There is evidence that exposure and response prevention (ERP) is efficacious in the treatment of obsessive-compulsive disorder (OCD). As an alternative to ERP metacognitive therapy (MCT) was developed. Two previous randomized controlled trials (RCTs) did not find significant differences between MCT and ERP. However, from non-significant results, non-inferiority of a treatment cannot be concluded. For this purpose, non-inferiority studies are required. Exner and colleagues carried out such a non-inferiority study whose results were recently published in this journal. The authors concluded from their results that MCT is a viable alternative treatment with efficacy similar to the standard ERP. However, this study raises several concerns, among them problems of transparency and of non-inferiority testing. These issues are critically discussed here in more detail. Taking all of these issues into account, the conclusions that can be drawn from the available studies are less clear. Further research is needed to decide whether MCT can really be considered as non-inferior to ERP or even as efficacious at all. Future studies need to fulfill the criteria of non-inferiority trials, that is (a) a priori define and (b) empirically justify a non-inferiority margin, (c) a preregistered sample size calculation ensuring a sufficient statistical power to confirm non-inferiority of the test treatment and (d) include a non-active control condition against which the standard and the test treatment are tested. Recommending a treatment prematurely as non-inferior to a standard treatment may prevent patients from receiving the most efficacious treatment.