Comment on: Predicting return to work among patients with colorectal cancer

Editor We read the den Bakker et al.1 article with great interest. The study analysed two prediction models for return to work (RTW) with 1 year and 2 years after the start of sick leave as the outcomes. Here, we offer some discussion points regarding their conclusions. We agree with the authors’ 1and 2-year RTW outcomes because the Netherlands’ social system guarantees that 70 per cent of employees’ wages are paid during the first 2 years of sick leave. Previous studies have reported that social support influenced RTW2,3 and, therefore, a country’s social welfare system might strongly influence the results. We believe that this information would be very helpful for developing a study protocol and that bias could be minimized by setting a point in the outcome measure that considers the social system. The authors appropriately chose the models’ predictors of RTW by relying on the results of previous studies; however, we propose that some other predictors might be almost as important. Previous studies have reported that type of treatment (surgery, radiation therapy, chemotherapy, etc.), financial factors and the severity of symptoms are important predictors of RTW3–5, and we would therefore include those variables. Additionally, the authors indicated that their data were poor quality because they were not originally collected for the research, and might have lacked some possible predictors of RTW. We hope that a prospective observational study based on a study protocol is conducted in the future.