Abstract

Background

Back pain and occupational stress are known risk factors for absenteeism and presenteeism. In addition, the relationship between back pain (BP) and psychosocial stressors has been examined in numerous studies.

Aims

To examine the prevalence of BP and perceived stress among employees of different occupational status and to investigate their combined impact on work ability.

Methods

A comprehensive survey combining questionnaire data and medical examination offered in one division of a major chemical company in Germany. Self-rated health and work ability were assessed using the Work Ability Index (WAI). A synergy index was used to assess a potential interaction between both exposures under an additive model.

Results

Of 867 voluntary participants, 653 returned complete questionnaires on BP and job stress perception. Although occupational stressors were perceived differently, there was no difference in the prevalence of BP between the occupational groups. Back pain and stress perception are correlated with each other and both are negatively associated with work ability. After adjustment for occupational status, demographic and lifestyle factors, we found a synergy effect of BP and stress perception with a modest to strong impact on declining WAI.

Conclusions

Corporate health promotion interventions to reduce the impact of BP and stress perception on work ability should target both physical and psychological dimensions. Such interventions may be initiated in the context of regular (occupational) medical examinations.

Introduction

For decades, musculoskeletal disorders, and more specifically back pain (BP), have been the leading cause of sickness absence in Germany, responsible for up to one quarter of all workdays lost [1]. BP is a common problem that most people experience at some point in their life, although to varying degrees [2]. Occasional mild episodes usually result in full recovery, but most people who experience activity-limiting BP go on to have recurrent episodes [2,3]. It is therefore unsurprising that recurrent BP has been described as a factor significantly influencing work ability [4].

BP and psychosocial stress are known risk factors for both absenteeism and presenteeism [5,6]. The relationship between BP and psychosocial stressors has been examined in numerous studies and reviews [7,8]. Most, but not all, have found evidence for some relationship between psychosocial factors and incidence or prevalence of BP, and some go as far as proposing the term ‘psychosocial pain syndrome’ to cover these aspects of BP [9]. Furthermore, the negative association between work-related stress and work ability has been reported in diverse occupational groups [10–12] and work ability has also been shown to have a significant predictive value for disability and early retirement [13]. Job-related stress can also affect employee productivity through increased absenteeism and presenteeism, imposing a direct economic cost on employers and society [14].

To date, we are not aware of a study addressing the combined effect of BP and perceived stress on work ability. The aim of this study was to examine the prevalence of BP and perceived stress across employees of different occupational status, to investigate their combined impact on work ability and to identify appropriate measures for future corporate health promotion activities.

Methods

This cross-sectional study was carried out in a major chemical company in Germany between 1 January 2008 and 31 March 2009 as a part of a health promotion campaign in a representative division of the company. Details of the programme have been described elsewhere [12]. The examination programme comprised a medical examination focusing on risk factors for coronary heart disease and diabetes mellitus as well as questionnaires on stress perception, BP and the Work Ability Index (WAI).

For the purposes of the study, the participants were classified according to job activity and responsibility as frontline operators (FO), skilled workers (SW) and professional and managerial staff (PM). This classification roughly represents the socio-economic status of the employees. The study was conducted and the data were collected in accordance with standard ethical procedures [15]. The aim and scope of the examination was stated at the beginning of the questionnaire, and a consent form was enclosed with the questionnaire. Participants were able to ask questions during their medical examination, which was conducted by a physician. Study participation was voluntary and could be declined at any part of the examination programme without a need for explanation. The study protocol had been approved in advance by the workers’ council.

The questions used to assess BP were taken from the standardized Nordic Musculoskeletal Questionnaires [16]. The questions provided a measure of common symptoms used in epidemiological studies on musculoskeletal disorders. A sum score was created from all items without weighting them, and the outcomes were then divided at quartiles into excellent (0 or 1), good (2 or 3), moderate (4, 5 or 6) and poor (≥7). The questions used to assess stress perception were taken from validated published instruments and have been used in the employees’ opinion survey in the company for several years [12]. The questions were aimed at ascertaining perception of safety in the workplace, self-rated health status, frequency of experienced stress symptoms, unrealistic job demands, time pressure and impaired work-life balance. Each question was answered on a five-point Likert scale. For the analysis, the lower two and the top two categories were grouped together so that three categories remained, with the lower two categories referring to the highest stress perception. The WAI [17] assesses a person’s perception of how well they are able to perform their work, compared with their lifetime best. A short version in German [18] was used. The WAI is calculated by summing the responses to each dimension, with a possible range of scores from 7 to 49. For a binary outcome, WAI was dichotomized at the median of sum scores into poor to moderate (7–41) and good to excellent (42–49).

