Cannabis Use in Psychiatry: Exploring Potency, Vulnerability, and Clinical Approach

Exploring Potency, Vulnerability, and Clinical Approach

Psychiatry · Seminar week 40 · released October 12, 2026 · includes a discussion video

With increasing prevalence and potency of modern cannabis, understanding its clinical implications is vital for psychiatry and emergency settings. Primary source P9 provides…

Learning Objectives

• Quantify cannabis exposure by product, THC concentration, route, dose, frequency, and timing.
• Recognize psychiatric risks associated with high-potency THC and individual vulnerability.
• Differentiate cannabis-induced psychosis from intoxication, delirium, bipolar disorder, and primary psychotic disorders.
• Diagnose and manage cannabinoid hyperemesis syndrome using evidence-informed acute-care protocols.
• Anticipate and treat cannabis withdrawal in psychiatric inpatient settings.
• Identify pharmacokinetic and pharmacodynamic interactions between cannabinoids and psychotropic medications.
• Apply motivational interviewing, harm reduction, and longitudinal follow-up to cannabis-related presentations.

Historical Evolution and Modern Potency of Cannabis

%%FIG0%% Duration: 10 Minutes

Cannabis has been used medicinally, ceremonially, and recreationally for centuries, but the exposure encountered in modern psychiatric practice is not equivalent to the plant preparations described in historical reports. Nineteenth-century Western pharmacopeias included cannabis extracts with poorly standardized cannabinoid content. Twentieth-century prohibition curtailed clinical use and research, while selective breeding, indoor cultivation, sinsemilla production, solvent extraction, vaping technology, and commercial legalization subsequently transformed both product potency and patterns of consumption. Contemporary “cannabis” may refer to dried flower, resin, a vaporized concentrate, an edible, a beverage, pharmaceutical THC, purified cannabidiol, an intoxicating hemp-derived cannabinoid, or a synthetic cannabinoid receptor agonist.

Teaching Point: A joint, vape cartridge, or “dab” is not a standardized dose. Clinicians should document the molecule, product, labeled potency, amount, route, frequency, and timing rather than simply recording “marijuana use.”

Potency generally denotes the percentage of a product’s dry weight composed of delta-9-tetrahydrocannabinol, or THC. Analysis of 38,681 cannabis samples obtained through the United States federal monitoring program found that average THC concentration in plant material increased from approximately 4% in 1995 to approximately 12% in 2014. Average cannabidiol, or CBD, content declined, shifting the THC-to-CBD ratio from roughly 14:1 to approximately 80:1 (ElSohly et al., PMID: 26903403). Seizure samples do not perfectly represent legal retail markets, and commercial labels may be inaccurate, but the direction of change is clear. Retail flower commonly exceeds 20% THC, while concentrates frequently contain 60–90% THC. A threshold such as 10% THC, considered “high potency” in several landmark European studies, now describes relatively ordinary commercial flower.

Potency is not the same as delivered dose. One-half gram of 20% flower contains 100 mg of THC before combustion and inhalational losses. A small amount of an 80% concentrate can deliver a similar nominal quantity in a few inhalations. Inhaled THC reaches the brain within minutes, which permits rapid titration but also rapid dose escalation. Oral THC has a delayed onset—often 30–120 minutes—and undergoes first-pass metabolism to psychoactive 11-hydroxy-THC. Effects may peak several hours after ingestion and persist for six to eight hours or longer. Patients who conclude that an edible “did not work” and redose at 30 minutes are at particular risk for panic, delirium-like confusion, severe anxiety, and psychotic symptoms.

Framework: Translate every history into five dimensions: product, potency, amount, frequency, and route. Then ask about age at initiation, recent escalation, time to first use after waking, longest abstinence, prior paranoia or vomiting, and the symptom the patient is trying to treat.

THC is a partial agonist at CB1 receptors, which are abundant in cortical, hippocampal, striatal, cerebellar, and limbic circuits. CB1 activation alters presynaptic regulation of glutamate and GABA, disrupting memory, attention, threat processing, salience attribution, and sensory integration. Controlled administration of THC can produce transient paranoia, perceptual alteration, negative-like symptoms, anxiety, and cognitive impairment even in healthy volunteers (D’Souza et al., PMID: 15173844). CBD has more complex pharmacology and may attenuate selected THC effects under some experimental conditions, but this finding is inconsistent and does not establish high-CBD cannabis as protection against psychosis.

