This meta-analysis compares depression rating outcomes among patients receiving control treatments in randomized clinical trials of psilocybin vs trials of selective serotonin reuptake inhibitors (SSRIs) and esketamine for depression.
Key Points
Question
Are there differences in depression outcomes among patients who receive control treatments in trials of psilocybin, selective serotonin reuptake inhibitors (SSRIs), or esketamine for depression?
Findings
In this meta-analysis of 17 acute-phase randomized clinical trials of psilocybin (n = 373), esketamine (n = 573), or SSRIs (n = 4014), patients receiving control treatments in psilocybin trials had significantly worse depression outcomes compared with those receiving control treatments in trials of esketamine or SSRIs. There were larger between-treatment differences for psilocybin than for esketamine or SSRIs.
Meaning
The poor control treatment outcomes in psilocybin trials suggest that it may not be as broadly effective for depression as estimated.
Abstract
Importance
Psilocybin has demonstrated rapid and sustained antidepressant efficacy, with acute-phase effect sizes often more than double those for conventional antidepressants. However, concerns have been raised that high rates of functional unblinding in combination with trial participants with positive expectations of psychedelic use might bias treatment outcomes.
Objective
To compare outcomes for patients receiving control treatments in randomized clinical trials of psilocybin for depression with control treatment outcomes from trials of selective serotonin reuptake inhibitors (SSRIs) and esketamine.
Data Sources
Two previous meta-analyses and 1 US Food and Drug Administration review published between March 2019 and December 2024 were used to identify double-blind trials on adult major depressive disorder (MDD) or treatment-resistant depression (TRD) that had a relevant control treatment arm and used the Montgomery-Åsberg Depression Rating Scale (MADRS) for symptom rating.
Study Selection
Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses reporting guideline, trials of psilocybin for MDD and TRD, esketamine for TRD, and a selective serotonin reuptake inhibitor (SSRI) for MDD were selected. Studies that included only individuals aged younger than 18 years or older than 65 years, used a crossover design, or had a duration less than 2 weeks were excluded.
Data Extraction and Synthesis
All authors assessed the 3 reviews for includable trials. Three authors independently extracted data for all trials, with disagreements resolved by consensus discussion. Data were pooled using random-effects models.
Main Outcomes and Measures
Standardized mean change (SMC) in MADRS scores from baseline to up to 6 weeks after randomization was used to assess within-group effect sizes, and standardized mean difference (SMD) was used to assess between-group effect sizes. Omnibus Test of Moderators (QM) was used to test whether the study population significantly moderated effect sizes.
Results
The study included 17 trials: 4 of psilocybin (n = 373), 2 of esketamine (n = 573), and 11 of SSRIs (n = 4014). Pretreatment to posttreatment SMCs (SEMs) were 1.21 (0.15) for psilocybin, 1.28 (0.06) for SSRIs, and 1.43 (0.15) for esketamine and were 0.50 (0.15), 1.00 (0.08), and 1.12 (0.17) for their respective control treatments. Study population was a significant moderator of between-group SMDs (QM, 10.7; df, 2; P = .005) and pre– to post–control treatment SMCs (QM, 10.4; df, 2; P = .005) but not of pre– to post–active treatment SMCs (QM, 1.21; df, 2; P = .55). MADRS response rates for control treatments in SSRI trials were 14 percentage points higher than in psilocybin trials and in esketamine trials were 23 percentage points higher than in psilocybin trials. Dropout rates for psilocybin (active treatment: 10 of 186 [5%]; control: 20 of 187 [11%]) and esketamine (active treatment: 43 of 349 [12%]; control: 18 of 224 [8%]) were similar and considerably lower than for SSRIs (active treatment: 866 of 2694 [32%]; control: 467 of 1320 [35%]).
Conclusions and Relevance
In this meta-analysis of control treatment outcomes in trials of psilocybin, SSRIs, or esketamine for depression, participants receiving control treatment in psilocybin trials had significantly less improvement in depression ratings than participants receiving control treatment in trials of SSRIs or esketamine. This might indicate that psilocybin’s antidepressant efficacy is overestimated compared with that of SSRIs and esketamine.
Introduction
Antidepressant efficacy is typically established by comparing a candidate antidepressant with some control treatment for 4 to 12 weeks. The response to control treatments in depression varies but is often substantial.1 Jones and coworkers2 reported the average pretreatment to posttreatment effect size across a range of control treatments for treatment-resistant depression (TRD) to be 1.05, and Cuijpers and colleagues3 found pretreatment to posttreatment effect sizes of 0.64 for care as usual and 0.37 for waiting list controls in a meta-analysis of psychotherapies for depression.
