Cannabis-Induced Psychotic Disorder (CIPD)
Cannabis-induced psychotic disorder (CIPD) is hallucinations or delusions severe enough to need clinical care, appearing during or soon after cannabis use (or, less often, after abrupt withdrawal), and not explained by delirium alone. It sits on a spectrum that runs from transient intoxication effects to an episode that looks—and sometimes later becomes—indistinguishable from schizophrenia.
The episode is a medical problem. It is also one of the strongest clinical warning signs we have that a longer-lasting psychotic illness may follow.
How the diagnosis is made
In DSM-5-TR, CIPD is a form of substance/medication-induced psychotic disorder. The core requirements are:
Delusions, hallucinations, or both
Evidence that the symptoms began during, or soon after, cannabis intoxication or withdrawal
Symptoms that are not better explained by a primary psychotic disorder (for example, they did not clearly precede cannabis use, and there is no established pattern of non-substance episodes)
The disturbance causes marked distress or impairs work, school, or relationships
DSM-5-TR treats psychotic symptoms that last more than about a month after intoxication or withdrawal ends as a reason to reconsider the diagnosis and look harder for schizophrenia-spectrum illness. In practice the line is messy. Heavy daily use, high-THC products, and poor recall make timing hard to pin down, and many people who later receive a schizophrenia diagnosis first present as “cannabis-induced.”
ICD-11 lists cannabis-induced psychotic disorder separately (code 6C41.6), alongside cannabis-induced delirium, mood disorder, and anxiety disorder. The diagnosis is reserved for symptoms that dominate the picture and require treatment in their own right.
How common is it?
Numbers depend on what is being counted.
Mild, short-lived psychotic-like experiences during intoxication are reported by a wide range of users—studies quote anywhere from about 5% to 50%, depending on how the question is asked. A full psychotic episode is much rarer. Pooled research across observational, experimental, and medical-cannabis studies puts a severe cannabis-associated psychotic episode at about 0.5% of people exposed to cannabis or THC—roughly 1 in 200. Among people who use cannabis, lifetime cannabis-induced psychotic symptoms have been estimated at about 0.47%.
That last figure is rising in several high-income countries. Hospitalizations and registry diagnoses of cannabis-induced psychosis increased substantially in Norway, Denmark, and Sweden between 2000 and 2016, and Canadian hospital data showed a further climb after 2016. Those trends track greater availability of high-THC products more closely than they track a sudden change in how psychiatrists code charts.
CIPD is not evenly distributed. It clusters in younger people, frequent users, and people using high-potency products. Experimental studies that give THC to volunteers produce clinically meaningful positive psychotic symptoms in a large minority of participants; medicinal-cannabis trials in older adults report far lower rates. The difference is dose, product, age, and who is being studied.
What it looks like
The typical episode is an acute paranoid psychosis.
Persecutory delusions are the most consistent feature: the conviction that one is being followed, recorded, poisoned, or plotted against. Auditory hallucinations, ideas of reference, grandiosity, agitation, conceptual disorganization, and loss of insight are common. Visual hallucinations occur more often than in many primary schizophrenic episodes, which can be a clinical clue but is not diagnostic. Anxiety, insomnia, and reduced appetite often travel with the psychosis, especially if withdrawal is underway.
Two time courses matter:
During or right after use. Symptoms may start within hours of a large THC dose. More than half of milder cannabis-associated psychotic presentations settle within 24 hours. Symptoms that last beyond a week, or that occur in someone with a prior psychotic episode, lead to hospitalization in a majority of cases (reported ranges around 54% to 76%).
After stopping. Abrupt cessation of heavy daily use can itself precipitate psychosis, usually in people who meet criteria for cannabis withdrawal (irritability, insomnia, restlessness, low appetite). In case series, frank psychosis often appears several days after the last use—median around day 6—not only while the person is still intoxicated. Sleep collapse is a frequent prodrome. This is easy to miss if clinicians assume cannabis can only cause psychosis while it is in the bloodstream.
Synthetic cannabinoid receptor agonists (“spice,” “K2,” and related products) are more likely than plant cannabis to produce delirium and severe acute psychosis and should be treated as a higher-risk exposure.
Causes and the argument about causality
Delta-9-tetrahydrocannabinol (THC) is a partial agonist at CB1 receptors. In the brain that system helps regulate dopamine, stress responses, and the salience of incoming information. A single THC dose, given in a laboratory, increases positive, negative, and general psychiatric symptoms with large effect sizes compared with placebo. That experimental fact is the cleanest evidence that THC is not merely correlated with psychosis—it can produce psychotic symptoms on demand.
Epidemiology adds a dose–response curve. Heaviest cannabis users have roughly a fourfold higher risk of schizophrenia or related psychotic outcomes than non-users. Daily use of high-potency cannabis (often defined as THC greater than 10%) has been associated with nearly fivefold higher odds of psychotic disorder compared with never-use. In the European EU-GEI first-episode study, differences in daily and high-potency use helped explain why new psychosis rates varied so widely across cities; researchers estimated that if high-potency cannabis were unavailable, about 12% of first-episode cases across sites—and as many as half in Amsterdam—might not have occurred, assuming the association is causal.
