Skip Navigation


Occupational Medicine Advance Access originally published online on December 10, 2008
Occupational Medicine 2009 59(1):32-37; doi:10.1093/occmed/kqn148
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
59/1/32    most recent
kqn148v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Alavinia, S. M.
Right arrow Articles by Burdorf, A
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Alavinia, S. M.
Right arrow Articles by Burdorf, A
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2008. Published by Oxford University Press on behalf of the Society of Occupational Medicine. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Determinants of work ability and its predictive value for disability

S. M. Alavinia1, A. G. E. M. de Boer2, J. C. van Duivenbooden3, M. H. W. Frings-Dresen2 and A Burdorf1

1 Department of Public Health, University Medical Center Rotterdam, Erasmus MC, The Netherlands
2 Coronel Institute of Occupational Health, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands
3 Arbouw Foundation, Amsterdam, The Netherlands

Correspondence to: A. Burdorf, Department of Public Health, Erasmus MC, PO Box 2040, 3000 CA Rotterdam, The Netherlands. Tel: +31 10 7038469; fax: +31 10 7038474; e-mail: a.burdorf{at}erasmusmc.nl


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 Conflicts of interest
 References
 
Background Maintaining the ability of workers to cope with physical and psychosocial demands at work becomes increasingly important in prolonging working life.

Aims To analyse the effects of work-related factors and individual characteristics on work ability and to determine the predictive value of work ability on receiving a work-related disability pension.

Methods A longitudinal study was conducted among 850 construction workers aged 40 years and older, with average follow-up period of 23 months. Disability was defined as receiving a disability pension, granted to workers unable to continue working in their regular job. Work ability was assessed using the work ability index (WAI). Associations between work-related factors and individual characteristics with work ability at baseline were evaluated using linear regression analysis, and Cox regression analysis was used to evaluate the predictive value of work ability for disability.

Results Work-related factors were associated with a lower work ability at baseline, but had little prognostic value for disability during follow-up. The hazard ratios for disability among workers with a moderate and poor work ability at baseline were 8 and 32, respectively. All separate scales in the WAI had predictive power for future disability with the highest influence of current work ability in relation to job demands and lowest influence of diseases diagnosed by a physician.

Conclusion A moderate or poor work ability was highly predictive for receiving a disability pension. Preventive measures should facilitate a good balance between work performance and health in order to prevent quitting labour participation.

Keywords      Construction workers; disability; predictive value; work ability


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 Conflicts of interest
 References
 
There is ample evidence that construction workers have a higher risk of work-related disability than workers in less physically demanding jobs [1–4]. Manual handling, awkward postures and repetitive movements are core determinants of physical load in the construction industry. With an ageing workforce, the proportion of construction workers at risk for disability is expected to increase in the near future [5]. Therefore, prevention of disability will become even more important in the construction industry.

In order to prevent workers from quitting the workforce due to (work-related) disability, the concept of measuring work ability has been developed as a valuable tool to tailor interventions at individual level. In order to assess work ability, Finnish researchers have constructed the work ability index (WAI). This combines subjective experiences of one's ability to cope with physical and mental requirements at work (performance at work) with information on diseases and consequent functional limitations, sick leave and mental resources at work (enjoying daily tasks, active and vital life, optimism about the future) [6]. The index is sensitive to changes in work conditions, health status and physical fitness [7].

A study by Finnish researchers on work ability has demonstrated clear effects of work-related physical and psychosocial factors and individual characteristics on work ability [8]. However, these studies have not addressed which aspects of work ability are affected most by work-related risk factors and individual characteristics. In addition, the relative contribution of performance, health and mental resources in the predictive power of work ability for future disability is largely unknown. This is clearly important for prioritizing interventions in the workplace. Thus, the aims of this study were to analyse the effects of work-related physical and psychosocial factors and individual characteristics on work ability and to determine the predictive value of work ability and its underlying dimensions for future work-related disability.


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 Conflicts of interest
 References
 
A prospective longitudinal study was performed on male Dutch construction workers who had participated in voluntary periodic occupational health examination and were not (partially) disabled, retired or working outside the construction industry at the time of the examination. The initial selection started with a random sample of 1000 workers aged 40 years and over, who had participated in the examination during a 3-month period from September 2002 until November 2002. An additional selection of 195 workers was based on workers who started a counselling and education programme during the period April to December 2002. The aim of the programme was the enhancement of individual work ability and comprised of an assessment interview, additional assessment of labour market capabilities, devising an individual education programme, executing the individual education programme and a follow-up activity. Since the source of the study population did not affect the statistical analysis due to the fact that this programme did not have effects on future work disability [9], this population was added to the current study population.