Psilocybin is a novel candidate antidepressant with a unique mechanism of action that has demonstrated substantial acute-phase efficacy in depression, with effect sizes often more than double those for conventional antidepressants.4 However, control treatment response rates have been noted to be lower than expected in some trials.5 Psilocybin in trials for depression has been given in psychedelic doses, which risks making functional unblinding both immediate and prevalent.6 Furthermore, since administration is usually supervised, unblinding may also extend to study personnel.7 If control treatment response rates are low across the board for psilocybin trials, it could indicate that efficacy is overestimated (eg, via functional unblinding and nocebo effects).
To assess whether low control treatment efficacy is a general phenomenon in trials of psilocybin for depression, we performed an indirect comparison of outcomes on the Montgomery-Åsberg Depression Rating Scale (MADRS) for 3 trial populations: (1) psilocybin trials on major depressive disorder (MDD) and TRD, (2) esketamine trials on TRD, and (3) selective serotonin reuptake inhibitor (SSRI) trials on MDD. Since the hypothesis was that control treatment outcomes may differ systematically depending on which active drug is trialed, there was no common treatment through which indirect evidence might be contrasted (ie, inert placebo might work differently in an SSRI trial than it does in a psilocybin trial). For this reason, the analyses were undertaken using conventional meta-analysis techniques rather than using network meta-analysis.8
Methods
Data Acquisition
In this meta-analysis, double-blind trials on adult MDD or TRD with a relevant control treatment arm (inert placebo, low-dose psilocybin, or niacin) and that had used the MADRS for symptom rating were included from 3 reviews published between March 2019 and December 2024.9,10,11 We followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline. All authors assessed the 3 reviews for includable trials. F.H., E.L., and H.W.S. independently extracted data for all trials, with disagreements resolved by consensus discussion. Studies exclusively including individuals aged younger than 18 years or older than 65 years, studies using crossover designs, and studies with a duration of less than 2 weeks were excluded. Psilocybin trials were collected from a meta-analysis by Freitas and colleagues,9 esketamine trials were taken from the corresponding US Food and Drug Administration review,10 and SSRI trials were included from a meta-analysis of individual participant data by Hieronymus and colleagues.11 Data were pooled using random-effects models.
Statistical Analysis
Within-group pretreatment to posttreatment effect sizes for control treatment and active treatment arms, respectively, and between-group (active treatment vs control treatment) effect sizes were calculated for all trials. Within-group effect sizes were calculated as standardized mean change (SMC) in MADRS scores using raw score standardization.12 Since SDs are usually larger at the end point than at baseline13 and since between-group effect sizes are normalized to the pooled end point SD, SMCs were normalized to the end point SD of the corresponding group. For 1 psilocybin study,14 baseline MADRS scores were not available. The pretest score for that study was imputed as the mean of all other psilocybin arms.
The correlation between preintervention and postintervention measures was set to 0 for all trials. When using raw score standardization, the pretreatment to posttreatment correlation is not used for calculating the effect size, but it affects the SE of the effect size. Specifically, SEs get larger the closer the correlation coefficient gets to −1 and smaller the closer the correlation coefficient gets to 1.12 Since preintervention and postintervention scores in depression are usually positively correlated (patients who begin with high scores tend to also end with high scores),13 assuming no correlation should yield conservative estimates. This SMC specification is functionally identical to calculating the standardized mean difference (SMD) between the preintervention and postintervention measurements using either only the control treatment arm SD or only the active treatment arm SD for normalization. SMDs normalized to the pooled SD were used for between-group effect sizes.
The 3 effect sizes (active treatment SMC, control treatment SMC, and active treatment vs control treatment SMD) were then meta-regressed with study population (study of psilocybin, esketamine, or SSRI) as the sole variable. The outcomes of interest were the pooled effect sizes for each study group and the result of the Omnibus Test of Moderators (QM) (ie, whether the study population significantly moderated effect sizes) for the 3 outcomes and the percentage of variance explained (R2) by models including the study population as a variable. We also calculated pooled MADRS response rates (≥50% reduction compared with baseline) and dropout rates. These are presented descriptively.
R, version 4.3.3 (R Project for Statistical Computing) was used for all analyses. Meta-analyses were performed using the metafor package, version 4.6.0, in R, and effect sizes (SMC using raw score standardization and SMD) were calculated using the escalc function in the metafor package; analyses were run using the rma function in the metafor package. Two-sided P <.05 was considered significant.
Results
Four trials of psilocybin (n = 373),5,14,15,16 2 trials of esketamine (n = 573),17,18 and 11 trials of SSRIs (n = 4014)19,20,21,22,23,24,25,26,27,28,29 were included. Study identification and screening are detailed in Figure 1, and baseline and end point scores for all included trials are provided in Table 1.
Figure 1. Study Identification and Screening.
