• bipolar disorder;
  • age-at-onset;
  • substance use disorders;
  • birth-cohorts;
  • cohort effect

Objective:  Substance use disorders and birth-cohort have been associated with an earlier onset in bipolar disorder (BD). This study aimed at evaluating the inter-relations of these factors in age-at-onset in bipolar illness.

Method:  Two-hundred and thirty patients with bipolar I disorder were cross-sectionally evaluated. Patients were categorized into four age groups for analysis. Lifetime comorbidity and age-at-onset were derived from the Structured Clinical Interview for DSM-IV.

Results:  There was a strong linear association between age group and age-at-onset. Lifetime alcohol and drug use disorders were also associated with age-at-onset. Illicit drug and alcohol use disorders and age group remained significant in the multivariate model. No interactions appeared.

Conclusion:  Both age group and dual diagnoses had strong and independent impacts on age-at-onset in out-patients with BD. Substance abuse may be partly accountable for earlier symptom onset, but other features of BD in younger generations are still in need to be accounted for.