Who Shouldn't Use Cannabis

Not a list of side effects. A list of people — and what the clinical guidance actually says about each one.

By Cannible Editorial

Clinical references and public-health guidance name specific groups who are advised not to use cannabis, or to use it only with a clinician involved: people who are pregnant, might become pregnant, or are breastfeeding; people with current psychosis or a history of schizophrenia; people with several named unstable cardiac conditions; young people; people with a history of substance use disorder; people who are immunocompromised; and people taking medications that interact with it.

Before any of that, one structural fact, because it changes how every sentence below should be read. For the cannabis products sold on a shelf, formal contraindications largely do not exist in the way they do for a prescription medicine. The clinical reference that this article leans on hardest states it directly: "limited standardized information exists regarding contraindications for non–FDA-approved medical cannabis products because composition and dosing vary widely. In contrast, contraindications for FDA-approved cannabinoid pharmaceuticals are clearly defined in product labeling."

So what follows is not a warning label. It is guidance, issued by clinicians and by state medical-cannabis programmes, built out of pharmaceutical labelling and population evidence, about who has reason to be careful and why. That is a weaker instrument than a label, and it is also the only instrument that exists. It is the question a dispensary counter is least equipped to answer, and the one most worth asking somewhere else.

Why "contraindication" means something different here

A contraindication is a specific reason not to give a specific product to a specific person, a line in a drug's official labelling, arrived at through the approval process, that a prescriber is expected to check. It attaches to a defined formulation at a defined dose.

That machinery exists for the handful of cannabinoid medicines that have been through drug approval. It does not exist for a jar of flower or a package of gummies, for the reason the clinical reference gives: composition and dosing vary too widely for a standardized statement to be made about them.

Two consequences follow, and they pull in opposite directions.

The first is that the categorical-sounding statements in this article, "should not use", are guidance, not labelling. They come from bodies that took the available evidence and issued a position, and a different body might have drawn the line elsewhere.

The second matters more. The absence of a label is not the absence of a risk. A product that has never been assessed has not been cleared; it has not been examined. Reading "no formal contraindications exist" as reassurance inverts what the sentence says.

Two kinds of statement therefore run through everything below: a labelled contraindication for an approved pharmaceutical, and a public-health caution about a plant product nobody has standardized. Telling them apart is most of what this article is for.

Who the guidance names

This is education, not medical advice. Cannible cannot assess anyone's situation and is not attempting to; every group below is a reason to raise the question with a clinician, not an answer about you.

Each of the following names who is meant, what the guidance actually says and who issued it, and how strong the evidence behind it is.

If you are pregnant, might become pregnant, or are breastfeeding

This is the plainest statement in the guidance and the best supported. Utah's Center for Medical Cannabis, in its general use guidelines, states that "women who are pregnant and women who are sexually active and not on a reliable form of contraception should not use cannabis or cannabis-based medical treatments", and separately that "women who are breastfeeding their infants should not use cannabis or cannabis-based medicines."

The clinical reference is equally direct: cannabinoid use in pregnancy "is unsafe and associated with adverse effects on the fetus, including small-for-gestational-age infants, preterm birth, and possible neurodevelopmental consequences". It also records that both THC and CBD have been detected in breastmilk, and that THC peaks in milk a few hours after use and accumulates with repeated exposure.

The evidence behind that is graded, and the grades are worth keeping separate. A 2024 review of the risks and medical benefits of cannabis describes "consistent evidence that regular cannabis use during pregnancy increases the risk of maternal anemia and neonatal problems such as reduced birth weight and the need for intensive care treatment", consistent evidence, in the reviewers' own words. For the longer-term effects the same review is deliberately weaker: "emerging preliminary evidence suggests that prenatal cannabis exposure may have subtle yet enduring effects on memory and achievement in children and adolescents." Those two grades are not interchangeable, and the review did not intend them to be.

If you have a history of psychosis or schizophrenia

Utah's guidance is categorical here: "individuals with current psychosis or history of schizophrenia and other psychotic disorders should not use cannabis or cannabis-based medicines with significant THC content (chemotypes I and II)", chemotypes I and II being the THC-dominant and mixed THC/CBD categories. The clinical reference frames the same population as one to treat cautiously, on the grounds that THC-containing products may exacerbate psychiatric symptoms.

Behind the guidance sits an association, and the distinction matters more here than anywhere else in this article. The 2024 review reports that "chronic patterns of cannabis use have been associated with multiple adverse outcomes that are of particular concern among adolescents and young adults, such as, disrupted learning, impaired cognitive performance, reduced educational attainment and an increased risk of CUD, psychosis/schizophrenia, mood and anxiety disorders and suicidal behaviors."