During the medical examination, a questionnaire was filled out on individual characteristics, working conditions, health problems and work ability. In February 2003, workers were sent a first follow-up questionnaire to their home address, on average ~5 months after attending the periodic health examination (follow-up 1). Respondents to this first questionnaire were sent two further follow-up questionnaires at fixed time intervals of ~9 months (follow-up 2 and 3), resulting in an average follow-up period of ~23 months. All questionnaires contained questions on work ability as well as work status including change of job to a job outside the construction industry, partial and full disability pension or (early) retirement.

In each questionnaire, workers were asked about their current economic status with five mutually exclusive categories: paid work, retired, unemployed, disabled or other. During the study period, disability was defined by receiving a formal disability pension which is granted to a worker who has been on sick leave for 52 consecutive weeks and whose functional limitations are too severe to be able to continue in their regular job. The eligibility criteria for such a disability pension, as stated in national legislation, further stipulated that a substantial loss of income must be demonstrated as a result of the fact that the worker cannot perform paid employment at all or holds a new, less strenuous job with a much lower salary.

The WAI was measured during the medical examination as well as determined in each follow-up questionnaire [8,10]. The answers on each scale were translated into a weighted score and the index was derived as the sum of these scores. The range of this summative index is 7–49, which is classified into poor (7–27), moderate (28–36), good (37–43) and excellent (44–49) work ability [10]. For the purpose of this study, the original seven scales were also used as separate dependent variables in the analysis.

The questionnaire during the medical examination comprised questions on physical load and psychosocial factors at work. Physical load at work was determined by dichotomous questions on regular exposure to awkward postures, kneeling or squatting, manually handling of materials, whole-body vibration and hand–arm vibration [11]. A sum score across these five questions was also calculated (0–5) in order to dichotomize around the median value the study population into high and low physical load. Psychosocial factors were measured according to the demand–control model defined by Karasek on job demands, skill discretion and job control [12]. An abbreviated version of the original questionnaire was used and job demands were assessed by means of a seven-item sum score, skill discretion by a six-item sum score and job control by a five-item sum score [13]. The median of each sum score was used to dichotomize the study population into workers with a high and a low psychosocial load at work.

During the medical examination, information on age and job title was collected by questionnaire. The information on job title was used to categorize subjects into white and blue-collar workers. Blue-collar workers were all workers who performed manual labour at a construction site and white-collar workers were all office workers and supervisors with primarily managerial tasks. The questionnaire during the medical examination also consisted of questions on mental and physical health. Mental health problems were ascertained using 11 questions on the presence (yes/no) of fatigue, sleep disturbance, gloomy feeling, nervousness, irritability, stress, being excited, memory and concentration problems and depression. These questions were largely similar to the EURO-D scale for depression diagnosis which defines a clinically relevant depression by a sum score >3 [14,15]. The sum score was used to dichotomize subjects to poor (>3 positive answers) and good mental health. Aspects of physical health were ascertained by dichotomous questions on the presence of regular chest pain, regular shortness of breath and four questions on regular pain in the back, neck, upper and lower extremities. The latter four questions were used to define the presence of musculoskeletal problems.

Multiple linear regression models were used to analyse the associations between individual characteristics and work-related risk factors with work ability at baseline and with its separate scales. In each model, a backward selection approach was used with a P-value of 0.10 or less for the initial selection of relevant variables, and only variables statistically significant at P < 0.05 were retained in the final models.

Kaplan–Meier curves were produced to describe the trend over time in proportion of workers without a disability pension relative to the time since inclusion in the cohort. The analysis was stratified by a poor, moderate and good/excellent work ability measured at baseline. A Cox regression analysis was performed with the hazard ratio (HR) as measure of association to study the relation between work ability and disability. The HR in a survival analysis is an approximation of the relative risk, expressing the likelihood that a particular risk factor predicts disability relative to the likelihood of disability in the absence of this risk factor. Disability pension, loss to follow-up because of early retirement and moving to another job outside the construction industry were considered censoring. All statistical analyses were carried out with SAS 8.2 and SPSS 15 statistical software package.