Information about general characteristics and health behaviour of participants was obtained from the medical information system used in the occupational medicine department. During the medical examination, further information (data not shown) such as smoking status, alcohol consumption, physical activity and information on the incidence and prevalence of chronic health conditions was obtained. Participants were classified according to their smoking status into current smokers, former smokers and non-smokers. Physical activity was classified according to the median time spent on physical exercise each week (2hr/week). Body mass index was used to define the subjects as normal (<25kg/m2), overweight (25–29kg/m2) or obese (≥30kg/m2).

Descriptive statistics (median, inter-quartile range and proportions) of the employees with different occupational status and the four WAI categories were calculated. The Chi-squared test was used to assess the homogeneity of the categorical variables. Within each occupational group, participants were stratified according to age (<30, 30–39, 40–49 and ≥50). Cochran–Mantel–Haenszel tests were used to compare the three employee groups, stratified according to age groups.

The outcomes of interest were the WAI sum score and the subscores of its seven dimensions as a function of perceived occupational stress and self-reported experience of BP. The WAI sum score and its subscores of the seven dimensions were dichotomized at the median and logistic regression models were used to estimate the combined impact of stress perception and BP. To this end, we formed groups of employees who reported either high levels of stress perception or high levels of BP or both. The employees with the lowest stress perception and the lowest level of BP were taken as reference. Only variables that reached significance level (P < 0.05) in the univariate analyses in relationship with WAI were included. Results are presented as odds ratios (OR) with 95% confidence intervals (CI). To measure if the combined presence of both risk factors exceeded the sum of the individual effects, a synergy index [19] was used. This measures the extent to which the risk ratio for both factors together exceeds 1, and whether this is greater than the sum of the extent to which each of the risk ratios considered individually exceed 1. The measure was calculated by substituting OR for relative risks [20]. An index >1 indicates a positive interaction, that is, a more than additive interaction. The analyses were performed with SAS version 9.2 (SAS Institute, Cary, NC, USA).

Results

At the time of the study, there were a total of 1250 employees in this division. Of these, 867 (70%) participated in the medical examinations and 653 returned completed questionnaires, which were eligible for the analysis, an overall response rate of 52%. Among the 653 participants included in the analysis, 11% (70) were PM, 39% (257) were SW and 50% (326) were FO. The median age of all participants was 43 (inter-quartile range: 36–48) and the median duration of work in the company was 20 (inter-quartile range: 14–27) years. The gender distribution was representative for the company; with almost no women in production, 20% in administration and 10% in management.

The prevalence of BP symptoms in relation to job levels is presented in Table 1. The self-reported lifetime prevalence of BP was 70% and was similar in all job types. About 30% had experienced BP lasting 1–7 days during the last 12 months, which was also similar in all job types. Reported reduced activity at work due to BP was statistically significantly different between occupational groups and the duration of reduced activity in the last 12 months showed a positive trend from job level PM to FO although this did not achieve statistical significance. The inter-relationship between stress perception and BP is shown in Figure 1. The frequency of stress symptoms (seldom, sometimes and often) was positively correlated with the degree of BP (slight, mild, moderate and severe), consistently for each item.

Table 1.

