Psychedelics, Mystical Experience, and Therapeutic Efficacy: A Systematic Review

The authors registered a pre-specified protocol with PROSPERO (CRD42021261752) and followed PRISMA guidance. Electronic searches were conducted in Embase, MEDLINE and PsycINFO for publications between January 1990 and August 2021 using a broad combination of terms for psychedelics and for acute or subjective/mystical experience; the MAPS bibliography was also searched using targeted keyword fields. From 744 retrieved records, 12 studies met inclusion criteria. Inclusion criteria required a clinical trial design, adult participants with psychiatric and/or addictive disorders who received at least one psychedelic dosing session in a laboratory or clinical setting, reporting of subjective data on mystical experience, and publication in English in a peer-reviewed journal. Two independent reviewers performed study selection, with disagreements resolved by consensus or a third reviewer. Quality appraisal used the Newcastle–Ottawa Modified Scale for uncontrolled open-label studies and the Revised Cochrane Risk of Bias Tool for randomized trials. Two reviewers independently assessed bias domains including randomisation, blinding, completeness of outcome data and selective reporting. Data extraction captured participant characteristics, substance and dose, follow-up duration, instruments for mystical and clinical measures, and analytic approaches, with emphasis on correlation, mediation, or prediction analyses linking mystical intensity to symptom change. Extracted trials were categorised by clinical indication; six were randomised and six were open-label, addressing cancer-related distress, depressive disorders (including treatment-resistant depression and major depressive disorder), and substance use disorders (alcohol, tobacco, cocaine). Information about psychotherapeutic support varied across studies; most (10/12) reported nondirective support during dosing and diverse pre/post-treatment models such as Motivational Enhancement Therapy or Cognitive Behavioural Therapy were used.

Twelve clinical trials met inclusion criteria: eight used psilocybin, and others used ayahuasca or ketamine. Study designs and outcome measures varied, and sample sizes were generally small. Overall, the authors report that most studies showed an association between mystical-type experience and therapeutic benefit for short- to medium-term outcomes, though long-term associations were less consistent. Cancer-related distress: Two randomized controlled trials of psilocybin for cancer-related psychiatric distress reported multiple significant correlations between MEQ-30 scores collected after the first session and reductions in anxiety and depression at 5–6 weeks. Griffiths and colleagues reported significant correlations for 18 of 20 self-rated primary outcome measures, with four p-values described as highly significant (p < 0.0001). Specific Spearman correlations reported included HADS total (r = 0.39; p = 0.04), BDI (r = 0.49; p = 0.01), STAI state (r = 0.42; p = 0.03) and STAI trait (r = 0.39; p = 0.04). Mediation analyses in these trials were also reported to support a role for mystical experience, although detailed statistics were not provided in the extracted text. A long-term follow-up (Agin-Liebes et al.) of the Ross trial, at a mean of 3.2–4.5 years post-treatment, found no statistically significant association between dosing-day MEQ-30 scores and current measures of anxiety or depression; attrition (roughly 50% of the original sample) and crossover design in the parent trial were noted limitations. Depressive disorders and treatment-resistant depression (TRD): In a psilocybin study of depressive disorder, a standardized beta of -0.649 was reported, predicting decrease in depressive symptoms measured by QIDS-SR (model R2 = 0.59, adjusted R2 = 0.54), with effects sustained to five weeks post-dosing. Participants scoring as having complete OBN (OBN > 0.6) showed improvements in trait anxiety, anhedonia, optimism/pessimism up to six months. Carhart-Harris and colleagues examined OBN subcomponents grouped as USB (unity, spiritual experience, blissful state) and insight; mean USB and insight scores during a 25 mg psilocybin session correlated with QIDS-SR16 change at week 5 (r = -0.49, p = 0.03 and r = -0.57, p = 0.01, respectively). An ayahuasca TRD trial did not find a significant overall correlation between HRS scores and change in MADRS from baseline to day 7 versus placebo, but reported strong correlations between MADRS change and the HRS perception subscale (r = 0.90; p = 0.002) and between MEQ-30 transcendence of time and space and outcome (r = -0.84; p = 0.009). An uncontrolled open-label ketamine inpatient study (MDD) found that derealisation-related dimensions (DED) of the 5D-ASC measured 4 hours after the first infusion were higher in non-responders, suggesting no positive correlation between classic mystical dimensions and response; post-hoc analysis showed openness to experience negatively correlated with acute anxiety (r = -0.54; p < 0.005). Substance use disorders: Five studies addressed substance problems: two alcohol use disorder (AUD), two tobacco cessation (from the same cohort analysed across different follow-ups), and one cocaine-dependence trial. Alcohol use disorder: Bogenschutz et al. (psilocybin) found strong correlations between HRS/MEQ scores and multiple drinking outcomes, for example Penn Alcohol Craving Scale (HRS r = -0.823, p = 0.006; MEQ r = -0.810, p = 0.008), percent drinking days (HRS r = -0.844, p = 0.004; MEQ r = -0.885, p = 0.002) and percent heavy drinking days (HRS r = -0.763, p = 0.017; MEQ r = -0.852, p = 0.004). Rothberg et al. (ketamine) reported a negative correlation between HMS and number of heavy drinking days (r = -0.466; p < 0.05) and additional subdimension correlations (for instance, ineffability and percentage heavy drinking days r = 0.624; p = 0.01). Tobacco addiction: In the Garcia-Romeu cohort, 12 of 15 participants remained abstinent by urine cotinine at 6 months and 10 at 12 months. Predictors of abstinence included higher scores on the States of Consciousness Questionnaire (p < 0.049), the spiritual significance rating (p < 0.047), and MEQ correlations (MEQ r = -0.55; p = 0.03). The study also reported that 13 of 42 dosing sessions (31%) met a threshold for complete mystical experience, most occurring at high dose. Cocaine dependence: A randomized ketamine infusion trial in cocaine-dependent subjects reported that HMS, dissociation (CADSS) and near-death experience phenomena increased after ketamine versus control; HMS was identified as the only significant mediator of decreased cocaine use/craving (β = 0.431; p = 0.0175). The trial modified some established questionnaires, which the authors flagged as a limitation. Adverse effects and moderators: Across studies, acute anxiety or distress (for example anxiety related to ego dissolution) was reported to reduce therapeutic benefit in some trials, though not to worsen long-term symptoms. Set and setting, therapeutic support models and personality traits (notably openness) were identified as factors that influence both mystical experience intensity and outcomes. Many trials were open-label, unblinded or had small, relatively homogenous samples, limiting generalisability and increasing risk of bias.

