Unwinding: Understanding where we are and agreeing on where we should go

With most states entering the final stages of their unwinding efforts – and several of their peers looking back after completing their renewals – there has been a lot of assessment around successes and shortcomings. Medicaid’s federal regulator, the U.S. Department of Health and Human Services, has continued to press states to be vigilant about their processes, to learn from data and experience and to course correct, particularly when it comes to continuity of coverage for eligible children.  The press has increased its reporting on ongoing issues. And advocates have continued to illuminate the challenges of addressing longstanding, structural issues in Medicaid.

Amidst all of this, Medicaid Directors and teams across the country continue to be relentlessly focused on examining data and experience to glean learning about the many, many interventions that have been put into place – targeted member outreach, use of multiple communication modalities, partnerships with managed care organizations (MCOs) and providers and community organizations, use of federal flexibilities, and other technology solutions, hard work to improve ex parte (automatic) renewal processes and others – and to keep responding and evolving in that effort.

In order to make the most of the learning we have gained from the Congressionally-mandated unwinding effort and make the near-term and permanent program improvements that are needed to address historical phenomena such as migration of people off and on the program, we urge the many interested parties who have joined in tracking and evaluating this effort to:

1. Recognize our shared priorities;
2. Test assumptions that are being made, based on early stages of the effort; and
3. Join us in calling for federal action and research that will help states to build on the extensive investment of time, effort and resources that they have made in unwinding efforts to date, aimed toward permanent improvements in eligibility processes and customer service for those served by the program.

Our shared priorities

We continue to agree on many aspects of the Medicaid unwinding. In particular, federal and state/territory governments, MCOs, community partners, and advocates have shared goals around:

• retention of coverage for all people who remain eligible for Medicaid, with a focus on children, older adults and people with disabilities;
• smoothing the administrative processes through which Medicaid eligibility renewals are conducted, particularly for individuals with limited English proficiency;
• agile transitioning of ineligible people to other sources of coverage, including the Children’s Health Insurance Program (CHIP), Medicare, federal or state marketplace coverage, and employer-sponsored insurance; and
• supporting continuity of care for people, especially those who have conditions that require consistent, timely, and uninterrupted access to health care services.

Perhaps most notable among these is our shared concern about eligible children having lost coverage. We share the sense of urgency around illuminating all means of restoring and preserving their Medicaid coverage immediately and as all state and territory programs ready themselves for implementation of the new national standards that take effect on Jan. 1, 2023 for children that require 12-month continuous eligibility for everyone under the age of 19 in Medicaid and CHIP.

Testing our assumptions and conclusions 

Where Medicaid Directors and advocates might depart from a shared worldview centers on the assumptions and conclusions being drawn about experience with unwinding to date. From the state program view, that story is much more nuanced than some acknowledge, and is far from complete. It’s inaccurate to presume:

• that early-stage experience with unwinding should be extrapolated over the entire period of the effort;
• that everyone who is affected by a procedural termination is otherwise eligible for the program and has lost coverage through administrative error; and
• that loss of Medicaid coverage is a terminal event that necessarily results in becoming uninsured for a significant period of time.

Early-stage experience in unwinding is not predictive of the ultimate impact on Medicaid enrollment.

A number of entities have begun to make dire predictions about the ultimate net enrollment in Medicaid, with some projecting that coverage loss will be much higher than anticipated by the federal government.

Through November of this year, more than 11 million people have been disenrolled from Medicaid coverage, 71 percent for procedural reasons. In its original projections around coverage experience, the Assistant Secretary for Planning and Evaluation (ASPE) Office of Health Policy estimated that 8.2 million people served during the pandemic would no longer be eligible for Medicaid, and that a further 6.8 million would migrate off the program related to “administrative churn.” ASPE also advised that due to their incidence on the program, “children and young adults will be impacted disproportionately, with 5.3 million children and 4.7 million adults ages 18 to 34 predicted to lose Medicaid/CHIP coverage.”

