pyrrolidine dithiocarbamate-inhibits-interleukin-6associated with-covid-19 3. pyrrolidine dithiocarbamate-inhibits-tnf protein, humanassociated with-covid-19 the mechanisms suggested here are similar to those observed for the previous drugs. pyrrolidine dithiocarbamate contains antioxidants and prevents inflammatory changes. it inhibits the expression of il-6 and tnf, and nf-îºb in the virus-infected chorion cells through its antiviral activity. it has been proposed for the treatment of influenza [107] and it may have potential as a therapeutic option for covid-19. butylated hydroxytoluene is a lipophilic compound useful for its antioxidant properties. it is widely used to prevent free radical-mediated oxidation in fluids and other materials and is generally recognized as safe as a food additive. it has been postulated in the past as an antiviral drug. open discovery identified the following relevant patterns: 1. butylated hydroxytoluene-inhibits-cd69 protein, human2. butylated hydroxytoluene-inhibits-free radicals-associated with3. butylated hydroxytoluene-inhibits-tnf protein, humanassociated with-covid-19 4. butylated hydroxytoluene-inhibits-hydrogen peroxideassociated with-covid-19 the first pattern indicates butylated hydroxytoluene as an inhibitor of cd69. studies have shown that the cd69+ cells were detected in the lung of patients with asthmatic and eosinophilic pneumonia, suggesting a crucial role for cd69 in the pathogenesis of such inflammatory diseases. cd69 is, potentially, a new therapeutic target for patients with intractable inflammatory disorders and tumors [108] . therefore, by inhibiting cd69, butylated hydroxytoluene may halt potential inflammatory responses in covid-19. however, cd69 does not appear to be a major player in the physiopathology of covid-19 (the query "cd69 and covid-19" did not return any results in pubmed). nonetheless this is noteworthy, because this pathway is suggested as a novel and important pathway for all immune responses [108] . the crucial role of free radicals in covid-19 has been acknowledged and an antioxidative therapeutic strategy for covid-19 has been suggested [109] . along these lines, patterns 2-3 point to antioxidant function of butylated hydroxytoluene by scavenging free radicals and inhibiting reactive oxygen species [110] . our approach relies on accuracy of the predications extracted by semrep. semrep precision is about 0.70 and its recall around 0.42 [73] . while the accuracy classifier helped us improve the accuracy of the predications used, the remaining errors were still significant, impacting the knowledge graph completion task. in addition, despite aggressive filtering, the graph formed by the relations in extended semmeddb is very large, making it difficult to apply computationally intensive models like stelp. in this study, we examined a sub-graph which, inevitably, results in a loss of information available to knowledge graph completion techniques. while we were still able to apply modeling techniques to a fairly large sub-graph focusing on drug repurposing, there exists a larger, complementary sub-graph that may provide further drug candidates. as noted above, the transe model benefited from hyperparameter tuning using a grid search method to find an optimal configuration. similarly, stelp would likely benefit from a similar tuning to find an optimal configuration. for example, a single linear layer was used on the pooled output from the biobert model to produce the logits when increasing the representational capacity of the linear layer, by depth or width, might allow for stelp to develop a richer model of the underlying space formed by the biobert contextualized embeddings. our methods were limited to knowledge from the literature. other types of biological data (e.g., protein-protein interactions, drug-target interactions, gene/protein sequences, pharmacogenomic and pharmacokinetic data) are likely to benefit identification of drug candidates, as shown to some extent by other studies [14] , as well as our prior work [53] . however, the computational resources needed for training models based on such massive data can be prohibitive. transe and similar methods seem more promising in that respect. lastly, with our in silico approach, we can of course only propose drug candidates for repurposing. to evaluate whether these drugs could indeed act as therapeutic agents for covid-19, clinical studies are needed. however, the fact that we were able to identify some drugs known to have some benefit for covid-19 (e.g., dexamethasone) via purely computational methods that rely only on automatically extracted literature knowledge is encouraging. in this study, we proposed an approach that combines literature-based discovery and knowledge graph completion for covid-19 drug repurposing. unlike similar efforts that largely focused on covid-19-specific knowledge, we incorporated knowledge from a wider range of biomedical literature. we used state-of-the-art knowledge graph completion models as well as simple but effective discovery patterns to identify candidate drugs. we also demonstrated the use of these patterns for generating plausible mechanistic explanations, showing the complementary nature of both methods. the approach proposed here is not specific to covid-19 and can be used to repurpose drugs for other diseases. it can also be generalized to answer other clinical questions, such as discovering drug-drug interactions or identifying drug adverse effects. as covid-19 pandemic continues its spread and disruption around the globe, we are reminded how the spread of infectious diseases is increasingly common and future pandemics ever more likely. innovative computational methods leveraging existing biomedical knowledge and infrastructure could help us plan for, respond to and mitigate the effects of such global health crises. drug repurposing is a key piece of this response, and our approach provides an efficient computational method to facilitate this goal. sb 203580-inhibits-interleukin-6 -causes sb 203580-inhibits-tnf protein sb 203580-inhibits-interleukin-1, beta-associated with sb 203580-inhibits-nf-kappa b-associated with sb 203580-inhibits-interleukin-1-causes sb 203580-inhibits-granulocyte-macrophage colony-stimulating factor -associated with sb 203580-inhibits-macrophage colony-stimulating factorassociated with similarly to paclixatel, all patterns involving sb 203580 point to a potential inhibition of the hyperinflammatory response in covid-19 of the protein kinases p38î± in inflammation and innate immunity was found when the compound sb203580 suppressed tumor necrosis factor (tnf) production in monocytes, and this resulted in inhibition of septic (infammatory) shock alpha 2-antiplasmin is a serine protease inhibitor responsible for inactivating plasmin draft landscape of covid-19 candidate vaccines placebo-controlled study of azd1222 for the prevention of covid-19 in adults statement on astrazeneca oxford sars-cov-2 vaccine, azd1222, covid-19 vaccine trials temporary pause safety and immunogenicity of an rad26 and rad5 vector-based heterologous prime-boost covid-19 vaccine in two formulations: two open, non-randomised phase 1/2 studies from russia researchers highlight 'questionable' data in russian coronavirus vaccine trial results dexamethasone in hospitalized patients with covid-19-preliminary report effect of hydroxychloroquine in hospitalized patients with covid-19: preliminary results from a multi-centre, randomized, controlled trial current status of covid-19 therapies and drug repositioning applications covid-19 drug repurposing: a review of computational screening methods, clinical trials, and protein interaction assays drug repurposing: progress, challenges and recommendations artificial intelligence in covid-19 drug repurposing, the lancet digital health a data-driven drug repositioning framework discovered a potential therapeutic agent targeting covid-19 networkbased drug repurposing for novel coronavirus 2019-ncov/sars-cov-2 a network medicine approach to investigation and population-based validation of disease manifestations and drug repurposing for covid-19 repurpose open data to discover therapeutics for covid-19 using deep learning network medicine: a networkbased approach to human disease literature based discovery: models, methods, and trends emerging approaches in literature-based discovery: techniques and performance review semmeddb: a pubmed-scale repository of biomedical semantic predications cord-19: the covid-19 open research dataset translating embeddings for modeling multi-relational data rotate: knowledge graph embedding by relational rotation in complex sspace semantic triple encoder for fast open-set link prediction exploiting semantic relations for literature-based discovery a sars-cov-2 protein interaction map reveals targets for drug repurposing discovery of sars-cov-2 antiviral drugs through large-scale compound repurposing analysis of therapeutic targets for sars-cov-2 and discovery of potential drugs by computational methods anti-hcv, nucleotide inhibitors, repurposing against covid-19 virtual screening and repurposing of fda approved drugs against covid-19 main protease using integrated computational approaches to identify safe and rapid treatment for sars-cov-2 fast identification of possible drug treatment of coronavirus disease-19 (covid-19) through computational drug repurposing study ribavirin, remdesivir, sofosbuvir, galidesivir, and tenofovir against sars-cov-2 rna dependent rna polymerase (rdrp): a molecular docking study drugbank: a knowledgebase for drugs, drug actions and drug targets chembl: a large-scale bioactivity database for drug discovery biogrid: a general repository for interaction datasets in silico discovery of candidate drugs against covid-19 predicting potential drug targets and repurposable drugs for covid-19 via a deep generative model for graphs network medicine framework for identifying drug repurposing opportunities for covid-19 fish oil, raynaud's syndrome, and undiscovered public knowledge., perspectives in biology and medicine graph-based methods for discovery browsing with semantic predications semantic medline for discovery browsing: using semantic predications and the literature-based discovery paradigm to elucidate a mechanism for the obesity paradox an interactive system for finding complementary literatures: a stimulus to scientific discovery using concepts in literature-based discovery: simulating swanson's raynaud-fish oil and migraine-magnesium discoveries using the literature-based discovery paradigm to investigate drug mechanisms exploring relation types for literature-based discovery context-driven automatic subgraph creation for literature-based discovery reflective random indexing and indirect inference: a scalable method for discovery of implicit connections finding schizophrenia's prozac emergent relational similarity in predication space embedding of semantic predications combining semantic relations and dna microarray data for novel hypotheses generation using literature-based discovery to identify novel therapeutic approaches predicting high-throughput screening results with scalable literature-based discovery methods exploiting literature-derived knowledge and semantics to identify potential prostate cancer drugs a new method for prioritizing drug repositioning candidates extracted by literature-based discovery literature-based discovery of new candidates for drug repurposing knowledge graph embedding: a survey of approaches and applications knowledge graph embedding by translating on hyperplanes a three-way model for collective learning on multi-relational data., in: icml embedding entities and relations for learning and inference in knowledge bases complex embeddings for simple link prediction holographic embeddings of knowledge graphs convolutional 2d knowledge graph embeddings modeling relational data with graph convolutional networks kg-bert: bert for knowledge graph completion a literaturebased knowledge graph embedding method for identifying drug repurposing opportunities in rare diseases modeling polypharmacy side effects with graph convolutional networks gredel: a knowledge graph embedding based method for drug discovery from biomedical literatures ttd: therapeutic target database the interaction