BACKGROUND Despite suggestions that worker perception might be the best predictor of return to work (RTW), there still is limited research on time to RTW in workers with lengthy non-work-related sick leave. METHODS Prospective cohort study of 663 workers with a current long-term non-work-related sick leave episode recruited during the first medical visit in a mutua (Spanish health insurance company) and followed until their sick leave episode ended. Workers completed a baseline questionnaire regarding their perceptions of sick leave episode and expectations of RTW (i.e., health status, work ability, expectations and time required to RTW, self-efficacy and self-perceived connection between health and job). Time to RTW was established based on the mutua's register. Cox regression models were used to examine the associations of worker perception and expectation of RTW with time to RTW within the study population as a whole as well as in three diagnostic subgroups (i.e., musculoskeletal disorders, mental disorders and other physical conditions). RESULTS As a whole, time to RTW was longer for workers reporting poor health [hazard ratio (HR) = 0.71, 95%CI 0.59-0.85], extremely reduced work ability (HR = 0.69, 95%CI 0.53-0.88), a longer period of time required to RTW (HR = 0.36, 95%CI 0.25-0.52) and lack of expectation of returning to the same job (HR = 0.13, 95%CI 0.06-0.31). Workers with musculoskeletal and other physical conditions showed a similar pattern to whole study population, while workers with mental disorders did not. CONCLUSIONS Self-required time and RTW expectations are important prognostic factors in sick listed workers by all types of health conditions certified as non-work-related. Questioning the workers on their perceptions and expectations of RTW during medical visits could help health care professionals to identify individuals at risk of long-term sickness absence and facilitate triage and management of the patient.