MEDICAID AND CHILDREN’S HEALTH INSURANCE

Disenrollments After COVID-19 Varied Across States and Populations

Report to Congressional Committees

June 2025

GAO-25-107413

United States Government Accountability Office

View GAO-25-107413. For more information, contact Michelle B. Rosenberg at RosenbergM@gao.gov

Highlights of GAO-25-107413, a report to congressional committees

June 2025

mEDICAID AND Children’s health insurance

Disenrollments After COVID-19 Varied Across States and Populations

Why GAO Did This Study

Continuous enrollment during the COVID-19 public health emergency, which largely paused disenrollments, contributed to Medicaid and CHIP growing from 71 million enrollees in February 2020 to 94 million enrollees in March 2023—an increase of more than 30 percent.

During unwinding, some people were expected to be disenrolled either because they were no longer eligible or because of procedural reasons, such as not returning information necessary to determine their eligibility. States are required to report data on redetermination outcomes during unwinding to CMS. CMS has used the data to monitor, for example, how many people were disenrolled or had their coverage renewed. 

The CARES Act includes a provision for GAO to report on the federal response to the COVID-19 pandemic. This report describes (1) the results of CMS’s analyses of unwinding outcomes, and (2) the extent to which selected states identified differences in unwinding outcomes across population characteristics.

GAO reviewed CMS analyses, such as analyses of state-reported data on eligibility redeterminations scheduled for completion from March 2023 through June 2024, some of which were completed through September 2024. GAO also interviewed CMS officials. GAO also reviewed information from five states—Arizona, Maryland, Montana, New York, and Wisconsin—selected to capture a mix of program size and geographic diversity.

What GAO Found

During the COVID-19 public health emergency, Congress provided temporary additional federal funding to states to keep Medicaid and most Children’s Health Insurance Program (CHIP) enrollees continuously enrolled. In April 2023, states began resuming full eligibility redeterminations for the millions of enrollees who had been continuously enrolled, including disenrolling those no longer eligible or who did not submit all required information—a process known as “unwinding.” States are to complete unwinding redeterminations by the end of 2025. 

The Centers for Medicare & Medicaid Services (CMS) found that, of the 89 million completed redeterminations by states, about 27 million individuals were disenrolled during the first year and a half of unwinding. Enrollment nationwide was around 79 million as of October 2024, about 10 percent higher than prior to the pandemic. CMS also found significant variation across states in the percentage of individuals disenrolled during unwinding, with a number of factors potentially contributing to those differences.

Note: For more details, see fig. 4 in GAO-25-107413.

CMS and selected states identified certain populations that may have been disenrolled more frequently than others. For example, CMS and two states found that certain young adults were the most likely to be disenrolled when compared to other groups. CMS officials said young adults could have become ineligible because they aged out of child-specific eligibility groups, which generally allow for higher income than eligibility groups for adults who are under the age of 65 and who do not have a disability.

 

 

 

 

 

CHIP	Children’s Health Insurance Program
CMS	Centers for Medicare & Medicaid Services
MAGI	Modified adjusted gross income

June 24, 2025

Congressional Committees

During the COVID-19 public health emergency, Medicaid and the Children’s Health Insurance Program (CHIP) played a key role in providing access to health care services.[1] Medicaid and CHIP are health care programs that are jointly financed by the federal government and states. In fiscal year 2023, total federal and state spending was about $890 billion for Medicaid and about $23 billion for CHIP. Congress temporarily provided additional federal funds to state Medicaid programs so that enrollees could maintain their health care coverage during the public health emergency.[2] This provision also helped maintain enrollment in CHIP in states that operate CHIP as an expansion of Medicaid.[3] To receive the temporarily increased federal funding, states were required to keep enrollees continuously enrolled in Medicaid, including children enrolled in states’ Medicaid-expansion CHIP programs.[4] In this report, we refer to this temporary continuous enrollment condition as “continuous enrollment.”

Typically, states redetermine enrollees’ eligibility annually and disenroll them if no longer eligible or if they do not complete the process. Under continuous enrollment, however, states were to pause disenrollments except in limited circumstances. This contributed to an increase in Medicaid and CHIP enrollment from a combined 71 million enrollees in February 2020 to 94 million enrollees in March 2023—an increase of more than 30 percent.

The Consolidated Appropriations Act, 2023, ended the continuous enrollment period effective March 31, 2023, and required states to resume full eligibility redeterminations, including disenrollments.[5] This transition from continuous enrollment is known as “unwinding.” As part of unwinding, states were redetermining eligibility for, and taking action to renew or disenroll, most of their enrollees. States were given flexibility in completing unwinding redeterminations, including how quickly they completed the process. Some states were continuing to complete unwinding redeterminations as of March 2025, according to officials from the Centers for Medicare & Medicaid Services (CMS), the federal agency that oversees Medicaid and CHIP.[6]

It was expected that some individuals who were continuously enrolled during the public health emergency were no longer eligible and would therefore be disenrolled during unwinding. For example, some individuals could have obtained new employment that increased their income above income thresholds or had another change in circumstance that affected their eligibility. It was also expected that some individuals would be disenrolled for procedural reasons, such as not submitting needed information, despite remaining eligible. Additionally, certain populations may be more likely to be disenrolled, despite remaining eligible, than others. For example, the federal government estimated that about 52 percent of Latino individuals disenrolled during unwinding would still be eligible for Medicaid compared to 40 percent of White non-Latino individuals.[7] Whether individuals who were disenrolled transition to other health care coverage, churn back to Medicaid or CHIP, or become uninsured could have implications for those individuals’ health outcomes and for federal and state spending.[8]

The Consolidated Appropriations Act, 2023, required states to report data on unwinding outcomes, including renewals and disenrollments, to CMS monthly.[9] States began reporting data on unwinding outcomes to CMS beginning with redeterminations completed in March 2023.[10] The agency has used these data to monitor states’ redetermination processes and has made the monthly data and related analyses public.[11] States were not required to report outcomes by enrollee population characteristics, but some states opted to analyze and publish such data.

The CARES Act includes a provision for us to report on the federal response to the COVID-19 pandemic.[12] This report describes

1.    the results of CMS’s analyses of the outcomes of unwinding, and

2.    the extent to which selected states have identified differences in unwinding outcomes across population characteristics.

To describe the results of CMS’s analyses of the outcomes of unwinding, we reviewed analyses published from July 2023 through January 2025. This included state-reported outcomes data CMS analyzed and published as well as agency calculations based on that data, such as calculations of total disenrollments and percentages of redeterminations that resulted in renewals and disenrollments. We focused our review on CMS data updated by states and published by CMS in January 2025, the most recent updated data available at the time of our review.[13] These data include unwinding redeterminations completed from March 2023 through September 2024.[14] We used those data to determine the percentage of completed redeterminations that resulted in (1) renewals or (2) disenrollments, removing the pending cases included in CMS’s calculations. (See appendix I for more information on these data.)

We also reviewed published analyses for which CMS used other data sources, including analyses of changes in net enrollment, redetermination outcomes by eligibility and age groups, and transitions to other health care coverage after disenrollment. In addition to reviewing published analyses, we reviewed some non-public CMS analyses. These analyses were limited by available data, and as a result, reflect different time periods during unwinding. We also interviewed agency officials about factors driving variation in outcomes across states and different populations and the agency’s plans for any additional analysis.

To describe the extent to which selected states have identified differences in unwinding outcomes across population characteristics, we reviewed documents and data and interviewed Medicaid officials from five states: Arizona, Maryland, Montana, New York, and Wisconsin. We selected these states to capture a mix of program size, geographic diversity, progress in completing unwinding eligibility redeterminations, and public reporting of unwinding outcomes by population characteristics. We reviewed publicly available state data and analyses of unwinding outcomes by age, geography, race and ethnicity, and other population characteristics, when available, as well as additional data states provided to us on redeterminations, renewals, and disenrollments, when available. We used the data to determine the percentage of completed redeterminations, by population characteristics, that resulted in (1) renewals or (2) disenrollments. When reporting on state-identified differences in unwinding outcomes by population characteristics, we used the categories the states used in their data and analyses.[15]

We assessed the reliability of CMS data on unwinding outcomes by reviewing agency documentation, interviewing CMS officials, and reviewing the data for internal consistency. We assessed the reliability of outcomes data from selected states by reviewing state documentation, interviewing state officials, and reviewing the data for internal consistency. We determined that the federal and state data included in this report were sufficiently reliable for the purposes of our reporting objectives.

We conducted this performance audit from February 2024 to June 2025 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives.

Background

Medicaid and CHIP are health care programs jointly financed by the federal government and states. States receive matching funds from the federal government for their Medicaid and CHIP programs.

Medicaid and CHIP Eligibility and COVID-19

The federal government sets minimum eligibility criteria for Medicaid and CHIP, and states have flexibility to expand eligibility to cover additional individuals. As a result, eligibility criteria vary across states.

To qualify for Medicaid coverage, individuals generally must fall within certain categories or groups of individuals such as children, pregnant women, or individuals with disabilities, and meet the eligibility criteria applicable within their state. Those criteria include, for example, having income below specified levels or having significant health care needs.[16] For certain groups, such as those needing long-term services and supports, states consider both income and an individual’s assets in assessing eligibility.[17] Some individuals are eligible for full Medicaid benefits and others for limited benefits. For example, benefits for certain individuals may be limited to certain services, such as family planning, or care provided for a specific condition like tuberculosis.

To qualify for CHIP, individuals must also fall into certain groups such as children under 19 or pregnant women and meet eligibility criteria applicable to the state. This includes having income below a specified level and, for certain CHIP programs, not having other health insurance coverage. In contrast to Medicaid, which has both full and limited benefit coverage, there is no limited benefit coverage in CHIP.

Typically, states must redetermine eligibility for Medicaid enrollees on an annual basis and disenroll those who are no longer eligible. During the COVID-19 public health emergency, however, states were generally required to keep individuals continuously enrolled in Medicaid to receive a temporary increase in federal matching funds. States that operated CHIP as an expansion of Medicaid were also required to keep individuals continuously enrolled.[18] During the continuous enrollment condition, states could continue redetermining eligibility, but they could not disenroll Medicaid enrollees, including those enrolled in states’ Medicaid-expansion CHIP programs, except in limited circumstances. These limited circumstances included if the individual asked to be disenrolled, moved out of state, or died. While most states continued redetermining eligibility during the public health emergency, other states paused those activities.