The 2022 systematic review by Petrilli and colleagues found that higher-potency cannabis, compared with lower-potency products, was associated most consistently with psychosis and cannabis use disorder; findings for anxiety and depression were less consistent (PMID: 35901795). These are predominantly observational data, with residual confounding from frequency, socioeconomic adversity, trauma, tobacco, other substances, and pre-existing psychiatric symptoms. Nevertheless, experimental THC effects, temporal associations, dose-response findings, and stronger associations with frequent high-potency use together support a clinically meaningful risk signal.

Nuance: THC is neither necessary nor sufficient to cause schizophrenia. A stress-diathesis formulation is more accurate: THC exposure interacts with developmental stage, genetics, sleep loss, trauma, and other biological or social vulnerabilities.

The highest-concern profile includes adolescent or early initiation, daily or near-daily use, concentrates, prior cannabis-related paranoia, attenuated psychotic symptoms, a personal or first-degree family history of psychosis or bipolar disorder, sleep deprivation, and stimulant co-use. The Colorado youth review by Hinckley and colleagues emphasizes the convergence of increased availability, changing risk perception, high-potency products, and developmental vulnerability (PMID: 35428897).

Decision Point: Following a psychotic episode, the safest recommendation is abstinence from THC. If the patient declines, prioritize eliminating concentrates and synthetic cannabinoids, reducing frequency and THC content, avoiding uncertain-dose edibles, protecting sleep, avoiding co-use, and establishing early warning signs. These are harm-reduction measures, not declarations of safety (PMIDs: 37450645, 37882050).

Audience Poll: Which exposure would concern you most: 5% flower once monthly, a 10 mg edible every weekend, 25% flower every night, or an 85% concentrate used throughout the day?

Cannabis-Induced Psychosis: Symptoms and Differentiation

%%FIG1%% Duration: 10 Minutes

Cannabis-induced psychotic disorder is defined by prominent delusions or hallucinations that develop during or soon after cannabis intoxication, cause clinically significant impairment, do not occur exclusively during delirium, and are not better explained by an independent psychotic disorder. Mild suspiciousness, altered time perception, derealization, or perceptual intensification with retained reality testing may occur during uncomplicated intoxication. The diagnosis of cannabis-induced psychotic disorder becomes appropriate when psychosis is the clinical focus, conviction is substantial, insight is impaired, or behavior becomes unsafe or markedly disorganized.

Presentations commonly include persecutory or referential delusions, auditory or visual hallucinations, severe anxiety, panic, insomnia, agitation, and disorganized thinking. Symptoms may begin after a large exposure, a switch to concentrates, repeated edible dosing, or several days of heavy use and sleep deprivation. Some patients improve within days after abstinence and restorative sleep; others remain psychotic for weeks. Resolution within one week supports a substance-related formulation but is not a diagnostic criterion.

MUST ACT: Do not use “it should resolve in a week” as a reason to defer treatment or follow-up. Persistence approaching or exceeding one month, symptoms predating heavy cannabis use, or recurrence during verified abstinence should shift the formulation toward a primary psychotic disorder.

No single symptom reliably separates cannabis-induced psychosis from schizophrenia. On average, substance-induced psychoses may show greater affective reactivity, anxiety, visual phenomena, and initial insight, with fewer negative symptoms or less premorbid deterioration. These are probabilistic observations, not bedside rules. Schizophrenia becomes more likely when social or academic decline, cognitive change, negative symptoms, unusual beliefs, or attenuated hallucinations began before cannabis escalation; when psychosis occurs without intoxication; or when symptoms persist despite sustained abstinence. A first-degree family history increases concern but does not determine diagnosis.

Framework: Build the differential from four elements: the exposure–symptom timeline, premorbid trajectory, observation during abstinence, and collateral information. Each is more informative than a positive urine cannabinoid screen.

Bipolar mania should be considered when decreased need for sleep is accompanied by elevated or persistently irritable mood, pressured speech, increased goal-directed activity, grandiosity, impulsive spending, or sexual risk. Severe insomnia caused by fear is not equivalent to decreased need for sleep. Delirium is suggested by impaired or fluctuating attention, disorientation, altered consciousness, abnormal vital signs, or a medical precipitant. Stimulants, hallucinogens, steroids, anticholinergics, and synthetic cannabinoid receptor agonists can produce similar syndromes. Synthetic cannabinoids are particularly important because they may cause severe agitation, seizures, hyperthermia, kidney injury, or autonomic instability while escaping standard toxicology screens.

