Purpose To evaluate the interventional capacity of problem based method groups (PBM) regarding mental health and work ability compared to cognitive behavioural therapy (CBT) for persons on sick leave due to common mental disorders. Methods In a randomised controlled design the experimental group received PBM and the control group received CBT. Outcomes were measured by the Hospital Anxiety and Depression Scale (HADS), the Stress and Crisis Inventory 93 (SCI-93) and the Dialogue about Working Ability instrument (DOA). Results Twenty-two participants in the PBM group and 28 in the CBT group completed intervention. Both groups showed significant lower scores on the two HADS subscales. Regarding stress the PBM group showed significant decrease in one (out of three) subscales of SCI-93. The CBT group showed significant decrease on all subscales of SCI-93. Regarding work ability the PBM group showed significant higher scores on one of five subscales of DOA. The CBT group showed significant higher scores on four of five subscales of DOA. Between groups there were significant differences to the favour of CBT on one of two subscales of HADS, all three subscales of SCI-93 and on two of the five subscales of DOA. Conclusion PBM seem to be able to reduce anxiety- and depression symptoms. CBT showed to be superior to PBM in reducing symptoms in all aspects of mental health, except for anxiety, in which they seem equally effective. Regarding work ability CBT showed to be superior, with significant effect on more aspects compared to PBM.