Medicaid and CHIP Eligibility Redeterminations During Unwinding

During unwinding, states were generally required to redetermine the eligibility of all individuals who had been continuously enrolled during the COVID-19 public health emergency, including both those with full and limited benefit coverage. According to CMS data published in January 2025, states had completed 89 million eligibility redeterminations as part of unwinding.[19]

The redetermination process results in one of two outcomes: renewal or disenrollment, the latter of which terminates coverage (see fig. 1). States must first attempt to use available reliable information to redetermine eligibility, a process called “ex parte.” For example, this could include using income information from the state’s Supplemental Nutrition Assistance Program. If there is sufficient information for the state to renew eligibility, the state must do so. If not, the state must seek additional information from the enrollee.[20] If the person does not respond within required time frames or is otherwise determined ineligible, the state must disenroll them, which terminates coverage.[21] Disenrollment due to lack of response is referred to as a procedural disenrollment. These disenrollments could include both individuals who may still be eligible for Medicaid and CHIP as well as those who were not.

Note: Marketplaces are state and federal health insurance exchanges established through the Patient Protection and Affordable Care Act. Some individuals who purchase health insurance coverage through a marketplace may be eligible for federal premium tax credits. “Other coverage” can include, for example, employer-sponsored insurance or, if eligible, Medicare.

After disenrollment, an individual could transition to another form of health care coverage or become uninsured. For example, they could enroll in employer-sponsored coverage; get coverage through a state or federal health insurance exchange (referred to as a marketplace); or churn back to Medicaid or CHIP (if eligible) after a temporary gap in coverage.[22] Prior to disenrollment, states are required to take certain steps to assess eligibility for and transfer information on certain individuals no longer eligible for Medicaid and CHIP to certain other health programs. For example, states may transfer information to marketplaces—which allow individuals to shop for and enroll in private health coverage—to determine individuals’ eligibility for federal premium tax credits.

Redetermining Medicaid and CHIP eligibility is complex for states to administer due to the different criteria that apply for different eligibility groups and the federal requirements that the eligibility determination process must meet. To protect enrollees from erroneous termination and reduce state administrative burden when redetermining Medicaid and CHIP eligibility, CMS provided states with temporary flexibilities during unwinding. For example, upon state request, CMS provided temporary waivers authorized under section 1902(e)(14) of the Social Security Act—also referred to as e(14) waivers—of certain eligibility verification and redetermination requirements.[23] CMS approved one or more flexibilities in nearly all states. Some of the most common flexibilities related to how states verify income and asset information and update enrollees’ contact information. However, the flexibilities sought varied across states.

We previously reported that the first year of unwinding revealed a number of issues with state redetermination processes, including areas where states were not complying with federal redetermination requirements, which sometimes resulted in erroneous disenrollments.[24] At times during unwinding, states have opted to delay procedural disenrollments or been required by CMS to pause procedural disenrollments while working to come into compliance with requirements.[25]

CMS Monitoring of Outcomes

Examples of Unwinding Monthly Data Reporting Requirements for States ·       Total number of renewals ·       Number of renewals made without additional information from the enrollee (i.e., ex parte renewals) ·       Total number of enrollees disenrolled ·       Number of enrollees disenrolled for procedural reasons, such as not providing information requested by the state Source: GAO summary of Centers for Medicare & Medicaid Services information.  |  GAO‑25‑107413

The Consolidated Appropriations Act, 2023, required states to report certain eligibility redetermination outcomes to CMS starting on April 1, 2023.[26] These outcomes included, for example, the number and type of renewals and disenrollments as well as transitions to marketplace coverage. While states had previously reported data on Medicaid and CHIP applications and enrollment, they began reporting new data on redetermination outcomes. During unwinding, CMS used the newly reported outcomes data to monitor states’ redetermination efforts. In addition to using the data to inform technical assistance to states, CMS used them to perform national analyses.

CMS began posting the unwinding outcomes data publicly in July 2023. During unwinding, the agency also posted a number of national snapshots and other analyses leveraging the data reported by states. Although the act’s reporting requirements ended on June 30, 2024, CMS issued guidance in May 2024, extending certain reporting requirements indefinitely and notifying states that the agency intended to continue monitoring and publicly sharing eligibility redetermination outcome data.[27] CMS has also published several analyses using those data and other data states report to CMS, such as eligibility and enrollment data reported through its Transformed Medicaid Statistical Information System.

State Monitoring of Outcomes

States generally have discretion in how they monitor unwinding outcomes. While states must report summary data on outcomes to CMS, such as the total number of renewals, CMS did not require states to analyze or report data on outcomes by age, race and ethnicity, geographic location or other population characteristics. However, some states have opted to do so and have made such outcomes data public.

CMS Analyses Showed Enrollment Nationwide Declined During Unwinding but Remained Higher than Pre-Pandemic Levels

Although about 27 million individuals were disenrolled during the first year and a half of unwinding, enrollment nationwide remained higher than before the COVID-19 pandemic, according to analyses conducted by CMS. In addition, the percentage of individuals disenrolled varied significantly across states from 8 percent to 53 percent of those for whom redeterminations were completed. CMS’s analyses also provided some insight into how unwinding outcomes varied by age and eligibility group.

CMS Data Indicate About 27 Million Individuals Were Disenrolled During Unwinding

CMS data indicate that of the 89.0 million redeterminations completed during unwinding, 26.8 million individuals (30 percent) were disenrolled from Medicaid or CHIP from March 2023 through September 2024.[28] About 62 million individuals (70 percent of those with completed redeterminations) had their coverage renewed. The 26.8 million disenrollments represent a significant increase from the preliminary data on disenrollments reported by CMS in September 2024. At that time, the total number of disenrollments reported was 20.8 million. The difference was due, at least in part, to a larger number of pending redeterminations in the preliminary data.

Although 26.8 million individuals were disenrolled during unwinding from March 2023 through September 2024, enrollment nationwide across Medicaid and CHIP declined by only 14.7 million, from 94.0 million to 79.3 million. In addition, enrollment nationwide remained higher than pre-pandemic levels as of October 2024.[29] (See fig. 2.)

Note: Medicaid enrollment only includes enrollees with full benefit coverage. It does not include those with limited benefit coverage. Pre-unwinding refers to the month prior to when disenrollments after the continuous enrollment period could begin taking effect. Disenrollments take effect in the month after they are determined (e.g., September disenrollments take effect in October). As a result, October 2024 enrollment reflects disenrollments through September 2024.

CMS attributed the difference between total disenrollments and the decline in enrollment nationwide to several different factors. For example, officials noted that enrollment nationwide reflects new enrollees, churn back into Medicaid or CHIP, and reinstatements for individuals erroneously disenrolled during unwinding.[30] In addition, certain policy changes may have also increased enrollment during this time period. For example, North Carolina expanded Medicaid eligibility to certain low-income adults in 2023, and its enrollment increased during unwinding. The difference between total disenrollments and the decline in enrollment nationwide is also due, in part, to data differences. Specifically, unwinding disenrollment data include enrollees with limited benefit coverage, while the enrollment data do not.[31]

CMS analysis indicates that the majority of disenrollments were for procedural reasons and the majority of renewals were done via ex parte where enrollees do not need to provide any new information.[32] CMS data indicate that about 70 percent of the disenrollments during unwinding were for procedural reasons while over 60 percent of the renewals resulted from the ex parte process (see fig. 3). Because CMS was not monitoring data on redeterminations prior to unwinding, CMS officials could not speak to how these percentages compared to outcomes prior to the COVID-19 public health emergency.

Notes: Data are for Medicaid and the Children’s Health Insurance Program. These data reflect the initial outcomes (renewal or disenrollment) of unwinding redeterminations for the monthly cohorts of redeterminations that were due to be completed from March 2023 through June 2024. For each monthly cohort, the data reflect redeterminations completed in the month due or any of the 3 subsequent months. Redeterminations completed more than 3 months after the month they were due are generally not included in the data. Thus, for example, data for the March 2023 cohort includes outcomes for redeterminations completed as of June 2023 and data for the June 2024 cohort includes outcomes for redeterminations that were completed as of September 2024. This analysis is focused on completed redeterminations and excluded those reflected as pending in the data.

The data generally do not reflect changes to an outcome after a redetermination was completed, such as state decisions to reinstate coverage for those initially disenrolled. The data may reflect outcomes from two redeterminations for some enrollees—the unwinding redetermination and the routine redetermination completed later in the period, which is generally 12 months later.

Most unwinding redeterminations had been completed as of March 2025, but the exact number remaining and in which states is unknown, according to CMS officials. CMS officials said they did not know what the outcomes were for 5.7 million redeterminations that were pending in the data at the time of our review and said it was likely that at least some unwinding-related redeterminations had not yet been initiated by states as of March 2025. States have until the end of December 2025 to complete all unwinding redeterminations.

Percentage of Individuals Disenrolled Varied Significantly Across States, Ranging from 8 Percent to 53 Percent of Completed Redeterminations

CMS data indicate that the percentage of completed unwinding redeterminations that resulted in a disenrollment from March 2023 through September 2024 ranged from 8 percent in Maine to 53 percent in Montana (see fig. 4). Five states disenrolled more than 50 percent of individuals from Medicaid or CHIP during this period. Six states disenrolled less than 20 percent of individuals from Medicaid or CHIP. (See appendix I for disenrollments, renewals, and pending redeterminations by state.)

Notes: Data are for Medicaid and the Children’s Health Insurance Program. They reflect the initial outcomes (renewal or disenrollment) of unwinding redeterminations for the monthly cohorts of redeterminations that were due to be completed from March 2023 through June 2024. For each monthly cohort, the data reflect redeterminations completed in the month due or any of the 3 subsequent months. Redeterminations completed more than 3 months after the month they were due are generally not included in the data. Thus, for example, data for the March 2023 cohort includes outcomes for redeterminations completed as of June 2023 and data for the June 2024 cohort includes outcomes for redeterminations that were completed as of September 2024. This analysis is focused on completed redeterminations and excluded those reflected as pending in the data.

