Primary Hypothesis: The primary hypotheses are that Bipolar Disease Program (BDP) intervention will significantly improve (1) manic and (2) depressive symptom scores, as well as Total Treatment Costs as compared to usual care of bipolar patients.

Secondary Hypothesis: Secondary hypotheses include significant improvement in BDP patients as compared to usual bipolar treatment care with respect to functional outcome, quality of life, intensity of somatotherapy, patient satisfaction, and provider attitudes.

Intervention: Usual (psychiatric) Care vs Bipolar Disorder Program

Primary Outcomes: The primary outcomes are: (1) Manic Symptom Score; (2) Depressive Symptom Score; and (3) Total Treatment Costs.

Study Abstract: Based on highly promising preliminary data, it is proposed to conduct a multi-site randomized controlled trial of a high-intensity ambulatory treatment program for bipolar disorder against standard office-based, physician-centered care. The major characteristics of this program are that it emphasizes (1) aggressive guideline-driven pharmacotherapy, (2) continuity of care with identified primary mental health nurse clinicians supported by psychiatrist back-up, and (3) patient education to improve treatment alliance and illness management skills.

Patients with bipolar disorder will be randomly assigned to either standard care or the high intensity ambulatory program for three years. Outcome variables will cover three distinct domains: disease-specific outcome (number, length, and severity of manic and depressive episodes), functional outcome (social and occupational role function and subjective quality of life), and total treatment costs (direct treatment costs and indirect costs of illness). Preliminary data indicate that these domains are related but not redundant. Each of these is relevant to patient well-being and to VHA management.

Patients with bipolar disorder. Index episode of manic, major depression or both requiring hospitalization on acute unit. At lease two hospitalizations, three or more months apart within the past 5 years.