Cannabis in the ER: hyperemesis and psychosis

A young man in his early twenties arrives at the emergency department in the early hours for the third time this month. He has been vomiting nonstop since the night before, has diffuse abdominal pain, and, a little embarrassed, says it only improves in a hot shower. At triage, someone writes "marijuana user" and the case seems solved before the physical exam. The same night, in another bed, a nineteen-year-old woman is brought in by her family because she has started saying the neighbors put cameras in her room; she has smoked cannabis every day for two years. Both scenes are fictional and were written for teaching. This text uses these two pictures as a setting for clinical reasoning: how to think about cyclic vomiting in someone who uses cannabis without letting what is serious slip by, and how to think about psychosis without deciding too early whether it is induced or primary.

NEXO is a daily medical diagnosis game: one clinical case a day, up to six progressive clues, and ICD-10 proximity feedback across five levels.

The text avoids being a protocol of conduct. Doses, tests and workflows follow each service's protocols, under supervision. Here the focus is the way of thinking: which clues raise or lower each hypothesis, which diagnoses must be ruled out before any label, and which biases push the reasoning to one side. If you play the case of the day in NEXO, you will recognize the move: the first clue draws attention, but the diagnosis depends on the whole set.

Key points

• Cannabinoid hyperemesis syndrome is a clinical diagnosis that requires regular and prolonged cannabis use, stereotyped episodes of vomiting, and improvement when use stops.
• Relief with hot bathing is a strong clue, but bowel obstruction, pancreatitis, ketoacidosis, pregnancy and other serious causes must enter the differential before the label.
• Stopping use is the most consistent treatment for hyperemesis; in the acute episode, a randomized trial compared haloperidol and ondansetron.
• Daily use and high-potency use are associated with a higher incidence of psychosis, and cannabis-induced psychosis can evolve into schizophrenia or bipolar disorder over time.
• Two biases alternate: closing the case upon hearing "cannabis" and, at the opposite extreme, forgetting to ask about use.

Why the "cannabis" label gets in the way in the emergency department

The word cannabis in the chart works as a very salient clue. It seems to explain everything: the vomiting, the agitation, the persecutory idea. This salience feeds premature closure, the tendency to accept the first plausible explanation and stop looking. Pat Croskerry describes this and other shortcuts among the most frequent cognitive errors in diagnosis (Croskerry, 2003). The post on examples of heuristics in clinical reasoning organizes the main shortcuts and shows when they help and when they get in the way.

The label also carries judgment. Kelly and Westerhoff showed, in a randomized study, that describing the same person as a "substance abuser" increased blame attribution and preference for punitive measures among mental health professionals (Kelly and Westerhoff, 2010). In the emergency department, blame becomes haste: the person receives less attention, less examination and less explanation. The post on stigma and language shows how replacing the label with a person-centered description changes conduct.

The opposite bias: forgetting to ask

The opposite error is as common as the first. Someone who arrives with recurrent vomiting often goes through several consultations, imaging tests and endoscopies until someone asks, calmly and without judgment, about cannabis use. The question must be asked routinely, as naturally as one asks about alcohol and tobacco. Tools such as ASSIST include cannabis among the substances screened, and the post on screening with AUDIT, CAGE and ASSIST explains how pre-test probability changes the interpretation of the answer.

Both biases have the same root: the clue of substance use weighs too much or too little. The way out is to treat use as one clue among several, with weight proportional to the consumption pattern and the history.

Cyclic vomiting: what characterizes cannabinoid hyperemesis

Cannabinoid hyperemesis syndrome has been described as recurrent episodes of nausea, vomiting and abdominal pain in people who use cannabis regularly and over a long period. A systematic review by Sorensen and colleagues gathered the published cases and found a consistent pattern: daily or weekly use, cyclic episodes of vomiting, relief with hot baths and resolution when use stops, which the authors point to as the most effective treatment (Sorensen and colleagues, 2017).

