Home > Cannabis Research > Cannabis and Psychosis
Cannabis and Psychosis
Research Evidence & Risk Factors
The definitive evidence-based guide to cannabis-related psychosis: dose-response data, genetic vulnerability, adolescent risk, THC potency effects, and clinical recommendations
The link between cannabis and psychosis is one of the most robust and consequential findings in modern psychiatry. This page presents the complete evidence base -- from population-level epidemiology to molecular genetics -- explaining how, why, and in whom cannabis increases psychosis risk, and why this matters more now than ever.
| Quick Answer: Cannabis and Psychosis | |
| Does cannabis cause psychosis? | Yes. Cannabis use is an established risk factor for both acute psychotic episodes and chronic psychotic disorders. Daily high-potency use increases risk 4-5x (Di Forti et al., 2019). The evidence supports a causal contribution model. |
| Who is at risk? | Adolescents, daily users, high-potency product users, those with COMT Val/Val or AKT1 C/C genotypes, people with family history of psychosis, and individuals with prodromal symptoms. |
| Conversion rate | 34-50% of cannabis-induced psychosis cases convert to chronic schizophrenia-spectrum disorders within 3-5 years -- the highest conversion rate of any substance-induced psychosis. |
| Why now? | THC potency has risen from 4% (1995) to 15-25% (current flower) and 60-90% (concentrates). Legalization has increased access. Adolescent brains are exposed to unprecedented potency levels. |
| Expert | Ryan S. Sultan, MD -- NIH NIDA K12 researcher, Columbia University. His PAWS project addresses cannabis-related psychosis prevention. |
On This Page
What Is Cannabis-Induced Psychosis?
Psychosis is a mental state in which an individual loses contact with reality. It is characterized by hallucinations (perceiving things that are not there -- most commonly auditory), delusions (fixed false beliefs that persist despite contradictory evidence), disorganized thinking (incoherent or tangential speech), and paranoia (intense, irrational suspicion that others are threatening or conspiring against you).
Cannabis-induced psychosis occurs when cannabis use triggers a psychotic episode. The DSM-5 classifies this as a Substance/Medication-Induced Psychotic Disorder when the psychotic symptoms developed during or soon after cannabis use, are more severe than what would be expected from intoxication alone, and are not better explained by a primary psychotic disorder.
However, the distinction between "cannabis-induced" and "primary" psychosis has become increasingly complex. Research now suggests that for many individuals, cannabis-induced psychosis is not simply a transient drug reaction but rather the unmasking or acceleration of an underlying vulnerability that may have eventually manifested regardless -- but would have done so later, less severely, or perhaps not at all without the cannabis exposure. This nuance is critical for understanding both individual risk and public health policy.
Symptoms of Cannabis-Induced Psychosis
- Paranoia: The most common psychotic symptom associated with cannabis. Intense, irrational fear that others are watching, following, or plotting against you. Occurs in up to 50% of high-dose THC exposures in experimental settings (D'Souza et al., 2004).
- Auditory hallucinations: Hearing voices or sounds that are not present. Voices may be threatening, commanding, or conversational.
- Visual hallucinations: Seeing things that are not there, though these are less common than auditory hallucinations.
- Delusions: Fixed false beliefs -- grandiose (believing you have special powers), persecutory (believing you are being targeted), or referential (believing neutral events are specifically directed at you).
- Disorganized thinking: Inability to organize thoughts logically, leading to incoherent speech, tangential reasoning, and difficulty communicating.
- Agitation and bizarre behavior: Unpredictable actions, extreme restlessness, or behavior that appears incomprehensible to observers.
- Flat or inappropriate affect: Emotional responses that are blunted (showing no emotion) or inappropriate to the situation (laughing at serious events).
Clinical Note: In emergency departments across the United States, cannabis-related psychosis presentations have increased dramatically in recent years. Many of these patients are adolescents or young adults using high-potency concentrates or vape cartridges. The acute episodes can be terrifying for both the patient and their family, and they carry significant implications for long-term mental health.