of domain knowledge and linguistic structure in natural language processing: interpreting hypernymic propositions in biomedical text broad-coverage biomedical relation extraction with semrep the unified medical language system the unified medical language system (umls): integrating biomedical terminology augmenting microarray data with literature-based knowledge to enhance gene regulatory network inference a reasoning and hypothesis-generation framework based on scalable graph analytics enabling discoveries link prediction on the semantic medline network are abstracts enough for hypothesis generation? investigating the role of interleukin-1 beta and glutamate in inflammatory bowel disease and epilepsy using discovery browsing keep up with the latest coronavirus research extending the log-likelihood measure to improve collocation identification, master's thesis mining biomedical literature to explore interactions between cancer drugs and dietary supplements evaluating active learning methods for annotating semantic predications bert: pre-training of deep bidirectional transformers for language understanding biobert: a pre-trained biomedical language representation model for biomedical text mining proceedings of the 2nd clinical natural language processing workshop transfer learning in biomedical natural language processing: an evaluation of bert and elmo on ten benchmarking datasets domain-specific language model pretraining for biomedical natural language processing adam: a method for stochastic optimization training knowledge graph embeddings at scale an overview of metamap: historical perspective and recent advances aggregating umls semantic types for reducing conceptual complexity visualizing data using t-sne a capsule network-based embedding model for knowledge graph completion and search personalization pharmacologic treatments for coronavirus disease 2019 (covid-19): a review pathophysiology, transmission, diagnosis, and treatment of coronavirus disease 2019 (covid-19): a review how taxol/paclitaxel kills cancer cells the trinity of covid-19: immunity, inflammation and intervention covid-19: coagulopathy, risk of thrombosis, and the rationale for anticoagulation the role of tlr4 in chemotherapy-driven metastasis is tolllike receptor 4 involved in the severity of covid-19 pathology in patients with cardiometabolic comorbidities? what goes up must come down: molecular basis of map-kap kinase 2/3-dependent regulation of the inflammatory response and its inhibition elevated plasmin (ogen) as a common risk factor for covid-19 susceptibility igy-turning the page toward passive immunization in covid-19 infection purification of human î±2-antiplasmin with chicken igy specific to its carboxy-terminal peptide antiviral function of pyrrolidine dithiocarbamate against influenza virus: the inhibition of viral gene replication and transcription a new therapeutic target: the cd69-myl9 system in immune responses tackle the free radicals damage in covid-19 understanding the chemistry behind the antioxidant activities of butylated hydroxytoluene (bht): a review we thank franã§ois-michel lang, leif neve, and jim mork for their assistance with processing the cord-19 dataset with semrep and providing updates to semmeddb. we acknowledge tom rindflesch for his encouragement with the project. key: cord-302305-xr067v2n authors: van aalst, jan title: distinguishing knowledge-sharing, knowledge-construction, and knowledge-creation discourses date: 2009-06-20 journal: int j comput support collab learn doi: 10.1007/s11412-009-9069-5 sha: doc_id: 302305 cord_uid: xr067v2n the study reported here sought to obtain the clear articulation of asynchronous computer-mediated discourse needed for carl bereiter and marlene scardamalia’s knowledge-creation model. distinctions were set up between three modes of discourse: knowledge sharing, knowledge construction, and knowledge creation. these were applied to the asynchronous online discourses of four groups of secondary school students (40 students in total) who studied aspects of an outbreak of severe acute respiratory syndrome (sars) and related topics. the participants completed a pretest of relevant knowledge and a collaborative summary note in knowledge forum, in which they self-assessed their collective knowledge advances. a coding scheme was then developed and applied to the group discourses to obtain a possible explanation of the between-group differences in the performance of the summary notes and examine the discourses as examples of the three modes. the findings indicate that the group with the best summary note was involved in a threshold knowledge-creation discourse. of the other groups, one engaged in a knowledge-sharing discourse and the discourses of other two groups were hybrids of all three modes. several strategies for cultivating knowledge-creation discourse are proposed. for two decades, carl bereiter and marlene scardamalia have been developing an educational model intended to make the processes experts use to advance the state of knowledge in their fields more common in education. the model was initially called "intentional learning" to emphasize that learning needs to be an intended goal rather than the by-product of activities (bereiter and scardamalia 1989) and then "knowledge building," suggesting that knowledge is the product of a constructive process (bereiter and scardamalia 1993) . but as constructivism has gained wide receiver if they comprehend its content and significance. examples are providing factual information to answer a query or uploading various kinds of information to an intranet. one thing that makes such interactions effective is that the receiver has already identified a need for the information. for example, someone new to editing digital video may need to be shown how to add music to the video, which will address an already meaningful goal. as a social practice, knowledge sharing is an accomplishment, especially in competitive environments; people are not naturally inclined to share what they know unless doing so is likely to enhance their own social position. the management literature indicates that knowledge-sharing practices can make organizations more effective, but they need to be cultivated (lencioni 2002) . in a community engaged in collaborative inquiry, knowledgesharing practices involve the introduction of information and ideas without paying extensive attention to their interpretation, evaluation, and development. the perceived lack of a need for interpretation and evaluation can be related to naïve realism, an epistemic position according to which data speak for themselves (science council of canada 1984) . a related epistemic belief is "quick learning," which has been linked to overconfidence in knowledge (schommer 1990 ). the ideas shared are not modified by the sharing interaction (bereiter and scardamalia 1987; pea 1994) , and knowledge sharing is not reflective. knowledge construction refers to the processes by which students solve problems and construct understanding of concepts, phenomena, and situations, considered within cognitive psychology. it is effortful, situated, and reflective, and can be individual or social (sullivan palincsar 1998) . the basic assumption of constructivism is that the student must make ideas meaningful in relation to his or her prior knowledge and to the situation in which the need for ideas arises (von glasersfeld 1995) . the cognitive processes are "situated" because they are mediated (enabled) by social interactions within the particular group that is working together and by the particular technologies used (brown et al. 1989; hutchins 1995) . knowledge construction is often associated with deep learning, which involves "qualitative changes in the complexity of students' thinking about and conceptualization of context-specific subject matter" (moore 2002, p. 27 ; also see biggs 1987) . dole and sinatra (1998) conceptualize the effort students invest in information processing as "engagement," ranging from simple processing that leads to assimilation (low), to deeper processing and some reflection that leads to knowledge restructuring (moderate), and on to substantially metacognitive processing (high). at moderate to high levels of engagement, knowledge construction can lead to the substantial restructuring of knowledge, which may include the invention of new concepts and enhanced meta-conceptual knowledge (e.g., knowledge about the hierarchical nature of networks of concepts). for example, students may initially consider the motion of an apple that falls from a tree to be unrelated to the motion of the earth in its orbit around the sun, but then come to realize that both can be described using the universal law of gravitation. this change would imply deeper insight into the nature of gravity and would lead to a restructuring of knowledge; the resulting knowledge structure would explain a greater range of observations and require fewer assumptions. more generally, synthesis that results in understanding phenomena on a higher plane and the creation of new concepts is an important form of knowledge advancement. for example, mendeleev's introduction of the periodic table of the elements accelerated progress in chemistry by predicting the existence of unobserved elements and the creation of new concepts to explain the partially observed patterns. scardamalia (2002) conceptualizes such advances as "rise-above," which she described as "working toward more inclusive principles and higher-level formulations of problems. it means learning to work with diversity, complexity and messiness, and out of that achieve new syntheses. by moving to higher planes of understanding knowledge [creators] transcend trivialities and oversimplifications and move beyond current best practices" (p. 79). although scardamalia proposes rise-above as a knowledge-creation principle, i regard it as a cognitive act whereby students articulate higher levels of understanding and not merely reorganize knowledge (gil-perez et al. 2002) ; nevertheless, the need for rise-above is greater when the need for synthesis is greater. knowledge construction involves a range of cognitive processes, including the use of explanation-seeking questions and problems, interpreting and evaluating new information, sharing, critiquing, and testing ideas at different levels (e.g., conjectures versus explanations that refer to concepts and/or causal mechanisms), and efforts to rise above current levels of explanation, including summarization, synthesis, and the creation of new concepts. however, educational approaches vary considerably in the extent to which they make it possible for students to engage in these processes. although most emphasize working with information and ideas (e.g. goldberg and bendall 1995; hunt and minstrell 1996; linn et al. 2003) , there may be limited opportunities for students to pursue problems they have identified themselves or to synthesize ideas and formulate new concepts. for example, in problem-based learning (hmelo-silver and barrows 2008), students are provided problems, although these are ill-structured and need considerable articulation. in other approaches, students may collaborate in small groups on relatively simple tasks that require little synthesis and reflection on progress. in the vast majority of approaches, knowledgeconstruction processes are directed at acquiring the reliable knowledge of a field (edelson et al. 1999; kolodner et al. 2003; krajcik et al. 2008) . knowledge construction, with its emphasis on building on students' prior ideas, concepts and explanations, and their metacognition, produces deeper knowledge in complex domains than does knowledge sharing (bransford et al. 1999; hmelo-silver et al. 2007 ). the term "knowledge creation" is used in the literature on expertise and innovation to describe how companies, organizations, and academic fields develop the ideas needed to sustain innovation (e.g. engeström 2001; gundling 2000; nonaka and takeuchi 1995) . knowledge creation depends on conditions in which creative work on ideas is valued and there are mechanisms for choosing the most promising ideas for further development, and rewarding creativity. these elements need to work together to create what gundling (2000) has called an "ecology of