AIM: To compare levels of self-efficacy among the general working population and employees with sickness absence from work, and to examine if general self-efficacy measured before occurrence of sickness absence predicted subsequent onset of sickness absence and Return-to-Work. METHODS: The study follows a cohort of 5357 working employees and 106 long-term sickness absent employees in Denmark. They were interviewed in 2000 regarding self-efficacy and various co-variates, and followed for 78 weeks in a national sickness absence register. Cox regression analysis was performed in order to assess the effect of self-efficacy on Return-to-Work after sickness absence. RESULTS: General self-efficacy was significantly lower among those with sickness absence compared to the general working population. Self-efficacy showed no statistically significant association with later onset of sickness absence or with Return-to-Work. CONCLUSION: The results may suggest that lower self-efficacy among employees with sickness absence is a result of the sickness absence itself rather than a precursor of it. This indicates a need to investigate the potential change in self-efficacy in relation to the employee's change in labor market status; this will help to focus Return-to-Work interventions where planning has to be attentive towards the change in self-efficacy that can occur after onset of disease and sickness absence.