Evidence Based Library and Information Practice Evidence Based Library and Information Practice 2011, 6.4 149 Evidence Based Library and Information Practice Evidence Summary Study of Limited Value in Exploring Irish Hospital Clinicians’ Information Behaviour and Attitudes Towards the Clinical Informationist A Review of: Flynn, M. G., & McGuinness, C. (2011). Hospital clinicians’ information behaviour and attitudes towards the ‘Clinical Informationist’: An Irish survey. Health Information & Libraries Journal, 28(1), 23-32. doi: 10.1111/j.1471-1842.2010.00917.x Reviewed by: Maria C. Melssen Head of Learning and Information Services Florida International University Miami, Florida, United States of America Email: mmelssen@fiu.edu Received: 15 July 2011 Accepted: 05 Oct. 2011 2011 Melssen. This is an Open Access article distributed under the terms of the Creative Commons-Attribution-Noncommercial-Share Alike License 2.5 Canada (http://creativecommons.org/licenses/by-nc-sa/2.5/ca/ ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly attributed, not used for commercial purposes, and, if transformed, the resulting work is redistributed under the same or similar license to this one. Abstract Objective – To determine the self-perceived information needs, information-seeking strategies, and skill levels of hospital clinicians, in addition to their opinions on the inclusion of a clinical informationist (CI) in their clinical teams. Design – Questionnaire survey. Setting – Two public, medium-sized teaching hospitals (200-250 beds) in Ireland. Subjects – Hospital clinicians. Methods – A 33-item questionnaire (pilot tested on nurses) was deployed using SurveyMonkey. Participants were recruited using snowball sampling and were emailed the link to the questionnaire in June, July, and August of 2008. Hard copies of the questionnaire were also distributed at one of the hospitals; the librarian at this hospital manually entered the responses into SurveyMonkey. Survey results were analyzed using descriptive statistics. Main Results – Of 230 eligible hospital clinicians (HCs), 22 participated in the survey. Of the HCs surveyed, 90% spend “over 21 hours per week engaged in patient care” (p. 26). During this time the HCs generate an average of 1-5 clinical questions each. The HCs surveyed frequently required information on the latest research on a specific topic, Evidence Based Library and Information Practice 2011, 6.4 150 treatment, or patient problem. Information on diagnosis, drugs, prognosis, new therapies and products, or the latest information on a disease area, was required less frequently, but still by at least one-third of participants. Not having the time to conduct searches was the greatest barrier to information seeking for HCs. HCs spend from 1-10 hours per week on investigating answers to clinical questions. Most of this information seeking occurs outside of working hours, either at home or during breaks at work. To answer their clinical questions, 90% of the HCs use published, medical literature. The resources used by HCs include textbooks (30%), journals (30%), the Internet (19%), colleagues (17%), and databases (4%). The most important factors that influence resource choices are access to electronic or Internet resources: 60% prefer electronic resources rather than print resources. Additional influential factors included whether the resource was evidence-based, if it provided concise summaries of the information provided, and if the information could be found in a paper copy. HCs in this study also consult colleagues regularly, and their proximity to a colleague for consultation was a factor. The HCs rated their search skills very high: over half (55%) rated their ability to find information as good, 15% believed they were very good, 25% felt they were average, and 5% rated themselves as poor. The HCs were also confident in finding information to meet their needs: 70% claim that they find the information they require more than half of the time. Of those HCs, 25% claim they are successful more than three quarters of the time. 65% of the HCs experience difficulties in keeping current with evidence based medicine. Evidence-based resources such as the Cochrane Collaboration are used less frequently (25%) than resources such as Medline (65%) and Google (75%). When HCs were provided with a definition of clinical informationists (CI), 68% were not familiar with the role of CIs and only 32% of clinicians were familiar with the term “clinical informationist.” The HCs were then asked their feelings regarding the idea of involving a CI in their hospital: 18 of the 19 responses were positive. Various suggestions for how CIs could be used and the benefits of CIs were provided by the participants. Only three disadvantages of CIs were noted. Of 18 responses, 72% perceived that the inclusion of CIs would have a positive impact on patient care while 27.7% were neutral. Conclusion – Overall, the HCs surveyed in this study ask fewer questions, have