Medicaid Unwinding Drops Coverage for Millions but Uninsured Rate Remains Steady

Even after more than 25 million people were dropped from Medicaid during the coverage “unwinding,” the uninsured rate stayed surprisingly steady as many re-enrolled or found other types of coverage, according to new research published in JAMA Health Forum.

According to study authors, including study lead Adrianna McIntyre, Ph.D., M.P.H., assistant professor of health policy and politics in the Department of Health Policy and Management at Harvard T.H. Chan School of Public Health in Boston, Medicaid unwinding was the large-scale review and removal of members from Medicaid that began in 2023 after pandemic rules that assured continuous coverage expired. This process caused millions to lose benefits as states resumed checking eligibility.

Over the past few years, Medicaid enrollment has swung dramatically, first surging during the COVID-19 pandemic and then falling quickly as states began unwinding continuous coverage in 2023. More than 25 million people, or nearly 8% of the U.S. population, lost their Medicaid coverage during this period, marking one of the largest drops in the program’s history. Although there were fears of widespread loss of insurance, federal data shows that the uninsured rate barely changed.

The JAMA study, published October 10, 2025, explains why.

Researchers found that many people who lost Medicaid either quickly re-enrolled, moved to other forms of coverage or were confused about their insurance status. These findings highlight how policy shifts, administrative processes and enrollee awareness all shaped the impact of Medicaid’s unwinding and may also inform future reforms aimed at reducing coverage gaps and improving the stability of public insurance programs.

Based on national survey and administrative data to track Medicaid coverage changes from 2019 to 2024, researchers analyzed enrollment data from CMS and compared it with self-reported coverage data from the National Health Interview Survey (NHIS) to identify differences between official records and what people reported.

These comparisons helped reveal confusion among enrollees and trends, where folks lose and later regain Medicaid. The study also looked at how often people applied for Medicaid before, during and after the pandemic’s continuous coverage period and how these patterns varied by state. The researchers also used statistical models to examine how coverage changes affected different groups based on factors like race, income and health status. Analyses covered all 50 states and Washington, D.C.

During the study period, Medicaid enrollment ranged from 72 million to 94.4 million people. While the continuous coverage rules were in place, enrollment grew, but surveys undercounted participants, especially children, where survey estimates were nearly 20 points lower than administrative records.

When states resumed eligibility checks during the unwinding, the share of people losing coverage varied widely, from under 15% in some states to more than 35% in others. Many people, however, re-enrolled or moved to other types of insurance, so the net loss was smaller.

Monthly Medicaid applications dropped about 30% during continuous coverage but then increased after unwinding, mainly in states with more terminations. Self-reported Medicaid coverage also fell across groups such as young adults, women and urban residents, but overall uninsured rates stayed mostly steady.

In addition, some groups reported more problems affording care, while women saw a small increase in private insurance, showing how coverage shifted without a major rise in the uninsured rate.

The study also highlights factors explaining why massive Medicaid terminations in 2023 didn’t increase uninsured rates, including enrollee confusion, re-enrollment and transitions to other coverage.

Based on study methods and results, study strengths include the combination of administrative and survey data and analysis across multiple subgroups. Limitations include unresolved terminations, a focus on low-income adults under 65 and an inability to establish a complete cause of terminations.

Authors suggest improving communication about coverage, monitoring patient turnover and evaluating policy changes like continuous eligibility and more frequent redeterminations to protect access and affordability.