How to solve the problem of low power in comparative outcome studies in psychotherapy? Commentary on Flückiger et al. (2026)

As illustrated in the paper by Flückiger and colleagues in this issue, non-inferiority margins for randomized controlled trials comparing different bona fide psychotherapies are highly relevant. A major problem of this type of research is, however, the low statistical power of almost all trials. Even meta-analyses cannot solve this issue. In this Commentary we illustrate the problem of low power in these trials, and provide some solutions, including different types of power calculations. The main solution is, however, to convince funders that comparative outcome trials can be one of the most important ways to understand who benefits from which treatment and that large trials are very much needed.