A new study published in JAMA Psychiatry (Orsini et al., 2026) delivers a sobering assessment of 112 randomized clinical trials (RCTs) on psilocybin, LSD, ketamine, MDMA, and DMT. This methodological review – a high-quality synthesis rather than a single trial – found that only 29.5% of these 'gold standard' studies even assessed blinding integrity. When assessed, blinding failed dramatically: more than 90% of the time in psilocybin, LSD, and DMT trials, 85% for MDMA, with ketamine performing somewhat better (17.9% failure) due to midazolam active placebos. The University Health Network researchers correctly conclude that results should be viewed skeptically. Yet the original MedicalXpress coverage stops short of connecting this to broader patterns of bias, financial conflicts, and historical parallels that threaten to repeat pharmaceutical missteps.

What the coverage misses is the compounding effect of unblinded therapists delivering the accompanying psychotherapy, creating double bias. Expectancy effects are powerful: participants enter trials inundated with positive media narratives about the 'psychedelic renaissance,' inflating perceived benefits beyond any pharmacological action. This aligns with a 2023 critical review in Frontiers in Psychiatry (Szigeti et al.) that used self-reported data from psilocybin trials to demonstrate that breaking blind strongly predicted positive outcomes, independent of actual drug dosage. Similarly, the 2024 FDA advisory committee rejection of Lykos Therapeutics' MDMA-assisted therapy for PTSD cited functional unblinding, inadequate controls for psychotherapy fidelity, and potential therapist allegiance effects – issues the JAMA paper reinforces but does not fully explore.

Conflicts of interest further erode trust. Many leading psychedelic researchers maintain equity stakes or advisory roles with companies like Compass Pathways, Atai Life Sciences, and MAPS Public Benefit Corporation. While the Orsini paper itself appears free of direct industry funding, the broader field shows patterns where positive trial results align with commercial interests. This mirrors early SSRI antidepressant research in the 1990s, where publication bias and weak blinding produced inflated efficacy claims later tempered by larger independent meta-analyses.

The ketamine exception is instructive: its shorter duration and availability of sedating placebos allow marginally better masking, yet even these trials often overlook long-term outcomes and dissociation effects that can unmask treatment. As regulatory pathways accelerate – with FDA breakthrough designations and state-level decriminalization efforts – insufficient mainstream skepticism risks widespread adoption of therapies whose effect sizes may be largely placebo-driven. Genuine innovation in trial design, such as using neurofeedback placebos or AI-assisted blinding assessment, remains underdeveloped.

This evidence demands tempered enthusiasm. While preliminary signals for treatment-resistant depression and PTSD warrant continued investigation, the current evidence base, compromised by these flaws, does not yet justify the scale of investment and policy shifts underway. Mental health treatments require rigorous, replicable science – not another cycle of hype, premature rollout, and eventual disillusionment.