A urine cannabinoid immunoassay establishes exposure, not the timing, dose, potency, impairment, or cause of the current mental state. THC metabolites can remain detectable for weeks in chronic users. Medical evaluation should therefore be driven by the syndrome: vital signs, glucose, oxygenation, attention, hydration, medication review, neurological examination, and targeted laboratory or imaging studies. New focal deficits, seizure, fever, catatonia, fluctuating consciousness, or an atypical age of onset warrants a broader neurological, infectious, endocrine, autoimmune, or toxicologic evaluation.

Epidemiological evidence supports a dose-related association. In the EU-GEI case-control study, daily cannabis use was associated with approximately threefold higher odds of psychotic disorder, while daily use of products containing at least 10% THC was associated with nearly fivefold higher odds compared with never use (Di Forti et al., PMID: 30902669). A dose-response meta-analysis likewise found the highest exposure group had substantially greater odds of psychosis than nonusers (PMID: 26884547). These estimates do not translate directly into an individual’s absolute risk and cannot eliminate confounding or reverse causation.

Acute management begins with a low-stimulation setting, verbal de-escalation, hydration, nutrition, and restoration of sleep. Assess suicide risk, violent intent, command hallucinations, access to weapons, capacity, and ability to care for basic needs. If oral treatment is feasible, first-episode doses such as risperidone 1–2 mg or olanzapine 5–10 mg orally are reasonable for persistent frank psychosis or significant agitation. Lorazepam 1–2 mg orally or intramuscularly may be considered for severe anxiety or agitation after assessing respiratory-depressant co-ingestion. Follow institutional protocols for imminent danger, monitor for QT prolongation and extrapyramidal effects, and avoid close temporal administration of intramuscular olanzapine with parenteral benzodiazepines.

Decision Point: Admit when there is dangerousness, grave disability, severe disorganization, catatonia, inability to ensure observation and abstinence, an uncertain medical cause, or inadequate support. Rapid improvement can inform the formulation but should not determine disposition by itself.

The prognosis deserves particular emphasis. A Danish registry study found that 47.4% of patients diagnosed with cannabis-induced psychosis subsequently received a schizophrenia-spectrum or bipolar diagnosis; 41.2% converted specifically to schizophrenia (PMID: 29179576). A separate meta-analysis estimated a 34% transition from cannabis-induced psychosis to schizophrenia (PMID: 31618428). These selected clinical cohorts should not be equated with every transient intoxication-related paranoid experience, but they demonstrate that “substance-induced” is not synonymous with benign. Early-psychosis referral, cannabis-use-disorder treatment, collateral history, and longitudinal reassessment over sustained abstinence are essential.

Teaching Point: Cannabis-induced psychosis is a provisional longitudinal formulation, not reassurance that the patient cannot later develop schizophrenia or bipolar disorder.

Audience Poll: Which finding most strongly changes your working diagnosis: a positive THC screen, complete remission after sleep, a six-month prodrome, or a first-degree relative with schizophrenia?

Cannabinoid Hyperemesis Syndrome: Diagnosis and Management

%%FIG2%% Duration: 10 Minutes

Cannabinoid hyperemesis syndrome, or CHS, is a recurrent vomiting disorder associated with prolonged, frequent cannabinoid exposure. The classic course includes a prodromal phase of morning nausea and abdominal discomfort, a hyperemetic phase with repetitive vomiting, abdominal pain, dehydration, and repeated bathing, and a recovery phase following cessation. Patients may paradoxically increase cannabis use during the prodrome because acute cannabis exposure previously relieved nausea, thereby reinforcing the behavior contributing to recurrent illness.

The American Gastroenterological Association’s 2024 clinical practice update describes a practical phenotype: stereotyped episodic vomiting occurring at least three times annually, cannabis use more than four times weekly for longer than one year before symptom onset, and resolution after sustained abstinence—typically six months or at least three usual vomiting cycles (PMID: 38456869). These criteria are a clinical framework, not a biomarker. Because durable remission cannot be demonstrated during one emergency visit, “suspected CHS” is often the most accurate initial diagnosis.

Teaching Point: Relief with hot showers is strongly suggestive but not pathognomonic. Hot bathing also occurs in cyclic vomiting syndrome, and some patients with CHS do not report it. Durable remission with abstinence is the most persuasive discriminator.

