17 SA JournAl of PhySiotherAPy 2010 Vol 66 no 2 The Association between Psychological Stress and Low Back Pain among District Hospital Employees in Gauteng, South Africa threats. Psychological stressors are subjective because the threat occurs because of the individual’s inter­ pretation of an event, rather than as a result of the event itself”. She defined the stress response as “the integrated and non­specific reaction of the body to demands or stressors, which is com­ prised of cognitive, behavioural and physiological components.” Work­related stressors were identified in various studies. In a South African study it was identified as unexpected events, dependence on others, negative perceptions of support, low job satisfac­ tion, time pressure and deadlines. These work­related stressors were found to be associated with LBP (Van Vuuren et al 2007). On the other hand, taking unscheduled breaks was found to be preventative in the development of LBP. Psychological job demand was defined by Karasek et al (1998) as “the effort required carrying out work”. They defined decision latitude as “the individual’s potential control over the performance of the job” and job insecurity as “perceived threat or reality of job termination and layoff faced by workers”. Psychological IntRoDuCtIon Low back pain (LBP) is a widespread health problem, prevalent not only in South Africa, but all over the world. The LBP point prevalence rate among South African steel plant workers is 35,8% with the lifetime prevalence rate being 63,9% (Van Vuuren et al 2005). According to Louw et al (2007) the mean point prevalence rate of LBP in Africa among adults is 32% and the lifetime prevalence rate is 62%. Hospital workers are not overseen in these LBP statistics. The prevalence of LBP among employees in an Italian hospital was found to be 58,8% (Folletti et al 2005) and in Denmark, Warming et al (2009) found that 64% of nurses suffered from LBP. It is has been proven that physical factors influence the presence of LBP, but there is increasing evidence and growing consensus that psychosocial factors also play a role in the precipita­ tion or worsening of LBP (Bernard 1997, Linton 2001). Psychological stressors have been defined by Cotton (1990, p 29) as “those threats which are attributable to the individual’s internal reactivity – thoughts, feelings, and concerns about perceived Correspondence to: Benita Olivier Physiotherapy Department Faculty of Health Sciences University of the Witwatersrand 7 York Road, Parktown, 2193 Email: benita.naude@wits.ac.za AbSTrAcT: The presence of low back pain (LBP) can be influenced by psychosocial stress experienced at work. The aim of this study was to determine the point prevalence for LBP and the psychological stress expe- rienced at work as a factor associated with the presence of LBP amongst staff employed at district hospital in Pretoria, Gauteng, South Africa. A self-administered questionnaire was completed by all participants. Results indicated that the point prevalence for LBP was 47.46%. Sixty five point five seven percent of employees who experienced stress at work all the time, suffered from LBP (p=0.001). Stress experienced at work all the time increased the risk of LBP (OR 3.47 CI 1.46 ; 8.23). A clinical recommendation resulting from this study is that healthcare providers need to include the provision of education, support and appropriate referral for patients who perceive themselves to have high levels of stress. KEY WORDS: LOW BACK PAIN, PSYCHOLOGICAL STRESS, AND HOSPITAL STAFF Olivier B, MSc1 Mudzi W, MSc1 Mamabolo MV, MPH1 Becker PJ, PhD2 1 University of the Witwatersrand 2 Medical Research Council of South Africa job demands and decision latitude were factors identified to play a role in the presence of LBP (Karasek et al 1998) while Hyeonkyeong et al (2007) found that high job insecurity was significantly associated with lower back work­related musculoskeletal disorders. Hyeonkyeong et al (2007) also found that flight attendants with LBP had higher perceived psycholo­ gical demands than those without LBP. In a study done in Brussels by Clays et al (2007) it was found that work­related psychosocial factors were associated with having LBP. They investigated the presence of psychosocial risk factors in a sample of 2556 middle­aged men and women. Participants were followed up after an average of 6.6 years while being employed at the same company. Psychosocial factors associated with Research Article 18 SA JournAl of PhySiotherAPy 2010 Vol 66 no 2 the presence of LBP included low deci­ sion latitude, high job strain, low social support at work, low wage satisfaction and feeling stressed at work. The strength of this association was not reduced by controlling for physical factors. These findings were supported by Linton (2001) who did a systematic review on occupational psychological factors and LBP. He concluded that there is strong evidence that low job satisfaction, monotonous work, work relations, job demands, stress