Wang · Molecular psychiatry 2024 · retrospective matched cohort study · n=681,268

Association of semaglutide with reduced incidence and relapse of cannabis use disorder in real-world populations: a retrospective cohort study.

Cited 81 times in the scientific literature.

Level 3 - non-randomized controlled study

Retrospective propensity-score matched cohort study using electronic health records.

PubMed 38486046 · doi:10.1038/s41380-024-02498-5 · record verified 2026-08-31

What was done

This retrospective cohort study examined electronic health records from the TriNetX Analytics Network (encompassing approximately 105.3 million patients across 61 US healthcare organizations) to evaluate the association of semaglutide with cannabis use disorder (CUD). The study analyzed 85,223 patients with obesity and 596,045 patients with type 2 diabetes (T2D). Using propensity-score matching, researchers compared semaglutide users to users of non-GLP-1RA anti-obesity or anti-diabetes medications over a 12-month follow-up period for both incident CUD (no prior history) and recurrent CUD (prior history).

What was found

In patients with obesity (mean age 51.3 years, 65.6% women), semaglutide compared to non-GLP-1RA anti-obesity medications was associated with a reduced risk of incident CUD (HR 0.56, 95% CI: 0.42–0.75) and recurrent CUD (HR 0.62, 95% CI: 0.46–0.84). In patients with T2D, semaglutide was associated with lower risk of incident CUD (HR 0.40, 95% CI: 0.29–0.56) and recurrent CUD (HR 0.66, 95% CI: 0.42–1.03) compared to non-GLP-1RA anti-diabetes medications. Reductions were consistent across gender, age, and race subgroups.

Why it matters

With no current FDA-approved pharmacotherapies for cannabis use disorder, these findings offer real-world evidence that GLP-1 receptor agonists may reduce both onset and recurrence of CUD. This provides a strong rationale for prospective randomized controlled trials.

Limits

The study is limited by its retrospective observational design, which is susceptible to residual confounding despite propensity-score matching. Reliance on electronic health record diagnostic codes often leads to underreporting of substance use, and granular data on cannabis consumption patterns or dosage were not captured.

Cited by