Distribution of reported BP in 653 respondents according to job levels FO, SW and PM

 FO, n = 326 (50%) SW, n = 257 (39%) PM, n = 70 (11%) P-valuea 
BP ever in life, n (%)    NS 
 No 88 (27) 79 (31) 18 (26)  
 Yes 238 (73) 178 (69) 52 (74)  
BP hospital treatment, n (%)    NS 
 No 297 (91) 244 (95) 64 (91)  
 Yes 29 (9) 13 (5) 6 (9)  
BP attributable to accident, n (%)    NS 
 No 314 (96) 254 (99) 69 (99)  
 Yes 12 (4) 3 (1) 1 (1)  
Job change because of low BP, n (%)    NS 
 No 319 (98) 255 (99) 70 (100)  
 Yes 7 (2) 2 (1) 0 (0)  
Duration of BP in the last 12 months, n (%)    NS 
 Never 109 (33) 92 (36) 20 (29)  
 1–7 days 87 (27) 75 (29) 26 (37)  
 8–30 days 60 (18) 52 (20) 15 (21)  
 ≥30 days 46 (14) 29 (11) 8 (11)  
 Everyday 24 (7) 9 (4) 1 (1)  
Reduced activity in job, n (%)    <0.05 
 No 269 (83) 221 (86) 66 (94)  
 Yes 57 (17) 36 (14) 4 (6)  
Reduced activity in private life, n (%)    NS 
 No 237 (73) 185 (72) 52 (74)  
 Yes 89 (27) 72 (28) 18 (26)  
Duration of reduced activity in the last 12 months, n (%)    NS 
 No 206 (63) 184 (72) 57 (81)  
 1–7 days 53 (16) 37 (14) 5 (7)  
 8–30 days 39 (12) 22 (9) 6 (9)  
 ≥30 days 28 (9) 14 (5) 2 (3)  
BP treated, n (%)    NS 
 No 177 (54) 151 (59) 46 (66)  
 Yes 149 (46) 106 (41) 24 (34)  
BP in the last 7 days, n (%)    NS 
 No 223 (68) 173 (67) 53 (76)  
 Yes 103 (32) 84 (33) 17 (24)  
 FO, n = 326 (50%) SW, n = 257 (39%) PM, n = 70 (11%) P-valuea 
BP ever in life, n (%)    NS 
 No 88 (27) 79 (31) 18 (26)  
 Yes 238 (73) 178 (69) 52 (74)  
BP hospital treatment, n (%)    NS 
 No 297 (91) 244 (95) 64 (91)  
 Yes 29 (9) 13 (5) 6 (9)  
BP attributable to accident, n (%)    NS 
 No 314 (96) 254 (99) 69 (99)  
 Yes 12 (4) 3 (1) 1 (1)  
Job change because of low BP, n (%)    NS 
 No 319 (98) 255 (99) 70 (100)  
 Yes 7 (2) 2 (1) 0 (0)  
Duration of BP in the last 12 months, n (%)    NS 
 Never 109 (33) 92 (36) 20 (29)  
 1–7 days 87 (27) 75 (29) 26 (37)  
 8–30 days 60 (18) 52 (20) 15 (21)  
 ≥30 days 46 (14) 29 (11) 8 (11)  
 Everyday 24 (7) 9 (4) 1 (1)  
Reduced activity in job, n (%)    <0.05 
 No 269 (83) 221 (86) 66 (94)  
 Yes 57 (17) 36 (14) 4 (6)  
Reduced activity in private life, n (%)    NS 
 No 237 (73) 185 (72) 52 (74)  
 Yes 89 (27) 72 (28) 18 (26)  
Duration of reduced activity in the last 12 months, n (%)    NS 
 No 206 (63) 184 (72) 57 (81)  
 1–7 days 53 (16) 37 (14) 5 (7)  
 8–30 days 39 (12) 22 (9) 6 (9)  
 ≥30 days 28 (9) 14 (5) 2 (3)  
BP treated, n (%)    NS 
 No 177 (54) 151 (59) 46 (66)  
 Yes 149 (46) 106 (41) 24 (34)  
BP in the last 7 days, n (%)    NS 
 No 223 (68) 173 (67) 53 (76)  
 Yes 103 (32) 84 (33) 17 (24)  

NS, not statistically significant.

aCochran–Mantel–Haenszel tests were used, stratified according to age groups.

Figure 1.

Interrelationship between stress perception and back pain.

Figure 1.

Interrelationship between stress perception and back pain.