The Dose-Response Relationship: More THC, More Risk
One of the most important findings in cannabis-psychosis research is the dose-response relationship: the risk of psychosis increases in a graded fashion with increasing frequency of use and increasing THC potency. This is a critical piece of evidence supporting a causal relationship, because true causal agents typically show dose-response effects.
Infrequent use of low-potency cannabis carries a modest but measurable increase in psychosis risk compared to non-users.
Regular (weekly or more) cannabis use approximately doubles the risk of psychotic experiences. Meta-analyses consistently confirm this finding (Marconi et al., 2016, Schizophrenia Bulletin).
Daily cannabis use, regardless of potency, triples the risk of psychotic disorder (Di Forti et al., 2019).
Daily use of high-potency cannabis (>10% THC) increases psychosis risk nearly 5-fold. This is the most critical risk category, and it describes a pattern that is increasingly common among young users.
The dose-response relationship has been confirmed by experimental studies. D'Souza et al. (2004, Neuropsychopharmacology) administered intravenous THC to healthy volunteers in a controlled setting and found that higher doses produced more severe psychotic symptoms in a linear fashion. Importantly, even some individuals with no prior psychiatric history developed transient psychotic symptoms at higher doses, demonstrating that THC has intrinsic psychotogenic properties -- it is not merely unmasking pre-existing illness.
The Landmark Lancet Study (Di Forti et al., 2019)
The Di Forti et al. (2019) study, published in Lancet Psychiatry, is the most important single study in the cannabis-psychosis literature. It was a case-control study conducted across 11 sites in Europe and Brazil, including 901 patients presenting with first-episode psychosis and 1,237 population controls. Its findings have fundamentally shaped our understanding of cannabis-related psychosis risk.
Key Findings
- Daily use of high-potency cannabis was associated with an odds ratio of 4.8 for first-episode psychosis compared to never-users -- nearly a 5-fold increase.
- Potency mattered independently: High-potency use increased risk even after controlling for frequency. The type of cannabis used was as important as how often it was used.
- Population-attributable fractions (PAFs): The study calculated what proportion of psychosis cases could be attributed to cannabis in each city. In Amsterdam, where high-potency cannabis dominated the market, the PAF was 50.3% -- meaning that eliminating daily high-potency cannabis use could theoretically prevent half of all new psychosis cases. In London, the PAF was 30.3%.
- City-level variation correlated with cannabis potency: cities where high-potency cannabis was more available had higher incidence of first-episode psychosis.
This study was transformative because it moved the conversation from "does cannabis increase psychosis risk?" (which was already established) to "how much of the psychosis in our communities is attributable to cannabis?" The answer -- up to 50% in some settings -- was sobering.
The Danish Population Study (Hjorthoj et al., 2023)
A landmark Danish study published in 2023 by Hjorthoj and colleagues examined national registry data covering the entire Danish population over several decades. They found that cannabis use disorder was associated with approximately 30% of schizophrenia diagnoses in young men aged 21-30. Furthermore, as cannabis use disorder prevalence increased in Denmark between 1972 and 2021, the proportion of schizophrenia cases attributable to CUD also increased. This study provided powerful population-level evidence that the cannabis-schizophrenia association is not merely a statistical artifact but reflects a genuine causal contribution.
The CAMH Evidence
Research from the Centre for Addiction and Mental Health (CAMH) in Toronto has documented a substantial increase in cannabis-related psychosis presentations in emergency departments coinciding with legalization in Canada (October 2018). Their data showed increases in both cannabis-related ED visits overall and psychosis-specific presentations, particularly among young adults aged 18-25. These real-world data corroborate the epidemiological findings from controlled studies.