A 2022 World Federation of Societies of Biological Psychiatry task force put the current scientific position carefully. Converging evidence supports a causal contribution of cannabis to psychoses that range from transient states to chronic illness. Risk rises with dose and with earlier age of first use. Cannabis is neither necessary nor sufficient: most users never become psychotic, and most people with schizophrenia have not had CIPD. Reverse causation (people in a prodrome using cannabis to self-medicate) and confounding (shared genes, childhood trauma, other drugs) have not been ruled out completely. The most accurate model is that cannabis is one causal component among several, and a modifiable one.
Vulnerability is not evenly shared. Family or personal history of psychosis, other mental disorders (bipolar disorder, depression, anxiety), adolescent onset of use, male sex in several cohort studies, and variants affecting catechol-O-methyltransferase (COMT), an enzyme that helps clear dopamine in the prefrontal cortex, have all been linked to higher risk. People with an established psychotic illness who keep using cannabis have more relapses, more hospitalizations, and more positive symptoms.
Who is at highest risk of CIPD itself
Risk is highest when several of the following stack:
Daily or near-daily use
High-THC flower, concentrates, or vapes
Starting in adolescence
Prior psychotic symptoms, even if brief
Family history of schizophrenia or bipolar disorder
Another psychiatric diagnosis
Cannabis use disorder
Abrupt cessation after heavy use
Concurrent use of stimulants or synthetic cannabinoids
Weekly or daily use is the frequency band where population risk of later psychotic disorder becomes clearly elevated; occasional yearly use does not show the same signal in recent dose–response analyses.
CIPD is often the first chapter, not the whole book
This is the statistic families need to hear. Among substance-induced psychoses, cannabis has the highest rate of later conversion to a primary psychotic disorder.
In a large Danish registry study, 47.4% of people with cannabis-induced psychosis later received a diagnosis of schizophrenia or bipolar disorder; 41.2% converted specifically to schizophrenia—about twice the conversion risk of non-cannabis substance-induced psychoses. A meta-analysis of transition after brief and substance-induced psychoses estimated that about one-third (34%) of CIPD cases later meet criteria for schizophrenia. Older reviews put the long-term rate of a schizophrenia-like illness after CIPD in a band from about one-fifth to one-half.
The absolute jump in risk is stark in emergency-department data. In a cohort of 9.8 million people in Ontario, an ED visit for cannabis-induced psychosis was associated with a 241-fold higher risk of a schizophrenia-spectrum diagnosis within three years compared with the general population. An ED visit for cannabis use without documented psychosis still carried a 14-fold increase. Those ratios describe a high-risk clinical population, not a typical recreational user.
Conversion is more likely when negative symptoms are prominent, premorbid functioning was already poor, symptoms do not remit within about a month, use started young, or there is a family history of psychosis. Rapid onset within a day or two of use, and equally rapid clearing with abstinence, point toward a more purely substance-driven episode—but they do not guarantee it.
Treatment
Immediate priorities are safety, a low-stimulation environment, medical clearance (other drugs, hypoglycemia, infection, head injury), and stopping further cannabis and synthetic cannabinoids. Agitation is often managed with a benzodiazepine plus a second-generation antipsychotic. First-generation agents such as haloperidol are still used in emergency settings when needed for rapid control.
Antipsychotics are the mainstay when symptoms are severe or last beyond intoxication. Olanzapine, risperidone, quetiapine, and aripiprazole are commonly used. Once the acute episode remits, a gradual taper can be considered if the working diagnosis remains purely substance-induced and abstinence is solid. Many clinicians keep treatment in place longer when risk factors for conversion are present, because CIPD is a poor place to gamble on a very short course.
Real-world Finnish data after a first CIPD diagnosis found that any antipsychotic use was associated with a 25% lower risk of later psychosis hospitalization. Long-acting injectable aripiprazole and olanzapine showed the largest reductions (on the order of 70% in that analysis), with signals also for clozapine and oral aripiprazole and olanzapine. Adherence is the practical problem: the same people at highest risk of relapse are often least likely to take a daily pill.
Sobriety is not a slogan; it is the intervention with the clearest effect on course. Continuing cannabis after an induced episode is associated with a far higher chance of another psychotic relapse (one withdrawal-psychosis series reported an odds ratio near 14). Motivational interviewing, contingency management, and treatment of cannabis use disorder belong in the same plan as the antipsychotic, not after it. Sleep must be protected; insomnia is both a withdrawal symptom and a relapse trigger.
Risk of a psychotic disorder after stopping cannabis appears to fall toward the risk of never-users after many months of sobriety—on the order of eight to nine months in one King’s College London analysis—though daily users of high-potency products may remain at elevated risk for longer. Stopping suddenly can briefly raise risk in heavy users, which is an argument for medically supported cessation rather than an argument for continuing use.
People with CIPD also have higher mortality than matched population controls in Scandinavian registries, a reminder that this is not a benign billing code. Suicide risk, accidents, and ongoing substance use all contribute. Follow-up should look like first-episode psychosis care: frequent contact in the first one to two years, family education, and a low threshold to restart medication if symptoms return.
An episode that meets criteria for CIPD should be treated as both an acute emergency and a high-risk marker. The two things that most reliably change the next chapter are staying off cannabis and staying on antipsychotics.
