2005 HSR&D National Meeting Abstract
1044 — Chronic Care Models for Chronic Mental Illness? Results of CSP #430
Bauer MS (Providence VAMC)
CSP #430 Study Team
Context: Chronic illness care models (CCMs) can improve outcome for chronic medical illnesses and depression treated in primary care, but depend heavily on patient collaboration and self-management skills. Can CCMs also improve outcome in common, chronic mental illnesses such as bipolar disorder?
Objective: This VA Cooperative Study was designed to determine impact of a CCM incorporating patient, provider, and system components on long-term clinical and functional outcome, health-related quality of life (HRQOL), treatment satisfaction, and direct treatment costs in bipolar disorder.
Design: 3-year randomized controlled trial of CCM vs. usual care using participant-level randomization, with a priori hypothesis that improvements would accrue in time over 3 years.
Setting: Mental health clinics of 11 VA medical centers (VAMCs).
Participants: Veterans in a major affective episode (manic, depressed, mixed) hospitalized on acute psychiatric wards, randomized at discharge to continued usual care (n=149) or CCM treatment (n=157).
Intervention: The CCM included clinic-based patient self-management enhancement via manual-based group psychoeducation; provider support via manual-based, simplified VA bipolar practice guidelines; and manual-driven access/continuity procedures via nurse care manager supplementing psychiatrist effort.
Main Outcome Measures: Affective status, social function, HRQOL, treatment satisfaction, and direct all-treatment costs from the VA perspective.
Results: The CCM reduced weeks in affective episode (p=0.041) and weeks manic (p=0.017), but not weeks depressed or mean symptom levels. CCM participants experienced 2.1 fewer weeks in episode annually. CCM improved overall (p=0.003), work (p=0.049), parental (p<0.001), and family (p=0.005), but not marital or leisure, function. Benefits accrued linearly as hypothesized. Mental (p=0.01), but not physical, HRQOL and treatment satisfaction (p<0.001) were significantly better over the entire 3 years. Results were not explained by increased mood pharmacotherapy. The CCM was cost-neutral.
Conclusions: 3-year CCM treatment resulted in broad-based improvements that accrued gradually as hypothesized.
The broad-based impact, low fixed start-up costs, and overall cost-neutrality of the CCM make a compelling case for its dissemination, and for the development of similar CCMs for other chronic mental illnesses. Importantly, this study demonstrates that individuals with a severe, chronic mental illness can participate successfully in highly collaborative CCMs, further breaking down conceptual barriers between “mental” and “physical” illness.