LAUSR

BACKGROUND This study aimed to explore the impact of age-at-onset on clinical outcomes in bipolar-spectrum disorders, including the development of alcohol or substance use disorder, suicidality, and psychiatric hospitalization. METHODS This population-based study enrolled newly-diagnosed bipolar-spectrum patients, including 4,367 patients with early-onset bipolar disorder (EOBD), 64,787 patients with adult-onset bipolar disorder (AOBD), and the same number of age-controlled comparison subjects without bipolar disorder, from Taiwan's National Health Insurance Research Database. Time-dependent covariate Cox regression models were used to estimate the effect of age-at-onset on clinical outcomes with adjustment for pre-existing psychiatric comorbid conditions and pharmacological treatment patterns. Sensitivity analyses using different definitions of study sample and age cutoffs were conducted. RESULTS The average follow-up duration was 5.7 years. After adjustment with time-dependent covariates and chronological age, there were no significant differences in the risks for developing new-onset alcohol or substance use disorders and psychiatric hospitalization between EOBD and AOBD patients. Although EOBD patients had a higher risk of hospitalization for suicide and self-harm than did AOBD patients in primary analysis, this finding did not replicated in the sensitivity analyses. LIMITATIONS The symptom profile and severity of bipolar disorder was not available in the NHIRD; therefore, surrogate indicators of clinical outcome might not be sensitive enough to detect the subtle differences. CONCLUSIONS EOBD and AOBD patients had similar risks for developing alcohol or substance use disorders. Their risk of psychiatric hospitalization was similar. Whether EOBD patients might have a higher risk of hospitalization for suicide and self-harm warrants further investigations.