Psychedelic‐assisted treatment for substance use disorder: A narrative systematic review

Abstract Background and aims This is the first systematic review of the extant literature on all major psychedelic‐assisted treatment for alcohol use disorder (AUD), tobacco use disorder (TUD) and other substance use disorders (SUD). We aimed to summarise the evidence for efficacy of psychedelic‐assisted treatment for AUD, TUD, and SUD; to evaluate its quality; and to offer recommendations for research.

Methods: This was a prospectively registered narrative systematic review of open‐label, randomised controlled trials (RCT), and observational studies of d‐lysergic acid diethylamide (LSD), mescaline, psilocybin, ayahuasca, ketamine, ibogaine and 3,4‐methylenedioxymethamphetamine (MDMA). Eligible studies had SUD outcome measures including craving, substance use, relapse, and remission. Study quality was evaluated using the Cochrane Collaboration Risk of Bias (RoB), and Cochrane Collaboration RoB in Non‐randomised Studies of Interventions tool. Certainty of evidence for RCTs was judged using the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) tool.

Findings: 37 studies (2035 participants) were reviewed: LSD (14; n = 1047); mescaline (1; n = 7); psilocybin (4; n = 135); ayahuasca (3; n = 101); ketamine (10; n = 579); ibogaine (5; n = 166); and MDMA (1; n = 14). There were no serious adverse events reported in any study. A two‐centre, placebo‐controlled, phase 2 superiority RCT of psilocybin for AUD, and a two‐centre, double‐blind, four‐arm, placebo‐controlled phase 2 RCT of ketamine for AUD yielded the best evidence of efficacy. Progression support to a phase 3 trials was secured from an open‐label phase 2 study of psilocybin for TUD and nine phase 2 RCTs of ketamine for AUD, cannabis use disorder, cocaine use disorder, and opioid use disorder (all nine with high‐RoB and low‐GRADE evidence certainty).

Conclusions: Psilocybin‐assisted treatment for alcohol use disorder appears to have the best evidence of efficacy among all major psychedelic‐assisted treatments for alcohol, tobacco, and other substance use disorders. Future research of psychedelic‐assisted treatment should report all safety events; screen for person‐level characteristics indicating that psychedelic‐assisted substance use disorders treatment is contraindicated; strive to mitigate blinding of participants to interventions; use factorial designs for drug and psychotherapy randomised controlled trials; and build consensus for a field‐specific Core Outcome Set.