Psilocybin & Addiction
48% complete abstinence at 8 months
In the only large double-blind RCT of psilocybin for alcohol use disorder, 48% of participants stopped drinking entirely by 8 months — double the placebo rate.
Bogenschutz et al., JAMA Psychiatry, 2022
93
participants in the definitive JAMA Psychiatry RCT
4 trials
completed across 2015-2025 with diverging single vs two-session results
2025
null result trial explained by single-dose vs two-session protocol difference
Bogenschutz et al., JAMA Psychiatry, 2022
View as image ↗Alcohol use disorder has a relapse problem. Existing medications — naltrexone, acamprosate, disulfiram — have real effect sizes but limited durability. The 2022 JAMA Psychiatry trial by Bogenschutz and colleagues was the first double-blind randomized controlled trial to test whether two psilocybin sessions could produce sustained sobriety, and it generated numbers that stood out: 48% of participants had stopped drinking entirely by 8 months, compared to 24% in the placebo group.[1]
The field has since become more complicated — a 2025 Swiss phase 2 trial found no significant primary outcome. Understanding what that null result means, and what it doesn't mean, requires looking at the full body of evidence on psilocybin alcohol use disorder long-term results together. This is that analysis. For the broader context of how psilocybin affects the addiction-relevant brain circuitry, see the psilocybin for alcohol addiction overview.
Key Takeaways
- The 2022 Bogenschutz JAMA Psychiatry RCT (n=93) found 48% complete abstinence vs 24% placebo at 8 months; heavy drinking days reduced 83% vs 51% (Bogenschutz et al., 2022)
- The 2015 open-label pilot (n=10) showed sustained gains through 36-week follow-up, establishing the initial proof of concept (Bogenschutz et al., 2015)
- A 2025 Swiss phase 2 RCT (n=37, single dose) found no significant difference from placebo on its primary outcome at 6 months, but reported positive secondary outcomes (Rieser et al., 2025)
- The likely explanation for the 2025 null result: the trial used one psilocybin session vs two in Bogenschutz 2022; preclinical evidence shows severe AUD blunts psilocybin's effect on the brain's self-referential network (Reinwald et al., 2023)
- A 2025 feasibility RCT in participants with both AUD and depression (n=30) showed preliminary efficacy and confirmed protocol feasibility (Luquiens et al., 2025)
- Two independent systematic reviews confirm consistent benefit signals for alcohol across psilocybin trials (van der Meer 2023, Goel 2022)
The 93-Person RCT: What the Bogenschutz 2022 Data Actually Shows
The defining study in this field is Bogenschutz and colleagues (2022), published in JAMA Psychiatry — the double-blind randomized controlled trial that enrolled 93 adults with alcohol use disorder and assigned them to either two psilocybin sessions or two diphenhydramine (antihistamine) control sessions, both embedded in a 12-week psychotherapy program.[1]
The primary outcome was the percentage of heavy drinking days during the 32-week double-blind observation period. In absolute terms, heavy drinking days fell to 9.7% in the psilocybin group versus 23.6% in the control group (a 13.9 percentage-point difference, P=.01) — which corresponds to roughly an 83% relative reduction from baseline in the psilocybin arm versus about 51% in the control arm. The secondary outcome of complete abstinence was 48% vs 24% at the 8-month follow-up — the first large-trial demonstration that roughly half of treatment-seeking patients can achieve full sobriety with psilocybin-assisted therapy.
Several aspects of this design are worth understanding:
Why diphenhydramine as control? The control condition needs to produce some noticeable subjective effect to maintain blinding — a placebo capsule that does nothing would be unblinded immediately. Diphenhydramine causes drowsiness and mild sedation, making it a more credible active control than placebo. This is a methodological strength that many earlier psychedelic studies lacked.
The psychotherapy component. Participants received preparation sessions before each psilocybin administration and integration sessions afterward. The psilocybin was not administered in isolation. This design reflects the clinical consensus that psychedelic effects are context-dependent — the drug effect combined with therapeutic support is the treatment, not the drug alone.
The blinding question. Participants in psilocybin studies almost always know whether they received an active psychedelic — the subjective effects are unmistakable. This is a limitation the trial acknowledges, and it means expectancy effects cannot be fully ruled out. The active control and double-blind design reduce this confound without eliminating it.