The data generally do not reflect changes to an outcome after a redetermination was completed, such as state decisions to reinstate coverage for those initially disenrolled. However, Arkansas, California, Oklahoma, Pennsylvania, Rhode Island, Wyoming, and South Carolina included data for some months on eligibility decisions that occurred after the initial redetermination, such as reinstatements of coverage for individuals that were initially disenrolled. The data may reflect outcomes from two redeterminations for some enrollees—the unwinding redetermination and the routine redetermination completed later in the period, which is generally 12 months later.

CMS officials and early CMS analyses indicated several factors may have contributed to the wide variation in the percentage disenrolled across states during unwinding. For example,

Flexibilities and mitigation measures. CMS officials said that state implementation of temporary flexibilities and mitigation measures that aimed to prevent erroneous disenrollments may have contributed to the variation in the percentage of individuals disenrolled across states. States adopted flexibilities during unwinding that made it easier to renew individuals and may have led to fewer disenrollments, according to CMS.[33] For example, 38 states requested and received approval to use older income information that demonstrated low or no income to renew a Medicaid or CHIP enrollee’s eligibility, when no other income information was listed on relevant data sources.[34] This streamlined the redetermination process for these states and enrollees and may have led to fewer disenrollments. In addition, some states implemented mitigation measures to help them address areas of non-compliance with federal redetermination requirements, which may have led to fewer disenrollments, according to CMS officials. For example, Maine paused procedural disenrollments from August 2023 through August 2024. As a result, Maine’s results for this time period mostly reflect individuals who were determined eligible or were determined ineligible. The results do not reflect most individuals whose eligibility could not be determined because, for example, they did not return required information. This may partially explain why Maine had the lowest overall percentage of disenrolled (8 percent), according to CMS officials.

Ex parte renewals. CMS officials indicated that the extent to which states used ex parte renewals may have contributed to some of the state variation in overall disenrollments. Because ex parte renewals require no action from the enrollee to retain coverage, states with relatively higher ex parte renewals may have fewer disenrollments, according to CMS officials. States varied in their percentage of unwinding renewals that were completed through the ex parte process, from 7 percent to almost 100 percent from March 2023 through September 2024. CMS data show that states with the lowest percentage of enrollees disenrolled often had relatively higher percentages of renewals completed through the ex parte process. For example, out of the six states with the lowest percentage of disenrollments, five had ex parte renewals exceeding 65 percent of their total renewals.[35]

Differences in eligibility criteria. Early CMS analysis indicated that states that did not expand Medicaid eligibility to certain low-income individuals—referred to as non-expansion states—generally had a higher percentage of disenrollments than states that did. Our analysis of disenrollments found similar results. For example, during unwinding, the percentage of individuals disenrolled from Medicaid or CHIP averaged 36 percent in non-expansion states (10 states) compared to 30 percent in expansion states (41 states), from March 2023 through September 2024.[36] States that did not expand Medicaid also had a disproportionate share of total disenrollments. These states accounted for 22 percent of completed unwinding redeterminations but almost 30 percent of all disenrollments during this time period.

Although states frequently used procedural disenrollments, CMS officials said they did not know how they were associated with variation among states in the overall percentage of individuals disenrolled from Medicaid or CHIP during unwinding. The percentage of unwinding disenrollments that were procedural exceeded 50 percent in 46 states from March 2023 through September 2024. In addition, more than 75 percent of disenrollments were procedural in 19 states, including in several states with relatively lower percentages of disenrollments overall. For example, North Carolina disenrolled 13 percent of individuals with a completed redetermination, but the percentage of disenrollments that were procedural was 85 percent.

CMS officials told us they used the data on renewals and disenrollments to monitor state compliance with redetermination requirements and to provide technical assistance to states on their redetermination processes. For example, CMS officials told us they reviewed data on renewals and disenrollments weekly to identify state outliers and determine if CMS needed additional information from the state before escalating an issue internally. In some cases, this data monitoring enabled CMS to find and address problems with states’ redetermination processes. CMS officials plan to continue collecting renewals and disenrollment data from states after unwinding is complete, due to the data’s importance in helping the agency monitor and oversee states’ redetermination processes, which they outlined in May 2024 guidance. This is consistent with our July 2024 recommendation to CMS on documenting and implementing oversight practices the agency learned during unwinding were needed to help ensure states’ compliance issues with redetermination requirements.[37]

CMS Analyses Provide Some Insight into Unwinding Outcomes by Age and Eligibility Group and Transitions to Other Health Care Coverage

CMS analyses of unwinding outcomes by age and eligibility group suggest that certain populations were more likely to be disenrolled. CMS indicated that these analyses were limited by what data were available and the time period reviewed.

·         Adult and child. CMS analyzed differences in the percentage decline in Medicaid and CHIP enrollment from March 2023 through May 2024 for adults compared to children. CMS found that the decline in enrollment was about 4 percentage points more for adults during this time period.[38] CMS officials said this may partly be a result of the higher eligibility income limits for children compared to adults.[39]

·         Eligibility group. CMS analyzed the percentage of individuals who were disenrolled from Medicaid or CHIP or transitioned to limited benefit coverage for different eligibility groups from April 2023 through December 2023.[40] This analysis suggests that young adults aged 19 to 23 in a child-eligibility group were the most likely to be disenrolled or transition to limited benefit coverage.[41] Specifically, over 50 percent of these young adults were disenrolled from full benefit Medicaid or CHIP or moved to a program with limited benefits.[42] Enrollees in a pregnancy-related eligibility group were the second most likely to be disenrolled or transition to limited benefit coverage (39 percent). In contrast, enrollees in a disability-specific eligibility group were the least likely (7 percent) to be disenrolled or transition to limited benefit coverage.

According to CMS, populations with relatively higher disenrollments may have moved into eligibility categories with lower income limits and as a result, likely were no longer eligible for Medicaid or CHIP coverage. For example, CMS officials said that young adults aged 19 to 23 may have aged out of child-specific eligibility groups, and the income limits for certain adult eligibility groups generally are lower than for children.[43] In addition, in states that did not expand eligibility to certain low-income adults, these individuals may no longer fall into a category that is eligible for Medicaid, according to CMS officials.

·         Home- and community-based services enrollees. CMS also analyzed the percentage of individuals who had been receiving home- and community-based services who were disenrolled from Medicaid or CHIP or transitioned to limited benefit coverage.[44] Specifically, CMS compared disenrollments and transitions to limited benefit coverage for those who had been receiving home- and community-based services to the general Medicaid and CHIP population for 36 states from April 2023 through December 2023.[45] This analysis suggests that individuals who had been receiving home- and community-based services may have had relatively fewer disenrollments when compared to other individuals. Specifically, 3 percent of individuals who had been receiving home- and community-based services were disenrolled from full benefit Medicaid or CHIP coverage or moved to limited benefit coverage.[46] This is significantly lower than the 18 percent who experienced those outcomes for the general Medicaid and CHIP population.

CMS officials said several factors may affect the relatively fewer disenrollments for home- and community-based services enrollees. For example, the eligibility factors for these individuals, such as their disability status, are less likely to change than the eligibility factors for other groups of enrollees. Individuals receiving home- and community-based services are also more likely to need and have help with renewal forms, which may reduce their likelihood of being disenrolled, according to officials.

CMS analyses of the extent to which those disenrolled during unwinding obtained other health care coverage found that certain populations may be more likely to transition to other coverage or return to Medicaid or CHIP. CMS officials said these analyses only give a partial picture on what happened to individuals after they were disenrolled from Medicaid or CHIP due to gaps in the data on transitions to other coverage.[47]

·         Transitions to marketplace coverage. From April 2023 through June 2024, CMS found that at least 2.7 million individuals disenrolled from Medicaid or CHIP during unwinding obtained health care coverage in states using the federal marketplace platform.[48] In addition, at least 830,000 individuals obtained health care coverage through state-based marketplaces using their own platforms.[49] CMS also analyzed enrollees who were disenrolled from April 2023 through December 2023 who transitioned to marketplace coverage by certain eligibility groups in states using the federal marketplace platform.[50] CMS found that certain groups were more likely to transition to marketplace coverage than others after disenrollment in these states. Specifically, among enrollees who were disenrolled from full benefit Medicaid or CHIP coverage, adults aged 19 to 64 and not in a disability-specific eligibility group were the most likely to obtain marketplace coverage (at least 14 percent). In contrast, enrollees over 65 and not in a disability-specific eligibility group were the least likely to obtain such coverage (1 percent).

CMS officials told us these differences may be related to who may be eligible for premium tax credits for marketplace coverage. For example. according to CMS analysis, enrollees over 65 and those in a disability-specific eligibility group often qualify for Medicare. As a result, CMS officials said they would often not be eligible for premium tax credits and may be less likely to transition to marketplace coverage compared to other groups.

·         Churn back into Medicaid or CHIP. CMS found that about 14 percent of individuals who were disenrolled from full benefit coverage from April 2023 through December 2023, churned back into Medicaid or CHIP by March 2024.[51] CMS also analyzed the level of churn by different eligibility groups and found children and those in a pregnancy-related eligibility group were the most likely to experience churn.[52] Specifically, over 15 percent of children under 19 and those enrolled in a pregnancy-related eligibility group who were disenrolled from March 2023 through December 2023 churned back into full benefit Medicaid or CHIP by March 2024. In contrast, enrollees over 65 and not in a disability-specific eligibility group were the least likely to churn back into full benefit Medicaid or CHIP (1 percent).

CMS officials noted that children may have been more likely to churn back into Medicaid or CHIP due to generally higher income limits for eligibility, which is also consistent with a past CMS analysis. Enrollees in a pregnancy-related eligibility group may have been more likely to churn back into Medicaid or CHIP due to states reinstating or reenrolling these individuals as a result of implementation of the 12-month postpartum coverage option, according to CMS.[53]

As of March 2025, CMS officials noted they are interested in doing more analysis on how unwinding outcomes are distributed across different age and Medicaid eligibility groups.

Selected States Found Relatively Higher Disenrollments for Certain Age, Race, and Other Population Groups, Though Findings Varied by State

Our five selected states found relatively higher Medicaid and CHIP disenrollments for certain population groups, though findings varied by state. States were not required to analyze unwinding outcomes, including disenrollments, by population characteristics—and the five states varied in the extent to which they did so (see table 1). (See appendix II for additional information on disenrollments by population characteristics for the five selected states.)