A case series from the Mayo Clinic, with almost one hundred people, helped propose diagnostic criteria and reinforced the importance of prolonged use and compulsive bathing behavior (Simonetto and colleagues, 2012). The Rome IV criteria for gastroduodenal disorders include cannabinoid hyperemesis among the nausea and vomiting syndromes and describe stereotyped episodes, similar to cyclic vomiting, in people who use cannabis for a long time and in excess, with relief after sustained cessation (Stanghellini and colleagues, 2016).

In 2024, the American Gastroenterological Association published a clinical practice update on the diagnosis and management of cannabinoid hyperemesis, which reviews the available evidence and organizes expert recommendations (Rubio-Tapia and colleagues, 2024). The common point among these sources is that the diagnosis is clinical and rests on the pattern over time, not on a single test.

Clues that raise the probability

• Regular cannabis use over a prolonged period, usually months to years before the episodes.
• Stereotyped episodes of vomiting, with intervals of well-being between them.
• Relief of symptoms with a hot bath or shower, often repeated several times a day.
• Laboratory and imaging tests with no other explanation for the picture.
• Sustained improvement when the person stops using for long enough.

Each clue on its own has limited value. The hot bath, for example, is suggestive, but it must be read together with the pattern of use and the exclusion of other causes. The strength lies in the combination and in the evolution.

The differential of vomiting that must be ruled out

Before writing cannabinoid hyperemesis, the reasoning must go through the causes of vomiting that change management in the next few hours. A person who uses cannabis can also have appendicitis, obstruction, pancreatitis or ketoacidosis. The hyperemesis label, when applied early, pushes these hypotheses off the list.

Hypothesis	Clues that favor it	What usually clarifies it	Risk of the hyperemesis label
Intestinal obstruction	Distension, absence of passing gas and stool, previous abdominal surgery	Serial physical examination and imaging	Surgical delay
Acute pancreatitis	Intense epigastric pain radiating to the back, alcohol, gallstones	Pancreatic enzymes and imaging	Delay in support
Diabetic ketoacidosis	Polyuria, polydipsia, ketotic breath, known diabetes or not	Blood glucose, blood gas and ketones	Severe dehydration and acid-base disturbance
Pregnancy and hyperemesis gravidarum	Missed period, childbearing age	Pregnancy test	Management inadequate for pregnancy
Inflammatory acute abdomen	Localized pain, fever, abdominal guarding	Physical examination and imaging	Perforation
Opioid withdrawal	Opioid use, mydriasis, yawning, piloerection	History and examination	Treatment of the wrong cause
Cyclic vomiting without cannabis	Similar episodes, migraine in the history	Rome IV criteria and evolution	Inadequate guidance

The table supports the reasoning, and it is not an exhaustive list. The question it organizes is simple: what, in this patient, can kill or worsen if I take too long? Someone who has been to the emergency department many times deserves the same physical examination as someone arriving for the first time, because today's visit may be different.

Dehydration and electrolytes

Even when the diagnosis of hyperemesis seems certain, repeated vomiting can cause dehydration, hypokalemia and acute kidney injury. Assessment of volume and electrolytes is part of the care, because the complication can be more dangerous than the syndrome. Repeated hot baths can also worsen dehydration, and it is worth asking how many times a day the person has resorted to them.

Other substances in the same picture

People who use cannabis may use other substances. Opioid withdrawal and intoxication from other drugs are part of the differential for vomiting and agitation. For stimulants, the post on stimulants in the emergency department organizes the differential and the approaches that often make the picture worse; this text avoids repeating that content.

Treating the acute episode and talking about use

In the acute episode, usual antiemetics often have a limited response in cannabinoid hyperemesis, according to case series. A Canadian randomized trial, HaVOC, compared haloperidol with ondansetron in people with cannabinoid hyperemesis in the emergency department. Haloperidol was superior for reducing abdominal pain and nausea and allowed earlier discharge from the emergency department; use of rescue medication was lower, without a statistical difference; there were two return visits for acute dystonia, both in the higher-dose group (Ruberto et al., 2021). The trial was small, and the choice of medication follows each department's protocol.

