Self-rated work ability as a risk factor for disability retirement

Abstract

Background: Simple and efficient survey measures to predict staying in or leaving work are needed. We examined the association of single-item self-rated work ability (SRWA) with disability retirement in two large populationbased samples and compared the association of SRWA to two other scales, work ability score (WAS) and self-rated health (SRH), used earlier in studies. Methods: The study population comprised 6034 participants aged 35-58 from the population-based Health 2000 and FinHealth 2017 cohort studies, pooled together. SRWA, WAS and SRH were all classified in three categories: poor, limited and good. A 36-month follow-up for disability retirement via linkage to electronic records was included in the analysis. Results: Of the participants, 195 retired during the follow-up. All three measures strongly predicted disability retirement. Hazard ratio (HR) for poor SRWA (vs. good) was 8.48 [95% confidence interval (CI) 5.41-13.28], WAS 7.99 (95% CI 5.62-11.37) and SRH 5.96 (95% CI 4.17-8.51). HR for limited SRWA (vs. good) was 4.35 (95% CI 3.21-5.91), WAS 3.54 (95% CI 2.49-5.04) and SRH 2.27 (95% CI 1.59-3.23). Taking into account gender, age, education and mental health narrowed the gap between poor and limited vs. good work ability as predictors of disability retirement, but the differences remained clear. Conclusions: Limited or poor self-rated work ability or health are strong predictors of disability retirement. The SRWA measure is a useful survey-measure of work ability in community-based surveys.