State	Age	Geographic location	Race and ethnicity	Language	Tribal affiliation
Arizona	✔	✔	✔	—	—
Maryland	✔	✔	✔	—	—
Montana	✔	✔	—	—	✔
New York	✔	✔	✔	✔	—
Wisconsin	✔	✔	✔	✔	—

Legend:

CHIP: Children’s Health Insurance Program

✔: State analyzed disenrollments by characteristic

Source: GAO analysis of state information.  |  GAO‑25‑107413

Differences in Disenrollments by Age

Information from all five selected states indicated relatively higher disenrollments among enrollees of certain ages. For example, in the two states that analyzed disenrollments for young adults aged 19 to 20 (Maryland and Montana), state data indicated that a higher percentage of enrollees were disenrolled in this age group than in all other age groups. Figure 5 shows disenrollments by age group in one of the states. Officials from this state said they expected young adults to experience higher disenrollments than children because of the different eligibility requirements for these groups in the state.[54]

Notes: “Percentage disenrolled” is the percentage of Medicaid and Children’s Health Insurance Program enrollees with completed redeterminations that resulted in disenrollment as of June 2024. At that time, the state had not completed about 8 percent of redeterminations due, including redeterminations for 13 percent of enrollees under age 1, 19 percent of those aged 65 or older, and 5 to 8 percent of enrollees in other age groups. The state data used to calculate percentages reflect the initial redetermination decisions made during unwinding and do not reflect changes in outcomes that may have occurred during the state’s 90-day reconsideration period after a procedural disenrollment. The data include enrollees who went through the state’s standard eligibility redetermination process; other enrollees, such as those Medicaid-eligible due to foster care status, are not included.

Similarly, data from three states (Arizona, Maryland, and Montana) indicated that older age groups (adults) had relatively higher disenrollments than children aged 18 and under. For example, in Maryland, 26 percent of enrollees aged 21 to 44 with completed redeterminations were disenrolled as compared to 19 percent of children aged 5 and under.[55] Similarly, in Arizona, adults aged 19 and older were a larger share of disenrollments when compared to their overall share of the enrollee population. Specifically, adults accounted for 66 percent of disenrollments during unwinding but were 58 percent of the state’s total Medicaid and CHIP population in May 2024.[56] Arizona and Montana officials said they expected relatively higher disenrollments among adults than among children aged 18 and under because of generally lower income eligibility limits for adults in these states.

In one selected state, officials told us that age differences in disenrollments had informed their outreach efforts. Specifically, after Wisconsin found that a relatively high percentage of adults aged 19 to 35 were disenrolled, they prioritized new outreach efforts to help young adults retain coverage, according to state officials.

Differences in Disenrollments by Geographic Location

Information from all five selected states indicated relatively higher disenrollments in certain zip codes, counties, or regions. For example,

·         Maryland. State data indicated that the suburban Washington, D.C. and western Maryland regions had the highest percentage of completed redeterminations that resulted in disenrollments (25 percent and 24 percent, respectively), while Baltimore City had the lowest percentage of disenrollments (19 percent).[57]

Geographic differences in disenrollments during unwinding led the state to increase outreach in certain areas, officials said. To better understand local disenrollment differences, officials analyzed demographic data for the 60 zip codes with the highest percentage of redeterminations that resulted in procedural disenrollment. In doing so, officials said they analyzed procedural disenrollments by race and ethnicity in those zip codes, which allowed the state to identify communities with relatively higher procedural disenrollments and provide targeted outreach. For example, after identifying relatively higher procedural disenrollments in parts of suburban Washington, D.C., particularly among Hispanic or Latino individuals, state officials increased outreach in those areas via Spanish language radio.

·         New York. Available data indicated that New York City and two other regions had the highest percentage of completed redeterminations that resulted in disenrollment (30 percent).[58] Officials said they had seen higher-than-expected percentages of procedural disenrollments in New York City, and the state targeted outreach in that area to help individuals renew their coverage.

Differences in Disenrollments by Race and Ethnicity

Of the four selected states that analyzed outcomes by race and ethnicity, data from Maryland and Wisconsin indicated relatively higher disenrollments for certain racial and ethnic groups. Data limitations in the two other states—Arizona and New York—made comparisons challenging.

·         Maryland. State data across 10 categories related to race and ethnicity indicated that individuals who were Pacific Islander, White, or of “other” or unknown race and ethnicity had the highest percentage of completed redeterminations that resulted in disenrollment (about 25 percent or more). In comparison, individuals who were American Indian or Alaska Native, both Black and White, or two or more other races had the lowest percentages of completed redeterminations that resulted in disenrollment (less than 20 percent).[59]

State data also indicated that individuals who were Black or African American; both Black and White; or of unknown race and ethnicity had the largest percentage of disenrollments that were procedural. Specifically, for each of these groups, more than 64 percent of disenrollments were procedural. In comparison, individuals who were Asian, Pacific Islander, White, or of “other” race and ethnicity had the lowest percentage of procedural disenrollments (less than 58 percent of disenrollments).

·         Wisconsin. State data across eight categories related to race and ethnicity indicated that Native Hawaiian or Pacific Islander, White, and Hispanic or Latino individuals had the highest percentages of completed redeterminations that resulted in disenrollment.[60] (See fig. 6.) State data also indicated that American Indian or Alaskan Native, Hispanic or Latino, and Native Hawaiian or Pacific Islander individuals had the largest percentage of disenrollments that were procedural (about 67 percent).

Notes: “Percentage disenrolled” is the percentage of Medicaid and Children’s Health Insurance Program enrollees with completed redeterminations that resulted in disenrollment. Data are based on enrollees’ self-reported race and ethnicity and reflect the categories the state used in its analysis. Data reflect redeterminations completed from June 2023 through June 2024, which account for all redeterminations due during unwinding. Data also reflect changes in outcomes that occurred during the state’s 90-day reconsideration period after a procedural disenrollment. The state data used to calculate percentages include enrollees who went through the state’s standard eligibility redetermination process; other enrollees, such as those Medicaid eligible due to foster care status, are not included.

^aEnrollees could choose not to report their race. Enrollees who did not report their race accounted for 12 percent of unwinding redeterminations.

Wisconsin officials told us they created an unwinding outreach dashboard to share outcomes data by race, primary language, and other population characteristics with partner organizations, such as provider groups. Feedback on the data from those organizations helped the state adapt outreach to certain populations.

Differences in Disenrollments by Other Population Characteristics

State data also indicated relatively higher disenrollments by other population characteristics. For example,

·         Language. Two selected states (New York and Wisconsin) analyzed disenrollments by enrollees’ language. For example, in New York, available data across 28 written language categories, including English, indicated that enrollees who selected French Creole, Nepali, or Japanese as their preferred written language had the highest percentage of completed redeterminations that resulted in disenrollment (more than 33 percent). In comparison, enrollees who selected Arabic, Swahili, or Somali as their preferred written language were among those with the lowest percentage of redeterminations that resulted in disenrollment (16 to 18 percent).[61] Individuals must select their preferred written language when applying for coverage through the eligibility system that manages most of the states’ Medicaid and CHIP enrollees.[62]

·         Tribal affiliation. One of our selected states—Montana—analyzed disenrollments by tribal affiliation. State data indicated relatively more disenrollments among enrollees who did not report being tribally affiliated than those who did. Forty-six percent of non-tribally affiliated enrollees with a completed redetermination had been disenrolled as of June 2024, as compared to 36 percent of tribally affiliated enrollees with a completed redetermination.[63] State officials said the relatively lower disenrollments among tribally affiliated enrollees could reflect many of these enrollees tending to live in areas with fewer economic opportunities. In addition, state officials worked with the federal Indian Health Service and tribal groups to help tribally affiliated enrollees retain coverage, if eligible. They did so by identifying which enrollees were at risk of disenrollment due to not returning renewal forms and then sharing that information with those external groups to facilitate outreach, state officials said.

Agency Comments

We provided a draft of this report to the Department of Health and Human Services for review and comment. The agency provided technical comments, which we incorporated as appropriate.

We are sending copies of this report to the appropriate congressional committees, the Secretary of Health and Human Services, and other interested parties. In addition, the report is available at no charge on the GAO website at http://www.gao.gov.

If you or your staff have any questions about this report, please contact me at RosenbergM@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made key contributions to this report are listed in appendix III.

Michelle B. Rosenberg
Director, Health Care

List of Committees

The Honorable Susan Collins
Chair
The Honorable Patty Murray
Vice Chair
Committee on Appropriations
United States Senate

The Honorable Mike Crapo
Chairman
The Honorable Ron Wyden
Ranking Member
Committee on Finance
United States Senate

The Honorable Bill Cassidy, M.D.
Chair
The Honorable Bernard Sanders
Ranking Member
Committee on Health, Education, Labor, and Pensions
United States Senate

The Honorable Rand Paul, M.D.
Chairman
The Honorable Gary C. Peters
Ranking Member
Committee on Homeland Security and Governmental Affairs
United States Senate

The Honorable Tom Cole
Chairman
The Honorable Rosa L. DeLauro
Ranking Member
Committee on Appropriations
House of Representatives

The Honorable Brett Guthrie
Chair
The Honorable Frank Pallone, Jr.
Ranking Member
Committee on Energy and Commerce
House of Representatives

The Honorable Mark E. Green, M.D.
Chairman
The Honorable Bennie G. Thompson
Ranking Member
Committee on Homeland Security
House of Representatives

The Honorable James Comer
Chairman

The Honorable Stephen F. Lynch
Acting Ranking Member
Committee on Oversight and Government Reform
House of Representatives

The Honorable Jason Smith
Chairman
The Honorable Richard Neal
Ranking Member
Committee on Ways and Means
House of Representatives

States reported certain unwinding outcomes—such as the number and type of renewals and disenrollments—for Medicaid and Children’s Health Insurance Program (CHIP) enrollees to the Centers for Medicare & Medicaid Services (CMS) beginning with redeterminations completed in March 2023. CMS has publicly shared preliminary data reported by states as well as updated data, which has generally lagged the preliminary data by five or more months. The preliminary data include millions more pending redeterminations—redeterminations that were incomplete and still being processed—compared to the updated data. We are reporting on numbers and percentages using the updated data that were published in January 2025.