Table 1. Baseline and End Point MADRS Scores for Individual Studies.
| Study | Time from baseline to evaluation, wk | Participants, No. | MADRS score, mean (SD) | Years conducted | ||||
|---|---|---|---|---|---|---|---|---|
| Active treatment | Placebo | Baseline | End point | |||||
| Active treatment | Placebo | Active treatment | Placebo | |||||
| Psilocybin | ||||||||
| Carhart-Harris et al,14 2021 | 6 | 30 | 29 | 30.6 (5.7)a | 30.6 (5.7)a | 16.2 (9.3) | 23.4 (9.2) | 2019-2020 |
| Goodwin et al,16 2022 | 6 | 79 | 79 | 31.9 (5.4) | 32.7 (6.2) | 20.4 (13.3) | 26.4 (13.3) | 2019-2021 |
| Raison et al,5 2023 | 6 | 51 | 53 | 35.5 (5.7) | 35.0 (4.5) | 16.4 (13.1) | 28.2 (13.7) | 2019-2022 |
| von Rotz et al,15 2022 | 2 | 26 | 26 | 24.3 (5.1) | 24.1 (7.1) | 11.2 (9.2) | 20.7 (10.1) | 2019-2021 |
| Esketamine | ||||||||
| S300117 | 4 | 233 | 113 | 37.6 (5.2) | 37.5 (6.2) | 18.7 (14.0) | 22.7 (15.1) | 2015-2018 |
| S300218 | 4 | 116 | 111 | 37.0 (5.7) | 37.3 (5.7) | 16.2 (13.3) | 20.5 (13.1) | 2015-2017 |
| SSRI | ||||||||
| GSK/00119 | 6 | 25 | 25 | 29.9 (4.5) | 30.1 (4.2) | 17.1 (7.0) | 19.2 (7.4) | 1984-1985 |
| GSK/00220 | 6 | 170 | 171 | 27.9 (5.2) | 28.2 (5.3) | 16.5 (10.0) | 21.4 (10.0) | 1985-1987 |
| GSK/00321 | 6 | 241 | 244 | 31.1 (5.3) | 31.4 (5.4) | 20.1 (10.4) | 25.7 (10.2) | 1985-1986 |
| GSK/00922 | 6 | 421 | 53 | 28.7 (6.1) | 27.5 (6.5) | 17.3 (10.2) | 17.8 (10.4) | 1985-1986 |
| GSK/78523 | 6 | 403 | 105 | 31.1 (4.8) | 31.2 (4.8) | 17.9 (10.8) | 19.9 (10.5) | 2001 |
| LB/8930325 | 6 | 134 | 66 | 33.0 (7.4) | 33.3 (8.3) | 15.9 (13.7) | 17.6 (12.9) | 1989-1990 |
| LB/8930626 | 6 | 185 | 88 | 32.0 (7.0) | 33.1 (6.8) | 15.0 (12.5) | 17.6 (14.1) | 1989-1991 |
| LB/9120627 | 6 | 521 | 129 | 27.5 (5.6) | 27.1 (6.0) | 15.7 (10.4) | 17.7 (10.2) | 1992-1993 |
| LB/9900128 | 6 | 188 | 189 | 29.2 (4.2) | 28.7 (3.7) | 14.6 (8.9) | 18.0 (9.2) | 1999-2000 |
| LB/9900329 | 6 | 314 | 154 | 29.1 (4.2) | 28.7 (4.0) | 15.9 (8.5) | 17.7 (9.2) | 1999-2000 |
| PZ/88124 | 6 | 92 | 96 | 27.9 (3.5) | 27.2 (3.9) | 14.7 (10.0) | 16.1 (9.2) | 1989-1990 |
Abbreviations: GSK, GlaxoSmithKline; LB, H. Lundbeck; MADRS, Montgomery-Åsberg Depression Rating Scale; PZ, Pfizer; SSRI, selective serotonin reuptake inhibitor.
Mean and SD imputed as the arithmetic mean from the other psilocybin and psilocybin control treatment arms.
Pretreatment to posttreatment effect sizes (SMC [SEM]) were 1.21 (0.15) for psilocybin, 1.43 (0.15) for esketamine, and 1.28 (0.06) for SSRIs and were 0.50 (0.15), 1.12 (0.17), and 1.00 (0.08), respectively, for their corresponding control treatments (Figure 2). Pretreatment to posttreatment SMC differences were thus 0.71 for psilocybin, 0.29 for esketamine, and 0.28 for SSRIs, which corresponded with the actual between-group effect sizes (SMD [SEM]): 0.70 (0.12), 0.30 (0.12), and 0.27 (0.05), respectively. Study population (psilocybin, SSRIs, or esketamine) was a significant moderator of between-group effect sizes (QM, 10.7; df, 2; P = .005) and of pre– to post–control treatment effect sizes (QM, 10.4; df, 2; P = .005) but not of active treatment effect sizes (QM, 1.21; df, 2; P = .55). Models including study population as a moderator explained 40.9% of the variance (R2) in pre– to post–control treatment outcomes, 34.8% of the variance in between-group outcomes, and 0% of the variance in pre– to post–active treatment outcomes.
