4113 — Differential effects of Medicaid expansion on VA and Medicaid dual use of outpatient services
Lead/Presenter: Patrick O'Mahen,
COIN - Houston
All Authors: O'Mahen PN (Center for Innovations in Quality, Effectiveness and Safety, Houston), Liaou, D (Center for Innovations in Quality, Effectiveness and Safety, Houston), Petersen, LA (Center for Innovations in Quality, Effectiveness and Safety, Houston)
To examine how pre-Affordable Care Act state-level Medicaid expansions affect dual use of outpatient services in the Veterans Health Administration (VA) and non-VA care funded by Medicaid among VA enrollees.
We used difference-in-difference to determine association between Medicaid expansions and health care utilization. We studied two states that expanded Medicaid to childless adults in 2001 and paired them with demographically similar non-expansion states (New York vs. Pennsylvania, Arizona vs. New Mexico and Nevada) to isolate changes in outpatient usage from expansion. We used fractional logistic regression to calculate effects of expansion on proportions of outpatient visits at VA facilities and ordered logit to determine the proportion of Veterans using both VA and Medicaid outpatient services (dual users) in a given year. We estimated effects for Veterans with priority 5 (eligible by financial hardship) and non-priority 5 status. We controlled for age, gender, race, distance to the nearest VA, and disease burden. Data sources were VA Corporate Data Warehouse and Medicaid Analytic Extract files. The study population was 807,573 VA-enrolled Veteran ages 18-64 living in a study state between 1999-2006.
For low-income Veterans, Medicaid expansion was associated with a 2.86 percentage-point reduction in the proportion of VA outpatient visits (99% CI 2.47. - 3.25) and a 0.94-point reduction (0.74 - 1.13) for other Veterans; and a shift of 107,155 visits (90,074 - 124,237) from the VA to Medicaid annually. The proportion of annual dual-user Veterans increased by 2.11 points (1.83-2.39) for low-income Veterans and 0.84 points (0.68-1.00) for other Veterans; we estimated expansion added 5,004 (4,281-5,787) dual users annually.
State-level Medicaid expansion is associated with a lower proportion of outpatient visits at VA facilities and a greater percentage of Veterans using both VA and Medicaid services. These trends are more pronounced for low-income Veterans.
Medicaid expansion is associated with dual use of VA and Medicaid-funded services, particularly for low-income Veterans, which has significant budgetary and care-coordination implications for VA planners. This information helps to project usage for both the ACA and the VA MISSION Act, which expand Veteran access to non-VA care to those who did not have prior access outside access to care.