It is very understandable that the number of Medicaid disenrollments has attracted attention. That said, many states staged their renewals to complete eligibility redeterminations for members who were likely ineligible based on information from administrative eligibility reviews conducted throughout the COVID-19 public health emergency, as well as third-party liability data on other sources of members’ health insurance coverage first. For that reason alone, the incidence of people who have already lost coverage is very likely to reflect high numbers of actually ineligible people. It is not appropriate, on that basis, to extrapolate rates at which those people are leaving the program in the early months to later stages of the unwinding process.

Early and static conclusions are drawn amidst a dynamic health care enrollment landscape. As of July, the overall enrollment in Medicaid and CHIP remained higher than prior to the public health emergency and there was some uptick in early summer months of new applications to the program. Also significant, albeit not determinative, is that the national uninsured rate between April and June 2023 (7.2 percent) was lower than it was December 2022 (8.3 percent), before unwinding began. This trend holds true for children 0-17 as well (4.2 percent to 3.7 percent).

We know from years of experience in Medicaid that there is necessarily a long tail associated with examining net enrollment after events that were far less significant or far-reaching than the renewal of each and every person on the program. While we share the urgency advocates and others feel around coverage loss, we must caution against drawing early conclusions that could distract programs from ongoing course corrections that can do the most good for Medicaid members.

Procedural termination does not automatically mean that coverage was lost inappropriately or that coverage won’t be regained.

We know that some procedural terminations are undoubtedly due to the fact that government processes and IT systems can be hard for members to navigate. Renewal form language can be complex and that can especially be the case for people with low literacy, a language of origin other than English or a disability. This is coupled with the fact that, despite best efforts, state system errors do occur. These are real factors in procedural terminations that state programs have readily acknowledged and worked to remedy throughout the unwinding.

But lost in much of the discussion around procedural terminations are essential pieces that are critical to note if we want to see and understand the whole picture:

1. Procedural terminations undoubtedly include people who did not respond to renewal communications because they already have other sources of health insurance. Preliminary reports from NAMD’s membership suggest rates of between 20 percent and 30 percent of people procedurally terminated have other coverage. This is based on analysis of third-party liability for these individuals.
2. It is very difficult to get a real-time view of specific, individualized rates of successful transitions to employer-sponsored health insurance or the federal and state marketplaces. While overall rates of uptake in the federal Marketplace to date have been modest, early aggregate data shows a higher than typical number of people who are transitioning from Medicaid are applying for coverage under a Qualified Health Plan. Further, we are right in the middle of the national open enrollment period and can expect these numbers to increase.
3. Because of complex life circumstances and human nature, some Medicaid members just won’t respond to notices and reminders until they or their family members need health care. At that point eligible members will be reenrolled into Medicaid. Many states are backfilling that coverage gap with retroactive coverage.

States have invested unprecedented resources into outreach to their Medicaid members (initial campaigns around updating contact information, staging of member-specific outreach around renewal packets, hard work to improve language access, historic investments in call center and website enhancements). Certainly, this hasn’t aways created the results states have wanted.  Because there are limited additional resources to invest, however, it’s critical that states understand what strategies are most effective and material to preventing loss of coverage to otherwise eligible people.

Medicaid eligibility processes are not single, point-in-time determinations that otherwise foreclose coverage.

Perhaps the most troubling aspect of national coverage of unwinding is its tendency to describe loss of Medicaid as a cliff over which people fall, without any net. This is highly problematic because it may actually chill otherwise eligible people from coming back onto the program when they again need it.

While no one wants an otherwise eligible person to lose Medicaid coverage, for all the reasons stated above, that does happen. That said, there are three important means through which peoples’ coverage can and is being restored:

• reinstatement during the 90-day reconsideration period, which many states are observing not only for parents and children, but for other coverage groups – this can include reinstating coverage back to the day a person has lost coverage as well as addressing out-of-pocket costs that they may have incurred;
• restoration of coverage through presumptive eligibility – 31 states use presumptive eligibility through providers such as hospitals and federally-qualified health centers for one or more groups (most commonly for pregnant women and children) thereby permitting real time coverage, enabling care when it is needed, and preventing gaps; and
• if no other means has enabled it, a new application because unlike private health insurance and Medicare outside of Special Enrollment Periods Medicaid supports rolling enrollment of eligible people at any point.