The AGA clinical practice update reviews the evidence on managing the acute episode and reinforces guidance on stopping use (Rubio-Tapia et al., 2024). The clinical decision here is less about which antiemetic to choose and more about recognizing that the long-term solution depends on stopping use. Meanwhile, hydration, electrolyte correction and serial reassessment of the abdomen remain a central part of care.

The conversation that changes the outcome

Stopping use is the only treatment that consistently resolves the syndrome, according to the review by Sorensen et al. (Sorensen et al., 2017). The conversation about this in the emergency department has an added difficulty: many people use cannabis precisely to relieve nausea and anxiety, and the idea that cannabis causes the vomiting seems counterintuitive.

Some phrases often help:

• "Your vomiting pattern is very characteristic of a syndrome linked to regular cannabis use; can I explain how it works?"
• "Many people feel that cannabis relieves nausea at first, and then it starts causing episodes like this one."
• "Improvement usually appears after a few weeks without use; what do you think about trying it and coming back so we can talk?"
• "If stopping is hard, we can think together about ways to reduce and about support for that."

Motivational interviewing helps when there is ambivalence: the person wants to stop vomiting and fears losing relief from anxiety. Exploring both sides, without pressure, increases the chance that the person returns for follow-up.

Psychosis in people who use cannabis: induced, primary or another cause

The second scene of the opening brings another challenge. A young person with persecutory ideas who uses cannabis every day may have substance-induced psychosis, a first episode of schizophrenia or bipolar disorder, psychosis from another substance or a medical cause, such as encephalitis, a metabolic disorder or a structural lesion. Cannabis use is a clue, not a diagnosis.

The EU-GEI multicenter study compared first-episode psychosis cases with controls in several European cities. Daily use and high-potency cannabis use were associated with a higher chance of psychotic disorder, and differences in use patterns helped explain the variation in psychosis incidence between sites (Di Forti et al., 2019). A meta-analysis found a dose-response relationship between the level of cannabis use and the risk of psychosis, and the authors stress that causality has not yet been established (Marconi et al., 2016).

Induced psychosis also requires follow-up

Cannabis-induced psychosis is often treated as a passing episode that resolves when use stops. The evidence suggests caution. A Danish registry study followed people treated for cannabis-induced psychosis and found a substantial share who later received a diagnosis of schizophrenia spectrum disorder (Arendt et al., 2005). Another registry study, with longer follow-up, showed that substance-induced psychosis, especially cannabis, frequently converted to schizophrenia or bipolar disorder, and that follow-up needs to be long to detect that conversion (Starzer et al., 2018).

In practice, this evidence changes how cases are recorded and referred. A diagnosis of induced psychosis is a working hypothesis, and the person needs follow-up with reassessment, even when symptoms improve quickly.

Clues for the differential of psychosis

• Temporal relationship between the start of use, the dose increase and the symptoms.
• Course after stopping: rapid improvement favors an induced condition, persistence favors a primary disorder.
• Prodromal symptoms before use, such as withdrawal and decline in performance.
• Family history of schizophrenia or bipolar disorder.
• Signs of a medical cause: fever, lowered level of consciousness, focal neurological change, seizure.
• Use of other substances, including synthetic cannabinoids, which can cause more severe conditions.

The biases that close the case too early

The two conditions in this text share a trap: the clue of cannabis use is so striking that it tends to end the reasoning. Some biases appear often.

Premature closure. The explanation "it's the marijuana" comes first and ends the search. In vomiting, it lets pancreatitis or obstruction slip by. In psychosis, it lets encephalitis or a first episode of a primary disorder slip by.

Anchoring. A diagnosis of hyperemesis made on an earlier visit becomes a fixed point, and each new visit is read from it, even when the pain has changed location or intensity.

Diagnostic overshadowing. A strong label, such as substance use or psychiatric disorder, hides physical symptoms. A person with psychosis and fever may have an infection of the central nervous system.

