1036 — Using System Dynamics Tools to Integrate Evidence into VA Stroke Care
Beadles CA (Center for Health Services Research in Primary Care, Durham VAMC), Kaufman MA
(Center for Health Services Research in Primary Care, Durham VAMC), Homer J
(Homer Consulting, Voorhees, NJ), Cheng E
(VA Greater Los Angeles Healthcare System), Williams LS
(Center of Excellence on Implementing Evidence-Based Practice; Richard L. Roudebush VA Medical Center), Bravata DM
(Center of Excellence on Implementing Evidence-Based Practice; Richard L. Roudebush VA Medical Center), Matchar D
(Duke-NUS Graduate Medical School Singapore), Bosworth HB
(Center for Health Services Research in Primary Care, Durham VAMC), Hassmiller-Lich KM
(Center for Health Services Research in Primary Care, Durham VAMC)
Stroke is a significant burden upon the quality of life of veterans. Numerous interventions exist for improving stroke prevention, acute care and rehabilitation services, but it is difficult to compare their relative effectiveness in a complex system. We used System Dynamics methodology to engage stakeholders in considering VA healthcare factors, patient factors, and intervention effects data to model the systems involved in stroke prevention and treatment within the VA.
Stroke clinicians, researchers, and administrators worked iteratively to identify VA and non-VA data describing current VA care, veteran characteristics, stroke interventions, and stroke outcomes. System Dynamics tools were implemented to model VA stroke care with inputs including degree of intervention implementation, heterogeneous treatment effects, aging population, and trends in health risk behavior through accumulation and feedback loop processes. We simulated 5, 10, and 20 year intervals to evaluate the effects of interventions on outcomes, including: disability adjusted life years (DALYs) and stroke-related fatalities.
The 20-year base case scenario predicted a cumulative total of 2.55 million DALYs lost and 98,099 stroke-related fatalities. Maximally implementing all 15 interventions (reflecting an upper bound for comparing intervention strategies) predicted 300,000 fewer DALYs lost and 15,201 fewer stroke-related fatalities at 20 years. The most effective single intervention strategy was enhancement of hypertension and atrial fibrillation (AF) management among veterans with cardiovascular disease (5,200, 22,000, and 72,000 fewer DALYs lost and 355, 1,282, and 3,360 fewer deaths at 5, 10, and 20 years). The most effective combination of strategies, employed 4 interventions (risk factor management for those with diabetes or cardiovascular disease, post-TIA, and post-stroke management) and predicted 10,469, 42,000, and 129,000 fewer DALYs lost and 1,030, 3,028, and 7,550 fewer deaths at 5, 10, and 20 years.
System Dynamics modeling provides a valuable framework to identify effective intervention strategies in VA stroke care.
Creating a framework to critically evaluate policy options given resource constraints and multiple stakeholder viewpoints can improve communication and shared goal setting between QUERIs and operational leaders. The model facilitates informed strategic agenda setting and identification of actionable priorities to improve VA stroke care.