The univariate and multivariate analyses to estimate the impact of BP on WAI are summarized in Table 2. Lifetime BP prevalence varied from 81% among participants with a lower WAI to 64% among those with a better WAI. Hospital treatment was seldom needed, suggesting a lower severity of BP. Accident was a rare cause of BP in this population and change of job was generally not necessary as a consequence. The duration of BP during the last 12 months was strongly associated with WAI and a stronger relationship was evident with an increasing duration of BP. The prevalence of reduced activity at work was less than that in private life. Even after adjusting for age and job level in the multivariate analysis, a stronger impact of reduced work activity on WAI was found. In addition, the duration of reduced activity in the last 12 months was found to be independ ently associated with a decreased WAI, and a gradient of impact was evident with increasing duration. Treated BP and short-term BP events were independently associated with WAI. In summary, a BP sum score lower than median was associated with a 2.75-fold (95% CI 1.95–3.86) risk of a lower WAI.

Table 2.

Univariate and multivariate analyses of impact of BP on the risk of a lower WAI

 WAI < 42,n = 288 (44%),n (%) WAI ≥ 42,n = 365 (56%),n (%) Model 1aOR (95% CI) Model 2bOR (95% CI) 
BP ever in life 
 No 55 (19) 130 (36) 
 Yes 233 (81) 235 (64) 2.21 (1.52–3.22) 2.14 (1.46–3.14) 
BP hospital treatment 
 No 259 (90) 346 (95) 
 Yes 29 (10) 19 (5) 1.65 (0.88–3.10) 1.47 (0.78–2.77) 
BP attributable to accident 
 No 279 (97) 358 (98) 
 Yes 9 (3) 7 (2) 1.34 (0.48–3.77) 1.12 (0.39–3.17) 
Job change because of low BP 
 No 281 (98) 363 (99) 
 Yes 7 (2) 2 (1) 4.24 (0.85–21.12) 3.93 (0.78–19.85) 
Duration of BP in the last 12 months 
 Never 71 (25) 150 (41) 
 1–7 days 71 (25) 117 (32) 1.38 (0.91–2.10) 1.31 (0.85–2.02) 
 8–30 days 66 (23) 61 (17) 2.29 (1.44–3.64) 2.13 (1.32–3.44) 
 ≥30 days 52 (18) 31 (8) 3.68 (2.14–6.33) 3.73 (2.14–6.51) 
 Everyday 28 (10) 6 (2) 9.01 (3.51–23.15) 8.04 (3.11–20.78) 
Reduced activity in job 
 No 216 (75) 340 (93) 
 Yes 72 (25) 25 (7) 4.04 (2.45–6.64) 3.68 (2.21–6.14) 
Reduced activity in private life 
