Cannabinoids for treating psychiatric disorders in youth: a systematic review of randomized controlled trials

Cannabinoids have been of increasing interest mainly due to their putative efficacy in a wide array of psychiatric, psychosomatic, and neurological conditions.

This systematic review aims to synthesize results from randomized placebo-controlled trials regarding the efficacy and the dosage of cannabinoids as therapeutics in psychiatric disorders in children, adolescents, and young adults.

All publications up to June 30th, 2024, were included from PubMed and Embase. Eligibility criteria in accordance with the PRISMA-guidelines was applied. RCTs providing pre- and post-treatment parameters on cannabinoid therapies for mental disorders in comparison to controls in an age range from 0 to 25 years were included. Effect sizes were calculated as Hedges' g for primary outcomes, and a multilevel random-effects meta-analysis was conducted to account for dependent outcomes from same study populations.

We identified 7603 records, of which 8 independent clinical trials (reported in 9 publications) met the pre-established eligibility criteria, comprising 474 unique participants (245 treatment, 229 control). Analysis of 13 primary outcomes (of 7 clinical trials) revealed a modest positive overall effect for symptom improvement or normalization of brain physiology (Hedges' g = 0.308, 95% CI: 0.167, 0.448). Autism spectrum disorder studies showed the most consistent evidence (g = 0.264, 95% CI: 0.107, 0.421), while other conditions showed wider confidence intervals. Age-stratified analysis showed that adult populations (mean age 23.3 years, n = 5 outcomes) demonstrated higher effect sizes (g = 0.463, SD = 0.402) compared to pediatric populations (mean age 11.8 years, n = 8 outcomes; g = 0.318, SD = 0.212). Whole plant preparations (g = 0.328, 95% CI: 0.083, 0.573) and pharmaceutical cannabinoids (g = 0.292, 95% CI: 0.069, 0.515) showed comparable effects. CBD dosages ranged from 17.5 mg to 600 mg per day, with no significant correlation between dosage and effect size (ρ = -0.014, p = 0.963). Mild to moderate side effects were reported, but no serious adverse events. Risk of bias assessment ranged from low (n = 3) to high (n = 5).

While meta-analysis of effect sizes for primary outcomes revealed modest positive effects, particularly for autism spectrum disorders, the current evidence remains insufficient to broadly recommend cannabinoids for treating mental disorders in youth populations. Larger, controlled studies with standardized outcomes are needed to establish definitive clinical recommendations.