This commentary critiques a published analysis by Williams et al. that compared efficacy data from (mostly double-blind) randomized clinical trials (RCTs) of psychedelic therapy (PT) against open-label trials (OLTs) of traditional antidepressants (TADs), arguing the comparison rests on flawed methodological assumptions.
The authors identify multiple systematic biases favoring TADs in Williams et al.: (1) pretrial expectancy does not predict PT response (unlike TADs), undermining the core rationale for the comparison; (2) TAD OLTs selectively recruited known TAD responders, sometimes represcribing the same medication; (3) recent TAD discontinuation (≤1 week washout) can worsen symptoms by up to 6 points on the 17-item Hamilton Depression Rating Scale; (4) TAD OLTs allowed concurrent psychotherapy, anxiolytics, and financial incentives; and (5) per-protocol analysis (more common in TAD trials) combined with higher TAD dropout rates would further inflate TAD effect sizes.
This is a commentary/letter with no new primary data; claims are based on the authors' interpretation of existing literature and the Williams et al. article, which is not fully reproduced here.
Clinicians interpreting head-to-head comparisons of psychedelic therapy vs. antidepressants should scrutinize whether trial designs are truly comparable — particularly blinding status, washout periods, patient selection, and analytic approach — before drawing conclusions about relative efficacy.
Explore related topics
