Introduction Within the occupational health setting, somatoform disorders are a frequent cause of sick leave. Few validated screening questionnaires for these disorders are available. The aim of this study is to validate the PHQ-15 in this setting. Methods In a cross-sectional study of 236 sicklisted employees, we studied the performance of the PHQ-15 in comparison with the Mini International Neuropsychiatric Interview (MINI) as golden reference standard. We approached employees who were sick listed for a period longer than 6 weeks and shorter than 2 years for participation. This study was conducted on one location of a large occupation health service in the Netherlands, serving companies with more than 500 employees. All employees who returned the PHQ-15 were invited for the MINI interview. Specificity and sensitivity were calculated for optimal cut point and a receiver operating characteristic (ROC) was constructed. Results A total of 107 participants consented to participate in the MINI interview. A non-response analysis showed no significant differences between groups. According to the MINI, the prevalence of somatoform disorders was 21.5%, and the most frequent found disorder was a pain disorder. The PHQ-15 had an optimal cut point of 9 (patients scoring 9 or higher (≥9) were most likely to suffer from a somatoform disorder), with specificity and sensitivity equal to 61.9 and 56.5%, respectively. ROCs showed an area under the curve (AUC) of 0.63. Conclusion The PHQ-15 shows moderate sensitivity but limited efficiency with a cut point of 9 and can be a useful questionnaire in the occupational health setting.