Meta-Analysis: One-Third of First-Episode Psychosis Patients Use Cannabis, With Regular Use Starting 6 Years Before Psychosis
A meta-analysis of 37 studies found cannabis use prevalence of 33.7% at first psychosis onset, with regular use typically beginning 6.3 years before psychosis, and use declining by nearly half after treatment.
Quick Facts
What This Study Found
This meta-analysis pooled data from 37 studies to characterize the epidemiology of cannabis use in first-episode psychosis.
Three key findings emerged. First, the interval between starting regular cannabis use and psychosis onset averaged 6.3 years. This long delay is important for understanding the temporal relationship between cannabis and psychosis.
Second, approximately one-third (33.7%) of people experiencing first-episode psychosis were using cannabis at the time of onset. This is substantially higher than general population cannabis use rates.
Third, cannabis use declined after psychosis treatment, with odds of continued use between 6 months and 10 years after onset being 0.56 (meaning about 44% reduction). This suggests that treatment engagement and psychosis experience lead many to reduce or stop cannabis use.
Key Numbers
37 samples included. Interval from regular cannabis to psychosis onset: 6.3 years (10 samples, SMD = 1.56). Cannabis use prevalence at first episode: 33.7% (35 samples, 95% CI 31-39%). Odds of continued use after treatment: 0.56 (19 samples, 95% CI 0.40-0.79).
How They Did This
Meta-analysis of 37 observational studies identified through MEDLINE, EMBASE, PsycINFO, Web of Science, and CINAHL. Random-effects meta-analyses synthesized prevalence estimates, intervals between cannabis initiation and psychosis onset, and odds of continued cannabis use after treatment.
Why This Research Matters
This meta-analysis provides the most precise estimates to date of the cannabis-psychosis timeline. The 6.3-year average between regular cannabis use onset and psychosis emergence defines a potential intervention window during which reducing cannabis use might prevent or delay psychosis onset.
The Bigger Picture
The 6.3-year window between cannabis initiation and psychosis onset is clinically actionable. It suggests that early intervention programs targeting cannabis use in high-risk individuals could have years of lead time before psychosis develops. The post-treatment decline in use also suggests that psychosis itself motivates many users to change their behavior.
What This Study Doesn't Tell Us
Observational studies cannot prove causation. Heterogeneity across studies was noted. Recall bias in reporting cannabis use onset age. Different studies defined cannabis use differently. The meta-analysis cannot distinguish between causation and shared vulnerability.
Questions This Raises
- ?Would cannabis cessation during the 6.3-year window prevent psychosis in some individuals?
- ?Why do about 56% of cannabis users continue after psychosis despite evidence of harm?
- ?What predicts continued versus discontinued use?
Trust & Context
- Key Stat:
- 6.3 years between starting regular cannabis use and psychosis onset
- Evidence Grade:
- Meta-analysis of 37 studies with random-effects methodology. Strong synthesis of observational evidence, though underlying studies have inherent limitations.
- Study Age:
- Published in 2016. Additional studies have continued to refine estimates of the cannabis-psychosis relationship.
- Original Title:
- Cannabis use in first episode psychosis: Meta-analysis of prevalence, and the time course of initiation and continued use.
- Published In:
- The Australian and New Zealand journal of psychiatry, 50(3), 208-19 (2016)
- Authors:
- Myles, Hannah, Myles, Nicholas, Large, Matthew(2)
- Database ID:
- RTHC-01230
Evidence Hierarchy
Combines results from multiple studies to find an overall pattern.
What do these levels mean? →Frequently Asked Questions
How common is cannabis use among people experiencing first-episode psychosis?
About one-third (33.7%) of people experiencing their first episode of psychosis are using cannabis at the time, significantly higher than the general population rate.
How long before psychosis do people typically start using cannabis?
On average, regular cannabis use begins about 6.3 years before the onset of psychosis, providing a potential window for early intervention.
Read More on RethinkTHC
- THC-amygdala-anxiety-brain
- anandamide-weed-withdrawal
- cannabinoid-receptors-recovery-time
- cannabis-developing-brain-teenagers
- cant-enjoy-anything-without-weed
- dopamine-recovery-after-quitting-weed
- endocannabinoid-system-explained-simply
- endocannabinoid-system-withdrawal
- nervous-system-weed-withdrawal-fight-flight
- teen-weed-use-under-18-effects-brain
- thc-brain-withdrawal
- thc-prefrontal-cortex-brain-effects
- weed-cortisol-stress-hormones
- weed-memory-loss-recovery
- weed-motivation-amotivational-syndrome
- weed-nervous-system-effects
- weed-reward-system-brain
Cite This Study
https://rethinkthc.com/research/RTHC-01230APA
Myles, Hannah; Myles, Nicholas; Large, Matthew. (2016). Cannabis use in first episode psychosis: Meta-analysis of prevalence, and the time course of initiation and continued use.. The Australian and New Zealand journal of psychiatry, 50(3), 208-19. https://doi.org/10.1177/0004867415599846
MLA
Myles, Hannah, et al. "Cannabis use in first episode psychosis: Meta-analysis of prevalence, and the time course of initiation and continued use.." The Australian and New Zealand journal of psychiatry, 2016. https://doi.org/10.1177/0004867415599846
RethinkTHC
RethinkTHC Research Database. "Cannabis use in first episode psychosis: Meta-analysis of pr..." RTHC-01230. Retrieved from https://rethinkthc.com/research/myles-2016-cannabis-use-in-first
Access the Original Study
Study data sourced from PubMed, a service of the U.S. National Library of Medicine, National Institutes of Health.
This study breakdown was produced by the RethinkTHC research team. We analyze and report published research findings without making health recommendations. All interpretations are based solely on the published abstract and study data.