The mechanism remains incompletely established. THC is centrally antiemetic in many acute settings, yet chronic high exposure may create a paradoxical proemetic state through CB1-receptor desensitization, altered gastric motility, hypothalamic thermoregulatory effects, stress-system dysregulation, and transient receptor potential vanilloid-1, or TRPV1, signaling. Both heat and capsaicin activate TRPV1 and may alter substance P signaling, providing a plausible explanation for temporary relief. These remain hypotheses: gastric emptying may be delayed, normal, or accelerated, and no receptor assay, genetic test, serum concentration, or imaging finding confirms CHS.

MUST ACT: A cannabis history must not terminate the medical evaluation. Exclude dangerous alternatives according to the presentation, including pregnancy or ectopic pregnancy, bowel obstruction or perforation, pancreatitis, biliary disease, appendicitis, infection, diabetic or starvation ketoacidosis, adrenal disease, gastroparesis, renal colic, intracranial pathology, toxic ingestion, and medication effects.

Cyclic vomiting syndrome is the closest mimic. Migraine history, autonomic symptoms, stereotyped episodes, symptom-free intervals, and hot-shower behavior may occur in either disorder. Other clues demanding broader investigation include peritonism, gastrointestinal bleeding, fever, focal neurological findings, anemia, abnormal liver tests, an abdominal mass, unintentional weight loss, a changing bowel pattern, or a new vomiting syndrome later in life. In psychiatric patients, also consider eating disorders, self-induced vomiting, alcohol or opioid withdrawal, lithium toxicity, and somatic preoccupation without prematurely psychologizing the presentation.

Framework: During the hyperemetic phase, assess glucose, volume status, creatinine, sodium, potassium, chloride, bicarbonate, calcium, and magnesium. Add a complete blood count, liver tests, lipase, urinalysis with ketones, pregnancy testing, toxicology, or imaging as indicated. Replace volume with isotonic crystalloid and reassess; correct hypokalemia, hypomagnesemia, hypoglycemia, and significant acid–base abnormalities.

Obtain an ECG before dopamine antagonists when vomiting, electrolyte disturbance, congenital long-QT risk, or concurrent QT-prolonging medication is present. Ondansetron, metoclopramide, or promethazine may be tried, although CHS is often refractory to conventional antiemetics. Avoid reflexive opioid treatment, which may worsen nausea and gut dysmotility while introducing additional dependence risk.

The HaVOC randomized trial compared intravenous haloperidol at 0.05 or 0.1 mg/kg with ondansetron 8 mg. Haloperidol produced greater improvement in nausea and abdominal pain and shortened emergency-department length of stay; both return visits for acute dystonia occurred after the 0.1 mg/kg dose (PMID: 33160719). An evidence-linked approach is haloperidol 0.05 mg/kg, generally capped at 5 mg, intravenously or intramuscularly under local policy. Avoid it with a prolonged QTc, uncorrected electrolyte abnormalities, Parkinson disease, Lewy body dementia, or previous severe dystonia. Monitor for akathisia, dystonia, sedation, and hypotension.

Topical capsaicin is a useful non-opioid alternative. Apply a thin layer of 0.1% cream to the anterior abdomen using gloves; avoid damaged skin, eyes, mucosa, occlusive dressings, and external heat. A 30-patient randomized pilot found greater nausea reduction at 60 minutes and more complete relief with capsaicin than placebo, although skin irritation limited treatment in one participant (PMID: 32569429). Evidence remains preliminary, and lower concentrations commonly stocked in hospitals have less direct trial support.

Decision Point: Admit for persistent inability to tolerate oral intake, acute kidney injury, consequential electrolyte or acid–base abnormalities, uncontrolled symptoms, pregnancy complications, diagnostic uncertainty, or an unsafe psychiatric or addiction context.

Haloperidol, capsaicin, fluids, and hot showers treat an episode; they do not prevent recurrence. Complete cessation of cannabis—including concentrates, edibles, delta-8 THC, and other intoxicating cannabinoid products—is the definitive intervention. A systematic review found symptom resolution in 62 of 64 reported patients who stopped cannabis, although this estimate arose mainly from selected case literature (PMID: 28000146). Explain the paradox without blame, arrange treatment for cannabis use disorder, and proactively address the withdrawal symptoms likely to follow cessation.

Audience Poll: Which finding most strongly confirms CHS: a positive urine screen, temporary relief with heat, failure of ondansetron, or sustained remission after cannabinoid abstinence?

