Changes to Medicaid Enrollment Income Verification Logic in West Virginia Increased Application Approvals

• In Event: Administrative Burden in the Safety Net: Spillovers and Solutions

Abstract

CONTEXT: During the pandemic, West Virginia Medicaid and its vendor Optum, made major changes to improve the automated verification rate for Medicaid renewals (ex parte), increasing the rate from 13% in April 2023 to 59% by April 2024. However, automated processes for new Medicaid applicants remained onerous, with income verification representing a crucial compliance barrier to enrollment. Methods to automatically verify an applicants’ income often use commercial sources that are costly, time-consuming, and miss a large proportion of applicants. For those not verified, manual verification methods are necessary, which means clients must submit pictures of paystubs or other proof of income. The status quo of income verification leads to reduced case worker productivity, higher churn, and lower participation rates among the eligible. To reduce these barriers, starting in May 2024, the Income Verification as a Service (IVaaS) team at Nava PBC partnered with the West Virginia Department of Human Services Bureau for Medical Services and their vendor Optum, to pilot changes in their automated income verification processes for new Medicaid applicants. 

METHODS: In this study, we evaluate the impact that these IVaaS changes had on income verification, processing times, and Medicaid approval rates in West Virginia. The process changes included: 1) rather than querying three sources of income simultaneously, each source was queried in order of reliability. 2) Logic was relaxed to remove “reasonable compatibility” restrictions as long as the returned income was still below the relevant threshold. 3) For people reporting $0 income and no results from the three sources, consider them verified. 4) Logic was changed to allow children to be automatically income verified (given their higher eligibility threshold) even if their parent wasn’t verified. We tested the hypothesis that, by improving the logic to automatically verify income eligibility, Medicaid applications could be processed and approved faster than before. To identify the causal impact of these process changes, we use difference-in-differences methods. 

RESULTS: Our analysis found improved automated income verification rates and reduced processing time to eligibility determination, particularly for lower income Medicaid categories and children. For automated verification of income eligibility, we find an increase in rates of 1 to 2 percentage points for adults, 10 percentage points for children, and 31 percentage points for $0 income applicants. For application approvals, results show faster approvals by 5 days for most adults and a 12 percentage point increase in approval rates for the lowest-income children. Those with higher or stable incomes (such as SSI recipients) do not experience changes in verification or approval rates as their incomes were already comparatively easy to verify. 

CONCLUSION: This study shows that changes to West Virginia Medicaid’s income verification processes removed unnecessary frictions to accessing benefits, especially for those with lower or unstable incomes and children. Our findings illustrate the potential gains in policy administration through partnerships between Civic Technology, state governments, and academics to build rigorous evidence on ways to improve access to safety-net programs. Results also show the opportunity for IVaaS to reduce burdens across states and programs.

Authors

• Jeremy Barofsky, Georgetown University
  Presenting Author

• Eric Giannella, Georgetown University
  Non-Presenting Co-Author

• Pamela Herd, Georgetown University
  Non-Presenting Co-Author

• Sebastian Jilke, Georgetown University
  Non-Presenting Co-Author

• Donald Moynihan, University of Michigan
  Non-Presenting Co-Author