Figure 2. Within- and Between-Group Effect Sizes for Psilocybin, Selective Serotonin Reuptake Inhibitors (SSRIs), and Esketamine.
Error bars indicate plus or minus 1 SE. SMC indicates standardized mean change; SMD, standardized mean difference.
Pooled response rates for active treatment arms were 89 of 186 (48%) for psilocybin, 181 of 349 (52%) for esketamine, and 1245 of 2694 (46%) for SSRIs. The corresponding control treatment response rates were 35 of 187 (19%) for psilocybin, 94 of 224 (42%) for esketamine, and 433 of 1320 (33%) for SSRIs. Control treatment response rates were thus 23 percentage points higher for esketamine than for psilocybin and 14 percentage points higher for SSRIs than psilocybin. Dropout rates in psilocybin trials (active arms: 10 of 186 [5%]; control treatment arms: 20 of 187 [11%]) were similar to those in esketamine trials (active arms: 43 of 349 [12%]; control treatment arms: 18 of 224 [8%]), whereas SSRI trials had markedly higher dropout rates (active arms: 866 of 2694 [32%]; control treatment arms: 467 of 1320 [35%]). Response and dropout rates for individual trials are provided in Table 2.
Table 2. Response and Dropout Rates for Individual Studies.
| Study | Dropout rate, No./total No. (%) | Response rate, No./total No. (%) | ||
|---|---|---|---|---|
| Active treatment | Placebo | Active treatment | Placebo | |
| Psilocybin | ||||
| Carhart-Harris et al,14 2021 | 2/30 (7) | 0/29 | 19/30 (63)a | 6/29 (21) |
| Goodwin et al,16 2022 | 5/79 (6) | 10/79 (13) | 26/79 (33) | 16/79 (20) |
| Raison et al,5 2023 | 1/51 (2) | 9/53 (17) | 29/51 (57)a | 9/53 (17)a |
| von Rotz et al,15 2022 | 2/26 (8) | 1/26 (4) | 15/26 (58) | 4/26 (15) |
| Pooled | 10/186 (5) | 20/187 (11) | 89/186 (48) | 35/187 (19) |
| Esketamine | ||||
| S300117 | 25/233 (11) | 6/113 (5) | 112/233 (48)a | 42/113 (37)a |
| S300218 | 18/116 (16) | 12/111 (11) | 69/116 (59)a | 52/111 (47)a |
| Pooled | 43/349 (12) | 18/224 (8) | 181/349 (52)a | 94/224 (42)a |
| SSRIs | ||||
| GSK/00119 | 13/25 (52) | 11/25 (44) | 8/25 (32) | 11/25 (44) |
| GSK/00220 | 84/170 (49) | 95/171 (56) | 69/170 (41) | 38/171 (22) |
| GSK/00321 | 121/241 (50) | 151/244 (62) | 87/241 (36) | 42/244 (17) |
| GSK/00922 | 180/421 (43) | 23/53 (43) | 167/421 (40) | 17/53 (32) |
| GSK/78523 | 150/403 (37) | 31/105 (30) | 179/403 (44) | 38/105 (36) |
| LB/8930325 | 34/134 (25) | 20/66 (30) | 79/134 (59) | 34/66 (52) |
| LB/8930626 | 53/185 (29) | 25/88 (28) | 110/185 (59) | 50/88 (57) |
| LB/9120627 | 168/521 (32) | 43/129 (33) | 256/521 (49) | 43/129 (33) |
| LB/9900128 | 21/188 (11) | 24/189 (13) | 99/188 (53) | 69/189 (37) |
| LB/9900329 | 13/314 (4) | 9/154 (6) | 143/314 (46) | 55/154 (36) |
| PZ/88124 | 29/92 (32) | 35/96 (36) | 48/92 (52) | 36/96 (38) |
| Pooled | 866/2694 (32) | 467/1320 (35) | 1245/2694 (46) | 433/1320 (33) |
Abbreviations: GSK, GlaxoSmithKline; LB, H. Lundbeck; PZ, Pfizer; SSRI, selective serotonin reuptake inhibitor.
Corrected for intention to treat.
Post Hoc Analysis
We originally opted to exclude 3 esketamine studies conducted in patients with depression and acute suicidality (ASPIRE 1,30 ASPIRE 2,30 and SUI200131) since there were no psilocybin or SSRI studies in that population. When we performed a separate meta-analysis of those studies as a post hoc analysis, the mean MADRS decrease from baseline for esketamine control treatment in participants with depression and acute suicidality was 22.9 points, which corresponded to an SMC (SEM) of 1.87 (0.12).