And then, in the same review, the sentence that has to travel with every one of those outcomes: "There is debate about the extent to which cannabis use is a cause of these adverse outcomes." These are associations. The people who assembled them say plainly that whether cannabis causes them is not settled.

Both halves are true at once, and neither cancels the other. Nobody has established causation; the guidance is categorical anyway, because for a person who already has a psychotic disorder the question is not whether cannabis caused it but whether it makes their symptoms worse, and that is what the clinical caution addresses. The subject here is risk to a person, not risk from them.

If you have a heart condition

This section is a signpost, and deliberately short: Cannible covers the cardiovascular evidence in its own piece, and the risk magnitudes belong there rather than here.

What the guidance says is this. Utah's guidelines state that "cannabis and cannabinoids should not be used in patients with unstable vital signs, congestive heart failure, angina, myocardial infarction, known/suspected structural or vascular heart disease, or known cerebrovascular disease." The clinical reference is a degree softer and broader: cannabis products should be used cautiously in patients with clinically significant cardiovascular disease such as coronary disease, arrhythmias or unstable hemodynamics, because THC-containing products can alter heart rate and blood pressure.

One finding sits behind that, and it is reported here with the qualifier its source attaches to it: "acute marijuana smoking has been associated with a transient increase in the risk of myocardial infarction during the first hour after use, based on findings from observational trigger studies." Associated with, in an observational design. That is the whole of what this article says about cardiac risk.

If you are young

The age question is treated in the guidance as a developmental window rather than a bright line, and the threshold is a programme's, not a finding's. Utah's guidelines carry a section headed "Use in Children and Adolescents Under 26", stating that "use in this age category may result in altered brain development and function with possible long-term negative consequences, including negative mental health outcomes and long-term cognitive impairments."

Note what that is: a line drawn by one US state's medical cannabis programme, reviewed in 2022. It is not a scientific consensus about the age at which a brain finishes developing, and it should not be quoted as one. What it reflects is a real direction in the evidence, the 2024 review names adolescents and young adults as the group for whom chronic-use associations are "of particular concern", expressed as an administrative threshold.

One number gives the concern a size. The clinical reference records the prevalence of cannabis use disorder within one year as 10.7% among adolescents and 6.4% among young adults.

If you have a history of substance use disorder

Utah's guidance is that "medical cannabis should generally be avoided in persons with a history of substance use disorders, including alcohol use disorder, due to increased risk of developing cannabis use disorder (CUD)." The clinical reference makes the same point in more measured terms: individuals with a history of substance use disorder may be at increased risk of misuse.

This is information, not a verdict on anyone. The risk described is of developing a cannabis use disorder, a specific, recognised condition, and it is described as elevated, not certain. Someone with that history who is considering cannabis has a concrete reason to have the conversation with a clinician who knows their history, which is a different thing from being told no.

If you are older

Older adults are the group most often missing from consumer safety pages, and the clinical reference is specific about why they belong: they "may be more sensitive to adverse effects such as postural hypotension and neuropsychiatric symptoms", a drop in blood pressure on standing, and confusion or other neuropsychiatric effects.

Alongside that sits a trend rather than a risk estimate. Emergency-department visits involving cannabis among people aged 65 and over rose from 20.7 per 100,000 emergency department visits in 2005 to 395.0 per 100,000 in 2019. Read that as what it is: a rate among emergency visits over fourteen years, during which use in that age group also became far more common and far more legal. It is not a probability that anything will happen to any individual, and it cannot be converted into one.

If you are immunocompromised, allergic, or have a seizure disorder

Three groups that almost never appear on a consumer safety page, and one of them is a genuine contamination hazard rather than a pharmacological one.

Immunocompromised readers. The clinical reference states that "patients who are immunocompromised may be at risk from contaminated inhaled cannabis. Invasive fungal infections, including aspergillosis, have been reported in association with cannabis use, and contamination has been documented." This is a risk from what may be growing on the plant material, and it attaches specifically to inhalation.

Allergy. Cannabis allergy "has been documented, including cross-reactivity with certain plant foods in sensitized individuals", an allergic response to cannabis, sometimes in people already sensitised to particular foods. The evidence here is case reports: individual patients written up because something notable happened. A case report establishes that something can occur. It says nothing about how often.

Seizure disorders. Utah's guidance is that "high doses of THC should probably be avoided in individuals with seizure disorder, and THC-predominant cannabis (chemotype I) should be used with significant caution." Note the hedge in the source itself, "should probably", and that it is a caution about THC-dominant products specifically.

