Isotretinoin and Adverse Neuropsychiatric Outcomes

Retrospective Cohort Study Using Routine Data

Tapio Paljarvi; Tess McPherson; Sierra Luciano; Kimmo Herttua; Seena Fazel


The British Journal of Dermatology. 2022;187(1):64-72. 

In This Article

Abstract and Introduction


Background: Severe neuropsychiatric outcomes have been reported in individuals exposed to isotretinoin, but the evidence is inconclusive and complicated by several methodological limitations.

Objectives: To establish and quantify the association between isotretinoin use for acne and 1-year incident neuropsychiatric adverse outcomes.

Methods: A propensity score-matched cohort study of electronic medical records between the years 2013 and 2019 with patients followed up for 1 year after their index dispensed prescription was conducted. The database included over 12 million patients aged 12–27 years. We analysed data for individuals with acne in this age range with a dispensed prescription for isotretinoin or a control prescription. Outcomes included diagnoses of any incident sleep or mental health disorder, or nonfatal self-harm within 1 year of the index prescription.

Results: We included 30 866 patients prescribed isotretinoin for their acne, 44 748 prescribed oral antibiotics, 108 367 prescribed topical anti-acne agents and 78 666 patients with acne but without an anti-acne prescription. After propensity score matching for baseline confounders, the odds ratio (OR) for any incident neuropsychiatric outcomes in patients with acne exposed to isotretinoin was 0·80 [95% confidence interval (CI) 0·74–0·87] compared with those on oral antibiotics; 0·94 (95% CI 0·87–1·02) compared with those using topical anti-acne medicines; and 1·06 (95% CI 0·97–1·16) compared with those without a prescription for anti-acne medicines. Patients exposed to isotretinoin experienced significantly more incident physical symptoms than patients in any of the three comparison cohorts.

Conclusions: Isotretinoin was not independently associated with excess adverse neuropsychiatric outcomes at the population level. When monitoring potential adverse outcomes during isotretinoin treatment, clinicians should also consider the high mental health burden associated with treatment-resistant acne and the potential contribution of physical side-effects of prescribed medication on mental health.


Acne vulgaris (acne) is a highly prevalent long-term inflammatory skin condition with symptom onset typically in adolescence, with the highest prevalence between the ages of 14 and 19 years.[1–3] Acne is known to affect both physical and psychosocial health and wellbeing.[4] Oral isotretinoin is a vitamin A derivative, and it is indicated for the treatment of recalcitrant inflammatory acne. Isotretinoin can also be used off-label, including to treat and prevent certain types of skin cancer.[5] Many patients treated with isotretinoin experience adverse side-effects such as dry skin or mouth, headache or musculoskeletal pain.[6,7] The various mucocutaneous and central nervous system side-effects of isotretinoin are mostly dose related and some, such as dry skin and skin irritation, are common with lower therapeutic doses. Some patients also experience an initial worsening of acne symptoms (flare-up) after the start of treatment.[8] In addition, severe neuropsychiatric outcomes, such as anxiety, depression, suicidal ideation, self-harm and completed suicide, have been reported. However, the evidence base for severe neuropsychiatric outcomes is mixed and inconclusive.[9–22] For example, isotretinoin has been associated with both increased[14] and reduced[16] risk of depression.

There are several methodological limitations in the current evidence about the association between isotretinoin and severe neuropsychiatric adverse outcomes.[14] Much of the evidence is based on small cohorts of patients leading to a lack of precision in effect estimates. For example, a recent meta-analysis of 31 studies was able to include only 1411 patients with an assessment of depression before and after isotretinoin exposure.[16] More importantly, because acne severity in those patients exposed to isotretinoin and patients in the unexposed control groups have likely differed, confounding by indication explains part of the conflicting evidence. To address confounding by indication in relation to isotretinoin-associated adverse neuropsychiatric outcomes, we conducted a propensity score-matched retrospective cohort study to compare the incidence of neuropsychiatric outcomes in patients exposed to isotretinoin and patients receiving other types of acne treatment by using information from the electronic medical records of over 12 million patients aged 12–27 years from 56 healthcare organizations. We defined three different patient populations as control groups for the patients who had been exposed to isotretinoin to indirectly control for the severity of acne.