Addressing Cannabis Withdrawal in Psychiatric Inpatient Settings

%%FIG3%% Duration: 10 Minutes

Cannabis withdrawal follows abrupt cessation or substantial reduction after heavy, prolonged THC exposure. DSM-5 criteria require clinically significant distress or impairment and at least three symptoms emerging within approximately one week: irritability, anger, or aggression; nervousness or anxiety; insomnia or disturbing dreams; decreased appetite or weight; restlessness; depressed mood; or a physical symptom such as abdominal pain, tremor, sweating, fever, chills, or headache.

Symptoms usually begin 24–48 hours after the last use and peak on days two through six. Appetite and many physical symptoms improve over one to two weeks, while insomnia, vivid dreams, irritability, and dysphoria may persist for three weeks or longer in heavy users (PMID: 34791767). This delayed peak matters on inpatient units: a patient may appear calm on admission, become sleepless and irritable on day three, and be labeled as having refractory mania, akathisia, or worsening psychosis when withdrawal is contributing.

Nuance: Withdrawal from plant cannabis is uncomfortable and relapse-promoting but does not ordinarily cause seizures or delirium. Seizure, marked autonomic instability, fluctuating attention, or severe persistent vomiting should prompt evaluation for alcohol or sedative withdrawal, synthetic cannabinoids, intoxication, infection, metabolic illness, or CHS.

Chronic THC exposure causes reversible CB1-receptor downregulation, with receptor availability beginning to recover during abstinence. A meta-analysis of 47 studies involving more than 23,000 participants estimated cannabis withdrawal prevalence at 47%, with substantially higher rates in clinical and inpatient populations (PMID: 32271390). Daily use, high cumulative exposure, tobacco co-use, and additional substance use disorders increase likelihood and severity.

Framework: At admission, record product, route, approximate THC concentration, amount, frequency, last use, prior withdrawal, longest abstinence, and co-use of nicotine, alcohol, benzodiazepines, opioids, stimulants, or synthetic cannabinoids. Establish baseline sleep, appetite, anxiety, agitation, mood, psychosis, suicidality, aggression, oral intake, and vital signs. Reassess these domains daily through at least days two through six.

The Marijuana Withdrawal Checklist or Cannabis Withdrawal Scale can structure observation, but neither should be treated like a medication-triggering alcohol-withdrawal scale. A urine cannabinoid concentration cannot reliably date cessation or measure withdrawal severity. Differential diagnosis remains active: akathisia produces an observable and subjective urge to move; mania includes mood and goal-directed activation; antidepressant discontinuation may cause sensory and flu-like symptoms; nicotine withdrawal can produce irritability and poor concentration; alcohol and benzodiazepine withdrawal carry seizure and delirium risk.

MUST ACT: First-line treatment is supportive and anticipatory. Explain the expected timeline, normalize symptoms without minimizing them, maintain hydration and nutrition, provide daytime light and activity, reduce overnight interruptions, and establish a consistent sleep schedule. Teach brief craving-management and anger-regulation strategies. Treat simultaneous nicotine withdrawal with nicotine replacement when appropriate, and manage alcohol or sedative withdrawal under its own validated protocol.

Symptom-targeted medications may be used cautiously. Acetaminophen or an NSAID can address headache or muscle discomfort when medically appropriate. Ondansetron may help nonspecific nausea, but recurrent or severe vomiting requires evaluation for CHS. Melatonin is a reasonable low-risk initial option for sleep. Hydroxyzine or trazodone may be considered selectively, accounting for anticholinergic burden, QTc, orthostasis, and next-day sedation. Benzodiazepines and Z-drugs should not become automatic cannabis “detox” regimens; reserve them for another evidence-based indication or severe, time-limited agitation under observation.

No medication is FDA-approved specifically for cannabis withdrawal or cannabis use disorder. Dronabinol 20 mg twice daily reduced withdrawal symptoms and improved treatment retention but did not improve abstinence in a 156-person randomized trial (PMID: 21310551). Nabiximols reduced withdrawal in a small inpatient trial, but it is not FDA-approved in the United States and remains an agonist-substitution strategy rather than routine psychiatric practice (PMID: 24430917). Gabapentin 1,200 mg/day reduced cannabis use and withdrawal symptoms in a 50-person proof-of-concept trial, but attrition was substantial and definitive replication is lacking (PMID: 22373942).

