Does Accutane Cause Gut Problems?

If you ask a dermatologist whether Accutane can give you inflammatory bowel disease, most of them will say no and point you toward the research. Then you unfold the paper that came in the box, and there it is under the warnings: inflammatory bowel disease. People go to three different doctors over a few years, get three different answers, and end up deciding based on whichever one they heard last.

‍

Both sides of that contradiction are accurate, and the reason sits in one word of the warning.

‍

What the insert actually says

‍

The current US warning reads: isotretinoin "has been associated with inflammatory bowel disease (including regional ileitis) in patients without a prior history of intestinal disorders."

‍

Associated with, not causes. Two things showing up together in the same people is the start of a question, not the answer to one. The label says as much a few sections later, about the entire list of reactions that inflammatory bowel disease sits on: "it is not always possible to reliably estimate their frequency or establish a causal relationship to drug exposure."

‍

That sentence is also old. It's word for word identical in the June 2002 Accutane label, in a 2013 generic label, and in the labels revised in August 2026. It has sat in the document untouched for twenty-four years while the research underneath it moved. Health Canada's version, revised in September 2025, uses an even more careful phrase: "temporally associated," meaning that the two things happened in the same window of time.

‍

So your dermatologist and the insert aren't really arguing with each other.

‍

Where the warning came from

‍

In 1983, a few years after the drug launched, the manufacturer received reports of six to eight patients out of roughly 300,000 who had taken it and developed inflammatory bowel disease. Those reports prompted the first warning in 1984, and they were the entire evidence base at the time.

‍

Case reports are how medicine notices things, and they're useful for exactly that. What they can't do is tell you how often something happens, because there's no denominator and nobody to compare against. If a doctor files a report saying that a patient on Accutane developed colitis, that report doesn't know how many people took Accutane and stayed fine, or how many people who never touched it developed colitis that same year.

‍

A 2009 review by researchers at UNC Chapel Hill worked out that roughly 59 cases of inflammatory bowel disease would be expected among isotretinoin users every year even if the drug did nothing at all. Inflammatory bowel disease tends to appear in the teens and twenties, which is exactly when people take Accutane. A lot of people were always going to be diagnosed while on it.

‍

In 2006, a paper collected 85 of these reports from the FDA's complaint database and concluded that the drug "might serve as a trigger" in some patients. It's the study most often cited as proof, and it's a review of complaint forms with no comparison group.

‍

Then came the lawsuits

‍

Thousands of people sued, and the cases were consolidated in New Jersey. That litigation is why your parents have heard of this, and why the claim has lasted twenty years.

‍

It didn't go the way most people assume. On August 1, 2018, the New Jersey Supreme Court upheld the exclusion of the plaintiffs' expert witnesses, which ended 2,076 Crohn's disease cases at once. The court's reasoning was about how those experts had handled the evidence: they "disregarded eight of nine epidemiological studies and relied on case reports and animal studies." On why one animal study didn't help, the opinion says: "dogs cannot get inflammatory bowel disease."

‍

Two months later the same court dismissed a separate group of 532 ulcerative colitis claims, on the grounds that the FDA-approved warning had been adequate.

‍

No court ever found that the drug causes inflammatory bowel disease. The verdicts plaintiffs won along the way turned on whether the company had warned people properly, which is a different question.

‍

Accutane also wasn't pulled off of the market for safety reasons, whatever you have heard. Roche discontinued the brand in 2009, and the FDA published a formal determination in 2010 stating that it "has found no information that would indicate that this product was withdrawn from sale for reasons of safety or effectiveness." The drug never left. It's still prescribed today as isotretinoin, under several brand names, carrying the same warning.

‍

Who actually filed those FDA reports

‍

Researchers went through the FDA's adverse event database covering 2003 to 2011 and found 2,214 reports linking isotretinoin to inflammatory bowel disease. Then they checked who had filed them.

‍

Attorneys filed 1,944 of them, or 87.8%. Physicians filed 6.0%.