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 Conflicts of interest
 References
 
From this initial study population of 1195 subjects, two subjects were excluded due to an incomplete medical examination, eight were excluded due to receiving a partial disability pension and 31 were excluded as they were female (a too small group for analysis), resulting in 1154 eligible participants.

The response to the first follow-up questionnaire was 74% (n = 850), with non-response (n = 303) partly due to 41 (14%) incomplete questionnaires. The response to the second and third questionnaires, relative to the response to the first follow-up questionnaire, was 70% (n = 592) and 69% (n = 583). In total, 450 workers completed all three questionnaires. Loss-to-follow-up (n = 267) was partly determined by workers changing job to other parts of industry (n = 30) and workers taking (early) retirement (n = 13).

Table 1 shows the baseline characteristics of the study population. The mean WAI was 38.7 and the proportion of workers with a poor or moderate work ability was 5% and 24%, respectively. The seven scales within the WAI were highly correlated, with highest associations between scales 1 and 2 (r = 0.59), scales 2 and 7 (r = 0.43), scales 1 and 4 (r = 0.41) and scales 2 and 6 (r = 0.41). The average score on the scale on current diseases was 56% of the maximum score, whereas on all other scales the average score varied from 73 to 86% of the maximum score.


View this table:
[in this window]
[in a new window]

  		Table 1. Characteristics of the 850 workers who participated in a voluntary medical examination at the start of the study

 
The loss to follow-up during the consecutive measurements was not related to work ability, but respondents in the first and second follow-up reported a slightly higher physical load, lower work demands, lower job control and less skill discretion than construction workers who dropped out of the study.

In the univariate analyses, all work-related factors were significantly associated with a lower work ability at baseline (Table 2). In the multivariate model, all factors except job type remained statistically significant, albeit with a lower magnitude of the regression coefficient. These factors explained 15% of the variability in the WAI at baseline. Similar analyses for each dimension of the WAI showed that the highest age group had statistically significantly lower scores on all scales, except sick leave and mental resources. Work-related risk factors were more strongly associated with the two scales on work ability compared with the three scales on health.


View this table:
[in this window]
[in a new window]

  		Table 2. Effects of work-related risk factors and individual characteristics on the WAI in the 850 workers at the start of the study

 
During the follow-up period (with a mean of 22 months), 40 persons became partially or fully work disabled and were granted a disability pension. Figure 1 presents the Kaplan–Meier curve describing the proportion of workers without disability relative to time since follow-up for three categories of work ability. From the original population of 850 workers, 7 of 606 (1%) workers with a good/excellent work ability, 19 of 204 (9%) with a moderate work ability and 14 of 40 (35%) with a poor work ability at baseline became disabled. Table 3 shows that the workers between 45 and 50 years were more likely to become disabled compared with workers under 45 years (HR 3.1, 95% CI 1.3–7.5), but the risk on disability was no longer statistically significant among workers aged 50 year and over (HR 1.6, 95% CI 0.7–3.8). A poor and moderate work ability score were highly predictive for becoming disabled with HRs of 32 and 8, respectively. Adjustment for age did not influence the predictive power to a large extent. Interaction terms between work-related factors and work ability score were not statistically significant. The analyses for each separate scale of the WAI showed statistically significant HRs for all scales, with the highest predictive value for work ability in relation to demands at work (HR = 1.96 per point decrease) and the lowest predictive value for current diseases diagnosed by a physician (HR = 1.37 per point decrease).


Figure 1
View larger version (16K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Survival time to becoming partially or fully disabled among construction workers, stratified by category of WAI.

 

View this table:
[in this window]
[in a new window]

  		Table 3. Univariate and multivariate Cox regression analysis on prognostic factors for becoming work disabled, during an average follow-up period of 23 months

 

   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
Conflicts of interest
 References
 
This study found that the self-reported work ability of construction workers aged 40 and over strongly predicted receiving a disability pension during the 23-month follow-up period. Current work performance, health problems and associated consequences for functioning and sick leave and mental resources were all important predictive factors. Work-related physical and psychosocial factors were associated with a lower work ability at baseline, but had limited predictive value for work disability during the follow-up.

The strength of this study is its longitudinal design. However, a limitation is the substantial dropout during follow-up measurements, since only 49% of the selected construction workers completed the last questionnaire. The loss to follow-up was not influenced by work ability, but respondents in the first and second follow-up reported a slightly higher occurrence of physical and psychosocial factors at work than those workers who dropped out of the study. This differential selection in self-reported exposure may have resulted in some underestimation of the predictive value of work-related factors for disability.

