OBJECTIVES: To compare the length of nonwork-related sick leave among cases managed by an insurance company versus those managed by the National Institute of Social Security (NISS). METHODS: We performed a retrospective cohort study of 289,686 cases of sick leave lasting for more than 15 days that began in 2005 after certification by a primary care physician in Catalonia, were reported to the Catalonian Institute of Medical Evaluations, and were followed to term. Of the total, 156,676 cases were managed by the NISS. To account for repeat episodes (approximately 25% of the total), the Wang-Chang estimator was used to calculate the median duration and percentiles; comparisons were made using log-logistic regression with shared gamma frailty models, with calculation of time ratios (TR) and their corresponding 95% confidence intervals (95% CI). RESULTS: The median duration of sick leave was 43 days for cases managed by the NISS and 39 days for those managed by the insurance company. This difference was statistically significant both for men employed under contract (TR=0.87; 95% CI: 0.85-0.88) and for those who were self-employed (TR=0.78; 95% CI: 0.75-0.80) as well as for women under contract (TR=0.85; 95% CI: 0.84-0.87) and self-employed women (TR=0.84; 95% CI: 0.81-0.88). These differences persisted after adjustment was performed for age and health region. CONCLUSIONS: For sick leave lasting more than 15 days, these results confirm that cases managed by an insurance company ended earlier than for those managed by the NISS, both for contract and self-employed workers. Further research is needed to explore the reasons for these differences.

Este trabajo ha contado con la financiación de una beca (I. Torá) para la realización de un máster en Salud Laboral, concedida mediante convocatoria pública, fruto de un convenio entre la Universitat Pompeu Fabra y el Institut Català d’Avaluacions Mèdiques, con la finalidad exclusiva de financiar esta beca.