THC Potency: Why Today's Cannabis Is Different
The dramatic increase in THC potency over the past three decades is central to understanding the modern cannabis-psychosis crisis. Today's cannabis is a fundamentally different drug than what was available a generation ago, and the psychosis risk profile has changed accordingly.
| Era | Average THC | THC:CBD Ratio | Psychosis Implications |
|---|---|---|---|
| 1995 | ~4% | ~14:1 | Lower potency, some CBD content providing partial protection. Lower psychosis risk per exposure. |
| 2010 | ~10-12% | ~80:1 | Significant potency increase. CBD nearly bred out. Emerging concerns about psychosis risk. |
| 2020-Present (flower) | 15-25% | >100:1 | High potency is now standard. Negligible CBD. Substantial psychosis risk with daily use. |
| Concentrates (current) | 60-90% | Negligible CBD | Massive THC doses. Rapidly escalating tolerance. Highest psychosis risk category. Limited research data because these products are so new. |
The shift in THC:CBD ratio is as important as the absolute THC increase. CBD has demonstrated antipsychotic properties in clinical trials (McGuire et al., 2018, American Journal of Psychiatry). Historical cannabis strains with balanced THC:CBD ratios may have provided some inherent protection against psychosis. Modern strains, bred selectively for maximum THC with negligible CBD, have eliminated that natural safeguard.
Dr. Yasmin Hurd's laboratory at Mount Sinai has been at the forefront of translational research on CBD's neuroprotective potential, exploring how CBD modulates the endocannabinoid system and potentially counteracts THC's psychotogenic effects. Her work underscores both the promise of CBD-based interventions and the danger of THC-dominant products. For more on potency trends, see our THC potency analysis.
Adolescent Brain Vulnerability
Adolescents are disproportionately vulnerable to cannabis-related psychosis for biological, developmental, and behavioral reasons. The convergence of active brain development with increasing access to high-potency cannabis creates what may be the most concerning public health aspect of the cannabis-psychosis relationship.
The Developing Brain and the Endocannabinoid System
The endocannabinoid system (ECS) is not simply a passive receiver of THC -- it is an active participant in brain development. During adolescence, the ECS guides critical processes including synaptic pruning (eliminating unnecessary neural connections), myelination (insulating neural pathways for faster communication), and the maturation of the prefrontal cortex (which governs judgment, impulse control, and reality testing -- and does not fully mature until approximately age 25).
When adolescents flood this developmental system with exogenous THC, they disrupt these precisely timed maturational processes. The consequences are not merely "acute intoxication" -- they are developmental derailment. Animal studies consistently show that adolescent THC exposure produces lasting changes in prefrontal cortex structure and function, dopamine system sensitivity, and stress response circuitry that persist into adulthood, even after THC exposure ends.
This is why age of onset is such a critical risk factor for psychosis. The Dunedin Multidisciplinary Health and Development Study -- a landmark longitudinal study from New Zealand that has followed a cohort from birth to midlife -- found that cannabis use before age 15 was associated with a 4-fold increase in the risk of schizophreniform disorder at age 26 (Arseneault et al., 2002, BMJ). Use after age 18 carried less risk, though it was still elevated. For more on how cannabis affects the developing brain, see our cannabis and the teenage brain page.
Peak Age of Psychosis Onset
Psychotic disorders, including schizophrenia, have their peak onset during late adolescence and early adulthood (ages 16-25) -- the same period when cannabis use typically begins and escalates. This overlap is not coincidental. Cannabis appears to accelerate the onset of psychosis in vulnerable individuals. Studies consistently show that cannabis users who develop schizophrenia do so approximately 2-3 years earlier than non-users who develop the same disorder. In some cases, cannabis may trigger psychosis in individuals who would not have developed it at all, or who would have developed only subclinical symptoms.
Genetic Risk Factors: COMT, AKT1, and Beyond
Not everyone who uses cannabis heavily develops psychosis. Genetics play a major role in determining individual vulnerability. Several specific genetic variants have been identified that modulate the cannabis-psychosis relationship:
COMT (Catechol-O-methyltransferase)
The COMT gene encodes an enzyme that breaks down dopamine in the prefrontal cortex. The Val158Met polymorphism determines how efficiently this enzyme works:
- Val/Val genotype: High COMT activity, rapid dopamine breakdown, lower baseline prefrontal dopamine. Cannabis use in Val/Val individuals is associated with the highest psychosis risk -- because their prefrontal cortex is already dopamine-poor, the THC-induced dopamine surge produces a disproportionate effect.
- Met/Met genotype: Low COMT activity, slower dopamine breakdown, higher baseline prefrontal dopamine. Relatively lower psychosis risk with cannabis use, though not zero.