The Origin: Bogenschutz 2015 Open-Label Pilot
The 2022 trial built on a 2015 proof-of-concept study by the same lead investigator, which enrolled 10 adults with alcohol dependence and administered 2-3 psilocybin sessions alongside motivational enhancement therapy.[2]
The design was open-label — no control group, no blinding. The evidence quality is accordingly low. But the sustained results mattered: significant increases in abstinence weeks after psilocybin sessions, and gains maintained through 36-week follow-up. This was the signal that justified a full RCT.
All Four Trials Side by Side
| Study | Year | n | Protocol | Primary Outcome | Follow-Up |
|---|---|---|---|---|---|
| Bogenschutz et al. (pilot) | 2015 | 10 | 2-3 sessions + MET | Significant abstinence increase | 36 weeks |
| Bogenschutz et al. (RCT) | 2022 | 93 | 2 sessions + psychotherapy | 83% vs 51% heavy drinking day reduction | 8 months |
| Rieser et al. (RCT) | 2025 | 37 | 1 session + 5 psychotherapy sessions | No significant primary outcome | 6 months |
| Luquiens et al. (feasibility RCT) | 2025 | 30 | 2 sessions + psychotherapy | Feasibility confirmed; preliminary efficacy | 12 weeks |
The table points to one recurring variable: session count. The trials reporting robust or promising effects (Bogenschutz 2015 and 2022, and the two-session Luquiens 2025 feasibility trial) all used multiple psilocybin sessions, while the one trial with a null primary outcome (Rieser 2025) used a single session. This pattern is suggestive rather than conclusive — the trials differ in more than session count, including baseline population and statistical power — but session number is among the most plausible methodological explanations for the diverging results.
- Animal & mechanistic studiescompleted
Reinwald 2023 rat fMRI: psilocybin's DMN effects are blunted in alcohol-dependent brains
- Open-label pilotcompleted
Bogenschutz 2015 (n=10): proof of concept, gains held to 36 weeks
- Small / feasibility RCTscompleted
Rieser 2025 (n=37, single dose) null primary; Luquiens 2025 (n=30) feasibility confirmed
- Large double-blind RCTcurrent
Bogenschutz 2022 (n=93): 48% vs 24% abstinence at 8 months — the field's anchor
- AUD-specific meta-analysismissing
No pooled quantitative meta-analysis of AUD-only trials yet; current reviews are narrative
The evidence has reached one adequately powered double-blind RCT showing benefit, supported by a pilot, mechanistic animal work, and two 2025 RCTs. What is missing is replication at the large-RCT level and an AUD-specific quantitative meta-analysis.
Bogenschutz et al. 2015, 2022; Rieser et al. 2025; Luquiens et al. 2025; Reinwald et al. 2023
Why the 2025 Swiss Trial Found No Effect on Its Primary Outcome
Rieser and colleagues (2025), publishing in eClinicalMedicine (a Lancet journal), ran a phase 2 RCT in Switzerland enrolling 37 participants with alcohol use disorder who had recently completed detoxification.[3] Participants received 25mg psilocybin or placebo alongside 5 psychotherapy sessions. The primary outcome — abstinence duration at 4-week and 6-month follow-up — showed no significant difference between psilocybin and placebo.
Secondary outcomes were more positive: psilocybin participants showed reduced craving, reduced depression scores, and reported lower alcohol use on several secondary measures. But the primary endpoint was not met.
This is important data that honest coverage of the field cannot omit. Several interpretations apply:
The single-dose explanation. The Bogenschutz 2022 trial used two psilocybin sessions. Rieser used one. The psychedelic experience in psilocybin therapy is intense and often incomplete on a single administration — some participants do not achieve the depth of experience associated with the strongest outcomes (often called the "mystical experience" or "ego dissolution"). A second session may be important for reinforcing or completing the therapeutic work.
The population difference. Rieser's participants had recently completed medical detoxification — a different starting point from Bogenschutz's participants who were still actively drinking at enrollment. Detox-completers may have different trajectories, motivations, and neurobiological baselines.
The neuroscience explanation. Reinwald and colleagues (2023) showed in a preclinical study that psilocybin's effect on the default mode network — the brain's self-referential circuit, which tends to be overactive in rumination and self-focused thinking — is blunted in alcohol-dependent animals.[7] Psilocybin normally suppresses activity in this network, which researchers believe is partly responsible for the shift in self-perception that supports behavioral change. But in brains altered by chronic alcohol exposure, that suppression is diminished.
