A Population-Based, Nationwide Longitudinal Study of Bipolar Disorder With Incident Dementia in Taiwan

Objective: Affective disorders are associated with increased risk of dementia, yet most studies focus on the association of major depressive disorder or depressive episodes of bipolar disorder with increased dementia risk. The association of manic/mixed episodes of bipolar disorder with increased dementia risk is unclear. Participants: Between January 1, 2001 and December 31, 2009, 20,535 individuals aged 45-80 years with bipolar disorder and 82,140 age- and sex-matched comparisons were enrolled and followed up to December 2011 in Taiwan. Those who developed dementia (ICD-9-CM codes: 290.0 -290.4, 294.1-294.2, and 331.0-331.2) during the follow-up period were identified. Design and Methods: Cox proportional hazards models were used to examine the relationship between manic/mixed/depressive episodes of bipolar disorder and incident dementia. We also assessed the association between the frequency of psychiatric admissions (total, manic/mixed, and depressive episodes per year) for bipolar disorder and dementia risk. Results: Bipolar disorder was associated with increased risk of incident dementia (hazard ratio [HR]: 7.52, 95% confidence interval [CI]: 6.86-8.25). Greater frequency of manic/mixed (> 2/year: HR: 4.50, 3.50-5.79; 1-2/year: HR: 3.17, 2.31-4.36) and depressive episodes (> 2/year: HR: 7.84, 5.93-10.36; 1-2/year: HR: 2.93, 2.05-4.19) were associated with increased risk of incident dementia. Conclusions: Not only depressive episode of bipolar disorder, but manic/mixed episodes of bipolar play a role as a risk factor of incident dementia, especially for those patients with more than two manic/mixed episodes per year. These findings remind the clinicians the importance of preventing the relapse of bipolar disorder for the potential subsequent cognitive decline and disease.