A second limitation of the study is the use of self-reported disability pension as a proxy for work-related disability. The eligibility criteria for a disability pension only partly depended on a deterioration in health affecting the worker's ability to perform his regular job. The assessment of loss of income associated with the level of disability is also part of the formal decision process. In addition, we did not investigate other mechanisms of displacement from work, such as unemployment, retirement or change of job to other parts of industry, due to the small number of these events in the study population.

At baseline, age and work-related factors determined 15% of the variation in work ability among workers, which is in agreement with studies in other occupational groups [16–21]. It has to be noted that in this study, the best part of the variability in work ability could not be explained by physical and psychosocial workload. An important reason is most likely the crude assessment of workload based on dichotomous parameters. Another explanation is the importance of variables not accounted for in this study, most notably physical activity in leisure time [17,19], mental stress [18] and body mass index [19,22].

During the follow-up period, the average work ability changed little and the duration of follow-up may have been too short to notice a substantial decrease in work ability. In a randomized controlled trial on a physical activity programme, the work ability also remained stable during the 24 months of follow-up [23]. In addition, Tuomi et al. [24] have shown in an 11-year follow-up study that both the improvement and decline in work ability were associated more strongly with changes in work and lifestyle during the follow-up than with their initial variation.

Our study has confirmed that a poor or moderate work ability predicts the risk of a work-related disability pension [10]. The fact that work-related risk factors were associated with work ability at baseline, but not predictive for disability during the follow-up period, suggests that physical and psychosocial factors at work are especially important in the balance between physical and mental requirements of the job and the capabilities of the worker. Every separate scale of the WAI showed a statistically significant HR, and the overall index was highly predictive for future work disability. Age has been acknowledged as a risk factor for work-related disability [25]. The fact that the HR for receiving a disability pension among workers aged higher than 50 was not statistically significant in this study might partly reflect a healthy worker survival effect.

Given the predictive power of the WAI for work-related disability, it offers a suitable framework for preventive interventions. These interventions should address the determinants of work ability, such as physical and psychosocial factors at work. However, the low explained variance in the analysis of determinants of work ability at baseline suggests that interventions on working conditions may have only a modest impact on work ability in the short term. As a consequence, interventions may need to focus more on the imbalance between health problems and associated functional limitations and an individual's capabilities to cope with the physical and mental requirements of work rather than addressing working conditions or health independently.

In conclusion, among construction workers a moderate or poor work ability was highly predictive for receiving a disability pension. Preventive measures should facilitate a good balance between work performance, health and mental resources in order to prevent quitting labour participation.


Key points
  • Among construction workers, self-reported work ability proved to be a strong risk factor for future work-related disability.
  • Current work performance, health problems and associated consequences for functioning and sick leave and mental resources were important prognostic factors for future disability.
  • Work-related physical and psychosocial factors were associated with lower work ability at baseline, but had limited predictive value for future work disability during the follow-up.

 


   Conflicts of interest
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conflicts of interest
References
 
None declared.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conflicts of interest
 References
 

  1. Arndt V, Rothenbacher D, Daniel U, et al. Construction work and risk of occupational disability: a ten year follow up of 14,474 male workers. Occup Environ Med (2005) 62:559–566.[Abstract/Free Full Text]

  2. Turner JA, Franklin G, Turk DC. Predictors of chronic disability in injured workers: a systematic literature synthesis. Am J Ind Med (2000) 38:707–722.[CrossRef][Web of Science][Medline]

  3. Guberan E, Usel M. Permanent work incapacity, mortality and survival without work incapacity among occupations and social classes: a cohort study of ageing men in Geneva. Int J Epidemiol (1998) 27:1026–1032.[Abstract/Free Full Text]

  4. Brenner H, Ahern W. Sickness absence and early retirement on health grounds in the construction industry in Ireland. Occup Environ Med (2000) 57:615–620.[Abstract/Free Full Text]

  5. Ilmarinen J. Aging workers. Occup Environ Med (2001) 58:546–552.[Free Full Text]

  6. De Zwart BC, Frings-Dresen MH, van Duivenbooden JC. Test-retest reliability of the Work Ability Index questionnaire. Occup Med (Lond) (2002) 52:177–181.[CrossRef][Medline]

  7. Tuomi K, Ilmarinen J, Klockars M, et al. Finnish research project on aging workers in 1981–1992. Scand J Work Environ Health (1997) 23(Suppl. 1):7–11.