Decision Point: Cannabinoid agonists should not be improvised as routine inpatient treatment, particularly in active psychosis, CHS, pregnancy, sedation risk, or medication-seeking behavior. If considered in an exceptional setting, involve addiction expertise and use a defined monitoring and discontinuation plan.

Withdrawal management is not equivalent to treatment of cannabis use disorder. Motivational enhancement therapy, cognitive-behavioral coping-skills treatment, and contingency management have the strongest psychosocial support, particularly when combined and delivered over multiple sessions. Before discharge, identify the function cannabis served—sleep, trauma avoidance, anxiety relief, pain, social connection, or appetite—and offer safer treatment for that need. Provide an early follow-up appointment, coping plan for evening cravings, instructions for insomnia and mood deterioration, and explicit return precautions for suicidality, psychosis, or inability to maintain hydration.

Teaching Point: The principal clinical danger of uncomplicated cannabis withdrawal is often not medical collapse but relapse, behavioral escalation, premature discharge, or destabilization of a co-occurring psychiatric illness.

Audience Poll: On which inpatient day would you expect cannabis withdrawal to be most easily mistaken for worsening psychiatric illness?

Pharmacokinetic Interactions and Risk Assessment

%%FIG4%% Duration: 10 Minutes

Cannabis–medication interactions are frequently oversimplified. “Cannabis” may mean THC-dominant flower, an oral edible, a concentrate, low-dose retail CBD, or several hundred milligrams of pharmaceutical CBD. Clinically meaningful interaction risk depends on the cannabinoid, dose, route, frequency, smoking behavior, co-medications, and organ function.

Framework: Interpret exposure on four axes: molecule, dose, route, and time. Ask for the package or a photograph when possible. Record THC and CBD content, number of inhalations or servings, frequency, last use, recent changes, and co-use of tobacco, alcohol, sedatives, stimulants, or psychedelics.

Inhaled THC reaches peak blood concentrations within minutes, followed by rapid redistribution even while impairment persists. Oral THC has delayed and variable absorption and produces proportionally more active 11-hydroxy-THC. THC is highly lipophilic, extensively protein-bound, and metabolized principally through CYP2C9 with CYP3A contribution. Its terminal elimination is prolonged in chronic users because of tissue redistribution. This explains why urinary metabolites may remain detectable long after intoxication has ended; it does not mean that a patient remains continuously impaired.

Nuance: A positive urine cannabinoid immunoassay supports exposure, not current intoxication, dose, route, or causation. Standard screens may also miss synthetic cannabinoid receptor agonists.

CBD warrants special attention because high doses can inhibit CYP2C19, CYP2C9, CYP3A, and other pathways. Controlled interaction studies indicate that a high-CBD oral product can increase exposure to several CYP probe substrates, whereas THC alone at more typical experimental doses has shown fewer consistent inhibitory effects (PMID: 37313955). In-vitro enzyme inhibition should not automatically trigger a medication change, but it should prompt closer assessment when the patient uses high-dose CBD or takes a narrow-therapeutic-index drug.

A well-established example is clobazam: CBD inhibits metabolism of its active metabolite, N-desmethylclobazam, increasing exposure and the risk of sedation. Pharmaceutical CBD combined with valproate also increases the risk of transaminase elevation even without a proportionate rise in valproate concentration. Case reports and small studies describe interactions with warfarin, tacrolimus, citalopram, escitalopram, and other CYP substrates, but evidence quality and cannabinoid doses vary. Retail CBD is not pharmacokinetically inert, and actual content may differ from its label.

MUST ACT: Separate the effect of cannabis smoke from the effect of THC. Polycyclic aromatic hydrocarbons produced by combustion—not THC or nicotine—induce CYP1A2. Regular tobacco or cannabis smoking can therefore lower concentrations of CYP1A2 substrates such as clozapine and olanzapine. Abrupt smoking cessation on a nonsmoking unit may raise concentrations over the following days.

Clozapine or olanzapine toxicity after combined tobacco and cannabis cessation has been reported (PMID: 11981356). For a clozapine-treated patient whose smoking changes, document tobacco and cannabis separately, obtain a baseline steady-state trough concentration when feasible, and monitor sedation, confusion, orthostasis, tachycardia, sialorrhea, constipation, myoclonus, and seizure risk. Many smoking-cessation algorithms anticipate that a roughly 30% clozapine dose reduction may ultimately be required, but the adjustment must be individualized through clinical assessment, trough concentrations, and pharmacy support. If smoking resumes after discharge, clozapine exposure may fall again and psychosis may recur. Noncombusted vaping and edibles do not produce the same smoke-mediated CYP1A2 induction.

