Six Months into the Medicaid Unwinding: What Do the Data Show and What Questions Remain?

Six months into the unwinding of the Medicaid continuous enrollment provision, KFF tracking shows states are reporting outcomes for over 28 million renewals, accounting for three in ten people who were enrolled as of March 2023 when continuous enrollment ended. This policy watch examines the latest data and key questions as the unwinding continues to unfold.

What do the data show?

Nationally, states have reported renewal outcomes for three in ten of all people who were enrolled in Medicaid in March 2023 and for whom states will need to redetermine eligibility during the unwinding period. KFF tracking shows that, to date, an estimated 16.0 million people have had their coverage renewed, representing 17% of all Medicaid enrollees who were enrolled at the start of the unwinding period and 8.8 million or 9% of all Medicaid enrollees have been disenrolled (Figure 1).

States started the unwinding in different months contributing to significant variation in where they are in the process of renewing coverage for all enrollees. The data show significant variation across states in the share of renewals they have processed, ranging from 69% in Arkansas to 10% in Massachusetts (Arkansas announced in early October that, as required by state law, it had completed renewals for all enrollees whose coverage had been extended because of the continuous enrollment provision but that leaves 31% of enrollees who still need to undergo a redetermination) (Figure 2). Although some of this variation is because of limited available data on renewal outcomes for some states, it also reflects that some states began processing renewals earlier than others. While some states began disenrollments in April or May, the majority of states did not start disenrolling people until June or July, including a number of states with a large share of Medicaid enrollees (like California, Illinois, New York, and Texas). But, even among states that resumed disenrollments in the same month, the pace of processing renewals differs. Some states are processing renewals more rapidly, while other states are starting more slowly and are processing fewer renewals in the first few months of unwinding.

The variation in the share of people being disenrolled across states reflects differences in both timing as well as state approaches to the unwinding. Some states have prioritized renewals of individuals they had flagged as likely ineligible. While this factor alone does not explain all the variation, it may contribute to higher disenrollment rates early in the process for some states. Another factor is how states have responded to high procedural disenrollment rates (a procedural disenrollment occurs when a renewal could not be completed to confirm ongoing eligibility). Nationally, nearly three in four of all disenrollments are due to procedural reasons. States adopted a range of strategies to help reduce procedural disenrollments and some states have paused procedural disenrollments for one or more months, while others have extended deadlines for enrollees to respond to renewal notices, which has slowed the timeframe for making final eligibility determinations in these states. KFF and other national estimates from last year anticipated an overall disenrollment rate of about 18% nationally, but also expected significant variation across states.

What questions remain?

How will mitigation strategies to address unwinding problems affect the overall pace of completing renewals and the number of people renewed and disenrolled in the months ahead? On September 21 CMS announced that 30 states would need to reinstate coverage for about 500,000 individuals (mostly children) who were erroneously disenrolled due to an issue in processing ex parte renewals. These states are also required to pause procedural disenrollments for individuals who may be affected by the problem until a mitigation strategy is in place. It is unclear how reinstatements of coverage will affect the overall numbers of renewals and disenrollments states have reported, but it is clear that pausing procedural disenrollments will affect the pace of unwinding going forward. CMS has already acknowledged that the process of completing redeterminations of all Medicaid enrollees may take longer than the anticipated 14 months, at least in some states.

What will unwinding mean for overall Medicaid enrollment? While data on disenrollments during the unwinding period tell part of the story, the net change to Medicaid enrollment will differ because total enrollment changes reflect not just people who are disenrolled but also people newly enrolling in the program as well as some people who reenroll after losing Medicaid, a process often referred to a “churn.” Consequently, monitoring overall Medicaid enrollment data and trends in conjunction with the data on unwinding renewal outcomes will provide a more complete picture of the effect of unwinding on Medicaid enrollment nationally and at the state level.

What will unwinding mean for broader coverage trends? There is also limited data about transitions in coverage and how the unwinding will affect the number of people who are uninsured. In September, data for 2022 showed the national uninsured rate had dropped to 7.9%, the lowest level on record. While the unwinding is likely to contribute to increases in the number of people who are uninsured and the uninsured rate, it is unclear what the magnitude of that increase will be. Federal survey data will not be available to understand these trends until at least one year after the unwinding period ends, and surveys that rely on self-reported health insurance status have significantly undercounted Medicaid enrollment. In the meantime, state and national administrative data on Medicaid and Marketplace enrollment can help to inform what is happening with coverage. While the data lag by several months, national Medicaid and CHIP enrollment data from June 2023 show that total enrollment declined by 1.6%, or 1.5 million people, compared to March 2023, the month in which Medicaid continuous enrollment ended. During the same time period, just over 400,000 people previously enrolled in Medicaid newly enrolled in Marketplace or Basic Health Plan coverage.