and perceived ability to work were related to future development of LBP. LBP was the most common and serious complaint among employees studied by Chen et al (2005). Five hundred and sixty one employees at an offshore oil company in South China participated in their study. Physical environment (noise, vibration, heat), safety, interface between job and family/ social life and ergonomics were found to be occupational stressors. According to Chen et al (2005) physical environ­ ment and ergonomics should not only be considered as physical factors but also as psychological stressors. Eating beha viour was identified as the most prevalent coping style and an impor­ tant predictor of musculoskeletal pain (Chen et al 2005). Gender differences in coping with psychosocial factors and its impact on LBP were also noted. Females were found to be at greater risk for LBP associated with psychosocial factors than males (Chen et al 2005, Tsuboi et al 2002, Yip et al 2001). Smedley et al (1997) did a longitudinal study on an all­female nursing population to esta­ blish predictors of LBP. They found that frequent low mood increased the chances of LBP leading to absence from work by 3.4 times. They argued that loss of time from work may be as a result of more severe LBP or a lack of ability to cope when the symptoms occurred. Among 141 nurses in Switzerland, Elfering et al (2002) found that a lack of control over stressful events at work, as well as lack of time control, may render an indivi­ dual vulnerable to musculoskeletal pain. In a study done among flight attendants, also a female­dominated occupational group, it was shown that job insecu­ A vast number of different psycho­ logical stressors have been found to be associated with LBP by various researchers. These stressors cause physio logical changes and reactions which may render an individual vul­ nerable to LBP. Although LBP may be experienced as a result of psychosocial factors or stress may be experienced as a result of chronic suffering from LBP, it is clear that perceived stress and LBP are associated. The aim of this article is to emphasise the presence of psycholo­ gical stress at work, and its association with LBP in a population of hospital employees, in Pretoria, South Africa. By identifying the role playing factors associated with LBP, effective LBP prevention and management programmes can be put in place in the occupational environment. Method This cross­sectional study was con­ ducted in 2007 using a self administered questionnaire. The population for the study comprised of all health care and support staff members employed perma­ nently at a district hospital in Pretoria, Gauteng. Only staff members who were permanently employed were included. Hospital staff members who were not willing to participate, students and casual workers were excluded from the study. A self­administered questionnaire was developed for the data collection. Arrangements were made to conduct the study on a date and time when most staff members from each department were available to participate. The total number of participants was 354 which was 77.80% of the total number of permanently employed hospital employ­ ees. The reasons for non­participation in the study included not being available as a result of leave, absence from work and also refusal to participate. Known risk indicators for LBP as described by Kwon et al (2006) was used as a foundation for the development of the self­administered questionnaire. The questionnaire contained questions under the following topics: demographics, recreation, occupation, perceived stress experienced at work, general health and the presence of LBP. The frequency of perceived stress at work was assessed rity in particular was associated with increased LBP presence (Heyonkyeong et al 2007). Among females in admini­ strative positions, feeling depressed was significantly associated with LBP (Clays et al 2007). Psychosocial factors may cause increased muscle tension which may in turn lead to altered spinal load­ ing. As a result of the latter, nutrition of the intervertebral discs, nerve roots and other spinal tissues is affected (Bergenudd and Johnell 1991, Bongers et al 1993). It was postulated that raised plasma cortisol levels may leave muscles vulnerable to injury due to mechanical loads and hence increased susceptibility to LBP (Theorell et al 1993). It is also believed that pain tole­ rance may be decreased due to stress among people living in poor psycho­ social environments, and those affected may be inclined to take more sick leave due to LBP (Burton and Erg 1997, Nachemson 1992). Self­reporting of LBP on a four­point scale in self­administered questionnaires has been found to have good reliability in test­retest analysis (Walsh and Coggon 1991). This method was also used by Warming et al (2009). They gained information from 148 nurses employed at a university hospital in Copenhagen, by completing daily logbooks on physi­ cal and psychosocial factors during three working