These data reflect the outcomes for monthly cohorts of redeterminations due for completion from March 2023 through June 2024. For each monthly cohort, the updated data generally include redeterminations completed in the month due or any of the 3 subsequent months. Thus, data for the March 2023 cohort includes outcomes for redeterminations completed as of June 2023 and data for the June 2024 cohort includes outcomes for redeterminations that were completed as of September 2024. Redeterminations completed more than 3 months after the month they were due are generally reflected as pending in the data and not included in our calculations of the percentage of disenrollments or renewals. In addition, the data generally do not reflect changes to an outcome after a redetermination was completed, such as state decisions to reinstate coverage for those initially disenrolled. Finally, the data may reflect outcomes from two redeterminations for some enrollees—the unwinding redetermination and the routine redetermination completed later in the period, which is generally 12 months later.

CMS data published in January 2025 included the number of completed redeterminations as well as how many enrollees were renewed and disenrolled from March 2023 through September 2024 (see table 2).

	Number of Medicaid and CHIP enrollees				Percentage of all completed redeterminations
State	Completed redeterminations	Renewed	Disenrolled		Renewed		Disenrolled
Alabama	1,324,475	886,970	437,505		67.0%		33.0%
Alaska	145,461	95,930	49,531		65.9%		34.1%
Arizona	2,912,469	2,318,188	594,281		79.6%		20.4%
Arkansas	1,036,666	661,734	374,932		63.8%	36.2%
California	11,660,941	9,539,870	2,121,071		81.8%	18.2%
Colorado	1,773,941	964,368	809,573		54.4%	45.6%
Connecticut	1,389,391	1,141,133	248,258		82.1%	17.9%
Delaware	284,628	200,884	83,744		70.6%	29.4%
District of Columbia	310,421	237,222	73,199		76.4%	23.6%
Florida	5,474,605	3,433,559	2,041,046		62.7%	37.3%
Georgia	2,258,892	1,310,900	947,992		58.0%	42.0%
Hawaii	454,399	357,597	96,802		78.7%	21.3%
Idaho	445,744	221,774	223,970		49.8%	50.2%
Illinois	3,481,913	2,609,772	872,141		75.0%	25.0%
Indiana	1,970,524	1,438,297	532,227		73.0%	27.0%
Iowa	992,736	690,750	301,986		69.6%	30.4%
Kansas	481,856	312,537	169,319		64.9%	35.1%
Kentucky	966,082	725,384	240,698		75.1%	24.9%
Louisiana	1,973,454	1,298,656	674,798		65.8%	34.2%
Maine	408,443	375,230	33,213		91.9%	8.1%
Maryland	1,702,861	1,277,379	425,482		75.0%	25.0%
Massachusetts	1,699,566	1,010,933	688,633		59.5%	40.5%
Michigan	3,091,089	1,970,952	1,120,137		63.8%	36.2%
Minnesota	1,438,010	983,378	454,632		68.4%	31.6%
Mississippi	684,397	524,124	160,273		76.6%	23.4%
Missouri	1,221,511	860,231	361,280		70.4%	29.6%
Montana	274,759	128,222	146,537		46.7%	53.3%
Nebraska	393,078	274,577	118,501		69.9%	30.1%
Nevada	1,107,978	718,230	389,748		64.8%	35.2%
New Hampshire	227,042	156,195	70,847		68.8%	31.2%
New Jersey	1,735,638	960,401	775,237		55.3%	44.7%
New Mexico	865,705	631,925	233,780		73.0%	27.0%
New York	6,744,956	4,716,821	2,028,135		69.9%	30.1%
North Carolina	2,560,336	2,229,624	330,712		87.1%	12.9%
North Dakota	157,682	90,795	66,887		57.6%	42.4%
Ohio	4,126,025	3,154,553	971,472		76.5%	23.5%
Oklahoma	1,490,558	724,683	765,875		48.6%	51.4%
Oregon	1,345,402	1,149,623	195,779		85.4%	14.6%
Pennsylvania	3,771,362	2,733,324	1,038,038		72.5%	27.5%
Rhode Island	317,371	225,032	92,339		70.9%	29.1%
South Carolina	1,195,123	760,613	434,510		63.6%	36.4%
South Dakota	125,106	62,828	62,278		50.2%	49.8%
Tennessee	1,510,063	899,112	610,951		59.5%	40.5%
Texas	4,891,717	2,322,485	2,569,232		47.5%	52.5%
Utah	471,555	221,327	250,228		46.9%	53.1%
Vermont	168,969	124,137	44,832		73.5%	26.5%
Virginia	2,083,864	1,808,974	274,890		86.8%	13.2%
Washington	1,946,119	1,445,212	500,907		74.3%	25.7%
West Virginia	575,604	346,508	229,096		60.2%	39.8%
Wisconsin	1,291,994	815,987	476,007		63.2%	36.8%
Wyoming	71,739	55,044	16,695		76.7%	23.3%
Total	89,034,220	62,203,984	26,830,236		69.9%	30.1%

Source: GAO analysis of Centers for Medicare & Medicaid Services (CMS) data.  |  GAO‑25‑107413

Notes: Data are for Medicaid and the Children’s Health Insurance Program. They reflect the initial outcomes (renewal or disenrollment) of unwinding redeterminations for the monthly cohorts of redeterminations that were due to be completed from March 2023 through June 2024. For each monthly cohort, the data reflect redeterminations completed in the month due or any of the 3 subsequent months. Redeterminations completed more than 3 months after the month they were due are generally not included in the data. Thus, for example, data for the March 2023 cohort includes outcomes for redeterminations completed as of June 2023 and data for the June 2024 cohort includes outcomes for redeterminations that were completed as of September 2024. This analysis is focused on completed redeterminations and excluded those reflected as pending in the data.

The data generally do not reflect changes to an outcome after a redetermination was completed, such as state decisions to reinstate coverage for those initially disenrolled. However, Arkansas, California, Oklahoma, Pennsylvania, Rhode Island, Wyoming, and South Carolina included data for some months on eligibility decisions that occurred after the initial redetermination, such as reinstatements of coverage for individuals that were initially disenrolled. The data may reflect outcomes from two redeterminations for some enrollees—the unwinding redetermination and the routine redetermination completed later in the period, which is generally 12 months later.

CMS also published state data on the number of renewals completed without the state requesting additional information from the enrollee, i.e., through the “ex parte” process, and the number of disenrollments that were based on procedural reasons, such as enrollees not providing requested information to state Medicaid agencies (see table 3).

	Percentage of all renewals			Percentage of all disenrollments
State	Renewed based on new information  from enrollee	Renewed via ex partea		Determined ineligible	Disenrolled for  procedural reasonsb
Alabama	44.6%	55.4%		14.9%	85.1%
Alaska	38.3%	61.7%		50.5%	49.5%
Arizona	10.8%	89.2%		32.2%	67.8%
Arkansas	27.5%	72.5%		24.1%	75.9%
California	26.2%	73.8%		25.9%	74.1%
Colorado	51.6%	48.4%		35.0%	65.0%
Connecticut	18.8%	81.2%		30.8%	69.2%
Delaware	53.7%	46.3%		38.1%	61.9%
District of Columbia	20.3%	79.7%		10.3%	89.7%
Florida	66.1%	33.9%		36.2%	63.8%
Georgia	31.0%	69.0%		30.7%	69.3%
Hawaii	28.7%	71.3%		21.1%	78.9%
Idaho	48.9%	51.1%		24.7%	75.3%
Illinois	24.8%	75.2%		28.3%	71.7%
Indiana	28.1%	71.9%		23.8%	76.2%
Iowa	48.7%	51.3%		27.7%	72.3%
Kansas	56.6%	43.4%		40.0%	60.0%
Kentucky	14.9%	85.1%		44.5%	55.5%
Louisiana	20.3%	79.7%		29.0%	71.0%
Maine	72.6%	27.4%		81.3%	18.7%
Maryland	25.9%	74.1%		25.0%	75.0%
Massachusetts	34.4%	65.6%		33.3%	66.7%
Michigan	41.8%	58.2%		19.6%	80.4%
Minnesota	43.4%	56.6%		21.7%	78.3%
Mississippi	70.1%	29.9%		26.2%	73.8%
Missouri	20.9%	79.1%		19.7%	80.3%
Montana	50.4%	49.6%		31.9%	68.1%
Nebraska	46.8%	53.2%		45.7%	54.3%
Nevada	26.1%	73.9%		7.7%	92.3%
New Hampshire	24.1%	75.9%		19.8%	80.2%
New Jersey	56.9%	43.1%		26.4%	73.6%
New Mexico	39.2%	60.8%		17.0%	83.0%
New York	48.1%	51.9%		53.6%	46.4%
North Carolina	0.5%	99.5%		15.3%	84.7%
North Dakota	52.8%	47.2%		30.9%	69.1%
Ohio	31.8%	68.2%		30.1%	69.9%
Oklahoma	66.4%	33.6%		18.3%	81.7%
Oregon	19.6%	80.4%		54.5%	45.5%
Pennsylvania	92.5%	7.5%		47.0%	53.0%
Rhode Island	9.0%	91.0%		24.7%	75.3%
South Carolina	33.8%	66.2%		34.6%	65.4%
South Dakota	73.9%	26.1%		51.5%	48.5%
Tennessee	29.5%	70.5%		24.4%	75.6%
Texas	88.6%	11.4%		32.1%	67.9%
Utah	50.9%	49.1%		10.0%	90.0%
Vermont	25.2%	74.8%		28.7%	71.3%
Virginia	33.3%	66.7%		45.9%	54.1%
Washington	11.2%	88.8%		15.4%	84.6%
West Virginia	63.2%	36.8%		24.3%	75.7%
Wisconsin	61.1%	38.9%		42.9%	57.1%
Wyoming	46.8%	53.2%		25.7%	74.3%
Total	38.4%	61.6%		31.0%	69.0%

Source: GAO analysis of Centers for Medicare & Medicaid Services (CMS) data.  |  GAO‑25‑107413

Notes: Data are for Medicaid and the Children’s Health Insurance Program. They reflect the initial outcomes (renewal or disenrollment) of unwinding redeterminations for the monthly cohorts of redeterminations that were due to be completed from March 2023 through June 2024. For each monthly cohort, the data reflect redeterminations completed in the month due or any of the 3 subsequent months. Redeterminations completed more than 3 months after the month they were due are generally not included in the data. Thus, for example, data for the March 2023 cohort includes outcomes for redeterminations completed as of June 2023 and data for the June 2024 cohort includes outcomes for redeterminations that were completed as of September 2024. This analysis is focused on completed redeterminations and excluded those reflected as pending in the data.