Because data on restoration of coverage through reconsideration isn’t a required element of the federal reporting on unwinding, it has been difficult to capture this information nationwide. However, evidence shows that Medicaid programs are restoring coverage for thousands of individuals during the reconsideration period, which continues to reduce the impacts for out-of-pocket expenses and gaps in care. For instance, in the first four months of unwinding in Connecticut, 40 percent of individuals who were disenrolled at renewal regained coverage 30-90 days later. In Arizona, the Medicaid agency reinstated coverage for more than 58,000 individuals during the reconsideration period. Notably, that state found that 39 percent of individuals who were procedurally disenrolled completed the renewal form within the reconsideration period were indeed no longer eligible for the program.

A call for further federal action as well as research, to inform the future 

While much can and should be done at the state level, Medicaid leaders would also benefit from further federal action, as well as research, to inform their work going forward.

NAMD supports more national discussion around the federal levers that could build on all the work that has been done at the individual state and territory level to improve access to the Medicaid program and to ensure continuity of coverage. This discussion should focus not only on the merits of potential new policies, but also the practical implementation and resource-related challenges that may influence state/territory uptake of options. Three non-exclusive examples of important federal levers include:

• Continuous coverage options for additional populations.  In addition to the requirement under Section 5112 of the 2023 Consolidated Appropriations Act that requires states to provide 12 months of continuous eligibility for children under the age of 19 in Medicaid and CHIP, Congress could consider permitting states to elect to extend that approach to one or more additional populations without needing waiver authority.
• Standardization of notice/application/renewal form language.  No state or territory is a proponent of dense, unreadable language in their forms. The reality, however, is that there are myriad federal requirements around rights, responsibilities and protections for members that must be embedded in notice language. Notice language has also been made more complicated as a result of litigation on behalf of members’ rights. The Centers for Medicare and Medicaid Services (CMS) has not historically released or required model notice language or required or suggested standardization of terms and processes that must be observed nationwide. There are untapped opportunities for this kind of process improvement that could be undertaken through an applied federal, state, and territory work group.
• Clear signals about which of the (e)(14) flexibilities will be permitted to remain permanent. CMS has championed the use of a range of authorities that leverage existing integrated processes and enable renewals. Many states are using these waivers, but a factor that has undoubtedly influenced interest and capacity to retain and expand them, given the significant procedural lift and expense of implementing them in state IT systems, is lack of clear signals and guidance from CMS on which will be permissible ongoing. CMS should work to indicate which authorities can be made permanent options to inform long-term planning and systems changes in state and territory Medicaid programs, as well as potentially increase the rate of adoption of e(14) strategies for the remainder of the unwinding period.

Academic researchers can also be of instrumental support to Medicaid leaders by helping them understand more about the behavioral economics involved in connecting, communicating with and engaging people served by public programs. Parallel efforts to increase voter turn-out and the incidence of responses to public surveying may also helpfully influence a better understanding of where states and territories should place emphasis in their use of limited resources.

We owe it to Medicaid members to bring our best to the work

The unwinding of a Medicaid program that had grown to cover more than 90 million Americans was never going to be easy. But we owe it to the people served by Medicaid to continue to do this hard work toward better outcomes.

We must continue to outline all of the means of retaining or reconnecting eligible Medicaid members with their coverage, whether that is under the best case of fully automatic renewal or the most challenging circumstances of rectifying an administrative error that results in coverage loss. And we should exercise care in projecting ahead for the new normal in Medicaid enrollment, when so many factors continue to be actively in play. The story of unwinding is not yet a closed book. And those of us who can influence how the final chapters are written must work together to make sure the story is one of health and wellbeing for everyone who needs the program.