Constipation After Quitting Weed: What Is Documented, and What Is Not

There is a particular kind of search result that tells you something is wrong with the internet. Type “quitting weed constipation” and you get a cannabis-card retailer, two quit-coaching businesses, a rehab centre, an AI-generated medical site, a health forum — and a 2003 research paper about giving up cigarettes, ranking because nothing more relevant exists. Nearly every related search has “reddit” appended to it, which is what people do when they have decided the results are not worth reading.

So we went looking for the actual evidence. What we found is more interesting than another list of remedies: nobody has ever measured this. Not measured it and found nothing — simply never looked. And it is possible to say exactly why.

The gapThe question was never on the form

Research into cannabis withdrawal runs largely through one validated instrument, the Cannabis Withdrawal Scale, developed in 2011 and used across the field since. It has nineteen items. Here they are.

Three of those items are digestive: appetite, nausea, stomach ache. Not one asks whether the person filling it in has been able to go to the toilet. A search of the published literature for cannabis cessation and constipation returns zero results. The single paper that comes back for the search terms is a 2023 review which notes, in passing, that no such data exists.

This matters more than it might sound. A gap in the evidence is not the same as evidence of absence, and the difference is the whole article. Every page currently ranking for this question fills the space with a number — a percentage of people who get diarrhoea, a window in which digestion “normalises” — and none of those numbers has a source behind it. We checked several and could not trace one to any study at all.

What is documentedThe official list, and a correction

Cannabis withdrawal itself is real and well characterised. The standard clinical review sets out the diagnostic criteria: irritability or anger, nervousness or anxiety, sleep difficulty, decreased appetite or weight loss, restlessness and depressed mood — plus at least one physical symptom causing significant discomfort, from a list that includes abdominal pain, shakiness, sweating, fever, chills or headache.

Two corrections while we are here, because both errors are everywhere. Abdominal pain is on the official list — it is the only gastrointestinal item. Nausea and vomiting are not, despite being listed as criteria by several large health sites. And constipation appears nowhere in the criteria, nor anywhere in the review that explains them.

The timeline is better established: symptoms typically begin 24 to 48 hours after stopping, most peak between days two and six, and some persist up to three weeks or more among people who were using heavily. Appetite loss is among the earliest and most common.

How commonTwo very different numbers, and which one is yours

The figure in circulation is that roughly half of people who quit experience withdrawal. A meta-analysis of 47 studies covering 23,518 people put the pooled rate at 47 percent, and that is accurate as far as it goes.

It goes less far than it appears. The same analysis breaks the number down by setting, and the spread is enormous: 17 percent in general-population samples, 54 percent in outpatient groups, and 87 percent among inpatients. The pooled 47 percent is an average across populations so different that the statistical heterogeneity was near-total. For someone who is not in treatment and is simply stopping, the relevant number is the first one, not the headline. Withdrawal is likelier with daily use, with tobacco co-use, and with other substance use — all of which the analysis quantified.

For context on scale: federal survey data for 2024 put past-month cannabis use among 18-to-25-year-olds at 24.1 percent, about 8.4 million people. And an analysis of 1.6 million survey responses spanning 1979 to 2022 found that 2022 was the first year on record in which more Americans reported using cannabis daily or near-daily than reported drinking that often — 17.7 million against 14.7 million. That is a statement about frequency and nothing else; far more people drink. But the group most likely to encounter withdrawal is precisely the daily-use group, and it has grown roughly fifteenfold in three decades.

The inversionThe pharmacology predicts the opposite complaint

Now the genuinely strange part, and the reason this deserves proper study rather than guesswork.

Cannabinoids slow the digestive tract down. This is not disputed — the standard review from the Mayo group that ran the original trials states flatly that cannabinoids generally inhibit gastrointestinal motor function. The underlying studies are specific. In a double-blind trial of 52 people using a pressure-sensing assembly in the descending colon, a single dose of dronabinol increased colonic compliance, inhibited postprandial colonic tone and reduced both fasting and postprandial contractile pressure; the authors concluded it relaxes the colon and reduces motility after eating. In a separate double-blind study of 13 people, THC slowed gastric emptying in every single one of them.

Worth noting what those studies do not show. A 30-person trial found dronabinol retarded gastric emptying but produced no significant change in colonic transit — the stomach slowed, the colon did not measurably. And every one of these used oral pharmaceutical cannabinoids at controlled doses; no human study has measured what smoked or vaped cannabis does to gastric emptying. So the honest summary is that a brake exists, and it has been measured mainly on the stomach.