innovation" that produces "a dazzling variety of new products each year" (p. 14). at one level, knowledge-creation discourse involves the design and improvement of intellectual artifacts such as theories, explanations, and proofs (bereiter 2002) . drawing from popper's theory of objective knowledge, bereiter considers ideas to be real objects similar to bicycles or telephones. we may ask how a bicycle can be improved, and we can ask the same of an idea. this aspect of the discourse is known as "design-mode" (bereiter and scardamalia 2003) , with an emphasis on explanations, casual mechanisms, and the coordination of claims and evidence. however, knowledge creation is not just a rational effort. for example, the community periodically needs discourse to identify priorities and long-term goals, decide how to mentor newcomers, and evaluate knowledge advances. as studies of scientific practice have shown, the associated discourse tends to be more argumentative (feyerabend 1975; kuhn 1970; lakatos 1970; latour 1987) . to mention just a few examples, in science, good problems may not be investigated because they are not currently considered important (latour 1987) . researchers promote their own work and that of close colleagues by alerting the community to recent findings, and may ignore important new findings that they do not find appealing (reeves 2008) . other researchers may not make their insights public, for fear of attracting criticism (e.g., madame curie's reluctance to make public the health hazards associated with radium, see quinn 1995) . in other words, belief-mode discourse also plays an important role in knowledge creation. despite individual idiosyncrasies, members of a scientific field share a goal of innovation and the advancement of knowledge. commitment to shared goals within a team is also important in a variety of other innovative contexts (gundling 2000; lencioni 2002; nonaka and takeuchi 1995) . in bereiter and scardamalia's knowledge-creation model (bereiter 2002; bereiter and scardamalia 1996; scardamalia 2002; scardamalia and bereiter 2006) , a class of students is considered a community that shares a commitment to creative work on ideas and advancement of the state of knowledge in that community. ideas are considered intellectual artifacts of the community; they reside in the community's discourse rather than in people's minds. the community needs to be able to identify gaps in its collective knowledge, map out ways to fill those gaps, design and manage inquiries, manage social processes, and evaluate progress. thus, the community's goals are emergent. students are expected to make "constructive use of authoritative sources" (scardamalia 2002) such as books, websites, and experiments, treating them as potentially useful for informing their work. they are also expected to engage in progressive problem solving, reinvesting cognitive resources to deepen their understanding of problems and taking on more difficult problems over time (bereiter and scardamalia 1993) . one of the most important roles of the teacher in this process is to facilitate the development of an innovation ecology. important progress has been made in this direction by the development of a system of principles that describe the socio-cognitive and socio-technological dynamics of knowledge creation, including collective cognitive responsibility for knowledge advancement, real ideas/authentic problems, epistemic agency, improvable ideas, rise-above, and constructive use of authoritative sources (scardamalia 2002) . these principles provide a technical vocabulary that students, teachers, and researchers can use to reflect on the extent to which there is evidence of a knowledge-creation discourse. initial studies show that elementary and secondary school students are capable of engaging in the dynamics described by these principles (niu and van aalst in press; zhang et al. 2007 zhang et al. , 2009 ). however, more work is needed to characterize the innovation ecology, such as by determining the social practices that make collaboration possible, the overall school culture, and the community's experience at knowledge creation and its long-term goals (bielaczyc 2006; truong 2008) . knowledge creation requires discourse for maintaining social relations, setting goals, deepening inquiry, and lending support to ideas that are already understood by some in the community. for example, van aalst (2006) discusses how a grade 6 student referred to the scientist francis bacon to support an explanation he had proposed earlier that had not been accepted by the community. this move was directed less at improving understanding than at improving the impact of the student's own ideas. similarly, students who wish to further a line of inquiry need the ability to argue the case for doing so. these types of moves cannot be understood by examining short-term goals such as the problem students are currently attempting to understand, but require the consideration of higher level and longer term goals such as the diffusion of new insight throughout the community and progressive problem solving (hmelo-silver 2003) . in groups that work together for short periods, there is less need for such moves. there are important theoretical differences between knowledge construction and knowledge creation, although they involve similar processes such as posing questions, formulating conjectures and explanations, summarizing progress, and proposing rise-above ideas. these processes are interpreted within different psychological perspectives. knowledge construction corresponds to cognitive psychology, in which improved understanding is regarded as the emergence of more complex cognitive structures and schemata (novak and gowin 1984) . such views have been criticized for their cartesian split between the knower and what is known, and for treating knowledge as residing in the mind. proponents of sociocultural theories posit knowing as the ability to participate in cultural practices (lave and wenger 1991; roth and tobin 2002) . for example, roth and tobin argue that "knowing physics … means to participate in talking about relevant objects and events in the ways physicists do, using acknowledged words, sentences, gestures, inscriptions, and so forth …" (p. 152). these developments have given rise to a division between learning as the acquisition of mental representations and learning as participation; sfard (1998) argues that both views are needed for a complete understanding of learning. brownell and sims propose a pragmatic and relational view of understanding implied by the ability to "act, feel, or think intelligently with respect to a situation" (1946 ( , quoted in bereiter 2002 , which bereiter uses to argue that understanding is always mediated by the object to be understood. accordingly, understanding has an "out-in-the-world" character. drawing from bereiter's analysis and work on expansive learning and knowledge-creating companies (engeström 2001; nonaka and takeuchi 1995) , paavola et al. (2004) propose a "knowledge creation metaphor" that further articulates this view. thus, understanding and knowing are mediated by the objects that a community creates and shares, and the cartesian split appears to be avoided. rather than residing inside individual minds, ideas are regarded as cultural objects (or artifacts) that mediate knowing and understanding. in summary, knowledge sharing, knowledge construction, and knowledge creation correspond to different theoretical perspectives. however, this does not mean that a community will use a single mode of discourse. for example, we would expect students to use a knowledge-sharing discourse when it meets their needs, and for there to be individual differences in epistemic beliefs and conceptions of learning that make the identification of a single discourse mode difficult. nevertheless, we can examine which discourse mode, in the balance, is most consistent with the observed discourse. the three modes of discourse can be supported by a wide variety of educational tools and activity structures, including online discussion forums (synchronous and asynchronous), mobile devices, face-to-face conversations, and lessons. this paper focuses on the use of an online discourse environment, knowledge forum. from a cognitive perspective, knowledge forum is designed to support knowledge construction through the use of scaffolds, which are sentence starters such as "my theory" that keep the writer and reader focused on cognitive processes. knowledge forum also has a variety of features that support working with ideas after they have been posted including: (a) the ability to revise notes; (b) the ability to add a note as a reference to another note; (c) the ability to reuse a note introduced in one workspace in a later workspace created for a different purpose (a workspace in knowledge forum is called a view for "point of view"); and (d) the ability to create rise-above notes, which have a special icon and are used to take the discourse to a higher conceptual plane. the ability to link notes is useful for making visually evident the connections between ideas. knowledge forum also makes it possible to objectify ideasto share them and then allow the community to work on them. the above-mentioned features then support the work of improving such objects, reviewing progress, and synthesis. the remainder of this paper reports a case study of asynchronous online discourse in knowledge forum using a coding scheme based on the distinctions between the three discourse modes. the data are drawn from a design experiment (brown 1992; collins et al. 2004) in which the researcher and teacher collaborated to achieve two goals: to achieve a fuller implementation of the knowledge-creation model than in previous iterations, and to test a new assessment strategy (van aalst et al. 2005) . the assessment task was designed to extend our previous work on portfolio notes, in which students had used concepts describing collective aspects of knowledge creation individually (lee et al. 2006 ; van aalst and chan 2007) . the new task was intended to underscore that knowledge advancements are collective achievements in a knowledge-creation community; it asked students to collaborate to review whether knowledge advances had been made on the problems they investigated and, if so, to coauthor a collaborative summary note with all who had contributed to the collective advance. while the work students did together throughout the project involved both the division of labor (cooperation) and joint activity to understand the same problems and ideas (collaboration), the word "collaborative" in the name of the task signified that students were to work together to review and create these notes (for details see van aalst et al. 2005) . the study evaluated performance on the collaborative summary notes and related that to what students were doing in knowledge forum. the unit of analysis was a group of students that worked together in the same workspaces (views) in knowledge forum; there were four such groups in the study (groups a-d). the analysis proceeded in five parts: (1) several relevant independent variables were examined to check whether the groups could be considered to be equivalent. (2) two dependent measures, knowledge quality and significance of findings, based on the collaborative summary notes, were measured to assess advances in collective knowledge made by the groups. (3) to identify mechanisms that could explain observed between-group differences in the dependent variables, the group discourses (all the notes written by each group) were coded and analyzed using a new coding scheme with 7 main codes and 33 subcodes. statistical analysis was then performed on the main code frequencies to determine which main codes provided the greatest group separation. (4) the results were used to select several main codes for qualitative analysis to further elucidate what the groups were doing differently. (5) the observed patterns in the subcode frequencies were used to examine the fit of the four group discourses to the knowledge-sharing, knowledge-construction, and knowledge-creation discourse modes. 