Objective To evaluate the association between return to work self-efficacy and psychosocial factors at work among workers on long term sickness absence due to mental disorders.

Methods A cross-sectional study was conducted in São Paulo, Brazil, in 2014. Participants were 202 workers claiming for a social security benefit due to mental sickness and out of work for more than 15 days. Diagnoses were defined according to the report of medical experts who evaluate the claim for benefit. The participants answered a comprehensive questionnaire on sociodemographics and occupational features. Psychosocial factors at work were evaluated using the Effort-Reward Imbalance (ERI) questionnaire. The expectation about work after sick leave was assessed by the Return to work self-efficacy (RTW-SE) questionnaire. A multiple logistic regression was used to evaluate the association between RTW-SE and ERI dimensions.

Results Most participants were women (71.3%), mean age of 35.9 years (SD + 9.1), 12 or more years of education (50.5%), diagnosed as depression (52.5%), perception of ERI at work (55.9%), low overcommitment (62.9%) and low RTW-SE (51.0%). The final model showed the association between RTW-SE and psychosocial factors at work – ERI (OR 2.05; 95% CI: 1.10–3.85) and high overcommitment (OR 2.65; 95% CI: 1.39–5.08). This model was adjusted by sex, age and education.

Conclusion We observed a significant association between low return to work self-efficacy and perception of occupational exposure to adverse psychosocial stressors among workers during long term sickness absence due to mental disorders. Actions to improve psychosocial work conditions could increase workers RTW-SE in episodes of disabling mental illness. Then, the time to reintegration after a sick leave episode could be decreased. Funding sources: The Brazilian National Council of Scientific and Technological Development (CNPq) (grant #442051/2014-0), and Brazilian National Social Security Institute (INSS).