How Medicaid expansion affects demand for VA care

Last year, Adrianna illuminated the interplay between demand for Veterans Health Administration (VA) care and Medicaid expansion.

Medicaid expansion could relieve some pressures on the VA system. For some veterans, geographic access could be a serious barrier to care, since VA benefits are typically provided at VA facilities. Medicaid coverage has its own shortcomings, but dual eligibility would permit veterans to seek care wherever it was most readily available.

Of course, almost half of states are still refusing to expand the public program, affecting an estimated 258,600 uninsured veterans.

At the time of her writing, and until last week, nobody had quantified for all states the extent to which Medicaid expansion (or non-expansion) affects VA enrollment or utilization. A new paper in the journal Healthcare by me, Amresh Hanchate, and Steve Pizer does just this.

If the ACA’s Medicaid expansion had been implemented in all states, enrollment for VA health coverage, acute inpatient care (days), and outpatient visits would have been 9%, 6%, and 12% lower, respectively. In states that did not expand Medicaid in 2014, VA enrollment, inpatient days, and outpatient visits were, respectively, 10, 6, and 13 percentage points higher than they would have been otherwise. VA medical centers in states that did not expand Medicaid in 2014 are likely to have experienced a higher demand, and commensurately longer wait times.

We based our estimates on the historical relationship between prior, state-level Medicaid expansions and VA enrollment and utilization. To do so, we constructed a measure of state and year varying Medicaid eligibility policy that is not sensitive to economic, demographic, or health care utilization variation. The extent to which this Medicaid eligibility policy variable drives VA demand formed the basis of our simulation of the effect of the ACA’s Medicaid expansion (or non-expansion). State-level results are provided in the paper.

One limitation: we did not simulate the separate effects of the individual or employer mandates, insurance subsidies, or exchanges. Other limitations are noted in the paper. One that is not in the paper is that demographic changes independently affect VA demand; we did not consider any such changes that might have occurred between the latest data in the study (2008) and the implementation of the Medicaid expansion (2014).

Prior posts on the implications of the ACA for veterans are here, here, and here. I gave a webinar on the work described above, which is available here. (Results in that webinar are preliminary and may not match those in the paper in all cases.)


Hidden information below


Email Address*