Ko and colleagues interpret the assembled evidence as indicating that mystical-type subjective experiences during psychedelic dosing are often associated with clinical improvement. Most studies reviewed showed correlation, prediction or mediation linking mystical intensity to short- and medium-term symptom reduction; the authors report significant correlations in nine of twelve studies analysed for these timeframes. They highlight that some trials demonstrated mystical experience as an independent predictor or mediator of outcome, reinforcing the plausibility of a mechanistic role. The researchers note notable exceptions and qualifications. A long-term follow-up (3.2–4.5 years) did not find a persisting association between dosing-day mystical intensity and current symptoms, which the authors attribute to factors such as subsequent therapies, life changes, and attrition reducing power. Heterogeneity within mystical experience measures was also discussed: certain subdimensions (for example ineffability or the perception subscale) appeared particularly impactful in some studies, while other subcomponents did not correlate with outcomes. Ketamine studies showed mixed patterns, with some dissociative or derealisation features associated with poorer short-term response. Key limitations of the evidence base are emphasised. Most trials had small sample sizes and limited diversity in gender, ethnicity and socioeconomic status. Six of the 12 studies were open-label and uncontrolled, raising the risk of expectancy and observer bias. Difficulty achieving and maintaining effective blinding in psychedelic trials, and the confounding effects of concomitant psychotherapeutic support, were named as persistent methodological challenges. The authors also acknowledge conceptual debates about the metaphysical framing of mystical experience and point out that reliance on self-report instruments makes the phenomenon non-observable and potentially influenced by participants' worldviews. For future research, the paper recommends experimental designs capable of testing causality, larger and more diverse samples, standardised and possibly more secular measures of subjective effects, and investigations into predictors of both intensity and tolerability of mystical experience. Practical suggestions include optimising set and setting to reduce acute anxiety, exploring dose–response relationships or new compounds if intensity is causal for benefit, and identifying clinical or personality markers that predict who will derive most benefit from mystical-type experiences. The authors also note structural barriers to progress such as funding, legal restrictions and the screening demands of psychedelic trials, which currently constrain sample sizes and representativeness.