Discussion
This meta-analysis found that control treatment depression scores in psilocybin trials were lower than those usually seen in antidepressant trials2,3 and significantly lower than those for esketamine and SSRIs. In absolute terms, effect sizes in control treatment arms (0.50) in psilocybin trials were closer to those in waiting list control (0.37) and care as usual (0.64) arms in psychotherapy trials3 than to those in pill placebo arms in antidepressant trials (1.05).1,2 Conversely, pretreatment to posttreatment effect sizes in active treatment arms did not differ significantly between study populations (Figure 2). MADRS response rates for patients receiving control treatment were 14 percentage points lower in psilocybin trials than in SSRI trials and 23 percentage points lower in psilocybin trials than in esketamine trials (Table 2).
The poor performance of control treatments in psilocybin trials is likely caused by 1 of 2 factors: psilocybin trials have recruited patients who are unlikely to respond to control treatment or the psilocybin trialing process applied is less likely to induce control treatment response. In other words, assuming that the observed differences are not explained by chance, psilocybin trials must systematically differ from trials of SSRIs and esketamine either in the patient makeup or in some relevant methodologic aspect.
One potential systematic methodologic difference is that blinding is more difficult to maintain with psilocybin administration compared with SSRI or esketamine administration6,7 and that control treatment in psilocybin trials consequently functions more akin to a waiting list control than to a conventional placebo control in terms of expectancy effects. While this hypothesis aligns with a much discussed difference between the studied treatments,6,7 the analyses herein reported were indirect and, thus, can provide little guidance as to why the 3 trial populations differed. Any number of potential systematic differences (eg, regarding funding, publication year, number of study sites, inclusion and exclusion criteria, or concomitant treatments32,33,34) among the 3 studied trial populations may exist, and any one of those could be what underlies the differences in control treatment outcomes. Alternatively formulated, the present results showed that the sample average treatment effects (ATEs) differed between psilocybin control treatment and esketamine and SSRI control treatments. However, the data did not allow us to determine the specific factors that explained the observed sample ATE differences or the extent to which the sample ATEs generalized to the population ATE for the respective treatments.
Adding to the uncertainty, while unblinding has been shown to be prevalent in some psilocybin studies,7 rates of unblinding were not assessed in the included psilocybin5,14,15,16 and esketamine10 trials and likely were not assessed in any of the included SSRI studies. The extent to which unblinding occurred in the different study groups was thus not possible to discern from the available data. Esketamine, similar to psilocybin, is administered in the presence of health care professionals and has immediate adverse effects, which may lead to unblinding.10 That esketamine was associated with a degree of control treatment improvement similar to that of SSRIs thus suggests that psilocybin has a higher risk for unblinding than esketamine or that esketamine and psilocybin trials differ in some other methodologically relevant aspect.
When designing this study, we reasoned that the available trial data would not be sufficient to determine with certainty which factor or factors were causative of any putative differences in control treatment outcomes between study populations. We assumed that owing to the limited number of relevant studies available for psilocybin and esketamine, at least some potentially relevant confounders would cluster to make it impossible to separate them from one another. For example, most SSRI trials were undertaken during the 1980s and 1990s, whereas all psilocybin trials were published after 2020 (Table 1). Consequently, separating the effects of time from the effects of partaking in an SSRI trial is not possible, and the same likely holds true for many other confounders of potential importance.
While not much can thus be said about which specific factors cause the differences in control treatment outcomes, 1 factor that can be ruled out with some confidence is differential dropout.35 Esketamine trials were typically shorter than psilocybin trials and had comparable dropout rates but showed a larger control treatment response, whereas SSRI trials were of similar duration to psilocybin trials but had higher noncompletion rates; however, they also showed a larger control treatment response. Relatedly, the high SSRI dropout rates seen for both control treatment and active treatment are likely explained in part by SSRI trials being older (Table 1) and older trials in general having higher dropout rates.34 Another factor that may have contributed to differences in control treatment outcomes is that psilocybin and esketamine are both administered on site, whereas SSRIs are not.
While realizing that we would not be able to account for many important confounders, we aspired to not unnecessarily introduce additional ones. Thus, we excluded studies with short duration (<2 weeks) since there are almost no such SSRI studies; we similarly did not include studies involving only children or individuals aged older than 65 years since to our knowledge, there are no studies of psilocybin for depression in these populations. In the same vein, we did not include the 2 ASPIRE studies30 or SUI200131 of esketamine, which were conducted in patients with depression and acute suicidality, since patients with acute suicidality are typically excluded from both SSRI and psilocybin trials.