2 the participants were two classes of secondary school students, from a grade 10 course on career preparation and inquiry (n=21) and a grade 11 course focusing on computers and their impact on "global society" (n=19). the courses were taught concurrently by the same teacher at an inner city school in western canada. approximately 40% of the students had some experience with knowledge forum in previous grades, such as in discussing "problems of the week" in mathematics. however, these experiences did not last more than one or 2 weeks and were not integrated into a pedagogical approach based on knowledge-creation principles. the teacher had 10 years of experience teaching secondary school mathematics. he had recently completed a master's degree focusing on cognitive strategy instruction and was in his third year of using knowledge forum. the researcher and teacher met several times at the beginning of the school year to plan the project, deciding that the then recent outbreaks of severe acute respiratory syndrome (sars) and avian flu in 2003 and 2004 could provide a suitable area of inquiry for secondary school students. for example, students could build on their knowledge of science to study what was known about these phenomena, critique media attention, examine the economic impact, or form a position on how governments should have responded to the outbreaks. the grade 10 course provided a promising context for integrating a focus on such questions into the curriculum, as one of its main goals was learning how to conduct research. the grade 11 course also provided a good opportunity to engage in knowledge creation, as one of its main goals was for students to learn how information and communication technology could be utilized for learning in global societies. the second main topic on the grade 11 course syllabus was "computer viruses," which was added to sars and avian flu as a third main topic for inquiry with the aim of having the students examine the nature of viruses in both biological and non-biological systems and identify patterns across them. (however, the topic only accounted for 11.5% of the coded notes.) the two classes shared a knowledge forum database and worked on the same topics. to limit the number of notes they would encounter, the students were divided into four groups. each group had students from both classes, with an equal number of students from each class; the students could choose their own groups but the teacher made some minor changes. each group had its own views on knowledge forum and the groups were not expected to interact with each other during the inquiry. in the week before the project commenced, all students responded to an icebreaker topic. the researcher then introduced both classes to knowledge-creation principles, and students were reminded of these by means of posters in their classrooms. both classes had daily access to a computer lab (70-minute periods), but students had a number of other assignments to complete. during typical periods, the teacher would spend 10 to 20 min interacting with the whole class, and the students would then work on one of their assignments. most of the students worked on knowledge forum during class a few times per week, and after school hours. the teacher discussed the students' work in knowledge forum with them from time to time, but he only read 23% of their notes and posted 7 of his own. the researcher visited the classes four times, and occasionally the teacher emailed the researcher to ask for advice on issues that arose during conversations with the students. because the teacher and students had little experience with extended and collaborative inquiry, a three-phase inquiry model was employed. phase 1 developed a focus, phase 2 was the main inquiry phase, and phase 3 involved the students evaluating what they had learned. the researcher provided extensive instructions for the three phases as outlined below. phase 1: orientation (2 weeks) the goal of the first phase was to enable the students to identify problems and select the most promising inquiry foci. research into inquiry-based learning has shown that the nature of students' own questions constrains student-led inquiry (krajcik et al. 1998; lipponen 2000; polman 2000) . the students were thus asked to read widely and post notes in their group's view, summarizing the main points and raising questions and ideas. toward the end of phase 1, they were asked to propose problems of understanding, using a research question note format stating the question, its background (relation to earlier notes), and ideas for studying the question. finally, they were asked to select a few of the most promising problems for further research, considering: (a) the extent to which a question might lend itself to inquiry worth several weeks of effort, (b) whether they had ideas or resources for researching the question, and (c) the coherence among the questions that were under consideration. the researcher explained the rationale for these processes and related it to knowledge-creation principles. phase 2: research (4 weeks) the students were asked to create a view in knowledge forum for each research question. they were then expected to work within their groups to research their problems by reading additional information on the internet and from other sources. the students were encouraged to evaluate the credibility of the sources (e.g., the world health organization website would be a more trustworthy source than writing by a person who did not declare his or her credentials), and to examine the evidence used to support the claims made in the sources. they were encouraged to extend their inquiries after they developed preliminary answers to deepen their understanding. the researcher and teacher were less involved in scaffolding the inquiry than in phase 1. phase 3: evaluation of learning (2 weeks) as knowledge advancement is an important outcome of knowledge creation, each group was asked to create a collaborative summary note for the problems on which progress had been made by the end of phase 2. the students began their review face-to-face within their own group and class, and then created coauthored notes in knowledge forum; in the best examples, the coauthors then edited the notes to gradually improve them. the note format was similar to a brief scientific research report, with the groups asked to (a) state the problem on which they were reporting, (b) explain the problem's background, with links to their work in phase 1, (c) describe what they did to investigate the problem, (d) report the main findings, and (e) explain the significance of the findings and outline opportunities for further inquiry. the instructions also indicated that a student could be coauthor of several summary notes. the notes were designed as self-assessments of group accomplishment but were not used by the teacher for formal assessment. to guide their work, the students were provided a rubric showing several dimensions of the task (writing quality, identification of collaborators, organization, findings, and implications) with levels of performance for each (van aalst et al. 2005) . the following baseline data were collected to examine the extent to which the groups could be considered equivalent in terms of their opportunities to create knowledge: prior knowledge relevant to the inquiry topics, general indexes of participation in knowledge forum, and the research questions proposed. a short test with eight questions was administered at the beginning of the project to assess existing knowledge of sars and avian flu. the questions asked students to describe their knowledge of sars, the corona virus, and what measures had been taken to control it; one asked whether a nurse should enter a hospital ward with sars patients, and another asked what students knew about avian flu. each question was scored on a 0-3 scale, ranging from "no domain knowledge evident" to "at least two relevant points." for example, in a response that received a score of "3" for knowledge of avian flu, a student stated that "it was the same thing as bird flu," which she further explained as follows: "the birds get the flu because they have to live in small spaces where bacteria grow and become more dangerous." the scores were added to create a scale with a range from 0 to 24 points. the papers were scored by the researcher; 50% of the papers were also scored independently by a research assistant resulting in an inter-rater reliability of .88 (pearson correlation). general indexes of participation in knowledge forum-notes created, percentage of notes read, and percentage of notes linked-were obtained using the analytic toolkit (atk) for knowledge forum; these kinds of measures have been used in many studies of online discourse (guzdial and turns 2000; hsi and hoadley 1997; lee et al. 2006 ; van aalst and chan 2007; zhang et al. 2007) . while high values of all three measures are not necessarily indicative of knowledge construction or knowledge creation, the measures can be informative. for example, a low percentage of notes read would suggest a low level of awareness of ideas in the database. conversely, a high percentage of linked notes could indicate attempts to synthesize and integrate contributions. these measures are correlated with both performance on self-assessment tasks and knowledge advancement, although such effects are contingent on the discourse being explanation-driven (niu and van aalst in press). the potential for knowledge advancement is also influenced by the nature of the research questions posed. do they require explanations or will descriptive information suffice? do the students have relevant knowledge that they can apply? the research question notes posed in phase 1 were thus checked to determine whether all groups posed some explanation-seeking questions and questions that related to prior learning. two dependent variables were derived from the collaborative summary notes. the knowledge quality scale measured: (a) an epistemic position ranging from knowledge as a single factual claim to a fully integrated explanation in which several concepts and/or causal mechanisms were invoked (hakkarainen et al. 2002) ; and (b) the extent and correctness of knowledge from a single finding, possibly with evidence of misconceptions, to at least three findings without evidence of misconceptions. the significance of findings scale was intended to measure the students' ability to identify the significance of what they had learned, ranging from a brief restatement of their findings to a clear explanation of the significance, limitations, and potential for further inquiry. self-assessment of the significance of learning is a metacognitive ability needed for knowledge construction and knowledge creation, especially for setting new learning goals. the descriptors for each point on these two scales are shown in table 1 . all summary notes were scored independently by the researcher and a research assistant who had completed a course on knowledge creation but was not familiar with the database. the inter-rater reliability was .85 for knowledge quality and .82 for implications of findings (pearson correlation coefficients). a coding scheme was developed for analyzing the group discourses during phase 1 and phase 2. the goal of the analysis was to identify mechanisms that could explain betweengroup differences in the dependent variables. the scheme was intended to be general enough for use in analyzing discourse from a variety of perspectives within the computersupported collaborative learning field, particularly knowledge sharing, knowledge construction, and knowledge creation. it includes seven main codes: community, ideas, questions, information, links, agency, and meta-discourse. the community code describes the extent to which the social interactions within a group suggest a "sense of community," in which "people feel they will be treated sympathetically by their fellows, seems to be a first necessary step for collaborative learning" (wegeriff 1998, as quoted in kirschner and kreijns 2005, p. 176 ). indicators of a sense of community include commitment to shared goals, appreciation for the work of group members, identification with the group, and ways of getting things done that are specific to the group (wenger 1998) . discourse that involves risk-taking requires a stronger sense of community than other types of discourse (e.g., improving