Follow that logic through cessation and you get a prediction: take the brake off, and things should move faster. Looser, sooner. Which is the opposite of what people overwhelmingly describe — though not, interestingly, the opposite of what everyone describes. “Is it normal to have diarrhoea after quitting weed” is a live search sitting right alongside the constipation questions. Both directions are being reported by real people, and the mechanism only predicts one of them.

The precedentWhat happened when someone finally measured the other one

There is a close parallel, and it is instructive. Constipation after stopping tobacco was, for decades, a thing quitters reported and clinical criteria ignored — until somebody counted.

Researchers followed 1,067 people through a stop-smoking clinic, rating constipation before the quit date and weekly after. It rose significantly, peaked at around two weeks, and remained elevated at four. Net, about 17 percent were affected — roughly one in six — and 9 percent went from no constipation at all to describing themselves as very or extremely constipated. The paper’s conclusion was that descriptions of tobacco withdrawal should be updated to include it. Its title was simply: Stopping smoking can cause constipation.

That is a real symptom, widely reported, absent from official criteria, and confirmed the moment anyone bothered to look. It is not evidence about cannabis — it is tobacco, a different drug, and many cannabis users also use tobacco, which is a genuine confounder rather than a footnote. But it is a decent argument that “not in the criteria” and “not happening” are different claims.

The candidatesWhat could plausibly be going on

Given that nobody has measured it, the responsible thing is to label what follows as hypotheses. The most plausible one is also the least exotic, and it is the only link in the chain that is actually documented: appetite drops. That is in the diagnostic criteria, it is among the most commonly reported withdrawal symptoms, and it is one of the earliest to appear. Less food means less bulk and less fibre, and the consequences of that for stool are not mysterious.

The others are reasonable and unevidenced: disrupted sleep, which is firmly documented as a withdrawal symptom and has its own relationship with bowel rhythm; anxiety, likewise documented; and the loss of a daily ritual that, for some people, happened to sit next to a reliable bathroom routine. Any of these could be doing the work. None has been tested against this outcome.

What helpsOrdinary measures, honestly graded

Because the cause is unestablished, what is left is what helps constipation generally — and it is worth being straight about how strong that evidence is. A meta-analysis of 16 randomised trials in more than 1,200 people found fibre supplementation improved both response and stool frequency, with psyllium the most consistent performer across three trials. The joint AGA and ACG guideline does recommend fibre — as a conditional recommendation resting on low to very low quality evidence. Its strong recommendations are all prescription treatments, which is a conversation with a clinician rather than a shopping list.

Beyond that: the appetite mechanism, if it is the right one, points at eating regularly even when the appetite for it has not returned, which is easier described than done in the first fortnight. Movement and hydration are sensible and thinly evidenced. And the tobacco precedent suggests that if this follows the same shape, it has a peak and a downslope rather than being a new permanent state — around two weeks in that literature, which is at least a shape to expect.

One more thingThe condition where stopping is the treatment

There is a related condition worth knowing about calmly, because it is the one instance where the gut and cannabis have a well-documented relationship in the other direction. Cannabinoid hyperemesis syndrome involves cycles of nausea and vomiting, often with abdominal pain, and a characteristic pattern of relief from hot bathing. A systematic review of 183 published reports found that stopping cannabis resolved it in 96.8 percent of cases, and concluded that cessation appears to be the best treatment. Its true frequency is unknown — the literature is almost entirely case reports, with no denominator to calculate one from.

It is mentioned here for the simple reason that it is the one gut condition in this territory where the answer is already established, and where a clinician can help quickly. Persistent vomiting is worth a doctor’s time regardless of its cause.

The ledgerWhat is actually known

Sorting it. Cannabis withdrawal is real, well described, and includes abdominal pain and appetite loss but no bowel symptom in its official criteria. It affects roughly 17 percent of people in general-population samples, more among heavier and treatment-seeking users. Cannabinoids measurably slow the gut, mostly demonstrated on the stomach, which predicts faster transit after stopping rather than slower. What actually happens to bowel habit after quitting has never been studied, the numbers circulating online are unsourced, and the closest real evidence comes from tobacco, where one in six quitters was affected and it peaked at a fortnight.

That is a less satisfying answer than a percentage, and it is the true one. If this is happening to you, it is not imaginary and you are not the only person searching for it — the search volume is precisely why we looked. It is simply a question the research has not gotten to yet. Anything that persists, anything that changes sharply, and anything arriving with pain or bleeding belongs with a clinician, who can look at your situation rather than a literature gap. This article is educational, reports what the cited research measured and where it is silent, and is not medical advice.