In a meta-analysis of those trials,30,31 the mean MADRS decrease from baseline for control treatment was 22.9 points compared with 14.9 and 16.8 points for S300117 and S3002,18 respectively,10 corresponding to an SMC (SEM) of 1.87 (0.12). One possible explanation for why control treatment was associated with a large degree of improvement in these studies30,31 could be that patients who present with acute suicidality may be more likely to be anxious, agitated, and potentially more prone to endorse greater symptom severity. Thus, when the acute distress abates, symptom severity likely lessens considerably irrespective of which specific treatment is provided.
While there is considerable uncertainty regarding the factors contributing to the divergent control treatment outcomes, of note, of the 3 studied compounds, psilocybin is the divergent one that meshes poorly with results from similar trials of other compounds. Thus, the outcomes herein observed for SSRI control treatments (SMC, 1.00) and esketamine control treatments (SMC, 1.12) agree with results from previous assessments of pill placebo control treatments.1,2 They also correspond to the mean (SEM) SMC (1.11 [0.02]) for the 242 placebo control groups with useable data included in the 2018 network meta-analysis by Cipriani and colleagues (our analysis).36 The mean SMC of 0.50 for psilocybin control treatments in this study, however, was below that commonly observed for pill placebo groups in depression and is more in line with outcomes from waiting list– and care as usual–controlled trials of psychotherapy.3
Irrespective of the causes, the small improvement rate for psilocybin control treatments in this study’s data (collected up to 6 weeks after randomization) suggests that psilocybin may not be as broadly effective for depression as estimated. If psilocybin trials have recruited patients who are less likely to respond to control treatment, it remains to be shown that psilocybin is highly effective also in the presumably larger population that tends to respond to control treatment. If a methodologic factor common to psilocybin trials (eg, that control treatment acts more like a waiting list control than a pill placebo control7) results in poorer control treatment outcomes, the acute antidepressant effects of psilocybin may be overestimated also for the comparatively narrow population of patients with depression for whom it has been trialed.
Limitations
The primary limitation of the present study is that it can only conclude that control treatment outcomes differed between trial populations but could not inform the reasons for the difference. In addition, the available psilocybin literature is small and heterogenous, and the reference groups (esketamine and SSRIs) are not exhaustive. Since the control group results for esketamine and SSRIs were in line with those from other studies,1,2 the latter limitation is unlikely to have had a major impact. Similarly, the study by Carhart-Harris and colleagues14 compared 25 mg of psilocybin with combined treatment with 1 mg of psilocybin and the SSRI escitalopram. We included this study (1) since an SSRI should be at least as effective as an inert placebo and (2) since the study had the largest pre– to post–control treatment score change of the 4 psilocybin trials5,14,15,16 (Table 1). Excluding this study would hence yield a lower mean control treatment outcome for psilocybin. Sensitivity analyses omitting this trial did not alter the results.
Conclusions
This meta-analysis found that participants receiving control treatment in psilocybin trials had significantly less improvement in depression ratings than participants receiving control treatment in trials of SSRIs or esketamine. The low response to control treatment suggests that psilocybin may not be as broadly effective for depression as estimated. Future studies should strive toward better understanding of which factors moderate control treatment outcomes in psilocybin trials, such as by trialing multiple control treatments and/or by recruiting study participants with positive expectations of the control treatment also.
Data Sharing Statement
References