If you take prescription medication

Documented interactions between cannabis and several categories of prescription medication exist, and they are catalogued in the published literature. That subject has its own article, because doing it properly means naming medication categories and weighing how much evidence stands behind each, and doing it badly means handing readers a list they will misread as a verdict on their own prescription.

What belongs here is only the routing: if you take anything prescribed, the person to raise cannabis with is your pharmacist, who holds your whole medication list and whose job includes exactly this check.

Why the counter can't answer this

The place most people are standing when this question occurs to them is the worst-equipped place to ask it, and that is a matter of role rather than of anyone's competence.

A national survey of dispensary staff, fielded between February and October 2020, found that of 434 eligible respondents, "most were budtenders (40%) or managers (32%), and a minority were clinicians (18%)". Most people behind a counter are retail staff, because that is what the job is. They do not have your history, your medication list or your pregnancy status, and they are not positioned to weigh them.

The staff surveyed appear to agree. The same study reports that when managing cannabis dosing and safety in customers with medical and psychiatric comorbidity, "dispensary staff preferred involving individuals' traditional HCPs." The survey is six years old and self-report, and its authors describe a modest sample, but on this point it is describing people asking for exactly the routing this article recommends.

What the evidence does and doesn't establish

The honest shape of this evidence base has three features, and none of them is a reason to disregard it.

Much of it is association, not causation. That is not a hedge added here; it is the reviewers' own position, stated in the review that supplies the psychiatric and cognitive associations. Uncertainty about whether cannabis causes an outcome is not permission, particularly for someone who already has the condition in question, but it does mean these are risk signals rather than established mechanisms, and they should be described that way.

Much of it is extrapolated rather than measured on the products people buy. The contraindications that are genuinely rigorous belong to approved cannabinoid pharmaceuticals, specific formulations at specific doses. Applying them to an unstandardized shelf product is an inference, and a reasonable one, and still an inference.

Some of it rests on designs that cannot carry much. The cardiac finding above is from observational trigger studies. The allergy evidence is case reports. An emergency-department rate is a population trend. Each of those supports "this can happen and here is a reason for caution"; none supports "this will happen to you."

What survives all three qualifications is the article's actual claim, and it is narrower than a warning and more useful than a shrug: for these specific groups, named bodies looked at what exists and advised against use or advised clinical involvement. That advice is worth acting on, and acting on it means asking someone with your history in front of them.

What this article can and can't tell you

The distinction this piece gives you, labelled contraindication versus public-health precaution, travels well. It tells you what kind of statement you are looking at whenever a "should not use" appears, who issued it, and how much machinery sits behind it.

It does not tell you whether you should use cannabis. There is no version of this article that could, and every feature that would make it feel like it could, a risk score, a self-assessment, a decision tree, would be a worse article pretending to be a better one.

It does not settle where the age boundary sits, whether the psychiatric associations are causal, or whether risk differs by product format and potency for anything other than the contamination risk described above. Nobody has answered those; some of them are not currently answerable for products that are not standardized in the first place.

And it carries no US federal public-health framing at all. That is a gap in this article's sources rather than a gap in the world, and it is stated here rather than papered over.

A note on responsible use, which applies whether or not you are in any group above. Cannabis is for adults only. Do not drive or operate machinery while impaired. Store it locked and out of reach of children and pets. And if you are unsure whether something in this article describes you, that uncertainty is the reason to ask a clinician, it is not something to resolve by reading more.

Two questions the sections above will raise

Does this apply if I only use occasionally, not regularly?

Much of the evidence behind these cautions is specifically about chronic use. The 2024 review that supplies the psychiatric and cognitive associations describes them as attaching to "chronic patterns of cannabis use," and the pregnancy findings it grades as consistent concern "regular cannabis use during pregnancy." That is a real distinction and it is worth knowing. What it is not is a threshold: no source in this article defines a frequency below which any of these cautions stops applying, and the categorical guidance for pregnancy, breastfeeding, current psychosis and unstable cardiac conditions is not written with a frequency qualifier at all.

Does the format matter, is eating it safer than smoking it?

For one specific risk, yes, and only that one. The contamination risk to immunocompromised readers, invasive fungal infection from what may be on the plant material, is described in the clinical reference as attaching to inhaled cannabis. Nothing in this article's sources supports a format-based answer for any other group. Switching format does not change what a substance does once it is in the body, and the pregnancy, psychiatric, cardiac and substance-use cautions above are not written about smoking; they are written about cannabis.

Sources

View on Cannible →