  8. Ilmarinen JK, Tuomi K, Klockars M. Changes in the work ability of active employees over an 11-year period. Scand J Work Environ Health (1997) 23(Suppl. 1):49–57.[Web of Science][Medline]

  9. De Boer AGEM, Burdorf A, Van Duivenbooden C, et al. The effect of individual counseling and education on work ability and disability pension: a prospective intervention study in the construction industry. Occup Environ Med (2007) 64:792–797.[Abstract/Free Full Text]

  10. Liira J, Matikainen E, Leino-Arjas P, et al. Work ability of middle-aged Finish construction workers—a follow-up study in 1991–1995. Int J Ind Ergon (2000) 25:477–481.[CrossRef][Web of Science]

  11. Elders LA, Burdorf A. Interrelations of risk factors and low back pain in scaffolders. Occup Environ Med (2001) 58:597–603.[Abstract/Free Full Text]

  12. Karasek R. Job demands, job decision latitude, and mental strain: implication for job redesign. Adm Sci Q (1979) 24:285–308.[CrossRef][Web of Science]

  13. Pelfrene E, Vlerick P, Mak R, et al. Scale reliability and validity of Karasek ‘Job Demand-control-support’ model in Belstress. Work Stress (2001) 15:297–331.[CrossRef][Web of Science]

  14. Copeland JR, Beekman AT, Dewey ME, et al. Depression in Europe. Geographical distribution among older people. Br J Psychiatry (1999) 174:312–321.[Abstract/Free Full Text]

  15. Borsch-Supan AB, Jurges H, Mackenbach JP, et al. Health, Ageing and Retirement in Europe. First Results from the Survey of Health, Ageing and Retirement in Europe (2005) Mannheim: Mannheim Research Institute for the Economic of Ageing (MEA).

  16. Tuomi K, Eskelinen L, Toikkanen J, et al. Work-load and individual factors affecting work ability among aging municipal employees. Scand J Work Environ Health (1991) 17(Suppl. 1):128–134.[Web of Science][Medline]

  17. Pohjonen T. Perceived work ability of home care workers in relation to individual and work-related factors in different age groups. Occup Med (Lond) (2001) 51:209–217.[CrossRef][Medline]

  18. Sjögren-Rönkä T, Ojanen MT, Leskinen E, et al. Physical and psychosocial prerequisites of functioning in relation to work ability and general subjective well-being among office workers. Scand J Work Environ Health (2002) 28:184–190.[Web of Science][Medline]

  19. Tuomi K, Huuhtanen P, Nykyri E, et al. Promotion of work ability, the quality of work and retirement. Occup Med (Lond) (2001) 51:318–324.[CrossRef][Medline]

  20. Berg van den T, Alavinia SM, Bredt FJ, et al. The influence of psychosocial factors at work and life style on health and work ability among professional workers. Int Arch Occup Environ Health (2008) 81:1029–1036.[CrossRef][Web of Science][Medline]

  21. Alavinia SM, Duivenbooden van JC, Burdorf A. Influence of work-related factors and individual characteristics on work ability among Dutch construction workers. Scand J Work Environ Health (2007) 33:351–357.[Web of Science][Medline]

  22. Laitinen JS, Nayha S, Kujala V. Body mass index and weight change from adolescence into adulthood, waist-to-hip ratio and perceived work ability among young adults. Int J Obes (2005) 29:697–702.[CrossRef][Web of Science][Medline]

  23. Smolander J, Blair SN, Kohl HW. Work ability, physical activity, and cardiorespiratory fitness: 2-year results from Project Active. J Occup Environ Med (2000) 42:906–910.[CrossRef][Web of Science][Medline]

  24. Tuomi K, Ilmarine J, Aalto L, et al. Aging, work, life-style and work ability among Finnish municipal workers in 1981–1992. Scand J Work Environ Health (1997) 23(Suppl. 1):58–65.[Web of Science][Medline]

  25. Bergh H, Baigi A, Månsson J, et al. Predictive factors for long-term sick leave and disability pension among frequent and normal attenders in primary health care over 5 years. Public Health (2007) 121:25–33.[CrossRef][Web of Science][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
59/1/32    most recent
kqn148v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Alavinia, S. M.
Right arrow Articles by Burdorf, A
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Alavinia, S. M.
Right arrow Articles by Burdorf, A
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?