The data generally do not reflect changes to an outcome after a redetermination was completed, such as state decisions to reinstate coverage for those initially disenrolled. However, Arkansas, California, Oklahoma, Pennsylvania, Rhode Island, Wyoming, and South Carolina included data for some months on eligibility decisions that occurred after the initial redetermination, such as reinstatements of coverage for individuals that were initially disenrolled. The data may reflect outcomes from two redeterminations for some enrollees—the unwinding redetermination and the routine redetermination completed later in the period, which is generally 12 months later.

^aEx parte renewals are renewals that are done without additional information from the enrollee.

^bProcedural reasons for being disenrolled include, for example, not providing all information needed to redetermine eligibility.

Most states still had pending redeterminations for those due from March 2023 through June 2024 (see table 4). These states had initiated these redeterminations but had not completed them in the month they were due or generally in any of the subsequent 3 months, according to updated data that CMS published in January 2025.

	Number of Medicaid and CHIP enrollees
State	Redeterminations due through June 2024a	Pending redeterminationsb		Percentage of pending redeterminations out of all due
Alabama	1,325,451	976		0.1%
Alaska	199,929	54,468		27.2%
Arizona	2,913,465	996		0.0%
Arkansas	1,049,432	12,766		1.2%
California	13,796,970	2,136,029		15.5%
Colorado	1,786,893	12,952		0.7%
Connecticut	1,426,741	37,350		2.6%
Delaware	298,799	14,171		4.7%
District of Columbia	330,549	20,128		6.1%
Florida	5,476,033	1,428		0.0%
Georgia	2,337,840	78,948		3.4%
Hawaii	474,041	19,642		4.1%
Idaho	445,744	0		0.0%
Illinois	3,660,815	178,902		4.9%
Indiana	2,141,471	170,947		8.0%
Iowa	1,036,955	44,219		4.3%
Kansas	609,614	127,758		21.0%
Kentucky	1,168,102	202,020		17.3%
Louisiana	1,976,376	2,922		0.1%
Maine	466,494	58,051		12.4%
Maryland	1,743,727	40,866		2.3%
Massachusetts	1,701,833	2,267		0.1%
Michigan	3,099,634	8,545		0.3%
Minnesota	1,441,999	3,989		0.3%
Mississippi	748,715	64,318		8.6%
Missouri	1,455,929	234,418		16.1%
Montana	336,470	61,711		18.3%
Nebraska	445,700	52,622		11.8%
Nevada	1,107,978	0		0.0%
New Hampshire	227,072	30		0.0%
New Jersey	2,153,190	417,552		19.4%
New Mexico	919,411	53,706		5.8%
New York	6,856,535	111,579		1.6%
North Carolina	2,973,021	412,685		13.9%
North Dakota	158,065	383		0.2%
Ohio	4,229,596	103,571		2.4%
Oklahoma	1,504,474	13,916		0.9%
Oregon	1,352,565	7,163		0.5%
Pennsylvania	3,787,211	15,849		0.4%
Rhode Island	325,826	8,455		2.6%
South Carolina	1,318,036	122,913		9.3%
South Dakota	125,106	0		0.0%
Tennessee	1,585,807	75,744		4.8%
Texas	5,280,084	388,367		7.4%
Utah	471,555	0		0.0%
Vermont	176,335	7,366		4.2%
Virginia	2,396,673	312,809		13.1%
Washington	1,946,558	439		0.0%
West Virginia	599,989	24,385		4.1%
Wisconsin	1,291,994	0		0.0%
Wyoming	79,423	7,684		9.7%
Total	94,762,225	5,728,005		6.0%

Source: GAO analysis of Centers for Medicare & Medicaid Services (CMS) data.  |  GAO‑25‑107413

Note: Data were published in January 2025.

^aThe number of redeterminations due includes those that were scheduled for completion from March 2023 through June 2024.

^bThe number of pending redeterminations are those that were due from March 2023 through June 2024 but were not completed in the month they were due or generally in any of the subsequent 3 months. These pending redeterminations may include cases where a state is pausing procedural disenrollments temporarily. Alaska, Colorado, District of Columbia, Delaware, Kansas, Kentucky, Maine, Michigan, New Hampshire, New Mexico, and New York had paused at least some disenrollments that were due to procedural reasons, such as enrollees not returning required information, as of June 2024.

The five selected states included in our review (Arizona, Maryland, Montana, New York, and Wisconsin) varied in the extent to which they analyzed unwinding outcomes for Medicaid and Children’s Health Insurance Program (CHIP) enrollees, including disenrollments and renewals, by population characteristics.[64] In addition, the time frames of outcomes analyzed varied by state. The following tables provide information on unwinding outcomes by characteristics states analyzed.

Arizona

Table 5 provides information on disenrollments during unwinding in Arizona by age and county; the state did not similarly track renewals by age and county.[65]

			Percentage of all disenrollments
Characteristic	Number of enrollees disenrolled		Determined ineligible	Disenrolled for  procedural reasonsa
Age group (years)
Under 19	209,507		—	—
19-64	381,346		—	—
65 or older	29,218		—	—
County
Apache	8,283		38.8%	61.2%
Coconino	11,694		38.1	61.9
Cochise	9,959		44.2	55.8
Maricopa	350,507		39.2	60.8
Mohave	20,082		46.6	53.4
Navajo	13,019		41.7	58.3
Pima	86,448		41.9	58.1
Pinal	43,571		42.6	57.4
Santa Cruz	6,258		42.9	57.1
Yavapai	16,332		41.7	58.3
Yuma	24,388		47.9	52.1
Other county in state	10,871		38.2	61.8
Other county out of state	18,659		91.0	9.0
Total	620,071		42.1%	57.9%

Legend: — = Not available.

Source: GAO analysis of state information.  |  GAO‑25‑107413

Notes: Data reflect eligibility redeterminations completed from April 2023 through March 2024. Data include changes in outcomes during the state’s 90-day reconsideration period after a procedural disenrollment when those changes took place by May 6, 2024, according to Arizona officials. The state did not track the number of renewals by age and county, and as a result, we could not calculate the percentage of individuals disenrolled by age or county.

^aProcedural reasons for being disenrolled include, for example, not providing all information needed to redetermine eligibility.

Table 6 provides information on disenrollments and renewals during unwinding in Arizona for the subset of enrollees for whom the state analyzed outcomes by race and ethnicity. Specifically, of the roughly 2.6 million enrollees whose eligibility was redetermined during unwinding, there were about 630,000 enrollees (about one-fourth) for whom the state had received information during the COVID-19 public health emergency that indicated they no longer met eligibility criteria or would likely be procedurally ineligible, and for whom the state completed a redetermination during unwinding. The state analyzed outcomes by race and ethnicity for the share of those individuals who provided information on race or ethnicity.[66]

	Number of enrollees
Characteristic	Completed redeterminations	Renewed	Disenrolled
Racea
Asian Indian	3,344	1,845	1,499
Asian Unknown	648	355	293
Black	60,621	33,340	27,281
Caucasian or White	387,757	206,580	181,177
Chinese	1,459	710	749
Filipino	2,763	1,314	1,449
Guam or Chamorro	367	132	235
Japanese	628	299	329
Korean	956	482	474
Native American	51,109	29,267	21,842
Native Hawaiian	758	379	379
Native Hawaiian or Other Pacific Islander Unknown	120	48	72
Other Asian	2,805	1,463	1,342
Other Pacific Islander	1,277	659	618
Samoan	395	164	231
Unknown	4,934	2,722	2,212
Unspecified	6,067	3,318	2,749
Vietnamese	3,316	1,796	1,520
Ethnicitya
Another Hispanic, Latino/a or Spanish Origin	11,042	5,975	5,067
Cuban	1,758	786	972
Ethnicity Unknown	7,506	4,092	3,414
Ethnicity Unspecified	29	7	22
Hispanic or Latino/a Unknown	81,879	46,835	35,044
Mexican, Mexican American, or Chicano/a	203,797	113,110	90,687
Not of Hispanic or Latino or Spanish Origin	189,090	99,591	89,499
Puerto Rican	3,443	1,700	1,743

Source: GAO analysis of state information.  |  GAO‑25‑107413

Notes: Data reflect eligibility redeterminations completed from April 2023 through December 2023 for a subset of enrollees. Specifically, of the roughly 2.6 million enrollees whose eligibility was redetermined during unwinding, there were about 630,000 enrollees (about one-fourth) for whom the state had received information during the COVID-19 public health emergency that indicated they no longer met eligibility criteria or would likely be procedurally ineligible, and for whom the state completed a redetermination during unwinding. Data include the subset of those individuals who provided information on race or ethnicity.

Data include changes in outcomes during the state’s 90-day reconsideration period after a procedural disenrollment.

^aCategories listed under race and ethnicity are those Arizona used in its analysis based on enrollees’ self-reported information. Individuals do not have to report their race or ethnicity. Individuals can select one or more race and ethnicity categories; as a result, individuals may be counted in more than one category related to race or ethnicity. We did not calculate percentages because the outcome for the same individual could be counted in more than one category.

Maryland

Table 7 provides information on disenrollments and renewals during unwinding in Maryland by age, race and ethnicity, region, and eligibility group.

