A study suggests that psychedelics are no more effective than antidepressants and confirms that the studies are not double-blind

The article by Williams et al. analyzes the published scientific evidence through a rigorous meta-analysis on the treatment of major depression in outpatient settings, comparing the efficacy of commonly used conventional antidepressant drugs in clinical practice—such as selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine reuptake inhibitors (SNRIs)—with so-called psychedelic-assisted psychotherapy (PAP). This is not a common practice in Spain, but it has had strong support for more than five decades in other places, such as the United States, and has re-emerged in recent years as a novel treatment for depression. This type of therapy uses different psychoactive substances as tools to support psychological intervention, including amphetamine derivatives, LSD, plant and fungal alkaloids from Latin America (psilocybin, mescaline, peyote), and ayahuasca.

Previous studies had shown greater efficacy of PAP compared to placebo, but this article suggests that this difference may be partly explained by factors such as patient expectations or a reduced placebo response in psychedelic trials. This highlights—consistent with other recent work—the challenges in clinical trial design and bias control. For example, it is known that PAP trials often include more highly educated patients, who may be more receptive to these therapies (particularly in the United States), and that ethnic and racial minorities are underrepresented, which can bias results in favor of PAP. It could even be argued that psychedelic medicine receives more positive media coverage than antidepressant drugs, meaning that without proper double-blind design, results may be skewed toward greater apparent efficacy due to higher expectations surrounding PAP.

For all these reasons, this meta-analysis addresses the issue by comparing the effectiveness of both approaches under equivalent methodological conditions, in adult patients with major depression without comorbidity treated in outpatient settings. Overall, it is a study of good methodological quality, following PRISMA guidelines, and it combines Bayesian and frequentist statistical models, reinforcing its transparency and scientific rigor. It includes 24 clinical trials: 16 open-label antidepressant trials (7,921 patients) and 8 PAP trials (249 patients), of which only 2 are open-label (36 patients) and the other 6 use blinded methodology (213 patients). In addition, the authors explicitly account for several confounding factors, especially the issue of functional unblinding, which is particularly relevant in psychedelic studies due to the obvious subjective and adverse effects of these substances—even in formally blinded trials. This phenomenon can amplify differences between drug and placebo effects and effectively makes all PAP studies functionally open-label. Differences in baseline depression severity, treatment duration, and population type were also considered.

However, several important limitations remain, such as the small number of psychedelic trials and their heterogeneity: small and sometimes unrepresentative sample sizes, as well as methodological differences across studies included in the meta-analysis (e.g., follow-up duration or inclusion criteria). Moreover, the analysis focuses solely on the reduction of depressive symptoms, without considering key aspects such as adverse effects or long-term functioning. Finally, the authors relied exclusively on the PubMed database for their literature search, which, although comprehensive, may not include all relevant trials. All of this should be taken into account when interpreting the results.

Despite these limitations, the study’s conclusions appear to be supported by consistent data, confirming the absence of significant differences in depression improvement (using the 17-item Hamilton Depression Rating Scale and standardizing results across studies) between PAP and antidepressant use. This finding is particularly relevant because it contradicts the dominant narrative in the United States suggesting a clear and overly optimistic superiority of psychedelic-assisted therapy. In this regard, the authors convincingly argue that many previous studies may have overestimated the efficacy of PAP due to issues with study design and the integrity of blinding.

In terms of real-world clinical practice, this work suggests that traditional antidepressant drugs remain a valid first-line treatment option. The use of psychedelics may be promising, but there is still insufficient evidence to consider them superior to conventional antidepressant treatment. More robust studies are needed, with better control of blinding and bias, and with evaluation of clinically relevant long-term outcomes. Regarding existing evidence, this study does not deny that psychedelics may be effective, but it does question their apparent advantage over conventional treatments. This is highly relevant in a context of growing media and clinical interest in psychedelics, as it cautions against drawing premature conclusions. Additionally, in the Spanish context, the issue is less pressing, as the use of psychedelic-assisted psychotherapy remains very limited (although some clinical trials have been initiated), given that many of the substances involved are regulated and illegal in the country.

This systematic review and meta-analysis evaluates the effectiveness of psychedelic-assisted therapy (PAT) compared to traditional antidepressants (TAD) for the treatment of major depression. The authors argue that, because the intense subjective effects of psychedelics compromise blinding, PAT results should be compared with antidepressant trials conducted under open-label conditions, in order to match the degree of unblinding across interventions.

