No definitive or gold standard outcome measure exists to test the efficacy of the mental disorder treatments examined within randomized controlled trials. As a result, researchers often evaluate efficacy via multiple outcome measures administered within a single controlled trial. This practice commonly yields inconsistent findings as to a treatment's efficacy. To address the issue of inconsistent findings, increasingly (and paradoxically) controlled trials include designations of a single measure as a primary outcome and other measures as secondary outcomes. In this paper, we review recent work highlighting the limitations of this approach to testing efficacy. In discussing how these limitations outweigh the strengths of the primary outcome method, we argue that this method needs to be replaced with an approach that addresses its limitations. In doing so, we outline the basic principles of a research agenda to develop such a replacement approach. The approach (Standardized Replication Rate [SRR] Approach) would focus on the extent to which multiple outcome measures within a controlled trial yield replicable effects, relative to the characteristics of the outcome measures and the treatment(s) examined within the trial. A research agenda focused on developing the SRR Approach would increase accountability for both reporting and interpreting controlled trials evidence.
- Controlled trial
- Primary outcome measure
- Standardized replication rate