	Number of enrollees				Percentage of all completed redeterminations			Percentage of all disenrollments
Characteristic	Completed redeterminations	Renewed	Disenrolled		Renewed	Disenrolled		Determined ineligible	Disenrolled for procedural reasonsa
Age group (years)
0-5	197,844	159,305	38,539		80.5%	19.5%		38.5%	61.5%
6-14	322,776	270,915	51,861		83.9	16.1		26.6	73.4
15-18	133,162	107,162	26,000		80.5	19.5		35.7	64.3
19-20	55,037	36,638	18,399		66.6	33.4		48.9	51.1
21-44	521,419	383,907	137,512		73.6	26.4		38.5	61.5
45-64	249,910	183,079	66,831		73.3	26.7		45.0	55.0
65+	87,204	62,365	24,839		71.5	28.5		51.1	48.9
Race and ethnicityb
American Indian or Alaska Native	15,023	12,086	2,937		80.4	19.6		37.9	62.1
Asian	90,668	69,942	20,726		77.1	22.9		44.5	55.5
Black or African American	673,952	523,601	150,351		77.7	22.3		35.5	64.5
Black and White	22,611	19,804	2,807		87.6	12.4		33.7	66.3
Hispanic or Latino	276,652	213,072	63,580		77.0	23.0		36.2	63.8
Other	43,594	31,459	12,135		72.2	27.8		43.6	56.4
Pacific Islander	2,326	1,652	674		71.0	29.0		42.4	57.6
Two or more races	17,518	14,096	3,422		80.5	19.5		38.7	61.3
Unknown	12,071	7,231	4,840		59.9	40.1		32.8	67.2
White	412,937	310,428	102,509		75.2	24.8		45.3	54.7
Region
Baltimore City	245,718	198,586	47,132		80.8	19.2		32.9	67.1
Baltimore Suburban	475,007	364,336	110,671		76.7	23.3		40.6	59.4
Eastern Shore	141,406	108,965	32,441		77.1	22.9		41.6	58.4
Out of state	1,312	924	388		70.4	29.6		40.2	59.8
Southern Maryland	79,330	60,617	18,713		76.4	23.6		38.2	61.8
Washington Suburban	494,501	371,359	123,142		75.1	24.9		38.5	61.5
Western Maryland	130,078	98,584	31,494		75.8	24.2		44.5	55.5
Eligibility groupc
MAGI	1,467,515	1,119,418	348,097		76.3	23.7		62.0	38.0
MAGI-exempt	99,837	83,953	15,884		84.1	15.9		34.2	65.8
Total	1,567,352	1,203,371	363,981		76.8%	23.2%		39.2%	60.8%

Source: GAO analysis of state information.  |  GAO‑25‑107413

Notes: Data reflect eligibility redeterminations completed from May 2023 through April 2024. Data include enrollees who went through the state’s standard annual eligibility redetermination process; enrollees who did not go through that process (e.g., individuals who are Medicaid eligible due to eligibility for the federal Supplemental Security Income program or foster care) but instead went through a separate process were not included in the state data, according to Maryland officials. Individuals known to the state to be ineligible due to confirmed out-of-state residence or death also were not included in the data, according to Maryland officials.

Data reflect the initial redetermination decision made during unwinding and do not include subsequent changes in outcomes during the state’s 120-day reconsideration period after a procedural disenrollment, according to Maryland officials.

^aProcedural reasons for being disenrolled include, for example, not providing all information needed to redetermine eligibility.

^bThe categories listed under race and ethnicity are those the state used in its analysis. “Black and White” includes individuals who identified as both Black and White; “Two or more races” includes all other individuals who identified as two or more races; and “Other” includes individuals whose only reported race or ethnicity was “Other,” according to Maryland officials. Race and ethnicity reflect the state’s analysis of enrollees’ self-reported race and ethnicity as recorded in Medicaid and other state data.

^cTo be eligible for Medicaid or CHIP, individuals generally must have an income below a certain limit. The Patient Protection and Affordable Care Act requires states to calculate the income for most nondisabled, nonelderly individuals using modified adjusted gross income (MAGI). Other individuals’ income is not calculated using MAGI. These MAGI-exempt groups include, for example, individuals whose eligibility is determined on the basis of having a disability or being over age 65.

Montana

Table 8 provides information on disenrollments and renewals during unwinding in Montana by age, tribal affiliation, and eligibility group.

	Number of enrollees				Percentage of all completed redeterminations
Characteristic	Completed redeterminations	Renewed	Disenrolled		Renewed	Disenrolled
Age group (years)
Less than 1	488	303	185		62.1%	37.9%
1-5	17,512	11,139	6,373		63.6	36.4
6-18	74,735	45,118	29,617		60.4	39.6
19-20	8,358	2,118	6,240		25.3	74.7
21-64	140,200	75,076	65,124		53.5	46.5
65 or older	15,225	7,462	7,763		49.0	51.0
Tribal affiliationa
Yes	34,349	21,884	12,465		63.7	36.3
No	222,169	119,332	102,837		53.7	46.3
Eligibility groupb
MAGI	236,442	129,158	107,284		54.6	45.4
MAGI-exempt	20,076	12,058	8,018		60.1	39.9
Total	256,518	141,216	115,302		55.1%	44.9%

Source: GAO analysis of state information.  |  GAO‑25‑107413

Notes: Data reflect eligibility redeterminations completed from May 2023 through June 2024. As of June 2024, the state had not completed about 24,000 additional redeterminations (about 8 percent of redeterminations due).

Data include enrollees who went through the state’s standard eligibility redetermination process; enrollees who did not go through the standard process (e.g., individuals who are Medicaid eligible due to eligibility for the federal Supplemental Security Income program or foster care) but instead went through a separate process were not included in the state data.

Data reflect the initial redetermination decision made during unwinding and do not include subsequent changes in outcomes during the state’s 90-day reconsideration period after a procedural disenrollment.

^aTribal affiliation is based on enrollees’ voluntary self-reported affiliation; it is not based on enrollees’ self-reported race or ethnicity.

^bTo be eligible for Medicaid or CHIP, individuals generally must have an income below a certain limit. The Patient Protection and Affordable Care Act requires states to calculate the income for most nondisabled, nonelderly individuals using modified adjusted gross income (MAGI). Other individuals’ income is not calculated using MAGI. These MAGI-exempt groups include, for example, individuals whose eligibility is determined on the basis of having a disability or being over age 65.

Additional Montana outcomes data indicated the following:

·         Of the enrollees who were part of the state’s Medicaid expansion population and due for a redetermination during unwinding, 44 percent had been disenrolled, 50 percent had been renewed, and 6 percent did not have a completed redetermination as of June 2024.[67] By comparison, of the enrollees who were not part of the Medicaid expansion population, 37 percent had been disenrolled, 48 percent had been renewed, and 15 percent did not have a completed redetermination.

·         In five of 56 counties, less than 30 percent of all enrollees due for a redetermination during unwinding were disenrolled as of June 2024; in two counties, more than 50 percent of enrollees had been disenrolled.[68]

New York

Table 9 provides information on disenrollments and renewals during unwinding in New York. The state tracked these outcomes separately for individuals whose eligibility was redetermined centrally through the New York State of Health eligibility system and for those whose eligibility was redetermined by local departments of social services using legacy eligibility systems.[69] Most Medicaid and all CHIP enrollees had their eligibility redetermined centrally through the New York State of Health eligibility system, while certain Medicaid enrollees (e.g., individuals over age 65 or with a disability) had their eligibility redetermined by local departments of social services.

	Number of enrollees				Percentage of all completed redeterminations
Subset of enrolleesa	Completed redeterminations	Renewed	Disenrolled		Renewed	Disenrolled
Medicaid (New York State of Health system)	4,818,729	3,352,427	1,466,302		69.6%	30.4%
Medicaid (local departments of social services)	1,267,320	1,131,318	136,002		89.3	10.7
CHIP (New York State of Health system)	267,739	248,223	19,516		92.7	7.3
Total	6,353,788	4,731,968	1,621,820		74.5%	25.5%

Source: GAO analysis of state information.  |  GAO‑25‑107413

Notes: Data reflect eligibility redeterminations completed from June 2023 through December 2024. At that time, the state had not yet completed about 70,000 additional redeterminations for enrollees whose eligibility would be redetermined by local departments of social services; those redeterminations accounted for 1 percent of all unwinding redeterminations due. For enrollees who had their eligibility redetermined centrally through the New York State of Health system, data include changes in outcomes that occurred during the state’s 90-day reconsideration period after a procedural disenrollment, according to New York officials. Certain individuals, such as those who moved out of state, withdrew from coverage, or died before the state sent a renewal form, and individuals eligible for Medicaid due to receiving Supplemental Security Income, were not included in the data, according to New York officials.

^aMost Medicaid and all CHIP enrollees had their eligibility redetermined centrally through the New York State of Health eligibility system. Certain Medicaid enrollees (e.g., individuals over age 65 or with a disability) had their eligibility redetermined by local departments of social services using legacy eligibility systems

Table 10 through Table 13 provide information on renewals and disenrollments by age, region, written language, and race and ethnicity for the 5.1 million Medicaid and CHIP enrollees whose eligibility was redetermined through the New York State of Health eligibility system. Those enrollees account for about 80 percent of all completed unwinding redeterminations.[70]

	Number of enrollees				Percentage of all completed redeterminations
Age group (years)	Completed redeterminations	Renewed	Disenrolled		Renewed	Disenrolled
0-17	1,602,302	1,463,895	138,407		91.4%	8.6%
18-25	736,795	444,921	291,874		60.4	39.6
26-34	814,254	463,921	350,333		57.0	43.0
35-44	723,825	457,399	266,426		63.2	36.8
45-54	559,007	358,517	200,490		64.1	35.9
55-64	518,502	341,728	176,774		65.9	34.1
65+	131,783	70,269	61,514		53.3	46.7
Total	5,086,468	3,600,650	1,485,818		70.8%	29.2%

Source: GAO analysis of state information.  |  GAO‑25‑107413

Notes: Data reflect eligibility redeterminations completed from June 2023 through May 2024 for the 5.1 million enrollees whose eligibility was redetermined centrally through the New York State of Health eligibility system. Data include changes in outcomes that occurred during the state’s 90-day reconsideration period after a procedural disenrollment, according to New York officials. Data do not include 1.3 million enrollees whose eligibility was redetermined by local departments of social services, which use legacy eligibility systems (e.g., individuals over age 65 or with a disability). Certain individuals, such as those who moved out of state, withdrew from coverage, or died before the state sent a renewal form, were not included in the data, according to New York officials.