The landmark Caspi et al. (2005, Biological Psychiatry) study demonstrated this gene-environment interaction: cannabis use increased psychosis risk specifically in individuals with the Val/Val genotype.
AKT1
The AKT1 gene is involved in dopamine signaling downstream of D2 receptors. Individuals with the AKT1 C/C genotype who use cannabis daily have been found to have approximately a 7-fold increased risk of psychotic disorder compared to non-users without this variant (Di Forti et al., 2012, Biological Psychiatry). This is one of the strongest gene-environment interactions identified in psychiatry.
Polygenic Risk
Beyond individual genes, polygenic risk scores (PRS) for schizophrenia -- which aggregate the effects of hundreds or thousands of common genetic variants -- interact with cannabis use. Individuals with high PRS who also use cannabis have substantially elevated psychosis risk compared to those with high PRS who do not use cannabis, or those with low PRS who do. Family history of psychosis or schizophrenia remains the most clinically accessible indicator of genetic vulnerability.
First-Episode Psychosis: The Cannabis Connection
First-episode psychosis (FEP) studies -- which examine patients at the time of their initial psychotic episode -- provide critical data on the cannabis-psychosis relationship because they minimize confounding factors such as the effects of chronic illness, medication, and institutionalization.
Key First-Episode Psychosis Findings
- Cannabis is the most commonly used substance among first-episode psychosis patients. Studies consistently report past-year cannabis use rates of 40-60% in FEP samples, compared to 10-15% in the general population.
- Earlier onset: FEP patients with a history of cannabis use present an average of 2-3 years younger than FEP patients without cannabis use. This suggests that cannabis accelerates the onset of psychosis in vulnerable individuals.
- Continued use worsens outcomes: FEP patients who continue cannabis use after their first episode have higher relapse rates, more severe positive symptoms, poorer functional outcomes, and higher rates of rehospitalization compared to those who discontinue.
- Dose-response in FEP populations: Among FEP patients, those who used cannabis more frequently and at higher potency have more severe psychotic symptoms at presentation.
These data have direct clinical implications. In our practice, any patient presenting with a first psychotic episode receives thorough substance use assessment, with particular attention to cannabis. Cessation of cannabis is a non-negotiable recommendation for anyone who has experienced psychosis -- continued use dramatically worsens the prognosis.
Distinguishing Cannabis Psychosis from Schizophrenia
Clinically, distinguishing cannabis-induced psychosis from the onset of schizophrenia can be extremely difficult -- and the distinction has been increasingly questioned as research reveals how deeply intertwined these conditions can be.
| Feature | Cannabis-Induced Psychosis | Schizophrenia |
|---|---|---|
| Onset | Clearly linked to cannabis use | May or may not involve substance use |
| Resolution | Usually resolves within days-weeks of cessation | Persists independent of substance use |
| Negative symptoms | Less prominent | Prominent (flat affect, social withdrawal, avolition) |
| Insight | May be partially preserved | Often severely impaired |
| Visual hallucinations | More common | Less common (auditory predominate) |
| Conversion rate | 34-50% develop chronic psychosis within 3-5 years | N/A (is the chronic condition) |
The high conversion rate is what makes this distinction clinically treacherous. A young person presenting with "just" cannabis-induced psychosis has a roughly 1 in 3 to 1 in 2 chance of developing schizophrenia within 5 years. This is the highest conversion rate of any substance-induced psychosis -- higher than alcohol-induced (24%), amphetamine-induced (22%), or hallucinogen-induced (21%) psychosis (Starzer et al., 2018, American Journal of Psychiatry).
This means that cannabis-induced psychosis should never be dismissed as "just a bad trip." It is a clinical red flag indicating high risk for a chronic, debilitating psychiatric disorder. Every patient with a cannabis-related psychotic episode requires close psychiatric follow-up, substance use cessation support, and monitoring for the emergence of a primary psychotic disorder.
Recovery and Prognosis
Recovery from cannabis-induced psychosis is possible, but prognosis depends heavily on what happens after the acute episode. The single most important prognostic factor is whether the individual continues to use cannabis.
