[Predictors of the duration of non-work-related sick leave due to anxiety disorders].

Objectives: To analyze the duration of non-work-related sick leave due to anxiety disorders and to identify demographic, occupational and clinical variables that may contribute to its prediction.

Methods: We performed a prospective cohort study of 1,161 workers with an episode of non-work-related sick leave due to an anxiety disorder, belonging to the insured population of a mutual insurance company. We assessed the duration of non-work-related sick leave episodes and the main potentially related demographic, occupational and clinical variables. All non-work-related sick leave processes were followed-up until discharge. Cox regression analyses were conducted to establish the predictors of non-work-related sick leave duration.

Results: The median duration of non-work-related sick leave due to anxiety disorders was 83 days. In a multivariate analysis, the following factors were identified as being significantly associated with increases in the duration of sick leave (p <0.05): age of over 35 years, lower educational level (primary school studies, secondary school studies or high-school diploma vs. university degree), and the existence of comorbidity and unemployment occurring during the sick leave. In contrast, being separated or divorced was associated with an earlier return to work (p <0.05).

Conclusions: Anxiety disorders are associated with long periods of non-work-related sick leave compared with other disorders and standard time duration. Demographic, occupational and clinical variables collected at the initial assessment of the sick leave episode would help to identify groups with an increased risk of prolonged sick leave, requiring strategies to facilitate return to work.