 No 173 (60) 301 (82) 
 Yes 115 (40) 64 (18) 3.11 (2.15–4.50) 3.04 (2.07–4.47) 
Duration of reduced activity in the last 12 months 
 No 155 (54) 292 (80) 
 1–7 days 48 (17) 47 (13) 1.81 (1.15–2.87) 1.80 (1.11–2.90) 
 8–30 days 46 (16) 21 (6) 3.27 (1.84–5.78) 2.98 (1.67–5.32) 
 ≥30 days 39 (13) 5 (1) 12.93 (4.94–33.83) 12.09 (4.58–31.90) 
BP treated 
 No 136 (47) 238 (65) 
 Yes 152 (53) 127 (35) 1.93 (1.39–2.67) 1.87 (1.34–2.61) 
BP in the last 7 days 
 No 163 (57) 286 (78) 
 Yes 125 (43) 79 (22) 2.62 (1.85–3.75) 2.54 (1.77–3.65) 
BP sum score 
 Low 115 (40) 247 (68) 
 High 173 (60) 118 (32) 2.91 (2.08–4.05) 2.74 (1.95–3.86) 
 WAI < 42,n = 288 (44%),n (%) WAI ≥ 42,n = 365 (56%),n (%) Model 1aOR (95% CI) Model 2bOR (95% CI) 
BP ever in life 
 No 55 (19) 130 (36) 
 Yes 233 (81) 235 (64) 2.21 (1.52–3.22) 2.14 (1.46–3.14) 
BP hospital treatment 
 No 259 (90) 346 (95) 
 Yes 29 (10) 19 (5) 1.65 (0.88–3.10) 1.47 (0.78–2.77) 
BP attributable to accident 
 No 279 (97) 358 (98) 
 Yes 9 (3) 7 (2) 1.34 (0.48–3.77) 1.12 (0.39–3.17) 
Job change because of low BP 
 No 281 (98) 363 (99) 
 Yes 7 (2) 2 (1) 4.24 (0.85–21.12) 3.93 (0.78–19.85) 
Duration of BP in the last 12 months 
 Never 71 (25) 150 (41) 
 1–7 days 71 (25) 117 (32) 1.38 (0.91–2.10) 1.31 (0.85–2.02) 
 8–30 days 66 (23) 61 (17) 2.29 (1.44–3.64) 2.13 (1.32–3.44) 
 ≥30 days 52 (18) 31 (8) 3.68 (2.14–6.33) 3.73 (2.14–6.51) 
 Everyday 28 (10) 6 (2) 9.01 (3.51–23.15) 8.04 (3.11–20.78) 
Reduced activity in job 
 No 216 (75) 340 (93) 
 Yes 72 (25) 25 (7) 4.04 (2.45–6.64) 3.68 (2.21–6.14) 
Reduced activity in private life 
 No 173 (60) 301 (82) 
 Yes 115 (40) 64 (18) 3.11 (2.15–4.50) 3.04 (2.07–4.47) 
Duration of reduced activity in the last 12 months 
 No 155 (54) 292 (80) 
 1–7 days 48 (17) 47 (13) 1.81 (1.15–2.87) 1.80 (1.11–2.90) 
 8–30 days 46 (16) 21 (6) 3.27 (1.84–5.78) 2.98 (1.67–5.32) 
 ≥30 days 39 (13) 5 (1) 12.93 (4.94–33.83) 12.09 (4.58–31.90) 
BP treated 
 No 136 (47) 238 (65) 
 Yes 152 (53) 127 (35) 1.93 (1.39–2.67) 1.87 (1.34–2.61) 
BP in the last 7 days 
 No 163 (57) 286 (78) 
 Yes 125 (43) 79 (22) 2.62 (1.85–3.75) 2.54 (1.77–3.65) 
BP sum score 
 Low 115 (40) 247 (68) 
 High 173 (60) 118 (32) 2.91 (2.08–4.05) 2.74 (1.95–3.86) 