‍

Across the whole database, for every drug and every reaction in that period, attorneys filed 3.6% of reports. The authors called the result "a clear distortion." It's still distorting things, too: a pharmacovigilance study published in 2026 found that among all drugs associated with inflammatory bowel disease reports, isotretinoin "demonstrated the strongest signal intensity." That signal is substantially an artifact of lawsuits filed over a decade ago, still sitting in the database.

‍

None of that means the individual patients were lying. It means that the pile of reports got much taller for reasons that had nothing to do with biology, and anyone who counts the pile gets a wrong answer.

‍

What the controlled studies found

‍

The largest pooled analysis, published in 2023, combined eight studies covering 2,522,422 people and found no increased odds of inflammatory bowel disease among isotretinoin users (odds ratio 1.01). The authors also reported that the accumulated data were already enough to rule out an effect of the size anyone was looking for.

‍

Then there's the disease mix-up, which almost every version of this claim gets backwards. The popular version is "Accutane causes Crohn's." Crohn's has the least support of anything in this literature:

The 2010 study that fed the lawsuits found nothing for Crohn's (odds ratio 0.68)
A French national study covering 76% of the country found lower Crohn's risk (0.45)
A cohort study published in December 2025, following people for six to ten years, also found lower risk (0.69)
The pooled estimate across studies is 0.87

If there's a signal anywhere, it's in ulcerative colitis, not Crohn's. That 2010 study did find a strong ulcerative colitis result, an odds ratio of 4.36. Out of 30,021 people in that study, exactly 60 had ever taken the drug. Twenty-four of them had inflammatory bowel disease. The entire alarm rests on twenty-four people, which is why the confidence interval runs from roughly 2 to nearly 10.

‍

The part that's still open

‍

The question isn't closed, and you should be suspicious of anyone who tells you that it is.

‍

A 2024 study using South Korea's national insurance database, covering 107,434 people with inflammatory bowel disease, found a small association that survived adjustment for antibiotic use, with higher risk at higher cumulative doses. The odds ratio was 1.20, a 20% relative increase on a very small baseline. It's the largest study of its kind, it covers a population nobody had studied before, and dose-response is one of the things scientists look for when deciding whether something is causal. It also didn't adjust for how severe anyone's acne was, and it sits at odds with every Western dataset.

‍

Two independent teams have now seen the same second pattern: a bump in ulcerative colitis risk confined to the first six months after starting, which then disappears. The best-designed version compared acne patients on isotretinoin against acne patients on oral antibiotics, and put the difference at 5 additional ulcerative colitis cases per 10,000 people in those first six months, with the risk returning to normal afterward.

‍

Five in ten thousand, temporarily. That's the upper end of what the careful research supports.

‍

A bump that vanishes is also what you'd expect if the drug were doing nothing. If you're on this drug, you see a doctor every single month for bloodwork and pregnancy testing. That's a lot of chances to get diagnosed with something you already had. John Barbieri, a dermatologist at Brigham and Women's Hospital who co-authored the current national acne guidelines, put it plainly: patients on isotretinoin "may be more likely to be screened earlier and thus could be diagnosed sooner than those not on" it.

‍

What is associated: acne itself

‍

A 2023 study separated the drug from the condition it treats. Among more than 350,000 patients, acne itself was associated with inflammatory bowel disease (odds ratio 1.42). Neither the antibiotics nor the isotretinoin were.

‍

A Canadian study of 1.7 million people pointed the same direction from a different angle: people using topical acne creams showed the same signal as people taking the pill. Face cream can't plausibly cause bowel disease, which suggests that whatever is going on tracks the skin condition rather than the treatment.

‍

Both acne and inflammatory bowel disease involve inflammation, and both tend to arrive in the same years of life. That's an unfinished explanation rather than a neat one, and it hasn't been replicated much, so hold it loosely.

‍

What Accutane actually does to your stomach

‍

Everything above is about a rare disease. The ordinary gut effects are far more likely to happen to you.

‍

In a review of 370 patients, 3.7% reported abdominal pain and 1.6% reported nausea. Those numbers come from a chart review with no comparison group, and no placebo-controlled figure for these symptoms seems to exist anywhere, so treat them as rough.

