Isotretinoin's depression link vanishes in two big health databases

Isotretinoin is a powerful acne treatment, and reports had suggested a causal link between the drug and depression, psychotic symptoms, suicide and attempted suicide. For anyone weighing up treatment for severe acne, that suggestion is frightening, and acne itself is linked with distress, which muddies any simple story. Case reports cannot separate a drug effect from the background rate of mental health problems in young people with skin disease. This study applied a formal comparative design to the question, using two large national health databases.

The researchers ran large population-based cohort studies using the Canadian Saskatchewan Health Database and the United Kingdom General Practice Research Database. They compared 7195 isotretinoin users with 13,700 oral antibiotic users with acne in Canada, and 340 isotretinoin users with 676 antibiotic users in the UK. Everyone had computer-recorded medical histories from between 6 months and 5 years before their first prescription, and at least 12 months afterwards. Rates of neurotic and psychotic disorders, suicide and attempted suicide were compared between the drug groups, and within isotretinoin users before versus after treatment, with relative risks calculated by multiple logistic regression.

Relative risk estimates for newly diagnosed depression or psychosis were approximately 1.0 in both databases, meaning isotretinoin users fared no worse than acne patients on antibiotics or people exposed to neither drug. Comparing the period before isotretinoin use with the period after it, the estimates were also around 1.0. For suicide and attempted suicide, the relative risk was 0.9, with a wide confidence range of 0.3 to 2.4, when comparing current isotretinoin exposure with nonexposure. The authors conclude the study provides no evidence that isotretinoin is associated with an increased risk of depression, suicide or other psychiatric disorders.

This work directly challenges a proposed causal association that had been built largely on case reports. By using antibiotic-treated acne patients as the comparison group, it accounts for the possibility that acne itself, rather than the drug, drives distress. It cannot close the question entirely: the suicide estimate carries a wide range because such events are rare, and databases may miss milder mood problems that never reach a doctor. Individual susceptibility also cannot be ruled out by population-level averages, so the debate did not end here.

If warnings about isotretinoin and depression have shaped how you think about acne treatment, this is a reason to weigh that fear against large-scale data in which the link simply did not appear. It is not proof of safety for every individual, since the study speaks to averages across thousands of patients rather than any one person's response. Treatment decisions still belong in a conversation with a prescriber. Knowing that the population-level evidence here found relative risks around 1.0 is useful context to bring to that conversation.