Association of early- and late-life bipolar disorder with incident dementia. A Danish cohort study

Background: This study aimed to explore the association between bipolar disorder and the risk of developing dementia, and whether the risk varies with age at the onset of bipolar disorder. Methods: In this study, 37,084 individuals with a first-time diagnosis of bipolar disorder diagnosed between 1969 and 2018 and a reference population (n = 189,662) matched on sex, birth year and time of bipolar diagnosis (index date) were followed in nationwide registries for incident dementia until October 2020. Associations were analysed using Cox proportional hazard regression with adjustment for sex, education level, alcohol or drug abuse, traumatic brain injury, ischemic heart disease, stroke and diabetes mellitus. Results: In total, 6.6 % of individuals with bipolar disorder and 4.0 % in the reference population developed dementia during the mean follow-up of 13.1 years. Compared to the reference population, individuals with bipolar disorder had a higher incidence of dementia during follow-up after adjusting for potential confounders (HR: 2.66, 95 % CI [2.53-2.79]). The strength of this association did not vary among individuals diagnosed with bipolar disorder before and after age 45. Limitations: The higher risk of dementia identified for individuals with bipolar disorder could be influenced by detection bias and, despite a large cohort, some of the age-stratified analyses were still affected by lack of statistical power. Conclusion: Individuals with bipolar disorder have a higher risk of developing dementia compared to a reference population without bipolar disorder, independent of the age at the onset of bipolar disorder.