Long-term follow-up of psilocybin-facilitated smoking cessation

Smoking remains a major global public health problem, causing millions of deaths annually and projected to increase in coming years. Existing smoking cessation treatments typically fail to produce long-term abstinence for most people, creating a need to explore novel approaches. Earlier work by the investigators described an open-label pilot that combined psilocybin, a 5-HT2AR agonist, with cognitive behavioural therapy (CBT) and reported an unusually high biologically verified 6-month abstinence rate (80%) in 15 participants, together with an acceptable acute safety profile. This paper reports previously unpublished long-term follow-up data from that pilot trial. Johnson and colleagues set out to assess smoking cessation outcomes at 12 months and at a later long-term follow-up (mean 30 months post-target-quit date), and to describe persisting subjective effects attributed to the psilocybin sessions. The aim was to evaluate the durability of abstinence and to explore relationships between session-related subjective experiences and smoking outcomes over the longer term.

The study was an open-label pilot conducted at Johns Hopkins, approved by the institutional review board; all participants provided informed consent. The analytic sample comprised 15 adult smokers (10 male), mean age 51 years, who reported a mean of 19 cigarettes per day for about 31 years and an average of six prior quit attempts. Individuals with histories of severe mental illness were excluded. Participants completed a 15-week combined treatment that began with four weekly preparatory meetings integrating cognitive behavioural therapy, elements of mindfulness training, and guided imagery. Psilocybin dosing was scheduled with a moderate dose (20 mg/70 kg) in week 5 serving as the target-quit date (TQD), followed by a high dose (30 mg/70 kg) approximately two weeks later. An optional third high-dose session was available in week 13. For 10 weeks after the TQD participants provided weekly breath and urine samples, completed questionnaires, and met with staff; follow-ups were conducted at 6 and 12 months, and a retrospective interview was offered at a mean of 30 months post-TQD. Smoking status was assessed using biomarkers and self-report. Breath carbon monoxide (CO) was measured with a Bedfont Micro+ Smokerlyzer and urine cotinine was analysed by an independent laboratory; self-reported daily cigarette consumption was captured continuously via the Timeline Follow-Back (TLFB) method. Participants completed a 143-item Persisting Effects Questionnaire one week after each psilocybin session and a retrospective version at 12 months to measure sustained changes in attitudes, mood, behaviour, relationships and spirituality. Acute mystical-type experiences were measured with the validated 30-item MEQ30 approximately seven hours after each psilocybin session. Analyses defined abstinence as CO ≤6 ppm, urinary cotinine <200 ng/mL, and no self-reported smoking in the prior seven days. Missing long-term data for three non-attenders were imputed from their 12-month values; non-attendance otherwise counted as smoking. Repeated measures ANOVA tested TLFB changes across time points; planned paired t-tests compared intake with end of treatment, 6-month, 12-month, and long-term follow-ups. Pearson correlations examined associations between mean MEQ30 and personal meaning/spiritual significance scores and long-term change scores on CO, cotinine, and TLFB. Normality of datasets for correlational analyses was assessed using Dallal-Wilkinson-Lilliefor corrected Kolmogorov–Smirnov tests.

All 15 participants completed the 12-month follow-up; 12 participants (80%) returned for the long-term follow-up (mean interval 30 months, range 16–57 months). At 12 months, 10 of 15 participants (67%) were biologically confirmed abstinent, and 8 of those 10 reported continuous abstinence since the TQD. At the long-term follow-up, 9 of 15 participants (60%) were biologically confirmed abstinent, with 7 reporting continuous abstinence since the TQD. The three participants who did not attend the long-term visit had been confirmed daily smokers at 12 months and their 12-month biomarker and TLFB data were used for long-term imputation. Repeated measures ANOVA indicated a significant change in self-reported smoking on the TLFB across intake and subsequent follow-ups (end of treatment, 6 months, 12 months, long-term), although exact test statistics are not fully reported in the extracted text. Persisting Effects Questionnaire data showed substantially higher ratings of positive versus negative persisting effects across six domains (attitudes about life, attitudes about self, mood, relationships, behavioural changes, spirituality). Average negative-effect scores ranged from 3.2% to 8.1% of the maximum possible score, whereas average positive-effect scores ranged from 53% to 64% of maximum. At 12 months, 13 of 15 participants (86.7%) rated their psilocybin experiences among the five most personally meaningful experiences of their lives, and 13 (86.7%) rated them among the five most spiritually significant experiences. No participants reported increased bothersome visual disturbances at 12 months relative to baseline, and no spontaneous reports of clinically significant psychological sequelae were recorded at long-term follow-up. Acute safety findings from the earlier report were reiterated: physiological adverse effects were limited to mild post-session headache and modest transient elevations in blood pressure and heart rate. Six participants (40%) reported challenging acute session experiences; these resolved during the session day with staff support. Two participants sought outside counselling after sessions—one because of re-experiencing traumatic childhood memories who later reported benefit from additional counselling, and one for personal growth. Correlational analyses found that reductions in urinary cotinine from intake to long-term follow-up were significantly associated with mean ratings of personal meaning assessed one week after each session (r = -0.55, p = 0.04) and with mean MEQ30 scores from the end of each session day (r = -0.55, p = 0.03). The seven other correlations between psilocybin session attributes and smoking-related change scores did not reach significance (p range 0.10–0.36) but were directionally consistent with moderate effect sizes (r range -0.25 to -0.44).

Johnson and colleagues interpret the long-term findings as indicating that psilocybin may be a feasible adjunct to smoking cessation treatment when combined with a structured therapeutic programme. They note that the observed 60% biologically verified abstinence at more than one year compares favourably to typical abstinence rates for the most effective smoking cessation medications, which often fall below 31% at 12 months. At the same time, the investigators acknowledge important limitations that constrain causal inference: the small sample size, open-label design, absence of a control condition, and potential participant self-selection bias since volunteers were motivated to quit and willing to undertake a time-intensive, uncompensated experimental intervention. These limitations, they state, preclude definitive conclusions about efficacy and may contribute to type II errors in some correlational analyses. To address these issues, the paper notes an ongoing randomized comparative efficacy trial (ClinicalTrials.gov Identifier NCT01943994) that is comparing a single high dose of psilocybin with a standard nicotine replacement therapy course, with both arms receiving the same CBT intervention. The authors discuss the possibility that lasting psychological and behavioural shifts following psilocybin may relate to salient acute subjective effects—often described as mystical or transcendent—and that such experiences were associated, in this sample, with greater reductions in cotinine. They also observe that participants frequently attributed spiritual significance to their sessions, raising questions about the role of spirituality in substance use treatment; previous literature linking spirituality and improved outcomes in addiction recovery is noted by the investigators to contextualise this finding. Finally, the discussion situates the results within historical and recent research suggesting potential therapeutic utility of classic psychedelics for various substance use disorders and advocates for further controlled research into psychedelic-facilitated treatments and underlying neurobiological mechanisms.