Under this approach, the study finds that the superiority of PAT over traditional medication disappears when both treatments are administered in conditions where patients are aware of the intervention received. Likewise, the results suggest that the integrity of blinding significantly influences the reported outcomes of traditional antidepressants, but not those of PAT, pointing to a relevant methodological asymmetry between the two types of intervention.

In this context, the authors propose that part of the apparent superiority of psychedelics could be explained by a reduction in the placebo response in control groups (the “know-cebo” effect), rather than by a greater intrinsic therapeutic effect. This is a methodologically rigorous, preregistered, and comprehensive study that addresses one of the main challenges in psychedelic research: the difficulty (possibly insurmountable) of implementing conventional double-blind designs due to the nature of their subjective effects.

In this sense, the findings reinforce the need to explore alternative designs, such as pragmatic, real-world, or mechanism-focused studies, as well as to interpret the current evidence on their clinical efficacy with caution.

However, it is important to highlight a relevant limitation in the comparability of the included samples: while studies on traditional antidepressants mostly focus on patients with non-resistant depression, psychedelic trials predominantly include populations with treatment-resistant depression. This clinical heterogeneity introduces a potential bias that complicates direct comparisons between interventions, as these populations differ in prognosis and expected treatment response. Furthermore, it is difficult to compare or generalize results across different psychedelic studies due to inconsistencies in the psychotherapeutic approaches and psychological support used, as well as inefficient and inconsistent reporting of the context in which they are applied—pointing to the need for improved reporting of these aspects of the interventions.

Is the study of good quality? Are the conclusions supported by solid data?

“Yes, it is a systematic review and meta-analysis which, according to the hierarchy of scientific evidence, sits at the very top. Meta-analyses are statistical analyses that combine all the studies conducted on a given topic.”

How does this work fit with the existing evidence?

“It fits perfectly, as it evaluates the current evidence for treating depression with psychedelic substances. What’s novel is that it tempers the existing overenthusiasm about the benefits of psychedelic therapies compared to standard treatments.”

Have the authors taken confounding factors into account? Are there important limitations to consider?

“They have considered all possible confounding factors arising from the idiosyncrasies of clinical trials, such as the shorter duration of psychedelic studies, the lack of demographic diversity among participants, the variety of scales used across studies, and the issue of blinding (i.e., how easy it is to detect placebo in clinical trials).

In my opinion, there is one limitation they do not address: namely, that the limitations of these studies stem from epistemological issues rather than purely methodological ones. The methods are sound, but they reflect a bias in trying to fit the efficacy of psychedelics into a clinical trial model. Clinical trials are designed to evaluate drugs where there is a one-to-one relationship between treatment and disease—the prototypical case being infectious and viral diseases. This is why it is called the bacteriological model.

Trying to fit psychological treatments, such as psychedelic-assisted therapies—where the psychological experience is fundamental—into a bacteriological model introduces a foundational bias. Another bias relates to the rigidity of protocols. Psychedelic-assisted therapy involves administering the psychedelic when it is appropriate within the course of a psychological treatment. The clinical trial model, however, centers on the drug rather than the process, as it arguably should not.

There is also a final, unavoidable bias: placebo-controlled clinical trials aim to eliminate the effect of expectation. But expectation is precisely one of the strongest predictors of therapeutic success. Patients need to expect improvement in order to improve. If expectation is removed, one of the most powerful elements of psychological treatment is also removed.

So, although the study is methodologically impeccable, there is a foundational epistemological bias that makes it difficult to extrapolate the results to real clinical practice.”

What are the real-world implications?

“In real-world settings, clinicians work with patients’ expectations, the therapeutic alliance, and a process-focused approach rather than one centered solely on the drug. Unfortunately, depression studies—unlike those on post-traumatic stress disorder with MDMA—tend to focus heavily on eliminating expectation rather than on the therapeutic process.

Finally, the reasons why depression has been chosen as the primary focus for psychedelic treatments are not based on clear evidence. Psychedelics appear to work better for reactive conditions, such as existential distress in people with serious illness, grief, and other disorders related to life circumstances, rather than for more characterological issues.”