	Number of enrollees				Percentage of all completed redeterminations
Region	Completed redeterminations	Renewed	Disenrolled		Renewed	Disenrolled
Capital / Mid-Hudson / North Country	833,274	607,508	225,766		72.9%	27.1%
Central	639,385	444,833	194,552		69.6	30.4
Long Island	587,201	419,787	167,414		71.5	28.5
New York City	2,699,161	1,897,889	801,272		70.3	29.7
Western	327,447	230,633	96,814		70.4	29.6
Total	5,086,468	3,600,650	1,485,818		70.8%	29.2%

Source: GAO analysis of state information.  |  GAO‑25‑107413

Notes: Data reflect eligibility redeterminations completed from June 2023 through May 2024 for the 5.1 million enrollees whose eligibility was redetermined centrally through the New York State of Health eligibility system. Data include changes in outcomes that occurred during the state’s 90-day reconsideration period after a procedural disenrollment, according to New York officials. Data do not include 1.3 million enrollees whose eligibility was redetermined by local departments of social services, which use legacy eligibility systems (e.g., individuals over age 65 or with a disability). Certain individuals, such as those who moved out of state, withdrew from coverage, or died before the state sent a renewal form, were not included in the data, according to New York officials.

	Number of enrollees				Percentage of all completed redeterminations
Written language	Completed redeterminations	Renewed	Disenrolled		Renewed	Disenrolled
Albanian	750	565	185		75.3%	24.7%
Arabic	4,645	3,811	834		82.0	18.0
Bengali	2,779	2,160	619		77.7	22.3
Burmese	1,272	892	380		70.1	29.9
Chinese	144,751	115,380	29,371		79.7	20.3
English	3,901,415	2,695,561	1,205,854		69.1	30.9
French	5,002	3,642	1,360		72.8	27.2
French Creole	958	591	367		61.7	38.3
Greek	220	149	71		67.7	32.3
Haitian Creole	4,050	2,727	1,323		67.3	32.7
Hindi	466	339	127		72.7	27.3
Italian	564	388	176		68.8	31.2
Japanese	451	297	154		65.9	34.1
Karen	1,064	825	239		77.5	22.5
Korean	9,111	6,294	2,817		69.1	30.9
Nepali	636	419	217		65.9	34.1
Polish	1,696	1,142	554		67.3	32.7
Russian	34,423	24,751	9,672		71.9	28.1
Somali	313	259	54		82.7	17.3
Spanish	968,484	737,874	230,610		76.2	23.8
Swahili	415	347	68		83.6	16.4
Tagalog	95	65	30		68.4	31.6
Tigrinya	62	44	18		71.0	29.0
Traditional Chinese	894	689	205		77.1	22.9
TWI	30	25	5		83.3	16.7
Urdu	785	611	174		77.8	22.2
Vietnamese	772	538	234		69.7	30.3
Yiddish	365	265	100		72.6	27.4
Total	5,086,468	3,600,650	1,485,818		70.8%	29.2%

Source: GAO analysis of state information.  |  GAO‑25‑107413

Notes: Data reflect eligibility redeterminations completed from June 2023 through May 2024 for the 5.1 million enrollees whose eligibility was redetermined centrally through the New York State of Health eligibility system. Data include changes in outcomes that occurred during the state’s 90-day reconsideration period after a procedural disenrollment, according to New York officials. Data do not include 1.3 million enrollees whose eligibility was redetermined by local departments of social services, which use legacy eligibility systems (e.g., individuals over age 65 or with a disability). Individuals enrolling through the New York State of Health eligibility system must select their preferred written language on their application. Certain individuals, such as those who moved out of state, withdrew from coverage, or died before the state sent a renewal form, were not included in the data, according to New York officials.

	Number of enrollees
Characteristic	Completed redeterminations	Renewed	Disenrolled
Racea
African	12,592	9,254	3,338
American Indian or Alaskan Native	25,244	17,361	7,883
Asian Indian	102,467	71,323	31,144
Bangladeshi	13,816	10,553	3,263
Black or African American	686,571	468,734	217,837
Burmese	529	376	153
Chinese	220,642	171,265	49,377
Filipino	17,632	11,484	6,148
Guamanian or Chamorro	2,635	2,043	592
Haitian	6,115	4,158	1,957
Jamaican	7,468	5,450	2,018
Japanese	5,202	3,408	1,794
Korean	26,114	17,370	8,744
Middle Eastern or North African	13,726	11,112	2,614
Native Hawaiian	1,107	777	330
Other	425,915	316,955	108,960
Other Asian	143,040	104,658	38,382
Other Pacific Islander	9,351	6,476	2,875
Pakistani	3,354	2,684	670
Samoan	800	582	218
Taiwanese	366	270	96
Unknownb	1,975,702	1,390,834	584,868
Vietnamese	8,890	6,157	2,733
White	1,513,546	1,067,388	446,158
Ethnicitya
Hispanic – Yes	1,347,149	986,407	360,742
Hispanic – No	2,723,012	1,913,113	809,899
Hispanic – Unknownb	1,016,307	701,130	315,177

Source: GAO analysis of state information.  |  GAO‑25‑107413

Notes: Data reflect eligibility redeterminations completed from June 2023 through May 2024 for the 5.1 million enrollees whose eligibility was redetermined centrally through the New York State of Health eligibility system. Data include changes in outcomes that occurred during the state’s 90-day reconsideration period after a procedural disenrollment, according to New York officials. Data do not include 1.3 million enrollees whose eligibility was redetermined by local departments of social services using legacy eligibility systems (e.g., individuals over age 65 or with a disability). Certain individuals, such as those who moved out of state, withdrew from coverage, or died before the state sent a renewal form, were not included in the data, according to New York officials.

^aCategories listed under race and ethnicity are those New York used in its analysis based on enrollees’ self-reported information. Individuals do not have to report their race or ethnicity. Individuals can select one or more race categories; as a result, individuals may be counted in more than one race category. We did not calculate percentages because the outcome for the same individual could be counted in more than one category.

^bThe “Unknown” category includes individuals who indicated “Don’t know” or “Choose not to answer,” according to New York officials.

Wisconsin

Table 14 provides information on disenrollments and renewals during unwinding in Wisconsin by race and ethnicity.

	Number of enrollees				Percentage of all completed redeterminations			Percentage of all disenrollments
Characteristic	Completed redeterminations	Renewed	Disenrolled		Renewed	Disenrolled		Determined ineligible	Disenrolled for procedural reasonsa
Raceb
American Indian or Alaskan Native	22,992	16,654	6,338		72.4%	27.6%		33.2%	66.8%
Asian	46,751	33,876	12,875		72.5	27.5		37.9	62.1
Black or African American	197,636	147,940	49,696		74.9	25.1		34.1	65.9
Hispanic or Latino	105,305	75,453	29,852		71.7	28.3		33.4	66.6
Multi-Race	101,725	74,139	27,586		72.9	27.1		34.8	65.2
Native Hawaiian or Pacific Islander	1,135	787	348		69.3	30.7		32.8	67.2
Race not provided	163,694	123,198	40,496		75.3	24.7		46.8	53.2
White	780,527	557,738	222,789		71.5	28.5		37.2	62.8
Ethnicityb
Hispanic or Latino	165,339	118,460	46,879		71.6	28.4		33.9	66.1
Not Hispanic or Latino	1,245,551	910,335	335,216		73.1	26.9		36.8	63.2
Ethnicity not provided	8,875	990	7,885		11.2	88.8		79.3	20.7
Total	1,419,765	1,029,785	389,980		72.5%	27.5%		37.3%	62.7%

Source: GAO analysis of state information.  |  GAO‑25‑107413

Notes: Data reflect eligibility redeterminations completed from June 2023 through June 2024. Data include changes in outcomes that occurred during the state’s 90-day reconsideration period after a procedural disenrollment.

Data include enrollees who went through the state’s standard eligibility redetermination process; enrollees who did not go through the standard process (e.g., individuals who are Medicaid eligible due to eligibility for the federal Supplemental Security Income program or foster care) but instead went through a separate process were not included in the state data. About 77,000 enrollees who were asked to reapply for coverage instead of renew their existing coverage, due to an eligibility systems limitation, also were not included in the data.

^aProcedural reasons for being disenrolled include, for example, not providing all information needed to redetermine eligibility.

^bCategories listed under race and ethnicity are those Wisconsin used in its analysis based on enrollees’ self-reported information. Individuals do not have to report their race or ethnicity.

Wisconsin officials told us they also analyzed disenrollments and renewals during unwinding by age, geographic location, preferred language, and eligibility group. Officials said they identified some differences in outcomes by some of those population characteristics, but the available state data were not sufficient for our reporting purposes.

GAO Contact

Michelle B. Rosenberg, RosenbergM@gao.gov

Staff Acknowledgments

In addition to the contact named above, Susan Barnidge (Assistant Director), Dhara Patel (Analyst-in-Charge), Robin Burke, and Jasleen Modi made key contributions to this report. Also contributing were Laura Elsberg, David Jones, Eric Peterson, Jennifer Rudisill, and Jeffrey Tamburello.

GAO’s Mission

The Government Accountability Office, the audit, evaluation, and investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO’s commitment to good government is reflected in its core values of accountability, integrity, and reliability.

Obtaining Copies of GAO Reports and Testimony

The fastest and easiest way to obtain copies of GAO documents at no cost is through our website. Each weekday afternoon, GAO posts on its website newly released reports, testimony, and correspondence. You can also subscribe to GAO’s email updates to receive notification of newly posted products.

Order by Phone

The price of each GAO publication reflects GAO’s actual cost of production and distribution and depends on the number of pages in the publication and whether the publication is printed in color or black and white. Pricing and ordering information is posted on GAO’s website, https://www.gao.gov/ordering.htm.

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537.

Orders may be paid for using American Express, Discover Card, MasterCard, Visa, check, or money order. Call for additional information.

Connect with GAO

Connect with GAO on X, LinkedIn, Instagram, and YouTube.
Subscribe to our Email Updates. Listen to our Podcasts.
Visit GAO on the web at https://www.gao.gov.

To Report Fraud, Waste, and Abuse in Federal Programs

Contact FraudNet:

Website: https://www.gao.gov/about/what-gao-does/fraudnet

Automated answering system: (800) 424-5454

Media Relations

Sarah Kaczmarek, Managing Director, Media@gao.gov

Congressional Relations

A. Nicole Clowers, Managing Director, CongRel@gao.gov

General Inquiries

https://www.gao.gov/about/contact-us