Research has discredited claims that isotretinoin (Accutane) causes Crohn’s disease or IBD. So why the lawsuits?
Some initial case studies suggested a connection between IBD and isotretinoin. However, subsequent research examining a potential link between inflammatory bowel disease (Crohn’s disease and ulcerative colitis in particular) and Accutane have been inconclusive. New research even goes so far as to disprove the connection. Dr. Brian Feagan and Dr. Reena Khanna even concluded in 2013 with Gastroenterology and Hepatology that:
“A large amount of high-quality evidence now exists that disproves the relationship between isotretinion and the risk of IBD. Nevertheless, the legal tsunami continue unabated.”
While acne is a common condition that may impact upwards of 90% of adolescents, severe acne is not a trivial problem. It is associated with an increased risk of mental illness, poor social functioning, and suicide. When topical creams are ineffective, isotretinoin, a vitamin A analog that was approved by the FDA in 1982, may be prescribed by a doctor.
Isotretinoin is a powerful medication that can combat severe, disfiguring nodular acne. But, it also comes with side effects. Isotretinoin shouldn’t be taken during pregnancy, as it can cause birth defects. The FDA notes that the medication may cause:
- depression or other mental health conditions;
- liver problems;
- serious skin reactions; and
- inflammatory bowel disease
The claim above that isotetinoin treatment for acne may cause IBD is based on case studies and reports submitted to the FDA through the MedWatch system. These reports triggered a “tsunami of litigation directed towards physicians and pharmaceutical companies,” including payments made by Accutane creator, Roche. There was only one problem: The research that brought this potential problem to light was incomplete and poorly gathered.
What was wrong with the research?
The best type of research study, known as the gold standard, is a randomized controlled trial. These trials have clinical arms and control arms that may receive a placebo. This enables researchers (who are blind to the conditions) to isolate the variables that are being tested. This ensures that any changes observed are caused by the intervention (medications, treatment, etc.).
On the other end of the spectrum are case studies. Case studies are essentially anecdotes. There are no control groups so there is no methodology to determine causation. Because there are no matching control groups, it is easy to misattribute changes to confounding variables including age, race, or co-morbid conditions.
With regard to isotretinoin, a number of case studies emerged in the late 1980s suggesting a link between the medication and IBD. These case studies generated concern and additional studies were planned.
One of the subsequent studies looked at the FDA adverse events reporting data and used the Naranjo adverse drug reaction probably scale to evaluate the link between IBD and isotretinoin. 73% of the reviewed cases scored at least as “probable” for a link between IBD and isotretinoin. The researchers concluded that there may be a link between the two.
The problem: the data in the study doesn’t allow for such a broad conclusion. Adverse events data is gathered spontaneously when physicians report it. The data is susceptible to bias and does not allow for a control arm. All told, this research isn’t sufficient to prove an association — and it definitely wasn’t enough to prove causation. But, this information brought increased interest from attorneys looking to make claims.
Between 2003 and 2011, there were 2,214 adverse events recorded by the Food and Drug Administration Adverse Event Reporting System regarding IBD and isotretinoin. A study found that, amazingly, 87.8% of these cases were reported by attorneys. Only 6% were filed by physicians and 5.1% by consumers. For comparison, attorneys only filed 3.6% of the total reports filed during that period (2,451,314 reports of adverse drug reactions).
Exploring the additional studies
A study published in 2009 in the American Journal of Gastroenterology used registries to determine the rates of IBD and isotretinoin use. By using a registry of patients and medications, the researchers could match control subjects, accounting for age, sex, and geographic differences. The researchers found that isotretinoin was used by 1.2% of people with IBD and by 1.1% of controls (19,814 subjects). Even with these large sample sizes, that small difference was not statistically significant, suggesting that the medication does not cause IBD. If it did, then the IBD and isotretinoin group should be much larger. These results held up when looking specifically at both Crohn’s disease and ulcerative colitis as well.
Additional studies were conducted and, perhaps the final nail in the proverbial coffin, was a nested case-control study and meta-analysis. This study created risk ratios for IBD and control participants. The ratios for the IBD group and the control group were not statistically different:
IBD Group: Of 2,159 participants, only 10 used isotretinoin (0.46%)
Control Group: Of 43,180 participants, only 191 used isotretinoin (0.44%)
The researchers concluded that the, “the results of this study do not suggest an increase in the risk for IBD, including [ulcerative colitis] or [Crohn’s disease] with use of isotretinoin.”
Research in context
It’s easy to be frustrated by the progression above. Most drugs have side effects that are part of the risk assessment for Crohn’s disease treatments. But, the hope is that potential dangers are discovered during the research process so that you can make the most informed decisions possible about your health.
Rather, the above story can show the importance of reporting adverse events to doctors and how the research landscape is an evolving process. When done properly, case studies and adverse events reporting should bring about larger, cohort-style studies to definitively answer valid questions. This is what happened in this instance. The most dangerous thing that can happen is to reach conclusions that are broader than the underlying research methods support.