Objectives Investigate the incidence and risk factors of sickness absence (SA) among Brazilian public workers.
Method It is a dynamic cohort composed by all public workers of the Goiania municipality hired from January, 2005 to December, 2011 where certified SA data were analysed. Incidence density (ID) was calculated per 1.000 person-years in each ICD-10 category. We used multiple logistic regression analysis to estimate the association between SA and socioeconomic and occupational characteristics.
Results 18.450 workers contributed to 49.410 06 person- years and 7.044 (38.2%) employees had at least one certified episode of SA. The ID of sickness absence was 142.5 per 1.000 person-years. This rate was significantly higher among women (155.5). Diagnostic groups with the highest sick leave ID were for injury (32.3) and musculoskeletal diseases (15.3). The logistic analysis showed association between SA and the female gender (OR 1.46 CI95% 1.35–1.58), low education (OR 1.34 CI95% 1.18–1.52), low salary (OR 1.82 CI95% 1.61–2.05), multiple work contracts (OR 1.56 CI95%1.46–1.66) and was greater among professionals in the health field (OR 1.73 CI95% 1.55–1.92).
Conclusions On average, 14% of workers received a SA certificate each year, and the socioeconomic and occupational characteristics were associated with higher rate of certification. This study suggests an increased risk according to the professional category, the field of education was more associated with mental disorders, blue collar with musculoskeletal disorders and inspection with injuries. The findings provide relevant information for disease’s prevention and health promotion policies with priority to the most vulnerable occupational groups.