days. High levels of stress were found to be strongly associated with LBP (OR 4.0, CI 1.04 ; 15.36) in their study. In contrast to the findings of the above studies, Kwon et al (2006) found that the development of LBP was not depend­ ent on the level of stress. In their study, they divided the respondents into groups of: no stress at all, slight stress, moderate stress and a great deal of stress. They argued that mental symptoms such as depression concurred with chronic dis­ eases. For this reason they doubted that mental stress is a cause of LBP, but may be as a result of chronic suffering from LBP. Their study excluded people that showed the slightest sign or symptom of systemic disease, diagnosed or undi­ agnosed. This decreased the influence of chronic co­morbid diseases on the presence of LBP, and hence meant that a purer result could be obtained. 19 SA JournAl of PhySiotherAPy 2010 Vol 66 no 2 as part of the questionnaire by means of a four point scale (never, sometimes, often, all the time). The content of the questionnaire was validated by having it scrutinised by “physiotherapy experts” in the field of back care and management. The repeatability of the questionnaire was established using the test re­test method. Agreement on all the questions, that could not be changed from day to day, existed. Examples of questions on information that change from day to day are the number of exercise sessions per month, cigarettes smoked per day, hours spend sitting and standing, and number of units of alcohol consumed per week. The English questionnaire was trans­ lated into Tswana by three translators, and back translated into English again by two independent translators (Naude et al 2009). Ethical clearance was granted (Number M070359) by the University of the Witwatersrand Human Research Ethics Committee prior to commence­ ment of the study. Permission to con­ duct this study was obtained from the Superintendent of the hospital. Participants were told that refusal to take part in the study would not preju­ dice them in any way and were asked to voluntarily sign the consent form. Statistical software (Stata Release 8.0) was used to analyse data. Categorical parameters were summarised using frequencies, percentages and cross­ tabulations. Comparison between LBP categories (yes, no) with respect to cate­ gorical parameters employed Fisher’s exact and tests for trends in odds ratios employed Pearson’s chi­square test. Univariate analysis (independently) and odds ratios for potential risk fac­ tors for LBP were determined and tested for trend. Results The point prevalence of LBP was 47.46% (n=168). The majority of the population was between the ages of 26 and 40 years (n=216). Seventy two percent (n=255) of the participants were female. The distribution of perceived stress at work and LBP is shown in Table 1. Few participants (12.15%) never experienced stress while 65.57% of the 61 partici­ pants who experienced stress all the time had LBP. Perceived stress at work was found to be significantly associated with the presence of LBP (p=0.001). The risk of developing LBP was higher when a person had perceived stress at work all the time (OR 3.47, CI 1.46 ; 8.23) (Table 2). Discussion The point prevalence of LBP in this study (47.46%) is higher than what was found in two South African studies done by Uebel et al (2009) (13.1%) and Naidoo et al (2007) (44.33%). Uebel et al’s (2009) study only included nursing staff consulting for LBP at the staff clinic and injury­on­duty unit of the hospital where they were employed. Although Uebel et al (2009) states that the number of nurses seeking outside medical atten­ tion is small, this may be a possible explanation for the low prevalence of LBP. Human resources and associated productivity at work will be negatively impacted by high point prevalence rates (Van Vuuren et al 2005). This issue is also relevant when looking at essential staff required in a hospital setting. This decrease in productivity will negatively impact on direct and in­direct patient care in a district hospital. According to a number of studies, females are greatly at risk for LBP associated with psychological factors (Cole et al 2001, Tsuboi et al 2002, Yip et al 2001). The majority of the popu­ lation in this study was female which may be a contributing factor to the high Perceived work stress Low Back Pain n (%) No Low Back Pain n (%) Total n (%) Never 15 (4.24) 28 (7.91) 43 (12.15) Sometimes 89 (25.14) 113 (31.92) 202 (57.06) Often 24 (6.78) 24 (6.78) 48 (13.54) All the time 40 (11.30) 21 (5.93) 61 (17.23) Total n (%) 168 (47.46) 186 (52.54) 354 (100.00) Table 1: The distribution of perceived stress at work and low back pain (n=354) Table 2: The relationship between perceived stress at work and the presence of low back pain (n=168) lifestyle factor category low Back Pain n (%) low Back Pain n (%) (95% confidence interval) p-value Test for Trend Perceived stress at work Never 15 (8.93) 1.00 0.001 Some- times 89 (52.98) 1.47 (0.74 ; 