Sick leave diagnoses and return to work: a Swedish register study.

PURPOSE: To provide a detailed description of return to work (RTW) for different diagnoses.
METHODS: A sample of 617,611 cases of sick leave with minimum duration of 14 days started in 2009 and 2010 was followed for 450 days. Data on sick leave episodes, diagnosis (ICD-10 codes on 3-digit level) and background factors were retrieved from Swedish National Social Insurance registers. RTW was analysed with Cox proportional hazard regression with separate analyses for women and men.
RESULTS: The diseases with the lowest RTW rates were present within malign neoplasms, severe mental disorders and severe cardiovascular diseases. High-RTW rates were present for infectious respiratory diseases, viral infections and less severe cardiovascular diseases. There were distinct differences across diagnoses within all diagnosis chapters. This also holds for mental disorders and musculoskeletal diseases, which are the most common causes of sick leave. The results for women and men across different diseases and disorders were strikingly similar.
CONCLUSIONS: Systematic use of standard information such as detailed sick listing diagnosis could, at low-additional cost, pinpoint cases at risk for prolonged sick leave. The use of broad diagnosis categories in sick leave research may be questioned since there are distinct differences in RTW within ICD-10 chapters.