- 1.Kirsch I, Deacon BJ, Huedo-Medina TB, Scoboria A, Moore TJ, Johnson BT. Initial severity and antidepressant benefits: a meta-analysis of data submitted to the Food and Drug Administration. PLoS Med. 2008;5(2):e45. doi: 10.1371/journal.pmed.0050045 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Jones BDM, Razza LB, Weissman CR, et al. Magnitude of the placebo response across treatment modalities used for treatment-resistant depression in adults: a systematic review and meta-analysis. JAMA Netw Open. 2021;4(9):e2125531. doi: 10.1001/jamanetworkopen.2021.25531 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Cuijpers P, Miguel C, Harrer M, Ciharova M, Karyotaki E. The overestimation of the effect sizes of psychotherapies for depression in waitlist controlled trials: a meta-analytic comparison with usual care controlled trials. Epidemiol Psychiatr Sci. 2024;33:e56. doi: 10.1017/S2045796024000611 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Metaxa AM, Clarke M. Efficacy of psilocybin for treating symptoms of depression: systematic review and meta-analysis. BMJ. 2024;385:e078084. doi: 10.1136/bmj-2023-078084 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Raison CL, Sanacora G, Woolley J, et al. Single-dose psilocybin treatment for major depressive disorder: a randomized clinical trial. JAMA. 2023;330(9):843-853. doi: 10.1001/jama.2023.14530 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wen A, Singhal N, Jones BDM, et al. A systematic review of study design and placebo controls in psychedelic research. Psychedelic Med (New Rochelle). 2024;2(1):15-24. doi: 10.1089/psymed.2023.0028 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Matvey M, Kelley DP, Bradley ER, Chiong W, O’Donovan A, Woolley J. Modifying informed consent to help address functional unmasking in psychedelic clinical trials. JAMA Psychiatry. 2025;82(3):311-318. doi: 10.1001/jamapsychiatry.2024.4312 [DOI] [PubMed] [Google Scholar]
- 8.Spineli LM, Kalyvas C, Yepes-Nuñez JJ, et al. Low awareness of the transitivity assumption in complex networks of interventions: a systematic survey from 721 network meta-analyses. BMC Med. 2024;22(1):112. doi: 10.1186/s12916-024-03322-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Freitas RR, Gotsis ES, Gallo AT, Fitzgibbon BM, Bailey NW, Fitzgerald PB. The safety of psilocybin-assisted psychotherapy: a systematic review. Aust N Z J Psychiatry. 2025;59(2):128-151. doi: 10.1177/00048674241289024 [DOI] [PubMed] [Google Scholar]
- 10.Center for Drug Evaluation and Research, US Food and Drug Administration. Application number 211243Orig1s000: clinical review—Spravato (esketamine). March 5, 2019. Accessed May 28, 2025. https://www.accessdata.fda.gov/drugsatfda_docs/nda/2019/211243Orig1s000MedR.pdf
- 11.Hieronymus F, Lisinski A, Østergaard SD, Eriksson E. The response pattern to SSRIs as assessed by the Montgomery-Åsberg Depression Rating Scale: a patient-level meta-analysis. World Psychiatry. 2022;21(3):472-473. doi: 10.1002/wps.21029 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Becker BJ. Synthesizing standardized mean-change measures. Br J Math Stat Psychol. 1988;41:257-278. doi: 10.1111/j.2044-8317.1988.tb00901.x [DOI] [Google Scholar]
- 13.Hieronymus F, Lisinski A, Hieronymus M, Näslund J, Eriksson E, Østergaard SD. Determining maximal achievable effect sizes of antidepressant therapies in placebo-controlled trials. Acta Psychiatr Scand. 2021;144(3):300-309. doi: 10.1111/acps.13340 [DOI] [PubMed] [Google Scholar]
- 14.Carhart-Harris R, Giribaldi B, Watts R, et al. Trial of psilocybin versus escitalopram for depression. N Engl J Med. 2021;384(15):1402-1411. doi: 10.1056/NEJMoa2032994 [DOI] [PubMed] [Google Scholar]
- 15.von Rotz R, Schindowski EM, Jungwirth J, et al. Single-dose psilocybin-assisted therapy in major depressive disorder: a placebo-controlled, double-blind, randomised clinical trial. EClinicalMedicine. 2022;56:101809. doi: 10.1016/j.eclinm.2022.101809 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Goodwin GM, Aaronson ST, Alvarez O, et al. Single-dose psilocybin for a treatment-resistant episode of major depression. N Engl J Med. 2022;387(18):1637-1648. doi: 10.1056/NEJMoa2206443 [DOI] [PubMed] [Google Scholar]
- 17.Fedgchin M, Trivedi M, Daly EJ, et al. Efficacy and safety of fixed-dose esketamine nasal spray combined with a new oral antidepressant in treatment-resistant depression: results of a randomized, double-blind, active-controlled study (TRANSFORM-1). Int J Neuropsychopharmacol. 2019;22(10):616-630. doi: 10.1093/ijnp/pyz039 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Popova V, Daly EJ, Trivedi M, et al. Efficacy and safety of flexibly dosed esketamine nasal spray combined with a newly initiated oral antidepressant in treatment-resistant depression: a randomized double-blind active-controlled study. Am J Psychiatry. 2019;176(6):428-438. doi: 10.1176/appi.ajp.2019.19020172 [DOI] [PubMed] [Google Scholar]
- 19.A phase II, placebo-controlled, double-blind study of paroxetine in depressed outpatients. GlaxoSmithKline Study Register. Updated November 4, 2018. Accessed June 19, 2025. https://www.gsk-studyregister.com/en/trial-details/?id=29060/01/001