Affective bias. Discomfort with frequent returns or with the person's behavior reduces patience for the examination and the explanation.

If you want to review the calibration between confidence and accuracy, the post on calibration of confidence in diagnosis offers a simple exercise: record how much you trust the hypothesis and check it later.

How to reduce the risk

A useful strategy is the safety question: "what else could it be, and what can I not let slip by?". It forces you to list at least one serious alternative before closing the case. Another is to record the diagnosis as a hypothesis, not as a certainty, making explicit what would change the conclusion. The post on differential diagnosis with clues shows how to train this habit.

The person's role in the decision

Someone who arrives with recurrent vomiting or with a first psychosis has often already heard contradictory explanations. Explaining the reasoning, including the hypotheses that are still open, builds trust and increases the chance of follow-up. The person understands why they need to return and what to watch for.

What to record at discharge

The discharge record is the bridge to the next appointment, which often happens with another professional. For hyperemesis, it is worth recording the pattern of use, the duration of use, the response to hot bathing, the tests that ruled out other causes and the guidance on stopping. For psychosis, it is worth recording the temporal relationship with use, the signs that ruled out a medical cause, the risk to the person and to others and the referral for mental health follow-up.

The main items of a good record:

1. Pattern of use described in neutral language, with frequency, duration and route.
2. Serious diagnoses considered and how they were ruled out.
3. Diagnosis recorded as a hypothesis, with what would change the conclusion.
4. Guidance on stopping use and the expected time to improvement.
5. Red flags that justify immediate return.
6. Referral and date for reassessment.

A record like this also protects against anchoring at the next visit. Whoever sees the patient later finds the hypotheses already ruled out, the tests done and what remained open, and can notice when today's picture departs from the pattern. If the pain moved, if fever appeared or if the vomit contained blood, the previous record makes clear that this is a new clue, and not just another episode like the others. In psychosis, recording the temporal relationship with use lets the professional doing follow-up compare the course after interruption and reassess the hypothesis of an induced condition with data, and not with impression.

For someone who arrives intoxicated by alcohol and cannabis at the same time, the logic of avoiding a label too early is similar, and the reasoning needs to consider both substances. The post on alcohol beyond withdrawal discusses the risks of labeling too early the patient who seems only intoxicated.

How NEXO helps train this reasoning

In NEXO, the case of the day is fictional, with a real clinical presentation and up to six progressive clues. You propose a diagnosis by searching for the code in ICD-10 and receive proximity feedback at five levels, which shows whether the guess landed on the exact code, the same category, the same group, the same chapter or a different chapter. This format speaks to the theme of this text: a striking clue, such as the use of a substance, can pull the guess toward the wrong chapter, and the proximity feedback shows the size of the deviation.

To use the case of the day as training against premature closure, try three habits. Before each guess, write down the serious hypothesis you need to rule out. When a clue seems to explain everything, ask what it leaves unexplained. After the result, note which clue you overvalued.

Three exercises with cases

List what can get worse. In each case with vomiting, write three causes that change management in the coming hours before thinking about the most likely one.

Separate induced and primary. For a fictional case of psychosis with substance use, write what favors an induced condition, what favors a primary disorder and what would change your mind.

Rewrite the label. Take a triage sentence with a label and rewrite it in person-centered language, keeping the clinical information.

The post on case-based learning in clinical practice shows why training with cases helps consolidate this type of decision.

The limits on using the game are described on the NEXO economy page. If you research the teaching of clinical reasoning, the page for researchers explains how to get in touch. Professionals who write or review cases find information under curation and under creators, and institutions can learn about the proposal on the partnerships page.

Frequently asked questions

How do you confirm cannabinoid hyperemesis syndrome?

The diagnosis is clinical. It rests on regular and prolonged cannabis use, stereotyped episodes of vomiting, the exclusion of other causes and sustained improvement after stopping use. The Rome IV criteria help organize the assessment.