ideas versus only sharing them). although the knowledge-creation model refers to communities, the discussion in its literature has been limited to the socio-cognitive features of those communities. the next five main codes-ideas, questions, information, links, and agency-are based on research into a wide variety of cognitively oriented inquiry approaches (chan 2001; hakkarainen 2003; hakkarainen et al. 2002; hmelo-silver 2004; kolodner et al. 2003; linn et al. 2003) . this body of work has shown that a focus on explanation is more likely to lead to knowledge advancement than answering fact-seeking questions (hakkarainen 2003) . the idea code captures the ways in which students contribute to and work on ideas (e.g., opinions, conjectures, and explanations), with its focus on the nature of those ideas. in contrast, the information code focuses on the extent to which students interpret or evaluate the information they introduce. the agency code is intended to describe the ways in which students selfregulate their inquiries; the subcodes emphasize planning and reflection relating to logistics and the epistemic features of their inquiries. in terms of these codes, we would expect informationsharing discourse to be characterized by fact-seeking questions and limited evidence of ideation, interpretation of information, synthesis, and planning and reflection. in contrast, both knowledge construction and knowledge creation would be characterized by stronger evidence in these areas, with minor differences between the two modes of discourse. for example, although rise-above should occur in knowledge construction, it should occur more often in knowledge creation, which takes place over a longer period and has greater need for synthesis. the final main code, meta-discourse, describes a level of discourse beyond maintaining social relations and building understanding, and relates to the existence of long-range goals in a knowledge-creation community. scardamalia and bereiter (2006) suggest that this feature is lacking in most online discussions. examples of meta-discourse would be reviews of the state of knowledge in the community, work aimed at helping new insights diffuse through the community, making arguments for a new phase of inquiry, and establishing more difficult goals over time. although evidence of meta-discourse may not be strong in an inquiry of 8 weeks, there should be some examples. to capture the different ways the seven codes could be exemplified, 33 subcodes were identified and their relevance to each of the discourse modes estimated (see table 4 in the "results" section). for these estimates, a three-point rating scale was used (low, medium, high). for example, the subcode fact (under ideas) was rated high for knowledge sharing and low for both knowledge construction and knowledge creation. in this example, knowledge construction and knowledge creation are called degenerate to indicate that the scale for this code does not differentiate between them. major review (under meta-discourse) was rated low for knowledge sharing, medium for knowledge construction, and high for knowledge creation on the assumption that knowledge creation is generally more complex and requires more time than knowledge construction, so the need for major review is greater. all ratings were completed independently by the researcher and an independent second rater, leading to an inter-rater reliability of .82 (cohen kappa). the computer notes were entered into atlas-ti® qualitative data analysis (qda) software for coding; 399 notes were coded (approximately 60,000 words). each view in knowledge forum was entered separately, beginning with the first view of group a and ending with the last view of group d. most of the development of the coding scheme was done using the data from groups a and b. the researcher started with a small set of codes based on knowledgecreation principles and prior research into asynchronous discourse, and gradually expanded the set. he started by focusing on the text, and applied each code that seemed relevant to a given text segment; the amount of text varied from a sentence to a few notes depending on the code (hmelo-silver 2003). the process was then repeated focusing on the codes and working through the corpus checking for potential examples for small groups of codes. the researcher began by coding data from groups a and b, and reflexively improved both the code definitions and coding procedures. it soon became clear that coding was needed for both the nature of the idea (e.g., conjecture or explanation) and the extent to which the students processed new information. after three rounds of improving the code definitions and procedures in groups a and b, the codes were organized into main codes and subcodes and the remaining data were coded. as employing a second coder was not possible, to further ensure the accuracy of the coding the researcher returned to it after an absence of approximately 3 months. the qda software was then used to check the consistency of subcode allocations, with 12% of the quotes needing to be recoded. most changes were between subcodes of the same main code (e.g., switching from "opinion" to "conjecture"). the coding results were analyzed in three ways. first, a frequency analysis was conducted to examine the extent to which each main code could be used to separate the four groups. the goal of this analysis was to identify potential mechanisms that could explain betweengroup differences for knowledge quality and implications of findings. next, several of the main codes were selected for qualitative analysis to further elucidate what students were doing in knowledge forum. main codes were selected for this analysis based on the amount of group separation. finally, the alignment of the subcodes with the three discourse modes allowed the mapping of the four group discourses onto those modes. the goal of the first analysis was to determine whether the four groups could be considered equivalent in subsequent analyses. table 2 shows the results for the knowledge pretest and atk indexes. the pretest results show that prior content knowledge was not extensive and varied very little between the four groups; the group means varied from 45.4% (group c) to 50.0% (group b). the majority of students (55%) stated at least two substantive points about sars, but 75% stated they knew nothing about the corona virus, and 60% stated they knew nothing about avian flu. between-group differences were also relatively minor for the atk indexes, the most noticeable being that the students in groups c and d read fewer notes. overall, the amounts of note writing and reading were consistent with those in other studies of online discourse (guzdial and turns 2000; hsi and hoadley 1997) . in contrast, the amount of linking (40.9% to 50.4%) was less than in other studies using knowledge forum, in which it reached 80% (lee et al. 2006; yoon 2008) . each group posted approximately 10 research question notes, although group c required 8 days longer than the others to reach this point. each group's output included some explanation-seeking questions, such as "why is it children are less likely to develop sars?" there were, however, important differences in the extent to which the questions allowed the students to build on prior knowledge. for example, while discussing the question about sars and children, the students used their knowledge of viruses and infection, but in discussing "is killing chickens the only way to end avian flu" they resorted to exchanging opinions. in sum, these data suggest that the four groups were similar in terms of prior knowledge about the main inquiry topics, the extent to which they used knowledge forum, and their ability to formulate research questions. however, group c had fallen behind the other groups by the time it had generated its research questions, and the research questions varied in their potential for knowledge creation. the goal of the second analysis was to evaluate the advances in collective knowledge reported by those students who collaborated on summary notes. the students collectively submitted 32 summary notes; 81.0% of grade 10 students and 84.2% of grade 11 students were coauthors of at least one note. all of the summary notes were assessed for knowledge quality and implications of findings with the scales shown in table 1 . table 3 shows the group means and standard errors for knowledge quality and significance of findings for the 32 summary notes. some students did not realize that a group was required to write only one note on a given research question, resulting in duplicate notes for some questions; in such cases, only the best note from the group was considered in the calculation of group means. group a had a higher mean score than the other groups for knowledge quality (effect sizes ≥ 0.7, cohen's d); for most groups. the knowledge gained was factual and did not reach the level needed for a 3 or 4 on the scale. group c had the lowest mean knowledge quality score; its small number of notes is understandable because it needed more time to articulate its focus. the goal of the third analysis was to identify possible mechanisms for the between-group differences in the dependent variables by coding the group discourses leading up to the creation of the summary notes. the code and subcode frequencies are shown in table 4 . the total frequencies for all subcodes associated with a main code are shown in the first row of each section. before examining intergroup variation, it will be useful to consider the total frequencies over all groups (last column). in descending order of total frequency, the following patterns can be observed. first, although there were many linkages (f=206), there were few examples (8) in which features of knowledge forum such as adding a note as a reference to another note were used; the majority of links were to web pages (106), although some groups did link their ideas verbally to earlier contributions in knowledge forum (66). this finding suggests that the level of competence with features of knowledge forum designed to support linking ideas was low, and may explain the lower than expected atk index for linking (table 2) . second, although there were many instances of working with ideas (171) and information (124), the subcodes suggest that information sharing was a significant aspect of all group discourses. third, there were few instances of two codes: questions and meta-discourse (both 65). group a had substantially more code instances than the other groups (329, compared with 181, 165, and 165) reflecting that it invested more effort into the processes measured by the coding scheme, which may explain the better performance on the summary notes to some extent. however, a more interesting question is what group a did differently, controlling for the difference in overall effort. thus, a two-way analysis of the main code frequencies (code × group) was conducted. the results are shown in table 5 ; cohen's guidelines regarding effect sizes were used to arrange the codes in three groups from large to small effect size. accordingly, community and questions were very effective in separating the groups-especially group a from the other groups. the next two codes (ideas, information) provided statistically significant but more moderate separation. the last three codes (linking, agency, meta-discourse) provided limited or non-significant group separation. the relatively low frequencies for these codes indicate that these aspects of the discourse were generally not well developed. to gain additional understanding of the nature of the group discourses, three sets of main codes were selected for qualitative analysis based on the group separations shown in table 5 : community (large), ideas/information (moderate), and agency/argument (non-significant). in the knowledge-creation model students contribute ideas, on which the community works and which thus become its "intellectual artifacts" (bereiter 2002) . at the same time, students continue to own their ideas, and whether their ideas are appreciated and taken up by the community is important to the formation of students' identities as community members (wenger 1998) . the coding revealed between-group differences relating to aspects of this issue. for