- 20.Dunbar GC, Claghorn JL, Kiev A, Rickels K, Smith WT. A comparison of paroxetine and placebo in depressed outpatients. Acta Psychiatr Scand. 1993;87(5):302-305. doi: 10.1111/j.1600-0447.1993.tb03376.x [DOI] [PubMed] [Google Scholar]
- 21.Feighner JP, Cohn JB, Fabre LF Jr, et al. A study comparing paroxetine placebo and imipramine in depressed patients. J Affect Disord. 1993;28(2):71-79. doi: 10.1016/0165-0327(93)90035-I [DOI] [PubMed] [Google Scholar]
- 22.Dunner DL, Dunbar GC. Optimal dose regimen for paroxetine. J Clin Psychiatry. 1992;53(suppl):21-26. [PubMed] [Google Scholar]
- 23.A double-blind, placebo-controlled, fixed-dosage study comparing the efficacy and tolerability of paroxetine CR and citalopram to placebo in the treatment of major depressive disorder with anxiety. GlaxoSmithKline Study Register. Updated November 7, 2018. Accessed June 19, 2025. https://www.gsk-studyregister.com/en/trial-details/?id=29060/785
- 24.Doogan DP, Langdon CJ. A double-blind, placebo-controlled comparison of sertraline and dothiepin in the treatment of major depression in general practice. Int Clin Psychopharmacol. 1994;9(2):95-100. doi: 10.1097/00004850-199400920-00005 [DOI] [PubMed] [Google Scholar]
- 25.Montgomery SA, Rasmussen JG, Lyby K, Connor P, Tanghøj P. Dose response relationship of citalopram 20 mg, citalopram 40 mg and placebo in the treatment of moderate and severe depression. Int Clin Psychopharmacol. 1992;6(suppl 5):65-70. doi: 10.1097/00004850-199206005-00007 [DOI] [PubMed] [Google Scholar]
- 26.US Food and Drug Administration . Drug approval package—Celexa tablets (citalopram hydrobromide). July 17, 1998. Accessed June 19, 2025. https://www.accessdata.fda.gov/drugsatfda_docs/nda/98/020822a.cfm
- 27.Feighner JP, Overø K. Multicenter, placebo-controlled, fixed-dose study of citalopram in moderate-to-severe depression. J Clin Psychiatry. 1999;60(12):824-830. doi: 10.4088/JCP.v60n1204 [DOI] [PubMed] [Google Scholar]
- 28.Wade A, Michael Lemming O, Bang Hedegaard K. Escitalopram 10 mg/day is effective and well tolerated in a placebo-controlled study in depression in primary care. Int Clin Psychopharmacol. 2002;17(3):95-102. doi: 10.1097/00004850-200205000-00001 [DOI] [PubMed] [Google Scholar]
- 29.Lepola UM, Loft H, Reines EH. Escitalopram (10-20 mg/day) is effective and well tolerated in a placebo-controlled study in depression in primary care. Int Clin Psychopharmacol. 2003;18(4):211-217. doi: 10.1097/01.yic.0000074988.54339.7c [DOI] [PubMed] [Google Scholar]
- 30.Canuso CM, Ionescu DF, Li X, et al. Esketamine nasal spray for the rapid reduction of depressive symptoms in major depressive disorder with acute suicidal ideation or behavior. J Clin Psychopharmacol. 2021;41(5):516-524. doi: 10.1097/JCP.0000000000001465 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Canuso CM, Singh JB, Fedgchin M, et al. Efficacy and safety of intranasal esketamine for the rapid reduction of symptoms of depression and suicidality in patients at imminent risk for suicide: results of a double-blind, randomized, placebo-controlled study. Am J Psychiatry. 2018;175(7):620-630. doi: 10.1176/appi.ajp.2018.17060720 [DOI] [PubMed] [Google Scholar]
- 32.Dunlop BW, Thase ME, Wun CC, et al. A meta-analysis of factors impacting detection of antidepressant efficacy in clinical trials: the importance of academic sites. Neuropsychopharmacology. 2012;37(13):2830-2836. doi: 10.1038/npp.2012.153 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Munkholm K, Paludan-Müller AS, Boesen K. Considering the methodological limitations in the evidence base of antidepressants for depression: a reanalysis of a network meta-analysis. BMJ Open. 2019;9(6):e024886. doi: 10.1136/bmjopen-2018-024886 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Li F, Nasir M, Olten B, Bloch MH. Meta-analysis of placebo group dropout in adult antidepressant trials. Prog Neuropsychopharmacol Biol Psychiatry. 2020;98:109777. doi: 10.1016/j.pnpbp.2019.109777 [DOI] [PubMed] [Google Scholar]
- 35.Rutherford BR, Sneed JR, Roose SP. Does differential drop-out explain the influence of study design on antidepressant response? a meta-analysis. J Affect Disord. 2012;140(1):57-65. doi: 10.1016/j.jad.2012.01.031 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Cipriani A, Furukawa TA, Salanti G, et al. Comparative efficacy and acceptability of 21 antidepressant drugs for the acute treatment of adults with major depressive disorder: a systematic review and network meta-analysis. Lancet. 2018;391(10128):1357-1366. doi: 10.1016/S0140-6736(17)32802-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Sharing Statement
