Objective: Little is known about how exclusion criteria applied in randomized controlled trials (RCTs) affect the transfer of psychotherapy outcome research to naturalistic settings. This study evaluated the effects of naturalistic depression therapies and benchmarked them with published RCTs.
Method: Commonly used exclusion criteria were applied to n=338 depressive patients receiving cognitive-behavioral therapy. Outcomes of the resulting subsample eligible for RCTs were compared to those reported in RCTs.
Results: Treatment outcomes of the total sample (d=1.16) and the subsample eligible for RCTs (d=1.15) were highly similar. Therapy outcome was worse than in high-quality RCTs (d=1.39).
Conclusions: No systematic bias was demonstrated due to patient selection criteria that are typically applied in RCTs. The comparability of psychotherapies conducted in RCTs and in real-world settings might be underestimated. Conclusions concerning the improvement of therapies in naturalistic settings are discussed.