example, group a encouraged its members more often than the other groups (table 4): i think your ideas for groups are good … it would mean that we could get a start on all the topics right away. good job of actually getting things going! i really like [s's] idea of setting ourselves little mini-deadlines so that everybody will stay on task and finish the job more efficiently. there were also examples in which additional views were sought from students who had not yet contributed ideas. some students also felt a responsibility toward the group and apologized for failing to contribute to the discourse: sorry i've been away at a tournament for quite a while, so i'm just trying to get caught up. i don't know how much work you've all got done already, but the groups etc. sound pretty good ... i'll get onto researching as soon as i'm sure what's going on. another way in which group a promoted a sense that students belonged to a community was by instituting a democratic voting scheme for prioritizing research questions (11 of 60 code instances). while taking a vote can be a superficial process, it was accompanied in this group by considerable ideation, information processing, and linking. the voting process was also present in group c but was absent from groups b and d. group b appeared to harbor some tension between the grade 10 and 11 students arising from miscommunication. one student in the grade 11 class wrote: as of now, we have less than 1 week left and because your class have not been very active in this final phase, we've decided to go with these two questions above because we've a student from the grade 10 class responded as follows: yeah, alright. if the rest of our group wants to do it then i guess that's what's being done since "we have not been very active." i thought we were only supposed to research our own questions first. are those the only questions that we are doing then? we are sorry that you are not satisfied with the level of our commitment on kf. we weren't aware that we needed to pick from your questions as well as ours. sorry for the inconvenience. in summary, group a had a shared commitment to the task, a sense of belonging to the group, and an appreciation for all group members' contributions, all of which are indicators of communities (wenger 1998) . these social processes were also present to some extent in group c, but they may have had less effect. group a's discourse had most of the kinds of idea units needed for knowledge construction, particularly concepts and explanations; only 22.2% were coded as facts or opinions. however, only 2.4% of its idea units were classified as rise-above, suggesting that the discourse was not yet a well-developed example of knowledge creation. all of the other groups' discourses were more fact oriented, with percentages of idea units coded as facts or opinions ranging from 39.1% (group b) to 62.5% (group d). nevertheless, in group b, there were some examples of concepts and explanations, and group c had 10 explanations and a few rise-above units. this mix of conceptual and factual contributions is the main reason for the idea code providing only moderate group separation ( table 5 ). the information code revealed a tendency toward knowledge sharing in all groups: the many instances where information was presented without interpretation or evaluation (group c, 66.7%; group d, 61.2%; group a, 50.0%, group b, 35.4%). this prevalence of information sharing may be related to an epistemological understanding of inquiry as asking questions, finding answers, and reporting them, with information assumed to be self-explanatory (a realist position). instead of describing information, students need to be developing explanations and using information to support them. the difference is illustrated below: i found this information on: [web link]. "thailand, the world's seventh largest poultry producer, will suffer only 'modest losses' to its economy due to the h5n1 strain of the avian virus, it says. a 'complete decimation' of thailand's poultry industry would this note describes information but does not interpret or evaluate it. in contrast, in the following note, the student is providing an explanation and uses the web link at the end of the first paragraph as a reference for further information; in the second paragraph, the description of the second web link is integrated into the explanation. first off, let's look at the very definition of 'quarantine.' we're not talking maybe putting up some red tape around affected farms here, we're talking the slaughter and elimination of literally millions of chickens in the us, canada, and 10 asian nations including vietnam and china. the most recent outbreak, in hong kong, resulted in the slaughter of about 40,000 poultry; the birds were killed, then bagged for dumping in landfills. this outbreak alone cost the poultry industry the equivalent of us $10.26 million. [web link] the short-term costs incurred may seem rather drastic, but if the flu can be contained now, by eliminating all possibly infected birds, it will cost far less than trying to contain it later on if there are more outbreaks. i found the following website had some really valuable information concerning avian flu outbreaks in the past-there have been 21 large-scale epidemics all over the world, ranging from australia to pakistan, and this site talks about the economic impact of each-for example, "the 1983 pennsylvania (usa) outbreak took 2 years to control. some 17 million birds were destroyed at a direct cost of us$62 million. indirect costs have been estimated at more than us$250 million." [web link] neither agency nor meta-discourse provided statistically significant group separation. the agency code looked separately at planning and reflection relating to epistemic aspects of the inquiry and the completion of the project. there were not many instances of agency relating to the inquiry (33). reflection on learning is an important aspect of knowledge construction and knowledge creation, and the lack of reflection in the discourses provides separate evidence that most groups treated information as unproblematic. another important finding from the agency code is the many instances of project planning from group c, suggesting that it had considerable difficulty in self-directing its inquiry. project planning is an important aspect of knowledge-creation discourse, but it should not dominate the cognitive features. the meta-discourse code is conspicuous because it occurred infrequently, but there were attempts by all groups. for example, a group b student attempted to advance the inquiry to a new stage by suggesting a new question: … i guess the question now is how can we make the chickens less likely to develop serious symptoms, and to become more like the wild poultry. and maybe an effective method of keeping the chickens from getting sick and to stop the spread of the avian flu is by doing something to the wild fowl to make them unable to carry the virus. it raises some interesting questions that can probably be analyzed further! perhaps the suggestion came too late, but it was not taken up by the group. relating the group discourses to the discourse modes the goal of the fourth analysis was to map the group discourses onto the three modes of discourse. first, the subcode frequencies were classified as small (0 to 5 instances), moderate (6 to 10), and large (greater than 10). the results were then compared to the relevance ratings of the subcodes (table 4) to predict the discourse modes. for example, group a had 20 explanation-seeking questions (large), which corresponds to knowledgecreation discourse. group c had two instances of fact-seeking questions (small), which is consistent with knowledge-construction and knowledge-creation discourses (a degenerate prediction). group b had nine instances of opinion (moderate), which did not correctly predict any discourse mode. to sample the main codes evenly, the two subcodes that predicted the most complex discourse mode were selected for creating profiles. figure 1 shows the number of correct predictions of each discourse mode for the four groups. perfect agreement with a discourse mode would include 14 predictions of that mode; however, because there are many degenerate predictions, these would be accompanied by some predictions of the other modes. the profile of group a is most consistent with knowledge creation: it includes nine predictions of that mode, of which only one is degenerate. it also includes two predictions of knowledge sharing. the overall fit of the predictions to the discourse mode is best for this group. the profile of group d is almost the reverse: it has nine predictions of knowledge sharing (seven nondegenerate) but includes more predictions of the other modes than the group a profile. the profiles of groups b and c are more difficult to interpret because they include nearly equal numbers of predictions of all three of the discourse types. this could be caused by a variety of factors including the existence of smaller units of social organization that approach the discourse differently and contextual dependencies that cause the discourse on one problem to be qualitatively different from that on another. this possibility was explored for group b using inquiry thread analysis. (group b was chosen for this because we already know that group c fell behind in phase 1 and had less time for its inquiry in phase 2.) fig. 1 number of correct predictions of discourse types from sub-code frequencies. two sub-codes from each main code were used, leading to at most 14 correct predictions per group. however, because some subcodes did not uniquely predict a single discourse type and some did not correctly predict any type the number of predictions per group is generally different from 14 an inquiry thread is a temporally ordered sequence of notes on the same problem or topic. the notes need not be hyperlinked to be part of the same inquiry thread, and it also is possible that notes that are hyperlinked are not part of the same inquiry thread (for details on the method see zhang et al. 2007 ). nine inquiry threads were identified, of which six were active for more than a month. the longest thread (13 notes, 7 authors, and 12 readers) was active throughout the entire project and examined scientific mechanisms by which avian flu infection occurs; it included relatively many of the instances of concept, explanation, and deepening inquiry identified in the coding. a similar inquiry thread, but of shorter duration, began in the second half of phase 2, and focused on a causal explanation of why children may be less susceptible to sars (10 notes, 6 authors, 10 readers). threads that were more descriptive were general explorations of sars and avian flu in the first 3 weeks of the project, and somewhat argumentative discussions of how the media had handled the sars outbreak, the disposal of chickens infected by avian flu, and the prevention of avian flu. although deeper analysis would be useful, these results generally support the context-dependence hypothesis. group b engaged in more explanation-oriented discourse when relevant concepts were available, and less when exploring sars and avian flu in general and when concepts were not available. this paper seeks a clearer articulation of the nature of computer-mediated discourse needed for bereiter and scardmalia's knowledge-creation model. its main contributions are the conceptual framework for distinguishing between knowledge-sharing, knowledgeconstruction, and knowledge-creation discourses, an accompanying coding scheme, and the application of both to an evaluation of discourse in knowledge forum. this section reviews what has been accomplished, suggests several strategies for improving the alignment of online discourse to the knowledge-creation model, and outlines further development of the coding scheme. conceptual framework i have argued for a conceptual framework that contrasts three modes of discourse, which can be associated with different theoretical perspectives (transmission/naïve realism, cognitive psychology, and interactive learning mediated by shared objects). knowledge sharing is included because it remains a common discourse mode and is useful in some situations; knowledge construction is included