Determinants of functional outcome and healthcare costs in bipolar disorder: a high-intensity follow-up study

Abstract

Background: Review of published studies reveals few data regarding determinants of the poor functional outcome and high healthcare costs that are characteristic of bipolar disorder. In order to identify potential mechanisms, critical to designing optimal treatment strategies, this longitudinal study investigated (a) the degree to which disease outcome is correlated with functional outcome and direct treatment costs, and (b) whether similar demographic or clinical characteristics predict disease and functional outcome and healthcare costs.

Methods: Disease and functional outcome were assessed in bimonthly structured interviews over 48 weeks in 43 outpatient veterans with bipolar disorder. Direct mental health treatment costs from the VA perspective were determined from the VA database and patient interview. Regression analysis was used to determine association among the three outcome domains, and to identify clinical or demographic variables that predicted each of the three domains.

Results: Functional outcome was correlated with depressive, but not manic, symptoms during follow-up. Costs were not correlated with any measure of disease or functional outcome. Several demographic, but not clinical, characteristics predicted functional outcome. In contrast, several clinical, but not demographic, characteristics predicted symptom status. No predictors were associated with direct treatment costs.

Limitations: Subjects were predominantly male veterans of relatively homogeneous social class, followed prospectively for approximately one year in a clinic designed specifically to minimize barriers to care.

Conclusions: Data from this and prior studies indicate that ongoing depressive symptoms are strongly associated with functional outcome, although substantial variance remains unexplained. Optimal models to explain functional outcome and healthcare costs will need to address factors besides simply disease severity and chronicity. The authors present a heuristic paradigm for understanding both the research and therapeutic aspects of these findings.