Objectives To examine: (1) the relationships between aerobic capacity, work ability and sick leave; (2) the potential mediating effect of work ability in the relationship between aerobic capacity and sick leave; and (3) the influence of age on these relationships.

Methods Information on aerobic capacity (predicted VO_2max), age, gender, type of work, cardiovascular risk and body mass index was collected from 580 workers at baseline. Work ability was assessed with the Work Ability Index at first follow-up (mean 3.4±1.3 years after baseline). The second follow-up period was defined as the time between completing the Work Ability Index and the first registered sick leave episode. Mediation analyses were performed using linear and Cox regression models.

Results A lower aerobic capacity was found to be significantly related to sick leave (HR=0.98; τ=−0.018; 95% CI 0.970 to 0.994). There was a significant positive relationship between aerobic capacity and work ability (α=0.165; 95% CI 0.122 to 0.208). Also, lower work ability was significantly related to sick leave after controlling for aerobic capacity (HR=0.97; β=−0.033; 95% CI 0.949 to 0.987). The mediating effect of work ability in the relationship between aerobic capacity and sick leave was −0.005 (SE=0.002), and mediated 27.8% (95% CI 10.4 to 45.2) of the total effect of aerobic capacity on sick leave. Age did not influence the relationship between aerobic capacity and sick leave.

Conclusions Fit workers had better work ability, and both fit workers and workers with higher work ability were at lower risk of starting an episode of sick leave.