because it is what knowledge creation needs to be distinguished from most. knowledge creation is not a new example of constructivism (in the cognitive paradigm), but an example that reifies a new theory of mind that does not depend on a notion of the mind as a container (bereiter 2002) . however, due to the incommensurability of the underlying theories, i do not regard the discourse modes as stages in the development of a community's discourse, as gunawardena et al. (1997) have suggested for knowledge sharing and knowledge construction. the framework extends the distinction between learning and knowledge creation (bereiter and scardamalia 1996) by differentiating between learning by knowledge sharing and learning by knowledge construction. this differentiation makes it possible to bring into focus both differences and similarities between knowledge construction and knowledge creation. the treatment of the knowledge-creation model in the framework marks a departure from the extant literature. bereiter and scardamalia focus on ideas as improvable objects and the socio-cognitive and epistemic dynamics of improving them, as though that could happen without regard for the social context (bereiter and scardamalia 2003; scardamalia 2002; scardamalia and bereiter 2006 ). in the framework described here, the recommended use of design-mode discourse over belief-mode discourse, a distinction valid for the epistemic work of improving ideas but not for the additional work needed to prioritize goals, ensures that new ideas diffuse throughout the community and possible advances in knowledge are evaluated. one of the most apparent differences between examples of knowledge construction and knowledge creation is the discourse by which this additional work is achieved. paying more attention to the social context in which knowledge creation occurs is not only important for an adequate portrayal of knowledge creation for students, but it also reflects the conditions in classrooms. recently, interest in these aspects of knowledge creation has been mounting. bielaczyc (2006) develops a social infrastructure framework that emphasizes culture and practices, and hakkarainen (2009) introduces the notion of "knowledge practices" to combine epistemic and social practice elements of knowledge creation. it is hoped that the framework described here will stimulate further research to clarify the relationship between explanationoriented discourse, argumentation, and the advancement of collective knowledge. group a's discourse was identified as knowledge creation, providing the strongest evidence of a sense of community, explanation-seeking inquiry, interpreting and evaluating information, knowledge advancement, and insight into these processes. from the analysis of frequencies in table 5 , we know that the leading factor differentiating group a from the others was its sense of community, but it is likely that all of the observed effects are necessary. the relatively clear identification of knowledge-creation discourse and better knowledge advances are encouraging because they suggest that knowledge creation is feasible for secondary school settings. however, there is a need for caution because there was little evidence of rising above, meta-discourse, and use of the advanced features of knowledge forum, and there was still too much evidence of knowledge sharing. the relatively clear identification of group d's discourse as knowledge sharing is more disconcerting in a classroom generally oriented toward student centered and constructivist learning. nevertheless, my work with many teachers in the last decade suggests it is a common occurrence. perhaps in this case context dependence mattered less, and the results may point to deeply held beliefs such as quick learning (schommer 1990 ) and achievement motivation. indeed, group d's results on the summary notes were second to group a's. group c was also problematic. like group a, it expended much effort on maintaining its sense of community, but was relatively inactive in posing questions and working with information (see table 4 ), and created fewer summary notes. we also know the group had less time for its inquiry than the others because coming up with research questions took longer than planned. although motivation could have been a factor, this was probably an example of an inadequate level of guidance (hmelo-silver et al. 2007 ). nevertheless, cases like this, in which students are unable to manage inquiries very well, are also common in project-based science (krajcik et al. 1998; polman 2000) . cultivating an innovation ecology as the study described here was a case study, its claims pertain to the observed groups only. further studies investigating the phenomena using different methods and in different settings would lead to a fuller understanding of the generality of the claims (yin 2003) . nevertheless, some cautious recommendations for encouraging an innovation ecology can be made based on the findings. this subsection examines several conditions that constrain or enable knowledge creation-the nature of the task, the sense of community, idea-centered discourse, the use of technology, and meta-discourse-and discusses how they can be optimized. a common reason for the failure of efforts at knowledge creation in school is their lack of authentic problems (bereiter and scardamalia 2003) . asking students to investigate what interested them about sars and avian flu held some attraction from this point of view because there was a lack of knowledge, the topics were discussed in society, and students could pursue their own questions. these are considered good things by proponents of socio-scientific issues in teaching (walker and zeidler 2007 ). yet, the students' interests frequently took them into areas where they could not build on their initial knowledge. it is possible that the extensive reliance on knowledge sharing resulted in part from a general lack of knowledge that led the students to explore opinions and "chart the territory," and from a lack of concepts and perspectives that could be used to question sources. some students seemed to suggest that what they were doing lacked authenticity as schoolwork because the depth of knowledge was inferior to what was normally expected of them: it is important to understand that there is not one answer to this question. i am limited in my understanding because i get my information from news sources that may be biased. i can also not understand how sars spreads scientifically because i am not a scientist or a doctor. i am like the rest of the public that gets information from news sources. (group d summary note) these considerations have important implications for developing an innovation ecology. in school, it is important that students develop academic knowledge: for example, concepts, explanations, explanatory principles, inquiry methods, and meta-conceptual knowledge. social norms are needed in the classroom to keep these things in focus as students engage in knowledge creation. in other words, in getting a knowledge-creation experience started, the curriculum, students' prior academic knowledge, and their interests should be explored together to forge a closer connection to the curriculum and assess the potential of ideas for inquiry. in the study, students did explore the potential of their ideas but did so independently in their own groups, and the social norms were not developed. a closer connection to the curriculum would also be needed for scaling up knowledge creation in schools. many other researchers link inquiry to the attainment of national educational standards (krajcik et al. 2008) but do this in ways that undermine key goals of knowledge creation (e.g., epistemic agency, adding to the intellectual heritage of a community). further research into how knowledge-creation experiences can be integrated into the curriculum is much needed. the study identified social processes that constitute a sense of community (such as encouragement, giving credit, drawing in participants, and apologizing) as the distinction between group discourses. these processes have been studied extensively in education and social psychology (e.g. johnson and johnson 1989; slavin 1990 ) but have received little attention in computer-supported collaborative learning research, where social interactions tend to be studied in the context of problem solving from the perspectives of cognitive theory (de laat et al. 2007; roschelle 1992; suthers et al. 2007 ), intersubjective meaningmaking (koschmann and lebaron 2002; stahl 2006) , or are dismissed as off-task behavior (meier et al. 2007 ). the social interactions identified in the study were neither part of problem-solving sessions nor irrelevant, but were directed at maintaining and improving how the groups worked. cultivating the social dynamics identified in the study would be important for creating a safe environment for knowledge creation, and thus an important aspect of an innovation ecology. the considerable differences in how the groups functioned socially provide reasons for reexamining the social organization of the class for its knowledge creation. in the assessment literature, the inequities of group work have been noted (webb et al. 1998 ). if students work together for several months, inequities arising from individual differences in motivation, effort, and ability could lead to substantial disadvantages for some students. using flexible and opportunistic groups, in which students join a group for a short time to accomplish specific goals, would make students less susceptible to the potential inequities and would help them learn and work with many different students. recently, zhang et al. (2009) used social network analysis to compare three social configurations-fixed groups, interacting groups, and opportunistic groups-and found that opportunistic groups best diffused new knowledge. in the present study, the decision to form fixed groups was intended to limit the number of notes students would need to deal with, but this problem could also be addressed by encouraging more reflective discourse with greater attention paid to synthesis and rise-above ( van aalst 2006) . one thing that this cognitively demanding work does is slow down the growth of the database. the importance of idea-centered discourse is so well established in the literature that it does not require further amplification (e.g. bereiter 2002; hakkarainen 2003; scardamalia 2002) . nevertheless, in the study students frequently introduced information without generating ideas or questions. if this kind of discourse is widespread and consistent within a community, it may suggest naïve epistemic beliefs (schommer 1990 ). deeper reflection on what makes a valuable contribution to knowledge forum may lead students to interpret and evaluate information, and to elaborate by providing examples and counterexamples. students could do this even when exploring a new content area, provided that they have concepts that can provide a lens for interpretation. social norms about the quality of knowledge to be created could also help students to focus on developing explanations. to facilitate developing a set of coherent explanations, it seems important to cultivate rise-above as a prominent dynamic of the discourse. although it can be used late in an inquiry to articulate what has been learned, it may also be useful for scaffolding the discourse, suggesting how students can contribute next. in studies of portfolio notes in knowledge forum, the reflections needed to prepare the portfolio notes have also had such a scaffolding function (lee et al. 2006 ; van aalst and chan 2007) . rise-above can produce incomplete explanatory frameworks, which may lead to predictions and new inquiry goals. thus, rise-above needs to be a social norm that is in focus throughout the inquiry. this would allow an approach whereby students contribute new information and ideas and regularly look for opportunities to review progress and identify more new ideas and lines of inquiry. the approach taken to the use of technology was typical of what i have seen in work with many teachers. the basic features (creating and responding to notes) were demonstrated at the beginning of the project, and instructions for creating links between