Purpose Only a limited number of studies have investigated return to work of sick-listed workers with mental health problems, and more knowledge is needed about the influence of non-disorder-related factors. This study aimed to identify longitudinal associations between demographic, personality, disorder-related and work-related characteristics and sustainable return to work of sick-listed workers with a depressive or anxiety disorder. Methods We used data of a large Dutch cohort study to prospectively study longitudinal associations between biopsychosocial factors and sustainable return to work in 2 years. Associations were studied by means of univariable and multivariable logistic regression analysis. Participants who were sick-listed at baseline and had a lifetime diagnosis of a depressive and/or anxiety disorder were included in this study (N = 215). Results In 2 years, 51.6 % of the participants returned to work sustainably. Age, household income, extraversion, employment status, skill discretion and job security were significantly (P a parts per thousand currency sign 0.05) associated with sustainable RTW in 2 years in the univariable analyses. The multivariable analysis revealed significant associations between sustainable return to work and age (OR per 10 years = 0.67; 95 % CI 0.47-0.95), household income (OR per 100 Euro's a month = 1.04; 95 % CI 1.00-1.08) and being on sickness benefit versus being (self-)employed (OR 0.39; 95 % CI 0.20-0.77). Conclusions In the long-run not disorder-related factors, but an older age, the absence of a job and a low household income seem to complicate return to work. Policy and research should focus on facilitators and barriers for return to work of workers with these characteristics.