Aim Sundström et al. (BMJ 2010; 341: c5812) aimed to assess suicide risk before, during and after isotretinoin treatment.
Hypothesis People with severe acne may be at higher risk of attempted suicide regardless of exposure to isotretinoin.
Setting and design This retrospective cohort study from Sweden linked a patient register of isotretinoin users (1980–1989) or patients identified through specified pharmacies (1986–1989) to hospital discharges and a register of causes of death (1980–2001).
Study exposure The study exposure was isotretinoin, identified from a register of named patients given isotretinoin (special application to the Medical Products Agency) or patients prescribed isotretinoin in a dermatology clinic with ‘clinic approval’ to prescribe the medication without individual application.
Outcomes The outcomes were recorded suicide attempts (identified using ICD-8 and ICD-9 codes E950–E958, ICD-10 codes Y6–64); accidents with unclear intent and an ‘external cause’ (ICD-8 and ICD-9 codes E980–E988 and ICD-10 codes Y10–Y34) and deaths with underlying causes.
Primary outcome measure Standardized incidence ratio (SIR) of suicide rates compared with the general population 3 years before, during and up to 15 years after completion of treatment with isotretinoin.
Results The cohort included 5756 people. During the 21-year follow-up period, 128 (2·2%) had a hospital admission for a suicide attempt. For all attempts, the SIR [95% confidence interval (CI)] increased from 0·99 (0·65–1·44) 3 years prior to treatment to 1·57 (0·86–2·63) in the year before treatment, increasing to 1·78 (1·04–2·85) within 6 months after starting treatment and decreasing to 1·04 (0·74–1·43) 3 years following treatment. For first suicide attempts, the SIR (95% CI) increased from 0·89 (0·54–1·37) 3 years prior to treatment to 1·36 (0·65–2·50) in the year before treatment, increasing to 1·93 (1·08–3·18) within 6 months of treatment and decreasing to 0·97 (0·64–1·40) 3 years following treatment. Twelve (38%) of 32 patients who attempted suicide prior to treatment made a further attempt or committed suicide at some stage during follow up. Of the 14 who attempted suicide during or within 6 months of treatment, 10 (71%) made a further attempt during follow up.
Conclusions Sundström et al. (BMJ 2010; 341: c5812) conclude that there was an increased risk of suicide attempts up to 6 months after the end of treatment with isotretinoin and advise close monitoring for up to a year after completing a course of treatment. Patients with a history of suicide attempts before treatment made fewer new attempts at suicide than those where suicidality was observed in connection with treatment, suggesting that patients with severe acne with a history of attempted suicide should not automatically be refused isotretinoin treatment. The authors also state that suicide risk was already rising prior to treatment and that the additional risk cannot therefore be attributed to isotretinoin use.