If this translates to humans, it means severely alcohol-dependent patients may need higher doses or more sessions to achieve the same network disruption that psilocybin produces in non-dependent brains. Single-dose protocols may simply be insufficient for the most entrenched cases.
Comorbid Depression and AUD: The Luquiens 2025 Data
Alcohol use disorder commonly co-occurs with depression — often without separate diagnosis, since alcohol's depressant effects can mimic or mask depressive symptoms. Luquiens and colleagues (2025) ran a feasibility RCT specifically in this comorbid population: 30 participants with severe alcohol use disorder and significant depressive symptoms (Beck Depression Inventory-II ≥14), randomized 2:1 to two sessions of 25mg psilocybin (n=20) or two sessions of 1mg active placebo (n=10) spaced three weeks apart, added on to an inpatient relapse-prevention program.[4]
The primary aim was feasibility — confirming that the protocol was implementable and that 350 screened patients could yield 30 enrolled participants who met both diagnostic criteria. That was confirmed. The preliminary efficacy data showed positive trends on both alcohol and depression outcomes, though the trial was sized for feasibility rather than statistical power to detect a treatment effect.
This trial matters because most clinical trials exclude participants with comorbid conditions to keep the population clean. Real-world treatment populations have comorbidities. If psilocybin shows promise in the comorbid AUD+depression population — which represents a large share of people with AUD — that expands the potential clinical application considerably.
What Multiple Trials Together Suggest About Session Count
Van der Meer and colleagues (2023) reviewed psilocybin evidence across alcohol, tobacco, and cocaine studies and found consistent benefit signals for alcohol and tobacco, with the shared serotonergic mechanism suggesting the substance-general pattern is real rather than artifact.[5] Goel and colleagues (2022) reached similar conclusions in a broader systematic review of psilocybin therapeutic evidence across depression, anxiety, and addiction.[6]
The pattern emerging from the trial data suggests:
Two psilocybin sessions with comprehensive psychotherapy, in motivated treatment-seeking patients, produces sustained outcomes that exceed existing pharmacotherapy. The Bogenschutz 2022 data is the evidence anchor for this claim.
Single-session protocols, or protocols with reduced psychotherapy intensity, have produced mixed results. The Rieser 2025 null primary outcome is the clearest example. Whether this reflects insufficient dose, insufficient session count, or population differences is not yet resolved.
Psilocybin is still investigational for alcohol use disorder — it is not FDA-approved for this indication. Psilocybin remains a Schedule I substance in the United States. The clinical evidence is meaningful and growing, but it comes from research settings with specific protocols that are not currently replicable in standard clinical care.
Legal & Clinical Status
ModeratePsilocybin for AUD is investigational, not approved treatment
Concern
Psilocybin is a Schedule I controlled substance in the United States and is not FDA-approved for alcohol use disorder. The benefits described here come from supervised clinical trials with screening, preparation, in-session monitoring, and integration therapy — not from self-administration.
What the research says
The trial outcomes reflect psilocybin delivered inside a structured psychotherapy protocol by trained clinicians. Outside that setting the safety and efficacy data do not apply, and unsupervised use carries legal and psychological risk.
Particularly relevant for: People considering psilocybin for alcohol use disorder
What to do
This article is educational and is not medical advice. Discuss any treatment for alcohol use disorder with a qualified clinician, and look for regulated clinical trials rather than unsupervised use.
Bogenschutz et al., JAMA Psychiatry, 2022; US DEA Schedule I classification
The parallel evidence for psilocybin's 80% smoking cessation rate at Johns Hopkins adds weight to the substance-general mechanism argument, suggesting the effect on addiction is not specific to alcohol.
The Bottom Line
The psilocybin alcohol use disorder evidence base is real but still forming. The 2022 Bogenschutz JAMA Psychiatry RCT remains the strongest data point: 48% abstinence at 8 months vs 24% placebo, from a 93-person double-blind trial. The 2025 Swiss trial null result is important context — it used one session rather than two, and preclinical data suggests severe AUD blunts psilocybin's brain network effects. The evolving literature suggests session count and protocol intensity matter significantly for sustained outcomes.
Frequently Asked Questions
Sources & References
- 1. Source details for this reference are not yet available.
- 2. Source details for this reference are not yet available.
- 3. Source details for this reference are not yet available.
- 4. Source details for this reference are not yet available.
- 5. Source details for this reference are not yet available.
- 6. Source details for this reference are not yet available.
- 7. Source details for this reference are not yet available.