aModel 1: OR adjusted for age and job level.

bModel 2: OR adjusted for age, job level, body mass index and physical activity.

To examine the combined effect of stress perception and BP, the presence of stress or BP or both was evaluated with respect to the individual dimensions of WAI (Table 3). Employees reporting both higher levels of stress and BP had a consistently higher risk of a lower WAI than those who reported stress or BP alone or no exposure. A synergistic interaction was suggested for several dimensions of the WAI, but reached statistical significance only for Dimension 7, which represents a perceived lack of mental resources.

Table 3.

Joint effect of strong stress perception and severe BP on seven single dimensions of WAI by means of logistic regression, respectively, adjusted for job levels and age groups

Dependent variable Stress +/BP +, n = 227OR (95% CI) Stress +/BP –, n = 221OR (95% CI) Stress–/BP +, n = 64OR (95% CI) SI (95% CI) 
Dimension 1 < 8 pts 10.55 (5.07–21.95)* 6.17 (2.93–12.98)* 5.26 (2.20–12.61)* 1.01 (0.33–3.40) 
Dimension 2 < 8 pts 8.53 (4.52–16.09)* 5.95 (3.13–11.29)* 2.48 (1.08–5.65)* 1.17 (0.51–2.70) 
Dimension 3 < 7 pts 2.98 (1.87–4.74)* 2.45 (1.41–3.59)* 2.77 (1.47–5.21)* 0.61 (0.15–2.47) 
Dimension 4 < 6 pts 0.90 (0.21–3.77) 0.38 (0.06–2.36) 1.96 (0.40–9.59) NA 
Dimension 5 < 4 pts 1.69 (0.99–2.88) 1.07 (0.61–1.87) 1.11 (0.54–2.32) 3.83 (0.13–109.95) 
Dimension 6 < 7 pts 5.40 (1.23–23.82)* 3.50 (0.75–16.38) 1.64 (0.22–12.18) 1.40 (0.79–2.48) 
Dimension 7 < 3 pts 17.07 (7.92–36.77)* 14.02 (6.52–30.15)* 0.84 (0.21–3.29) 1.25 (1.23–1.27)* 
WAI sum score < 42 pts 13.33 (7.41–24.39)* 6.62 (3.68–11.91)* 5.03 (2.43–10.42)* 1.28 (0.21–7.76) 
Dependent variable Stress +/BP +, n = 227OR (95% CI) Stress +/BP –, n = 221OR (95% CI) Stress–/BP +, n = 64OR (95% CI) SI (95% CI) 
Dimension 1 < 8 pts 10.55 (5.07–21.95)* 6.17 (2.93–12.98)* 5.26 (2.20–12.61)* 1.01 (0.33–3.40) 
Dimension 2 < 8 pts 8.53 (4.52–16.09)* 5.95 (3.13–11.29)* 2.48 (1.08–5.65)* 1.17 (0.51–2.70) 
Dimension 3 < 7 pts 2.98 (1.87–4.74)* 2.45 (1.41–3.59)* 2.77 (1.47–5.21)* 0.61 (0.15–2.47) 
Dimension 4 < 6 pts 0.90 (0.21–3.77) 0.38 (0.06–2.36) 1.96 (0.40–9.59) NA 
Dimension 5 < 4 pts 1.69 (0.99–2.88) 1.07 (0.61–1.87) 1.11 (0.54–2.32) 3.83 (0.13–109.95) 
Dimension 6 < 7 pts 5.40 (1.23–23.82)* 3.50 (0.75–16.38) 1.64 (0.22–12.18) 1.40 (0.79–2.48) 
Dimension 7 < 3 pts 17.07 (7.92–36.77)* 14.02 (6.52–30.15)* 0.84 (0.21–3.29) 1.25 (1.23–1.27)* 
WAI sum score < 42 pts 13.33 (7.41–24.39)* 6.62 (3.68–11.91)* 5.03 (2.43–10.42)* 1.28 (0.21–7.76) 

CI, confidence interval; NA, not applicable; SI, synergy index.

The employees with low stress and slight or mild BP were taken as reference.

*P < 0.05.

Discussion

In this cross-sectional study, we found that there was a strong inter-relationship between BP and stress perception, that both BP and stress perception were negatively associated with work ability assessed by the WAI and that there was a synergic effect of both factors on WAI.

The classification of the participants according to their occupational status may constitute a strength of the study, as the occupational status can be considered to be a proxy measure of educational and occupational status, which are important factors influencing stress perception [21].

The study does, however, have a number of limitations. As the study population was selected on a voluntary basis rather than randomly, there is a possibility of potential selection bias. However, no systematic difference in participants’ age and gender was found across the different occupational groups compared with the total population at this site, so the results may well be representative of the total workforce across the occupational groups involved. Other potential limitations include the study’s cross-sectional nature, meaning that no causal relationship between BP and stress perception can be confirmed. Furthermore, as this study involved employees of only one company, its results cannot directly be extrapolated to other employed populations. However, the consistency of our findings with the existing literature on the influence of the single risk factors stress perception and BP on WAI suggests that our observation of an interaction between the two may be relevant to other populations as well.

In this study, the reported frequency of BP did not differ significantly across the occupational groups, except for ‘reduced activity in job’ and ‘duration of reduced activity in the 12 past months’. This could be due to physical loads encountered at work; for example, heavy lifting occurs more commonly among FO than managerial staff, although stress perception was more frequently reported among managerial staff [12]. Nevertheless, a strong correlation between BP and reported stress, unrealistic job demands and work–life imbalance was found. This accords with the findings of a previous study that a combination of high physical workload and high mental demand was predictive for low BP with an OR of 4.34 (95% CI 1.43–13.15) [22]. Our study also supports the findings of other studies that not only physical but also high mental demands may trigger the onset of BP [23,24].

