Predictors of Hospitalization for Manic Episode in Alzheimer's Dementia: Inputs From an Inpatient Case-Control Study

Objectives The correlates of manic episodes in dementia have not been systematically studied. The primary goal of our study is to compare the sociodemographic characteristics and psychiatric comorbidities in Alzheimer’s dementia (AD) inpatients with manic episodes versus without manic episodes, and to evaluate the demographic predictors and risk factors for manic episodes in AD inpatients. Methods We conducted a case-control study using the Nationwide Inpatient Sample of 34,285 AD patients (age ≥60 years). Subsequently, the cases i.e., AD inpatients with a manic episode (N = 1,035) and the controls (without a manic episode, N = 1,035), were extracted using propensity-score matching based on age. The cases did not have a past psychiatric history of bipolar disorders. We used the logistic regression model to evaluate the odds ratio (OR) of association between pre-existing psychiatric comorbidities and manic episodes and evaluate the demographic predictors of manic episodes in AD inpatients. Results A higher proportion of AD inpatients with manic episodes were females (63.8%), whites (85.2%), and from low-income families below the 50th percentile (63%). Females were more likely to be hospitalized for manic episodes (OR 1.33; 95% CI 1.09-1.64) than males. AD inpatients with manic episodes had a higher risk of presenting with suicidal behaviors (OR 1.88; 95% CI 1.23-2.86). A significantly higher proportion of AD inpatients with manic episodes had comorbid tobacco use (5.3% vs. 3.4%) and cannabis use (1.4% vs. 0%) compared to those without manic episodes. Conclusion Females with AD had a greater risk of being hospitalized for manic episodes. These patients have an 88% higher risk of suicidal behaviors during the manic presentation and have comorbid tobacco and cannabis use. Early diagnosis and management of manic episodes in at-risk AD patients are important to improve the quality of life (QoL) and outcomes.