[The analysis of sickness absence incidence by subgroups of duration]

We describe the usefulness of performing an analysis of sickness absence (SA) incidence, segmented by the duration of episodes, in comparison with the more common analysis that considers the overall incidence of SA without segmentation.

We used data from a health insurance company ("mutua") in Spain during 2011 (230,332 episodes, 752,906 workers) and non-work related SA as a case study. We compared the overall incidence of SA and incidence segmented by duration of episodes: short (≤ 15 days), medium (16-30 and 31-90 days) and long (> 90 days). The analyses were also performed by age, as an example of one of the multiple variables that affect SA incidence.

The overall incidence of SA was 30.6%, and declined steadily with increasing age. When SA incidence was analyzed by duration, we observed that: 1) the incidence of the episodes of short duration is the highest; 2) the overall excess observed in younger workers (<25 years) is driven mainly by short duration and 3) the pattern for long-term SA incidence was reversed, being more frequent among those ≥55 years of age relative to the youngest.

Examining SA incidence by duration is more informative than relying on overall incidence of SA.