‍

Some of it is probably your esophagus

‍

Doctors documented a case of a 28-year-old man who developed sudden painful swallowing; an endoscopy found two ulcers in his upper esophagus. He'd been taking his capsule immediately after dinner and going to bed. The capsule sat against the lining long enough to burn it. The label's instruction exists for this reason: "swallow the capsules with a full glass of liquid. Swallow capsules whole." The doctors who wrote up that case added the part the label leaves out, which is to stay upright for at least 30 minutes afterward. If swallowing hurts or your chest burns, tell your doctor, and know that it isn't inflammatory bowel disease.

‍

Why it gets taken with food

‍

In a study of 74 people, taking isotretinoin with a high-fat meal more than doubled both the peak concentration and the total amount that reached the bloodstream, with no change in how long it lasted. Food doesn't make the drug gentler on you; it makes more of it actually get in. Taking it on an empty stomach mostly wastes it. One caveat: some newer formulations (e.g., Absorica and Absorica LD) are built to work without food and are labeled that way, and you can't tell which one you have from the word "isotretinoin" on the bottle, so ask your pharmacist.

‍

Pancreatitis, the rare one that matters

‍

Pancreatitis is in the label, and it mostly isn't driven by high triglycerides, which is the opposite of what you'd assume. A systematic review of 25 published cases found only four that were likely caused by high triglycerides. The label confirms that it has happened at "either elevated or normal serum triglyceride levels." Severe pain in your upper abdomen, especially with vomiting or fever, is a same-day call rather than a wait-and-see.

‍

Triglycerides do rise on this drug, in about 25% of patients in trials, and that's one of the things the monthly bloodwork checks. Whether that monthly testing is necessary for healthy young people is actively debated among dermatologists right now, so if your doctor orders less of it than you expected, that's a defensible position rather than negligence.

‍

What if you already have IBD?

‍

If you have Crohn's or ulcerative colitis and you want acne treatment, you deserve a straight answer, and nobody has one.

‍

No major dermatology guideline says whether pre-existing inflammatory bowel disease is a reason not to prescribe it, a reason for caution, or irrelevant. The Australasian college lists its contraindications explicitly, and inflammatory bowel disease isn't among them. The only direct study anyone has run is a chart review of ten patients, which found no association with flares. Ten patients, average age 28, can't establish safety.

‍

So people in this situation get opposite answers from their own gastroenterologists, and neither doctor is being careless. The evidence to settle it hasn't been produced. If this is you, that's a conversation between your dermatologist and your GI doctor together, not a question to resolve from an article.

‍

When to stop and call someone

‍

The label's rule is deliberately broad, and it's the one to follow: stop the drug and contact your doctor for abdominal pain, rectal bleeding, or severe diarrhea. The FDA's patient guide adds trouble swallowing or painful swallowing, new or worsening heartburn, nausea and vomiting, and severe pain in your upper abdomen.

‍

That list will catch far more ordinary stomach bugs than anything serious, which is the point of it. What it won't do is miss something that matters. Don't stop taking it and tell nobody, and don't push through symptoms because you only have four weeks left in your course. Tell the person who prescribed it.

‍

If you want to know what an ongoing gut problem looks like when it isn't drug-related, we wrote a guide to the common causes.

‍

So, does Accutane cause gut problems?

‍

Not the one people are afraid of, as far as anyone can currently tell. The Crohn's version of the claim, which is the version everybody repeats, has the least evidence behind it of anything in this literature. The warning on the insert is a 1984 sentence about reports, kept in place ever since, and it never claimed to be more than that. The mountain of FDA reports was built mostly by lawyers.

‍

What survives is small and specific: a possible short-lived rise in ulcerative colitis risk in the first six months, on the order of 5 cases per 10,000 people, which may well be a diagnosis catching up to a disease that was already there. One large recent study still finds a modest association, and nobody has explained it away. Anyone who tells you the risk is definitively zero is going past the evidence in the other direction.

‍

The gut problems most likely to happen to you on this drug are the boring ones. Take the capsule with water, sitting up, with food, and not right before bed. Know what upper abdominal pain means. And if something feels wrong, say it out loud to your doctor instead of to a search bar at 1am.

‍

Sources

‍