Prognostic Factors
- Cannabis cessation: Complete discontinuation of cannabis after a psychotic episode is the strongest modifiable factor for preventing recurrence and conversion to chronic psychosis.
- Speed of treatment: Shorter duration of untreated psychosis (DUP) is associated with better outcomes. Early psychiatric intervention matters.
- Family history: Individuals with a family history of schizophrenia have higher conversion rates.
- Age at first episode: Younger age at first psychotic episode is associated with worse long-term outcomes.
- Severity of initial episode: More severe initial presentations carry somewhat higher conversion risk.
- Social support: Strong family support, engagement with treatment, and stable housing improve prognosis.
The treatment approach for cannabis-related psychosis involves: acute management (antipsychotic medication if needed, supportive care, ensuring safety), cannabis cessation (the non-negotiable cornerstone of treatment), ongoing psychiatric monitoring (regular follow-up to detect early signs of chronic psychosis), and psychosocial support (therapy, family education, relapse prevention for cannabis use). For patients at high risk of conversion, coordinated specialty care programs for early psychosis offer the best outcomes.
Why Cannabis-Related Psychosis Matters More Now Than Ever
Three converging trends have created what some researchers describe as a "perfect storm" for cannabis-related psychosis:
1. Unprecedented THC Potency
Today's cannabis products deliver THC doses that were unimaginable a generation ago. Average flower potency has increased 4-6 fold since 1995. Concentrates, which are increasingly popular among young users, deliver THC at 60-90%. Vape cartridges provide convenient, discreet access to high-potency THC. The human brain's endocannabinoid system did not evolve to handle these exposure levels.
2. Expanded Access Through Legalization
Legalization has increased access, reduced risk perception, and normalized cannabis use. While legalization has important benefits (reduced incarceration, regulated products, tax revenue), it has also removed social and legal deterrents to use, particularly among young people. Decreased perception of risk is one of the strongest predictors of increased use in adolescent populations.
3. Adolescent Exposure at a Critical Window
Despite legalization being restricted to adults (21+), adolescent access has not decreased in most jurisdictions. Adolescents are therefore being exposed to dramatically more potent products during the most vulnerable period of brain development. The combination of immature prefrontal cortex, active endocannabinoid system development, peak sensitivity to psychotogenic effects of THC, and unprecedented product potency creates a risk profile that simply did not exist in previous generations.
This is not an argument against legalization. It is an argument for evidence-informed policy that addresses potency regulation, youth access, public education about dose-dependent risk, and research funding for prevention and treatment. The current approach -- legalizing without adequate attention to potency, youth access, and mental health consequences -- is leaving the most vulnerable populations unprotected.
Dr. Sultan's PAWS Project: Preventing Cannabis-Related Psychosis
As an NIH NIDA K12-funded researcher at Columbia University, I am acutely aware of the gap between the evidence base on cannabis-related psychosis and the available tools to prevent and treat it. This gap is the driving force behind our PAWS digital therapeutic project.
PAWS is being developed as a digital intervention that addresses both cannabis use disorder and cannabis-related psychosis risk. The project recognizes that traditional treatment models have significant limitations -- including access barriers, stigma, low treatment completion rates (only 36.8% of adolescents complete CUD treatment), and the absence of FDA-approved medications. A digital therapeutic approach has the potential to reach patients who would never walk into a clinic, deliver evidence-based interventions at scale, and provide ongoing monitoring for psychosis risk indicators.
My research collaborators in this work include Dr. Yasmin Hurd at Mount Sinai, whose translational neuroscience research on the endocannabinoid system and CBD's therapeutic potential has been foundational to the field; Dr. Sharon Levy at Harvard, an expert in adolescent substance use and early intervention; Dr. Kevin Gray at the Medical University of South Carolina, who has led pivotal clinical trials for adolescent CUD treatment; and Dr. Carlos Blanco at NIDA, whose epidemiological work has quantified the scope of cannabis-related mental health burden at the population level.
Frequently Asked Questions About Cannabis and Psychosis
Q: Can cannabis cause psychosis?