WAI, an established tool for the assessment of a worker’s ability to cope with the demands of work [17], was defined as an outcome measure in this study. We found an association between stress perception and WAI in our previous study [12], in agreement with results in diverse occupational groups [10,11]. In addition, a consistent relationship between more severe BP and decreased WAI was found in this study. The strongest association was found for items such as the duration of BP within the last 12 months, reduced work activity due to BP and the duration of reduced activity in the last 12 months. These findings are in accordance with a study by Monteiro and Alexandre [4] showing that the presence of BP affected several aspects of work ability.

Furthermore, an interaction between stress perception and BP (i.e. a synergy effect on WAI) was studied. A synergy index >1 implies that the presence of both factors exceeds the sum of individual effects of both factors. In addition to the sum score of WAI, a detailed analysis of each individual dimension was performed to achieve more insight into the underlying construct of WAI in the workforce. This relationship was consistently observed in each dimension, except for Dimension 3, which measures the number of known diagnoses. The lack of interaction is therefore not surprising because of the objective character of this dimension. Dimension 7 works out the mental resource, which is strongly correlated with the presence of stress individually and finally exhibits a synergy effect.

A healthy workforce is the basis for productivity and profitability, and preventive measures to maintain work ability have become an important component of workplace health promotion. Prevention of BP is an important issue in view of the projected ageing of the workforce in Germany in coming years [25]. Occupational health also has to confront psychosocial risks in a changing world if it is to protect the health of employees.

Based on the results of this study, the employer has already adopted corporate health promotion activities. Questionnaires on BP, stress perception and work ability are incorporated into routine occupational medical surveillance examinations and regular health checks. Within these assessments, BP and job-related stress are addressed individually, and personal advice on further intervention is provided.

Key points
  • Back pain is still the leading cause of sickness absence, job-related stress has an impact on productivity and work ability has been demonstrated to have a significant predictive value for disability and early retirement.

  • Within our study population, back pain and job-related stress were inter-related, and both had an independent and possibly synergy effect on work ability.

  • Consequently, workplace health promotion interventions to reduce the impact of back pain and stress perception on work ability should target both physical and psychological dimensions.

Conflicts of interest

All the authors are employees of BASF SE, Germany, which operates the Ludwigshafen site. M.N. holds stock in BASF.

References

1.
Meyer
M
Weirauch
H
Weber
F
.
Krankheitsbedingte Fehlzeiten in der deutschen Wirtschaft im Jahr 2011
.
Fehlzeiten Report 2012
 
2012
:
291
467
.
2.
Hoy
D
Brooks
P
Blyth
F
Buchbinder
R
.
The epidemiology of low back pain
.
Best Pract Res Clin Rheumatol
 
2010
;
24
:
769
781
.
3.
Heymans
MW
van Buuren
S
Knol
DL
Anema
JR
van Mechelen
W
de Vet
HC
.
The prognosis of chronic low back pain is determined by changes in pain and disability in the initial period
.
Spine J
 
2010
;
10
:
847
856
.
4.
Monteiro
MS
Alexandre
NM
.
Work ability and low back pain among workers from a public health institution
.
Rev Gaucha Enferm
 
2009
;
30
:
297
302
.
5.
Burton
WN
Pransky
G
Conti
DJ
Chen
CY
Edington
DW
.
The association of medical conditions and presenteeism
.
J Occup Environ Med
 
2004
;
46
:
38
45
.
6.
Leineweber
C
Westerlund
H
Hagberg
J
Svedberg
P
Alexanderson
K
.
Sickness presenteeism is more than an alternative to sickness absence: results from the population-based SLOSH study
.
Int Arch Occup Environ Health
 
2012
;
85
:
905
914
.
7.
Hartvigsen
J
Lings
S
Leboeuf-Yde
C
Bakketeig
L
.
Psychosocial factors at work in relation to low back pain and consequences of low back pain; a systematic, critical review of prospective cohort studies
.
Occup Environ Med
 
