Characteristics of Patients With Bipolar Disorder Managed in VA Primary Care or Specialty Mental Health Care Settings

OBJECTIVES: This study compared the clinical characteristics, use of guideline-concordant pharmacotherapy, and outcomes of patients diagnosed as having bipolar disorder who were exclusively seen in Department of Veteran Affairs (VA) primary care settings with those of patients with bipolar disorder who received any VA mental health services. METHODS: Data from the 1999 Large Health Survey of Veterans were linked to VA data from the National Psychosis Registry to identify patients diagnosed as having bipolar disorder (N=14,643). Multivariable analyses adjusting for sociodemographic characteristics and clinical and severity factors determined whether exclusive primary care use versus any mental health care use was associated with poor clinical and services outcomes. RESULTS: Overall, 7.6% used primary care services exclusively. Compared with persons who used specialty care services, those who used primary care exclusively were more likely to be diagnosed as having cardiovascular disease (odds ratio OR=1.26, p<.05) or hypertension (OR=1.31, p<.01), less likely to receive guideline-concordant pharmacotherapy (OR=.18, p<.001), more likely to have an inpatient medical visit (OR=1.36, p<.01), and less likely to have an inpatient psychiatric visit (OR=.36, p<.001). Persons who received only primary care were more likely to have worse physical health and better mental health, as measured by the 36-Item Short-Form Health Survey. CONCLUSIONS: Patients with bipolar disorder treated in primary care settings were more likely than those who received some care in a mental health specialty setting to have comorbid general medical disorders. Optimal care settings for patients with bipolar disorder may require strategies that address gaps in general medical as well as psychiatric care.