OBJECTIVES: While numerous studies have reported higher prevalence of Type 2 diabetes mellitus in patients with bipolar disorder, few have examined the impact of comorbid diabetes on the costs associated with treating bipolar disorder patients in a commercial health plan. This study explored differences in health care costs between bipolar disorder patients with and without comorbid diabetes. METHODS: Administrative claims data from a large commercial health plan were used to identify prevalent bipolar disorder cases from medical claims based on a diagnosis code for bipolar disorder (ICD-9-CM diagnosis code 296.0x, 296.1x, 296.4x-296.7x) from 7/1/99 through 12/31/01. Medical and pharmacy costs were calculated for a 365-day period for continuously enrolled subjects. Costs were defined as bipolar-related if a bipolar diagnosis appeared in the primary diagnosis position. Bipolar-related pharmacy costs were calculated from claims for mood stabilizers and atypical antipsychotics. RESULTS: A total of 9459 bipolar disorder patients were identified for analysis. Of these, 468 (5%) had a diabetes diagnosis. Both median bipolar-related medical costs and median pharmacy costs were lower for patients without diabetes compared to patients with diabetes ($148 versus $205; $225 versus $504). All-cause medical and pharmacy costs were also higher among patients with comorbid diabetes. CONCLUSIONS: Comorbid diabetes appears to substantially increase the cost of providing both general medical care and bipolar-related health services. Several mechanisms could account for these differences, including clinical course of bipolar disorder, patient motivation for medical care, and physician practice patterns. Further exploration into these mechanisms is warranted.