2.93) Often 24 (14.29) 1.87 (0.79 ; 4.41) All the time 40 (23.81) 3.47 (1.46 ; 8.23) Total n (%) 168 (100.00) 20 SA JournAl of PhySiotherAPy 2010 Vol 66 no 2 management and relaxation techniques. Further research is needed to establish why certain psychological variables may be related to LBP as well as to establish the exact physiological mechanism behind this relationship. References Bergenudd H, Johnell O 1991 Somatic versus nonsomatic shoulder and back pain experience in middle age in relation to body build, physical fitness, bone mineral content, gamma­glutamyl­ transferase, occupational workload, and psychoso­ cial factors. Spine 16:1051­5 Bernard BP 1997 Musculoskeletal disorders and workplace flight attendants: a critical review of epidemiologic evidence for work­related mus­ culoskeletal disorders of the enck, upper extremity and low back. DHHS (NIOSH). Publication no 97­141. National Institute for Occupational Safety and Health, Cincinnati Bongers PM, De Winter CR, Kompier MA, Hildebrandt VH 1993 Psychosocial factors at work and musculoskeletal disease. Scand J Work Environ Health. 19:197­312 Burton AK, Erg E 1997 Back injury and work loss. Biomechanical and psychosocial influences. Spine 22:2575­80. Chen WQ, Yu I, Wong TW 2005 Impact of occupational stress and other psychosocial factors on musculoskeletal pain among Chinese offshore oil installation workers. Occup Environ Med 62:251­6 Clays E, De Bacquer D, Leynen F, Kornitzer M, Kittel F, De Backer G 2007 The impact of psy­ chosocial factors on low back pain: longitudinal results from the Belstress study. Spine 32:262­268 Cole DC, Ibrahim SA, Shannon HS, Scott F, Eyles J 2001 Work correlates of back problems and activity restriction due to musculoskeletal disorders in the Canadian national population health surey (NPHS) 1994­5 data. Occup Environ Med 58:728­734 Cotton DHG. Stress Management: An Integrated­ approach to therapy, Brunner/Mazel/Psychological Stress Series: Routledge; 1990, p 29 Elfering A, Grebner S, Semmer NK, Gerber H 2002 Time control, catecholamines and back pain among young nurses. Scand J Work Environ Health 28:386­93 Folletti I, Belardinelli V, Giovannini G, Cresta B, Fabrizi G, Tacconi C et al 2005 Prevalence and determinants of low back pain in hospital workers. G Ital Med Lav Ergon 27:359­361 Hartvigsen J, Lings S, Leboeuf­Yde C, Bakketeig L 2004 Psychosocial factors at work in relation to low back pain and consequences of low back pain; a systematic, critical review of prospec­ tive cohort studies. J Occup Environ Med 61:e2 ees with LBP was very high (65.57%). All of the above mentioned work related psychological components may play a role in precipitating a non­specific stress reaction. This stress reaction can pre­ dispose an individual to LBP. As these stressors are being perceived differently by different individuals, the following questions may be asked: what causes one individual to be more immune to stress than another and which psychological characteristics are present in those with a high stress threshold? The answer to these questions can be transformed into an educational tool in order to increase the individual’s threshold to work­related stressors and as a result reduce the chances of developing LBP. In the present study, perceived stress was not measured by a standardised out­ come measure. It was merely tested by a question aiming to gain information on the frequency of stress experienced by employees a hospital environment. It focuses the attention on the non­ negligible issue of work stress and LBP, and this paper was used to identify rele­ vant literature that confirms this finding. Further research is recommended in the association between psychological stress and LBP. Limitations of this study include self reporting of health and the presence of LBP my influence the accu­ racy of information given by partici­ pants. Health, LBP and perceived stress could have been underreported by staff for fear of negative repercussions from the employer with regards to job promo­ tion and being prejudiced against. Conclusion The aim of this article is to emphasise the presence of psychological stress and its association with LBP among the staff members employed at a district hospi­ tal in Gauteng, South Africa. The point prevalence of LBP among employees at 47.46% is high. Psychological stress experienced at work is associated with the presence of LBP. Clinical recommen­ dations include that healthcare provid­ ers need to incorporate the provision of education, support and appropriate referral for patients who perceive them­ selves to have high levels of stress. LBP prevention and management pro­ grammes should incorporate stress prevalence of LBP found in this study as well as the significant association of LBP and psychological stress. This study established that 40 (65.57%) of the 61 participants who experienced stress all the time, had LBP. The risk to develop LBP for this group was also elevated (OR 3.47: CI 1.46 ; 8.23) and a positive association