notes were given with the instructions for phase 3. that was not sufficient. we made an implicit assumption that learning about knowledge forum would not occupy very much instructional time-it was just "a tool." to the contrary, the use of knowledge forum needs to mediate the social and cognitive work of creation (cole and engeström 1993; hakkarainen 2009 ); students need to learn to coordinate use of its features with use of the concepts of the knowledge-creation model. for example, rise-above involves important social skills because the ideas contributed by different students are combined and the authors may disagree with how their ideas are used. technical skills such as the ability to create a private view (accessible to a subgroup) and annotations can be helpful for temporarily storing copies of notes that are being considered, as well as draft ideas and notes before the final result is made public. thus, rise-above is a social practice that can exist only because the technology makes it possible, and it is a practice that needs to be developed. it is also worth considering whether knowledge forum provides the best medium for creating knowledge. at least one group had difficulties using it to reach a consensus about priorities and goal setting in phase 1, and talking face-to-face may have been more effective. while asynchronous writing can support reflective thought, reading and writing notes is time consuming and should only be used when it provides advantages over more social ways of interacting. some researchers and teachers have developed practices such as poster presentations, gallery walks, and whole-class talk, whereby students report and discuss the ideas, questions, and challenges they are considering within their groups (kolodner et al. 2003; zhang et al. 2009 ). according to kolodner and colleagues, such practices become routines and rituals within the community, and students come to see why they are necessary. zhang et al. (2009) found that students requested "kb talks" to discuss the database with each other. these practices provide opportunities for students to be aware of their progress, suggest ways of addressing problems, and identify learning needs not otherwise recognized. in their absence, work on knowledge forum is disconnected from the educational culture of the class and feels like a special project. however, research into the role of the social infrastructure that supports knowledge creation is still in an early phase (bielaczyc 2006; truong 2008; zhang et al. 2009 ). the evidence of meta-discourse in the groups was limited, partly due to the short duration of the project but also because it was an intervention. had we not intended to study their work, the students would have been completing individual inquiries. when knowledge creation pervades the general approach students take to their schoolwork, long-range effects may become more evident. for example, students may discuss how to improve on previous efforts or evaluate the evolution of ideas over a substantial period such as an entire school year. before this can happen, their inquiries need to be connected more deeply to the curriculum, and the use of technology woven into a set of coherent practices aimed at knowledge creation. the study provides a coding scheme for analyzing asynchronous discourse, extending earlier schemes that emphasized the socio-cognitive aspects of online discourse and drawing from previous work on rating scales on levels of questioning and explanations (chan 2001; hakkarainen et al. 2002) . the main codes can serve as a general framework for coding to facilitate comparisons across studies of computer-supported collaborative learning. however, if the subcodes are considered as indicators of the phenomena intended with the main codes, further research is needed to improve and expand the current set. for example, it would be useful to add further subcodes for questioning, agency, and meta-discourse to provide more balance among the main codes and improve the usability of the coding scheme for a wider range of research questions. the currently limited set of subcodes for some main codes reflects the overall limited evidence of the underlying phenomena (i.e., additional subcodes could have been induced from the data had the evidence of these main codes been stronger). while it is not my intention to fully map the codes onto the knowledge-creation principles, the coding scheme may provide a complementary framework useful for elaborating several principles. for example, information is intended to describe different levels of information processing, ranging from uncritical sharing to evaluation of a collection of sources in the context of the problems under investigation. if one correlates frequencies for this code with those relating to working with ideas, a fuller understanding of the principle of the constructive use of authoritative sources could be achieved. this paper has elaborated distinctions between three modes of online discourse-knowledge sharing, knowledge construction, and knowledge creation-which correspond to theories of transmission/naïve realism, cognitive psychology, and interactive learning using shared mediating objects. the framework was applied to a case study of four groups of students who used knowledge forum as part of an attempt to create knowledge about sars and avian flu. through the use of a new coding scheme, one group discourse was indentified as a threshold case of knowledge creation, one as knowledge sharing, and two as hybrids of all three modes. the study revealed the importance of the social interactions needed for a sense of community as one of the leading factors separating the group discourses. education and mind in the knowledge age the psychology of written composition knowing, learning and instruction: essays in honour of robert glaser surpassing ourselves: an inquiry into the nature and implications of expertise the handbook of education and human development: new models of learning, teaching and schooling learning to work creatively with knowledge designing social infrastructure: critical issues in creating learningenvironments with technology student approaches to learning and studying how people learn: brain, mind, experience and school design experiments: theoretical and methodological challenges for creating complex interventions in classroom settings situated cognition and the culture of learning peer collaboration and discourse patterns in processing incompatible information distributed cognitions: psychological and educational considerations design research: theoretical and methodological issues investigating patterns of interaction in networked learning and computer-supported collaborative learning: a role for social network analysis reconceptualizing change in the cognitive construction of knowledge addressing the challenges of inquiry-based learning through technology and curriculum design expansive learning at work: toward activity-theoretical reconceptualization against method: outline of an anarchistic theory of knowledge: humanities defending constructivismin science education making the invisible visible: a teaching/learning environment that builds on a new view of the physics learner interaction analysis of a global on-line debate and the development of a constructivist interaction analysis model for computer conferencing the 3m way to innovation: balancing people and profit effective discussion through a computer-mediated anchored forum emergence of progressive-inquiry culture in computer-supported collaborative learning a knowledge-practice perspective on technology-mediated learning epistemology of inquiry and computer-supported collaborative learning analyzing collaborative knowledge construction: multiple methods for integrated understanding problem-based learning: what and how do students learn? facilitating collaborative knowledge building scaffolding and achievement in problem-based and inquiry learning: a response to kirschner productive discussion in science: gender equity through electronic discourse a cognitive approach to the teaching of physics cognition in the wild cooperation and competition: theory and research enhancing sociability of computer-supported collaborative learning environments problem-based learning meets case-based reasoning in the middle-school science classroom: putting learning by design into practice learner articulation as interactional achievement: studying the conversation of gesture inquiry in projectbased science classrooms: initial attempts by middle school students learning-goals-driven design model: developing curriculum materials that align with national standards and incorporate project-based pedagogy the structure of scientific revolutions falsification and the methodology of scientific research programmes science in action: how to follow scientists and engineers through society situated learning: legitimate peripheral participation students assessing their own collaborative knowlegde building the five dysfunctions of a team wise design for knowledge integration towards knowledge building: from facts to explanations in primary students' computer mediated discourse a rating scheme for assessing the quality of computersupported collaboration processes understanding learning in a postmodern world: reconsidering the perry scheme of intellectual and ethical development participation in knowledge-building discourse: an analysis of online discussions in mainstream and honours social studies courses the knowledge creating company: how japanese companies create the dynamics of innovation learning how to learn models of innovative knowledge communities and three metaphors of learning seeing what we build together: distributed multimedia learning environments for transformative communities designing project-based science: connecting learners through guided inquiry marie curie, a life in science a force of nature: the frontier nature of ernest rutherford learning by collaborating: convergent conceptual change college physics teaching: from boundary work to border crossing and community building collective cognitive responsibility for the advancement of knowledge knowledge building environments: extending the limits of the possible in education and knowledge work knowledge building: theory, pedagogy, and technology effects of beliefs about the nature of knowledge on comprehension epistemology and the teaching of science. ottawa, canada: science council of canada on two metaphors for learning and the dangers of choosing one cooperative learning: theory, research and practice group cognition: computer support for building collaborative knowledge social constructivist perspectives on teaching and learning beyond threaded discussion: representational guidance in asynchronous collaborative learning environments exploring social practices that support knowledge building in a primary school student-directed assessment of knowledge building using electronic portfolios exploring collective aspects of knowledge building through assessment radical constructivism a way of knowing and learning promoting discourse about socioscientific issues through scaffolded inquiry equity issues in collaborative group assessment: group composition and performance communities of practice, meaning, and identity case study research: design and methods an evolutionary approach to harnessing complex systems thinking in the science and technology classroom socio-cognitive dynamics of knowledge building in the work of 9-and 10-year-olds designs for collective cogniitve responsibility in knowledge-building communities acknowledgments the classroom work of this study was completed while i was at simon fraser university; it was supported by a discovery parks grant from that university and the social sciences and humanities research council of canada. i thank the students and teacher for their work. preliminary findings from the summary note task were reported at the 2005 cscl conference held in taipei, taiwan. i thank cindy hmelo-silver, jianwei zhang, carol chan, and three anonymous reviewers for their comments on a draft of the paper.open access this article is distributed under the terms of the creative commons attribution noncommercial license which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.