Health Care Resource Utilization and Costs in a Commercially Insured Population of Patients With Bipolar Disorder Type I and Frequent Psychiatric Interventions

Abstract

Background Bipolar disorder type I (BP-I) is one of the most expensive behavioral diagnoses in the United States. Characterizing patient populations that consume significant resources would be useful for designing and implementing additional resources and targeted interventions to reduce the costs of BP-I. Objective This analysis compared the characteristics, health care resource utilization, and costs of commercially insured patients with BP-I (indicating a history of manic or mixed episodes) and frequent psychiatric interventions (FPIs) versus those without FPIs. Methods This retrospective study used data from commercial insurance claims to identify adults with FPIs (≥2 clinically significant events [CSEs]) or without FPIs during a 12-month identification period (year 1). CSEs included emergency department (ED) visits or hospitalizations with a principal diagnosis of BP-I, the addition of a new medication to the observed treatment regimen, or a ≥50% increase in BP-I medication dose. Demographic and clinical characteristics were evaluated during the identification period, and health care resource utilization and costs were evaluated during a 12-month follow-up period (year 2). Results Data from 7620 patients with FPIs and 11,571 without FPIs were included (women, 67.1% and 59.9%, respectively; P < 0.001). Of patients with FPIs in the identification period, 22.2% continued to have FPIs in the follow-up period. In the follow-up period, the group with FPIs had a greater proportion of patients with psychiatric-related inpatient hospitalizations (14.6% vs 2.8%) and ED visits (11.74% vs 2.7%), a longer mean hospital length of stay (18 vs 8.24 days), and greater adjusted mean psychiatric-related costs ($6617 vs $3276) and all-cause health care costs ($14,091 vs $9357) compared with the group without FPIs (all, P < 0.001). The risks for a psychiatric-related hospitalization and an ED visit during the follow-up period were significantly greater in the group with FPIs compared with the group without (odds ratios, 4.86 and 3.76, respectively; both, P < 0.01). Conclusions In this retrospective analysis, FPIs were associated with a greater number of FPIs during follow-up, ∼2-fold the psychiatric-related costs, and 1.5-fold the all-cause health care costs compared with no FPIs. These data highlight the economic burden of FPIs and the potential for health care cost reductions from improved management options in these patients.