Work ability score and future work ability as predictors of register-based disability pension and long-term sickness absence: A three-year follow-up study

• Published in: Scandinavian journal of public health 2018-05, Vol.46 (3), p.321-330
• Main Authors: KINNUNEN, ULLA, NÄTTI, JOUKO
• Format: Article
• Language: English
• Subjects: Work disability
• ISSN: 1403-4948; 1651-1905
• DOI: 10.1177/1403494817745190

Aims: We investigated two single items of the Work Ability Index – work ability score, and future work ability – as predictors of register-based disability pension and long-term sickness absence over a three-year follow-up. Methods: Survey responses of 11,131 Finnish employees were linked to pension and long-term (more than 10 days) sickness absence register data by Statistics Finland. Work ability score was divided into poor (0–5), moderate (6–7) and good/excellent (8–10) and future work ability into poor (1–2) and good (3) work ability at baseline. Cox proportional hazard regressions were used in the analysis of disability pension, and a negative binomial model in the analysis of long-term sickness absence. The results were adjusted for several background, work- and health-related covariates. Results: Compared with those with good/excellent work ability scores, the hazard ratios of disability pension after adjusting for all covariates were 9.84 (95% CI 6.68–14.49) for poor and 2.25 (CI 95% 1.51–3.35) for moderate work ability score. For future work ability, the hazard ratio was 8.19 (95% CI 4.71–14.23) among those with poor future work ability. The incidence rate ratios of accumulated long-term sickness absence days were 3.08 (95% CI 2.19–4.32) and 1.59 (95% CI 1.32–1.92) for poor and moderate work ability scores, and 1.51 (95% CI 0.97–2.36) for poor future work ability. Conclusions: The single items of work ability score and future work ability predicted register-based disability pension equally well, but work ability score was a better predictor of register-based long-term sickness absence days than future work ability in a three-year follow-up. Both items seem to be of use especially when examining the risk of poor work ability for disability but also for long sick leave.