Is a hot shower enough to confirm the diagnosis?

A hot shower is a strong clue and deserves to be asked about in every history of recurrent vomiting. The diagnosis gains strength when this clue is added to the pattern of use, the stereotyped episodes and the exclusion of serious causes.

How long does it take to improve after stopping?

Case series describe improvement over weeks after stopping. Agreeing with the person on a test period without use and a return date helps assess the response.

Does cannabis-induced psychosis go away on its own?

Symptoms often improve after stopping, but registry studies show that a share of people later receive a diagnosis of schizophrenia or bipolar disorder. Follow-up with reassessment is therefore part of the care.

Which tests help in a first episode of psychosis in someone who uses cannabis?

The choice follows each service's protocol. In general, the assessment looks for medical causes, such as infection, metabolic disturbance and structural lesion, guided by the history and the physical examination.

How do you talk about use without pushing the person away?

Open questions, neutral language and explaining the clinical reasoning usually work better. Showing that the goal is to understand the picture, and not to judge, increases the chance that the person reports the real pattern of use.

Final checklist

• I asked about cannabis use with neutral language, including frequency, duration and potency.
• I listed the serious causes of vomiting and recorded how they were ruled out.
• I assessed volume and electrolytes, even with a diagnosis of probable hyperemesis.
• I explained the link between regular use and vomiting and agreed on a trial period without use.
• In psychosis, I looked for signs of a medical cause and of use of other substances.
• I recorded induced psychosis as a hypothesis and referred the patient for follow-up.

Want to practice your reasoning with progressive clues? Play today's case.

References

1. Arendt M, Rosenberg R, Foldager L, et al. Cannabis-induced psychosis and subsequent schizophrenia-spectrum disorders: follow-up study of 535 incident cases. Br J Psychiatry. 2005. PMID 16319402.
2. Croskerry P. The importance of cognitive errors in diagnosis and strategies to minimize them. Acad Med. 2003. PMID 12915363.
3. Di Forti M, Quattrone D, Freeman TP, et al. The contribution of cannabis use to variation in the incidence of psychotic disorder across Europe (EU-GEI): a multicentre case-control study. Lancet Psychiatry. 2019. PMID 30902669.
4. Kelly JF, Westerhoff CM. Does it matter how we refer to individuals with substance-related conditions? A randomized study of two commonly used terms. Int J Drug Policy. 2010. PMID 20005692.
5. Marconi A, Di Forti M, Lewis CM, et al. Meta-analysis of the Association Between the Level of Cannabis Use and Risk of Psychosis. Schizophr Bull. 2016. PMID 26884547.
6. Ruberto AJ, Sivilotti MLA, Forrester S, et al. Intravenous Haloperidol Versus Ondansetron for Cannabis Hyperemesis Syndrome (HaVOC): A Randomized, Controlled Trial. Ann Emerg Med. 2021. PMID 33160719.
7. Rubio-Tapia A, McCallum R, Camilleri M. AGA Clinical Practice Update on Diagnosis and Management of Cannabinoid Hyperemesis Syndrome: Commentary. Gastroenterology. 2024. PMID 38456869.
8. Simonetto DA, Oxentenko AS, Herman ML, et al. Cannabinoid hyperemesis: a case series of 98 patients. Mayo Clin Proc. 2012. PMID 22305024.
9. Sorensen CJ, DeSanto K, Borgelt L, et al. Cannabinoid Hyperemesis Syndrome: Diagnosis, Pathophysiology, and Treatment: a Systematic Review. J Med Toxicol. 2017. PMID 28000146.
10. Stanghellini V, Chan FK, Hasler WL, et al. Gastroduodenal Disorders. Gastroenterology. 2016. PMID 27147122.
11. Starzer MSK, Nordentoft M, Hjorthøj C. Rates and Predictors of Conversion to Schizophrenia or Bipolar Disorder Following Substance-Induced Psychosis. Am J Psychiatry. 2018. PMID 29179576.