Johnson · JAMA network open 2026 · open-label randomized controlled pilot trial · n=82

Psilocybin or Nicotine Patch for Smoking Cessation: A Pilot Randomized Clinical Trial.

Cited 10 times in the scientific literature.

Level 2 - randomized trial

Individual randomized clinical trial

PubMed 41805956 · doi:10.1001/jamanetworkopen.2026.0972 · record verified 2026-08-28

What was done

An open-label pilot randomized clinical trial evaluated 82 psychiatrically healthy adult cigarette smokers (mean age 47.6 years; 59.8% male) at a single academic medical center from 2015 to 2023. Participants were randomized to receive either a single high dose of psilocybin (30 mg/70 kg) or 8 to 10 weeks of FDA-approved nicotine patch treatment on their target quit date. Both groups received a 13-week manualized cognitive behavioral therapy (CBT) program. The primary outcome was intention-to-treat biochemically verified prolonged smoking abstinence at 6 months post-quit date; secondary outcome was 7-day point prevalence abstinence at 6 months.

What was found

Sixty-eight participants (82.9%) completed the 6-month follow-up. At 6 months, 17 participants in the psilocybin group (40.5%) achieved biochemically verified prolonged abstinence compared with 4 participants in the nicotine patch group (10.0%) (odds ratio, 6.12; 95% CI, 1.99–23.26; P = .003). For 7-day point prevalence abstinence, rates were 52.4% (22 participants) for psilocybin versus 25.0% (10 participants) for nicotine patch (odds ratio, 3.30; 95% CI, 1.32–8.70; P = .01). No serious adverse events were attributed to psilocybin or the nicotine patch.

Why it matters

This study provides comparative randomized evidence that a single psilocybin session combined with psychotherapy can yield substantially higher long-term tobacco abstinence rates than standard nicotine replacement therapy.

Limits

The trial had a small pilot sample size (n = 82) from a single academic center and an 8-year recruitment window. Participants and investigators were unblinded to treatment assignments, introducing possible expectancy and reporting bias. The sample was restricted to psychiatrically healthy smokers, limiting generalizability to broader smoking populations with psychiatric comorbidities.

Cited by