Antidepressant use and risk of bipolar disorder in children, adolescents and young adults with unipolar depression: a target trial emulation study.

Bipolar disorder (BD) typically manifests during adolescence and young adulthood, with unipolar depression (UD) being common initially. Concerns persist that antidepressants may increase the conversion risk from UD to BD. This study employed a target-trial-emulation framework, using an electronic-medical-record database of public-healthcare services in Hong-Kong, to examine BD-conversion risk, defined by two BD-diagnoses or one BD-diagnosis and a mood-stabilizer prescription, over 12-week and 52-week follow-up among individuals aged 6-30 years diagnosed with incident UD between 2002-2022. Participants initiating antidepressants within 90 days of UD were compared to those without antidepressant treatment, applying inverse-probability-weighting to balance baseline characteristics. We further identified factors associated with BD-conversion within both follow-up intervals using weighted-Cox-regression analyses with backward stepwise-selection. Among 10,801 individuals (mean-age=22.1±5.3 years; male=24.9%), 84.7% initiated antidepressants. The outcome occurred in 87 individuals by 12-weeks and in 257 by 52-weeks. Treatment group showed no significantly increased BD-conversion risk at 12-weeks compared to controls (hazard-ratio [HR]=2.94; 95% confidence interval (CI)=0.89-9.69), while a modestly increased risk at 52-weeks (HR=2.14; 95%CI=1.31-3.51). Presence of psychotic features and past psychiatric hospitalization were associated with BD-conversion. Findings were consistent in stratified analyses by age (6-17, 18-30 years) and secondary analyses repeated in a separate cohort of individuals aged >30 years. Our findings indicate no short-term risk of antidepressant-associated BD-conversion in people with UD regardless of age. Although a modest increase in risk was observed at extended 52-week follow-up, this association likely reflects the influence of risk factors rather than antidepressant exposure causally contributing to BD-conversion.