Background: Improving access to high-quality care is a top priority for VHA. However, access is difficult to measure, especially in the ever-changing U.S. health care landscape. VHA currently focuses on perceived satisfaction measured from survey questions, or wait times measured with administrative data. The wait time metric has received considerable focus in both the Choice and MISSION Acts, yet experts recognize that wait times are imperfect; they are not only challenging to interpret, but they are rarely available from community providers, hindering any VHA to non-VA comparisons. VHA leaders tasked with implementing the Choice and MISSION Acts desire better evidence-based access measures so they can evaluate their program’s impact. Our inability to measure access threatens the future of VHA as a health care provider. Without new metrics that track gaps or improvements in access, VHA is likely to invest in the wrong initiatives, fueling critics who will argue that privatization will fix the programs Significance/Impact: Our objective is to develop new measures of access. These new measures will provide causal information on gaps in VHA services, while also showing the potential impact that expanded access would have on Veterans’ health. Our study leverages natural experiments in the form of arbitrary administrative rules that enable Veterans to access care outside VHA in the forms of Medicare and VA Community Care (VACC). Veterans close to the rule thresholds are similar, yet some gain access based on the rule, while the others do not. We can leverage this information to understand how this added access changes health care utilization and health outcomes. Our results will be useful to national and local VHA leaders as they grapple with how to best improve access with a limited budget. Innovation: It is not feasible to perform a large-scale randomized clinical trial to find the effect of access on utilization and health outcomes. Correlational studies will miss important confounders, and as everyone knows correlation does not equal causation, which is what VHA leaders need most. Instead, we apply novel econometric techniques to take advantage of natural experiments and find the causal effects of increasing access. The results from this approach can then be used as a measure of access for both VHA and non-VA care. This is directly aligned with HSR&D's priorities on access to care, research related to the MISSION Act, and advancing health services research methods across conditions or care settings. Specific Aims: Aim 1: Find the causal impact of Medicare eligibility on Veteran utilization and health, and identify procedures and diagnosis groups that are most affected. Aim 2: Find the causal impact of VACC on Veteran utilization and health, and find the procedures and diagnoses most affected. Aim 3: Identify subgroup analyses that would give crucial information to VHA leaders. Methodology: In Aims 1 and 2, we will apply an econometric technique called regression discontinuity design. We will gather a near complete census of VHA and non-VA records for all recent Veteran VHA users, allowing for precise claims-based measures of utilization, health, and mortality. In Aim 1, this approach leverages the sharp change in Medicare eligibility that occurs at age 65 to find the effect of Medicare on utilization and health. Aim 2 uses the same approach, but instead of age 65, we use the driving distance rules to find the effect of VA Community Care on utilization and health. For these aims we will examine both overall effects and procedure and diagnosis specific effects. In Aim 3, we take a combined approach of working with operational partners and applying machine learning techniques for heterogenous treatment effects to identify and examine metrics and measures that can be used for policy formation. Next Steps and Implementation: By working with operational partners through Aim 3, we will identify opportunities and barriers to implementing measures and metrics derived from our results. This information will be key to setting policy related to VHA’s evolution into a provider and payer of Veteran care while ensuring that Veterans receive high-quality care in both VHA and non-VA settings.
External Links for this Project
Grant Number: I01HX003106-01
None at this time.
TRL - Applied/Translational
Models of Care, Outcomes - Patient, Outcomes - System, Utilization
None at this time.