Publication bias in psilocybin for depression: The real effect is smaller than you think

A 2026 study published in PsyArXiv suggests that the published effects of psilocybin on depression may be exaggerated by selective reporting, with the real impact being significantly smaller than previously thought. This isn’t just about numbers — it’s about how science gets told, and how the truth gets buried in the noise.

Bottom line: A new meta-analysis of psilocybin trials for depression found moderate evidence of publication bias, suggesting the real effect is about 41% smaller than what’s been reported — and the results may be influenced by expectancy-driven artefacts rather than selective reporting.

How a single study can shape a whole field

In 2023, a living systematic review in Nature Mental Health concluded that psilocybin reduced depression symptoms across 12 randomized controlled trials, with an overall effect size of g = 0.90. This is a large effect — comparable to the impact of antidepressants in some trials. The original analysis used Egger’s test to assess publication bias, but it came up non-significant. However, Egger’s test has limited power when there are few studies, and it can’t produce a bias-corrected estimate. That’s where the new study comes in.

The researchers re-analyzed the same dataset using Robust Bayesian Meta-Analysis (RoBMA), a method that simultaneously estimates the treatment effect, heterogeneity, and publication bias. What they found was striking: moderate evidence for publication bias (Bayes Factor = 3.89). After adjusting for selective reporting, the bias-corrected effect size dropped to g ≈ 0.53 — about 41% smaller than the original estimate. The credible interval for this corrected effect spanned to zero, meaning the real effect could be close to nothing.

This isn’t a complete reversal of the original conclusion — the study doesn’t say psilocybin is ineffective. Rather, it suggests that the original analysis underestimated the uncertainty, and that the published results may be inflated. The researchers also note that this evidence of bias disappears when open-label studies are excluded, suggesting that the effect might be driven by placebo expectations rather than the actual properties of psilocybin.

Why publication bias matters — and how it hides the truth

Publication bias is the tendency for studies with positive results to be more likely to be published than those with negative or neutral findings. It’s a silent force that shapes the scientific narrative without anyone noticing. In the case of psilocybin for depression, the original meta-analysis suggested a strong effect, but the new study shows that this may not be the whole story.

The Bayes Factor of 3.89 indicates that the evidence for publication bias is moderate — not overwhelming, but enough to raise questions. The key insight here is that the original analysis didn’t account for the possibility that some studies were never published, or that some were selectively reported. The Bayesian approach, by contrast, allows for a more nuanced understanding of how much the results might have been exaggerated.

What’s even more interesting is that the evidence for bias disappears when open-label studies are excluded. Open-label trials are those where both participants and researchers know who is receiving the active treatment. These studies are often criticized for being less rigorous, but they also tend to be more susceptible to placebo effects. The fact that the bias disappears when these are excluded suggests that the observed effect might be more about the expectations of participants and researchers than about the actual properties of psilocybin.

The bigger picture: How psilocybin stacks up against other treatments

To put this in context, it’s useful to compare psilocybin with other antidepressants. A 2023 systematic review and network meta-analysis of 69 randomized controlled trials found that six treatments had a higher response rate than placebo: electroconvulsive therapy (ECT), minocycline, theta-burst stimulation (TBS), repetitive transcranial magnetic stimulation (rTMS), ketamine, and aripiprazole. The odds ratios for these treatments ranged from 1.9 to 12.86 — with ketamine and ECT showing the strongest effects.

In comparison, the bias-corrected effect size for psilocybin (g ≈ 0.53) is lower than the average effect size for antidepressants in some trials. However, it’s important to remember that these are very different treatments — psilocybin is not a traditional antidepressant, and its mechanism of action is still not fully understood. The fact that it shows a moderate effect, even after accounting for bias, suggests it might have a unique place in the treatment landscape.

But the real takeaway is that the original estimates of psilocybin’s effectiveness may be inflated. This is not to say that psilocybin is ineffective — just that the evidence is less certain than previously thought. And that’s a crucial distinction in science.

Frequently asked questions

Q: Does this mean psilocybin is not effective for depression?
A: No. This study does not say that psilocybin is ineffective. It suggests that the published effect size may be exaggerated, and that the true effect could be smaller than what’s been reported.

Q: Why is the effect size so much smaller after adjusting for bias?
A: Because the original analysis didn’t account for publication bias. The Bayesian approach allows for a more accurate estimate of the true effect by considering the possibility that some studies were never published or selectively reported.

Q: What does it mean when the credible interval spans to zero?
A: It means that the real effect could be close to zero. The study suggests that the published effect size may be an overestimate, and that the true effect is less certain than previously thought.

Q: Why did the evidence for bias disappear when open-label studies were excluded?
A: Because open-label studies are more susceptible to placebo effects. The fact that the bias disappears when these are excluded suggests that the observed effect might be more about expectations than about the drug itself.

Q: What should we take away from this study?
A: That the published evidence for psilocybin’s effectiveness may be exaggerated. This doesn’t mean psilocybin is ineffective, but it does mean we need to be cautious about interpreting the results as definitive.

Sources

• ¶Publication bias in randomized trials of psilocybin for depression: a Robust Bayesian Meta-Analysis
• ¶Relative effectiveness of antidepressant treatments in treatment-resistant depression: a systematic review and network meta-analysis of randomized controlled trials
• ¶Current practices in modelling dose-response relationships in network meta-analyses of depression treatments: a meta-research study

Educational Disclaimer

This article is for informational and educational purposes only. It is not medical advice, mental health advice, diagnosis, treatment guidance, or a recommendation to use any substance, supplement, therapy, or protocol.

We review publicly available research and explain what the evidence may suggest. Some studies may be early-stage, observational, animal-based, lab-based, theoretical, or incomplete. Always consult a qualified professional before making health-related decisions.

If you or someone you know is struggling, you are not alone. In the US, call or text 988 (Suicide & Crisis Lifeline). Elsewhere, contact your local emergency or crisis service.

Researched and drafted by Spore, ShroomWire’s AI research assistant, and reviewed by the ShroomWire editorial team before publishing.