different information needs, and are more confident in their search skills than clinicians found in previous studies; however, the authors state that previous studies had been done with clinicians in office settings rather than clinicians in hospital settings. HCs in this study identified lack of time as their main barrier to researching clinical questions and when they do find the time to search for clinical questions, it is either during breaks in their day or after work at home. Their preferred resources are those found electronically. Though they value evidence- based resources, HCs rarely use them. These factors point to a need for information professionals to provide either remote access to electronic medical information resources from home, or provide a service that would allow hospital clinicians to quickly and easily find information during the work day. This is an area in which a CI might play a role. Though many HCs were not familiar with CIs, they were receptive to having a CI on their clinical team. The HCs provided various suggestions for where a CI could be involved as well as desired skills and qualifications of CIs. The only possible disadvantages that the clinicians could foresee was cost, the deskilling of clinicians’ own information- seeking skills, and medico-legal issues. The authors identified several limitations of this study which include the small sample size, the snowball sampling method and the possibility of bias in subject recruitment, and not Evidence Based Library and Information Practice 2011, 6.4 151 including other health care professionals in this study. Further research regarding the information behaviour, seeking and skills of other health professionals is needed, as well as research on training and accreditation of CIs. Commentary Though the data collected could provide evidence for the need for clinical informationists (CIs) in the two hospitals surveyed, there are several areas of concern. First, the snowball sampling method of selecting participants could present bias: the participants in the study were contacts of the authors. A second issue is the small sample size: roughly 10% of eligible hospital clinicians (HCs) participated in the study. The diversity of HC specialization, which ranges from orthopedics and plastic surgery to cardiology and rheumatology, is also problematic. Different specialists may have different information needs. It is hard to generalize the information needs of such a diverse group of clinical practitioners. Various terms used in the article also lack clear definitions. The reader is left to assume that “hospital clinicians” only include doctors and do not include other medical professionals, such as physician assistants. Also, the term “Internet” is used to describe one of the various resources HCs utilize to answer clinical questions, but it is not clear how the authors define “Internet.” “Internet” could include databases, search engines, or other electronic resources. An additional point of confusion is why the survey was first tested on nurses but then given to HCs. Nurses and HCs may have different clinical information needs and perceptions of CIs. The questions asked of nurses would be different from the questions that would be asked of HCs. Finally, there is no information regarding informed consent. The reporting of the findings is also problematic. Mathematical discrepancies are first found in the area of specialization for the HCs at Hospital B: the total percentage is over 100%. Another error was found in Figure 1: the total percentage for the number of clinical questions that HCs generate per week is over 100%. A third miscalculation was in the HCs self rating of their search skills: the number of responses analyzed by the authors (n=20) is less than the number of reported responses. The corresponding chart for this data (Figure 6) is also incorrect: “Average” should be 25%, not 5%. Such inaccuracies call into question the quality of this study’s findings. It is important to note that participants were able to select multiple responses for several questions, which is why the percentages for the questions regarding reasons why HCs do not pursue clinical questions, where HCs need additional clinical information, and Cochrane, Medline, and Google use preference add up to over 100%. Despite the drawbacks outlined above, the research methodology utilized for the group studied is appropriate and this study could be replicated. The authors detail areas of limitations with the study as well as discuss areas for further research. Future researchers can also learn from the errors in this study and be mindful of their own research methodologies and data analysis. Hospital librarians who are interested in CI services may find this study useful as a model for conducting a similar assessment in their own hospital. Also, this study may provide additional ideas for hospitals that currently provide CI services. / Evidence Based Library and Information Practice