Pharmacodynamic interactions are often more immediate than CYP effects. THC combined with alcohol, opioids, benzodiazepines, sedating antipsychotics, gabapentinoids, or antihistamines increases sedation, falls, cognitive impairment, and driving risk. Tachycardia and orthostasis may be amplified by antipsychotics, tricyclic antidepressants, stimulants, or antihypertensives. Anxiety, panic, and impaired judgment can undermine adherence even in the absence of a measurable pharmacokinetic interaction.

Volume depletion creates an important indirect interaction. Lithium is not metabolized by CYP enzymes, but CHS-related vomiting, poor intake, and acute kidney injury reduce lithium clearance. New tremor, ataxia, confusion, dysarthria, or worsening vomiting in a cannabis-using patient taking lithium should trigger an urgent lithium level, creatinine, electrolytes, medication hold assessment, and fluid evaluation rather than attribution to anxiety.

Risk assessment should also include cannabis use disorder. Ask about unsuccessful efforts to cut down, craving, time devoted to use or recovery, role impairment, hazardous use, continued use despite harm, tolerance, and withdrawal. DSM-5 severity is mild with two or three criteria, moderate with four or five, and severe with six or more. The CUDIT-R or ASSIST can support screening but does not replace diagnosis.

Decision Point: Escalate medication review when high-dose CBD is started or stopped, smoked tobacco or cannabis changes in a patient taking clozapine or olanzapine, a narrow-therapeutic-index medication is involved, organ function deteriorates, or toxicity follows a cannabinoid change. Use a pharmacist, drug concentrations when interpretable, ECG or laboratory monitoring guided by the syndrome, and serial clinical examination.

The Fischer review should be used to support reduction of psychosis-related cannabis risk, not CYP interaction claims (PMID: 37450645). The more important practical lesson is that route changes can be as consequential as dose changes.

Audience Poll: For a patient taking clozapine, which aspect of cannabis use most directly affects CYP1A2: THC concentration, CBD concentration, nicotine content, or combustion?

Case Studies and Practical Application in Clinical Settings

%%FIG5%% Duration: 10 Minutes

MUST ACT: Do not allow a cannabis-positive urine result to terminate diagnostic reasoning. Cannabis may be the precipitant, a perpetuating factor, an incidental exposure, or one component of a polysubstance syndrome.

Integrated Clinical Case: Paranoia After a Potency Escalation

A 19-year-old university student is brought to the emergency department after barricading his bedroom because he believes classmates placed cameras in the smoke detector. He has slept approximately two hours over three nights, intermittently hears his name whispered, and has stopped eating because he fears contamination. Six months earlier he began using cannabis flower on weekends. During the preceding month he progressed to a concentrate labeled 85% THC, initially each evening and then repeatedly throughout the day. He uses energy drinks but denies prescribed medication, methamphetamine, cocaine, or hallucinogens. A maternal uncle has schizophrenia.

His pulse is 118/min, but temperature, blood pressure, oxygen saturation, glucose, attention, orientation, and neurological examination are normal. He is frightened, guarded, and internally preoccupied without fluctuating consciousness. Urine immunoassay is positive for cannabinoids and negative for the limited substances included on the standard panel.

Framework: Work in parallel across syndrome, safety, substances, and substrate. The syndrome is psychosis with severe insomnia, not delirium. Safety assessment covers suicidal or violent intent, command hallucinations, access to weapons, capacity, oral intake, and ability to accept care. The substance history must include exact product, route, timing, concentrates, edibles, delta-8 THC, synthetic cannabinoids, stimulants, alcohol, and medications. The substrate includes age, family history, premorbid function, prior attenuated symptoms, mood episodes, trauma, and medical red flags.

The leading formulations are cannabis-induced psychotic disorder, a first primary psychotic episode precipitated or unmasked by cannabis, mania with psychotic features, stimulant intoxication not detected or disclosed, and a medical cause. Family history and age increase concern for primary psychosis, while the abrupt onset after escalating an extremely potent product supports a cannabis contribution. Neither factor resolves the diagnosis.

He is placed in a low-stimulation room, receives oral fluids, and accepts olanzapine 5 mg orally disintegrating tablet. Medication is reassessed before additional dosing. Admission is appropriate because he cannot maintain nutrition, remains highly paranoid, and cannot participate in a safe outpatient plan. Collateral history reveals no prior mania but several months of declining class attendance, which raises concern that cannabis escalation may have occurred during an emerging prodrome.

