States have spent millions of dollars building Medicaid eligibility systems that promise automated, real-time, and accurate eligibility decisions. This is a worthy goal because unnecessary paperwork burdens cause eligible individuals to lose coverage for failure to complete forms. The promise of automated eligibility determinations, however, is often not realized as states across the country—such as Ohio, Rhode Island, Tennessee, and Georgia—have struggled with complex, error-prone eligibility systems.
CMS oversight is inadequate
Oversight from CMS in recent years has focused more on deterring erroneous enrollment and fraud than on reducing paperwork or erroneous denials of eligible enrollees. For instance, one of the government’s primary measures–the “Payment Rate Error Measurement” (PERM)–focuses exclusively on identifying improper payments – that is, situations where states enrolled an individual who was not eligible (or for whom the state did not sufficiently document the verification of their eligibility). States who have PERM rates higher than 3% can face financial penalties. The PERM does not, however, evaluate whether individuals were improperly denied coverage.
In fact, only one CMS audit—the Medicaid Eligibility & Quality Control—evaluates whether states improperly denied or terminated Medicaid coverage. But the required sample sizes for review are extremely small—just 200 Medicaid cases per state per year—and, in contrast to the PERM, no financial penalties are attached to these results.
This limited review of eligibility denials cannot capture the numerous, ongoing problems with state systems. Worse, the greater emphasis on fraud has led to new state policies requiring additional verification and paperwork, increasing the number of eligible individuals who lose coverage.
Fortunately, eligibility and enrollment systems can be improved while ensuring sufficient safeguards and due process protections for enrollees. States can:
Ensure critical populations are included in ex parte renewals
A central feature of the new eligibility and enrollment systems is supposed to be their ability to conduct automatic renewals (often called “ex parte” renewals) that rely on electronic data sources, rather than asking enrollees to submit paper documentation to verify their continued eligibility.
Although ex parte renewals are required by federal law, they simply are not happening often enough in many states. For instance, in Pennsylvania, advocates discovered that any individual who was receiving both Medicaid and SNAP benefits had to have benefits renewed manually by a caseworker. Other states exclude non-MAGI populations (generally individuals eligible on the basis of age or disability) from ex parte renewals, even though their circumstances are unlikely to change because many rely on a fixed income from social security benefits.
Account for information that is unlikely to change
States do not need to re-verify information that is unlikely to change. However, we have witnessed problems when states re-verify all immigration statuses at each renewal, even though the status has not changed. This caused immigrants to lose Medicaid coverage erroneously.
Some states also automatically re-verify an individual’s social security number at renewal. This can lead to mismatch errors for individuals who may have changed their name and/or gender markers if this information is different in state and Social Security Administration records and cause denials of eligibility and critical health services
Stop requiring verification to prove a negative
Some states will not automatically renew eligibility for someone who has no income, even when the data sources show no income. They needlessly require individuals to document their lack of income, which can be particularly difficult and stressful for instance for individuals who are unbanked or recently lost work.
Likewise, asset verification systems used by the Social Security Administration have put the burden on beneficiaries to prove they do not own real estate identified by the data match (which returns a match based only on first and last name). These data matches require minimal standards of accuracy and result in a disproportionate number of people of color and immigrants losing benefits due to erroneous matches. In some states, Medicaid agencies rely on SSA’s verification of assets; others use similar asset verification systems to independently verify assets.
Minimize requests for information in between scheduled renewals
Federal regulations limit renewals for MAGI populations to once every 12 months. The prior administration, however, strongly encouraged states to conduct periodic data matching in between scheduled renewals. Through this process, states check various electronic databases to evaluate whether an individual’s circumstances have changed. If the data matching reveals a possible change, states send a notice requesting that an individual explain or document the change. This process can result in frequent, unnecessary, and burdensome requests for information, for instance requiring individuals to explain a change in employer (e.g., from Walmart to Target) or explain a short-term increase in hours (e.g., a worker with extra shifts around the holidays), even when the individual’s income remains below the financial eligibility thresholds. Moreover, if the quality or age of data the state uses is poor, the resulting requests for information can be confusing, duplicative, or contradictory. These processes may result in terminations from coverage because of failure to return information or meet the documentation requirements, even though the person remained eligible.
End undeliverable mail policies
Many states terminate Medicaid coverage for individuals if a notice is returned as undeliverable. An undeliverable notice can result in termination even if the subject matter of the notice did not require an individual to take action. Predictably, these policies disproportionately impact individuals with housing instability—a population which has significantly increased during the COVID-19 pandemic, and which Congress sought to protect: federal law requires states to ensure that Medicaid coverage is “available to an eligible individual who does not reside in a permanent dwelling or does not have a fixed home or mailing address.”
Ensure rigorous oversight of automatic terminations
Finally, it is important to note that, while increasing the number of cases that can be automatically renewed to continue coverage is an important priority, automated decisions terminating coverage need more oversight. Automated decisions with no human review can be hard to stop, and in our experience disproportionately impact historically marginalized groups including immigrants, people with disabilities, and those with limited English proficiency. The harm from erroneous terminations is often compounded by flaws in the very due process protections that should protect against harms from wrongful terminations.
States and CMS should examine eligibility policies to minimize burden on enrollees and strengthen oversight to place equal attention on erroneous denials and terminations of coverage as they do on program integrity.
Additional Resources
Race-Based Prediction in Pregnancy Algorithm Is Damaging to Maternal Health
Preventing Harm from Automated Decision-Making Systems in MedicaidCommon
Demanding Ascertainable Standards: Medicaid as a Case Study
Q&A: Using Assessment Tools to Decide Medicaid Coverage
Ensuring that Assessment Tools are Available to Enrollees
Medicaid Assessments for Long-Term Supports & Services (LTSS)
Evaluating Functional Assessments for Older Adults
Opportunities for Public Comment on HCBS Assessment Tools – National Health Law Program
A Promise Unfulfilled: Automated Medicaid Eligibility Decisions