which was statis­ tically significant (p=0.001) was found between stress at work and the presence of LBP in this study. A similar significant effect of work related stress on LBP was also found by various other researchers (Hartvigsen et al 2004, Karasek et al 1998, Van Vuuren et al 2007, Warming et al 2009). This study did not establish various factors that can be identified as work related stressors, but research has shown that unexpected events, lack of control over stressful events, dependence on others, time pressures and lack of con­ trol over time are work­related stres­ sors (Van Vuuren et al 2005, Elfering et al 2002). These stressors may also be present in the understaffed hospitals. Due to budget constraints, in November 2008 the department of health had 21 923 unfilled posts, including admini­ stration, support and management posts (Sibongakonke 2008). Unexpected and stressful events are also an integral part of daily patient care and the effect may be worsened by the lack of human resource. Van Vuuren et al (2007) esta­ blished that unscheduled breaks are protective against LBP however hospital employees may not be able to have these breaks if there is understaffing. Stressful situations may also be present in a staff member’s personal life and may influence the stress experienced at work and in general. Clays et al (2007) found that nonwork­related factors like feel­ ing depressed and low satisfaction with private life were associated with LBP. What could not be derived from this study is whether it was stress that was experienced at work which increased LBP, or if increased stress was experi­ enced as a result of LBP. From the literature it becomes clear that psychological stressors are perceived threats (Cotton 1990) and that these are subjective. In our study the frequency of perceived stress among hospital employ­ 21 SA JournAl of PhySiotherAPy 2010 Vol 66 no 2 Heyonkyeong L, Wilbur J, Kim MJ, Miller AM 2007 Psychosocial risk factors for work­related musculoskeletal disorders of the lower­back among long­haul international female flight female attendants. JAN: Original Research, The Authors, Journal Compilation, Blackwell Publishing Ltd. Karasek R, Brisson C, Kawakami N 1998 The job content questionnaire (JCQ): an instrument for internationally comparative assessments of psychosocial job characteristics. J Occup Health Psychol 3:322­355 Kwon MA, Shim WS, Kim MH, Gwak MS, Hahm TS, Kim GS, et al. 2006 A correlation between LBP and associated factors: a study involving 772 patients who had undergone gen­ eral physical examination. J Korean Med Sci 21: 1086­91 Linton SJ 2001 Occupational psychological factors increase the risk for back pain: A system­ atic review. J Occup Rehabil 11:53­66 Louw QA, Morris LD, Grimmer­Somers K 2007 The prevalence of low back pain in Africa: a systematic review. BMC Musculoskelet Disord 1:105 Nachemson AL 1992 Newest knowledge of low back pain. A critical look. Clin Orthop Rel Res 279:8­20. Naidoo R, Coopoo Y 2007 The health and fitness profiles of nurses in KwaZulu­Natal. Curationis 30:66­73 Naude B, Mudzi W, Mamabolo VM 2009 Low back pain among hospital employees in Gauteng, South Africa: point prevalence and associated factors. Occup Health SA 15:24­30 Sibongakonke Shoba: Business Day, Gauteng health stops filling vacant posts ­ 3 November 2008, BHF Global, [cited 2008 Nov]. Available from: http://www.bhfglobal.com/gauteng­health­ stops­filling­vacant­posts­3­november­2008. Smedley J, Egger P, Cooper C, Coggon D 1997 Prospective cohort study of predictors of incident low back pain in nurses. BMJ 314:1225­8. Theorell T, Nordemar R, Michelsen H 1993 Pain thresholds during standardized psychological stress in relation to perceived psychosocial work situation. J Psychosom Res 37:299­305 Tsuboi H, Takeuchi K, Watanabe M, Hori R, Kobayashi F 2002 Psychosocial factors related to low back pain among school personnel in Nagoya, Japan. Ind Health 40:266­271 Uebel KE, Rae W, Joubert G, Hiemstra L 2009 Reported low back pain amongst nurses at a dis­ trict hospital: incidence, profile and risk factors. Occup Health SA 2009 15:8­15 Van Vuuren B, Zinzen E, Van Heerden HJ, Becker PJ, Meeusen R 2007 Work and family support systems and the prevalence of lower back problems in a South African steel industry. J Occup Rehabil 17:409­421 Van Vuuren BJ, Becker PJ, Van Heerden HJ, Zinzen E, Meunisen R 2005 Lower back problems and occupational risk factors in a South African steel industry. Am J Ind Med 47:451­7 Walsh K, Coggon D 1991 Reproducibility of histories of low back pain obtained by self­admin­ istered questionnaire. Spine 16:1075­1077 Warming S, Precht DH, Suadicani P, Ebbohoj NE 2009 Musculoskeletal complaints among nurses related to patient handling tasks and psy­ chosocial factors – based on logbook registrations. Appl Ergon 2009 40: 569­76. Epub 2007 Sept 11 Yip Y, Ho SC, Chan SG 2001 Socio­ psychological stressors as risk factors for low back pain in Chinese middle­aged women. J Adv Nurs 36:409­416