After 72 hours of abstinence, sleep, structured care, and low-dose antipsychotic treatment, the hallucinations abate and he begins questioning the surveillance belief.

Nuance: Rapid improvement strengthens a cannabis-induced formulation but does not prove it. Sleep restoration and antipsychotic treatment can also improve an early episode of schizophrenia or mania. The preceding functional decline and family history require longitudinal observation.

The safest recommendation is complete THC abstinence. Early-psychosis follow-up should occur within days, with collateral assessment, substance-use treatment, medication review, and monitoring during sustained abstinence. The antipsychotic duration should be individualized rather than stopped automatically at discharge. Written warning signs include renewed insomnia, social withdrawal, suspiciousness, referential thinking, voices, reduced eating, or resumption of high-potency products.

Decision Point: When a patient is unwilling to abstain, provide harm reduction without implying that continued use is safe: stop concentrates and synthetic cannabinoids, reduce frequency and THC exposure, avoid uncertain-dose edibles, protect sleep, avoid stimulants and alcohol, never drive while impaired, and seek care promptly for paranoia or hallucinations. Collaborative harm-reduction approaches are particularly relevant in clinical-high-risk youth, although evidence that they prevent conversion remains limited (Kapler et al., PMID: 37882050).

Motivational interviewing starts by asking permission: “Could I share how the timing looks to me?” Elicit both perceived benefits and harms: “What does cannabis do for you, and what has become less good?” Reflect the patient’s reasons, connect the potency escalation and sleep loss with the psychosis without overstating certainty, and ask what the patient makes of that connection. An importance or confidence ruler can evoke change talk: “Why are you at four rather than zero?” Translate the response into one observable step, such as discarding cartridges, involving a roommate, or attending an addiction appointment.

Two brief variations illustrate common pitfalls. A clozapine-treated patient stops smoking tobacco and cannabis on admission and becomes sedated, confused, constipated, and myoclonic on day four. The team should suspect CYP1A2 de-induction, assess for infection and interacting inhibitors, obtain a clozapine level, and adjust treatment with pharmacy support. A lithium-treated patient presents with cyclic vomiting, compulsive hot showers, tremor, and ataxia. Before labeling the episode as anxiety or uncomplicated CHS, obtain electrolytes, creatinine, and an urgent lithium level because dehydration may have converted a stable dose into toxicity.

Teaching Point: Every presentation requires two plans: a syndrome plan for tonight and an exposure-change plan for the next transition of care.

Audience Poll: In the primary case, which finding most strongly affects immediate disposition: rapid symptom improvement, family history, access to weapons, or willingness to reduce cannabis?

Tonight on Shift

1. Name the exposure precisely. Record THC versus CBD, product, labeled potency, route, amount, frequency, last use, recent escalation, source, and tobacco or other-drug co-use. “Uses marijuana” is not an adequate medication history.

1. Stabilize before attributing. Check vital signs, glucose, oxygenation, hydration, attention, suicidality, violence risk, oral intake, and neurological red flags. Order ECG, electrolytes, renal function, pregnancy testing, creatine kinase, imaging, or expanded toxicology when the syndrome warrants them.

1. Treat the syndrome, not the urine result. A positive cannabinoid screen does not prove current intoxication or causation, and a standard panel may miss synthetic cannabinoids. Manage psychosis, delirium, hyperemesis, withdrawal, or medication toxicity according to its immediate risks.

1. Recognize the high-yield interactions. Ask about changes in smoked tobacco and cannabis for patients taking clozapine or olanzapine; high-dose CBD with CYP2C19 substrates; additive sedation with alcohol, opioids, benzodiazepines, or gabapentinoids; and vomiting or dehydration in anyone taking lithium.

1. Anticipate days two through six. Monitor sleep, appetite, irritability, aggression, mood, suicidality, and psychosis as withdrawal peaks. Before discharge, plan for medication levels or dose changes if smoke exposure will change again, and arrange early-psychosis and cannabis-use-disorder follow-up.

1. Deliver one brief motivational intervention. Ask permission, elicit benefits and costs, offer a personalized link between exposure and symptoms, assess readiness and confidence, and agree on one concrete abstinence or harm-reduction step. Document the patient’s own goal and the next appointment.

Read this seminar as Markdown · All seminars · Lecture library · Question bank