A: Yes. Cannabis can cause both acute psychotic episodes and increase the long-term risk of chronic psychotic disorders. Daily use of high-potency cannabis increases psychosis risk 4-5x (Di Forti et al., 2019, Lancet Psychiatry). The evidence supports a causal contribution model.
Q: What is cannabis-induced psychosis?
A: A psychotic episode triggered by cannabis use, featuring hallucinations, delusions, paranoia, and disorganized thinking. It usually resolves when cannabis is cleared from the system, but 34-50% of cases convert to chronic schizophrenia-spectrum disorders within 3-5 years -- the highest conversion rate of any substance-induced psychosis.
Q: Does cannabis cause schizophrenia?
A: Cannabis does not "cause" schizophrenia in isolation, but it is a significant contributing cause. A 2023 Danish population study found that cannabis use disorder was associated with approximately 30% of schizophrenia diagnoses in young men. Cannabis use accelerates onset by 2-3 years and may trigger the disorder in genetically vulnerable individuals who might not otherwise develop it.
Q: How does THC potency affect psychosis risk?
A: THC potency is one of the strongest modifiable risk factors. Daily high-potency cannabis use (>10% THC) increases psychosis risk nearly 5-fold. In cities where high-potency cannabis dominates, up to 50% of new psychosis cases may be attributable to cannabis (Di Forti et al., 2019). THC has risen from 4% (1995) to 15-25% in flower and 60-90% in concentrates.
Q: Who is at greatest risk?
A: Adolescents (especially those using before age 15), daily users, high-potency product users, those with COMT Val/Val or AKT1 C/C genotypes, people with family history of psychosis, and individuals already showing subclinical psychotic experiences.
Q: Can you recover from cannabis-induced psychosis?
A: Many people recover fully if they discontinue cannabis. However, 34-50% develop chronic psychotic disorders within 3-5 years. Cannabis cessation is the single most important factor for recovery. Continued use after a psychotic episode dramatically worsens prognosis.
Q: Does CBD protect against psychosis?
A: Preliminary evidence suggests CBD has antipsychotic properties (McGuire et al., 2018). However, modern cannabis products contain negligible CBD. The historical natural protection from balanced THC:CBD ratios has been eliminated by selective breeding for maximum THC.
Q: Why does cannabis psychosis matter more now?
A: Three converging factors: unprecedented THC potency (4-6x increase since 1995, concentrates at 60-90%), expanded access through legalization (with decreased risk perception), and adolescent exposure to high-potency products during critical brain development. The result is measurably increasing cannabis-related psychosis presentations across North America and Europe.
Related Resources
Cannabis & Psychosis Overview
Shorter overview of the cannabis-psychosis connection for quick reference.
Quick Overview →Cannabis & the Teenage Brain
How cannabis affects developing brains and why adolescents are uniquely vulnerable.
Teen Brain →PAWS Digital Therapeutic
Dr. Sultan's intervention for CUD and cannabis-related psychosis prevention.
PAWS Project →Cannabis & Mental Health Hub
Complete cannabis research hub with links to all topics.
Cannabis Hub →Concerned About Cannabis and Psychosis?
Dr. Ryan Sultan provides evidence-based evaluation for cannabis-related psychosis risk. As an NIH NIDA-funded researcher at Columbia University, he brings clinical expertise and cutting-edge research to every consultation.
Schedule a ConsultationTelehealth appointments available | Board-Certified Psychiatrist | Columbia University Faculty
Cannabis Research
Cannabis Hub
Cannabis Use Disorder
Is Cannabis Addictive?
Cannabis Withdrawal
Brain & Health
Cannabis & Teen Brain
Brain Effects
THC Potency
Marijuana vs Alcohol
Dr. Sultan
Contact
Schedule Consultation
ADHD Resources
Columbia University Irving Medical Center
© 2026 Ryan S. Sultan, MD | Assistant Professor of Clinical Psychiatry, Columbia University Irving Medical Center
The information on this page is for educational purposes only and does not constitute medical advice. Always consult with a qualified healthcare professional for diagnosis and treatment of any medical condition.