2004
;
61
:
e2
.
8.
Lang
J
Ochsmann
E
Kraus
T
Lang
JW
.
Psychosocial work stressors as antecedents of musculoskeletal problems: a systematic review and meta-analysis of stability-adjusted longitudinal studies
.
Soc Sci Med
 
2012
;
75
:
1163
1174
.
9.
Kikuchi
S
.
New concept for backache: biopsychosocial pain syndrome
.
Eur Spine J
 
2008
;
17
:
421
427
.
10.
Sampaio
RF
Coelho
CM
Barbosa
FB
Mancini
MC
Parreira
VF
.
Work ability and stress in a bus transportation company in Belo Horizonte, Brazil
.
Cien Saude Colet
 
2009
;
14
:
287
296
.
11.
Bethge
M
Radoschewski
FM
Müller-Fahrnow
W
.
Work stress and work ability: cross-sectional findings from the German sociomedical panel of employees
.
Disabil Rehabil
 
2009
;
31
:
1692
1699
.
12.
Yong
M
Nasterlack
M
Pluto
RP
Lang
S
Oberlinner
C
.
Occupational stress perception and its potential impact on work ability
.
Work
 
2013
;
46
:
347
354
.
13.
Alavinia
SM
de Boer
AG
van Duivenbooden
JC
Frings-Dresen
MH
Burdorf
A
.
Determinants of work ability and its predictive value for disability
.
Occup Med (Lond)
 
2009
;
59
:
32
37
.
14.
Béjean
S
Sultan-Taïeb
H
.
Modeling the economic burden of diseases imputable to stress at work
.
Eur J Health Econ
 
2005
;
6
:
16
23
.
15.
Portaluppi
F
Touitou
Y
Smolensky
MH
.
Ethical and methodological standards for laboratory and medical biological rhythm research
.
Chronobiol Int
 
2008
;
25
:
999
1016
.
16.
Kuorinka
I
Jonsson
B
Kilbom
A
et al
.
Standardised Nordic questionnaires for the analysis of musculoskeletal symptoms
.
Appl Ergon
 
1987
;
18
:
233
237
.
17.
Tuomi
K
Ilmarinen
J
Jahkola
A
Katajarinne
L
Tulkki
A
.
Work ability index
.
Helsinki, Finland: Finnish Institute of Occupational Health
 ,
1994
.
18.
Karazman
R
Kloimüller
I
Karazman-Morawetz
I
.
From work ability index to human work index
. In:
People and Work. Research Reports
 .
Helsinki, Finland: Finnish Institute of Occupational Health
,
2002
.
19.
Rothman
KJ
(ed.).
Modern Epidemiology
 .
Boston, MA
:
Little, Brown and Company
,
1986
.
20.
Zou
GY
.
On the estimation of additive interaction by use of the four-by-two table and beyond
.
Am J Epidemiol
 
2008
;
168
:
212
224
.
21.
Skakon
J
Kristensen
TS
Christensen
KB
Lund
T
Labriola
M
.
Do managers experience more stress than employees? Results from the Intervention Project on Absence and Well-being (IPAW) study among Danish managers and their employees
.
Work
 
2011
;
38
:
103
109
.
22.
Pluto
R
Nolting
HD
Nasterlack
M
Zober
A
.
Risk factors for low back pain – results of a questionnaire survey at BASF Aktiengesellschaft, Ludwigshafen. PREMUS 2004
.
Fifth International Scientific Conference on Prevention of Work-Related Musculoskeletal Disorders
.
Zurich, Switzerland,
2004
.
23.
Macfarlane
GJ
Pallewatte
N
Paudyal
P
et al
.
Evaluation of work-related psychosocial factors and regional musculoskeletal pain: results from a EULAR Task Force
.
Ann Rheum Dis
 
2009
;
68
:
885
891
.
24.
Mendelek
F
Kheir
RB
Caby
I
Thevenon
A
Pelayo
P
.
On the quantitative relationships between individual/occupational risk factors and low back pain prevalence using nonparametric approaches
.
Joint Bone Spine
 
2011
;
78
:
619
624
.
25.
Buck
H
Kistler
E
Mendius
HG
.
Demographic Change in the World of Work
 .
Stuttgart
,
Germany
:
ISSA
,
2002
.