Study reveals coverage gaps and health access challenges post-Medicaid unwinding

In a survey of low-income adults across Arkansas, Kentucky, Louisiana, and Texas, one in eight respondents who were enrolled in Medicaid at some point since March 2020 reported no longer having Medicaid coverage by late 2023, with nearly half of that pool reporting being currently uninsured, according to a study by Harvard T.H. Chan School of Public Health.

The researchers conducted the survey from September to November 2023, about six months after the start of “Medicaid unwinding”-;the process by which states rechecked Medicaid enrollees’ eligibility after the expiration of COVID-19-era coverage protections.

We know from government statistics that, of the more than 90 million people whose health coverage was in jeopardy amid Medicaid unwinding, more than 23 million were removed from the program. But those statistics don’t tell us what happened to those people, or why they lost coverage. Our study is one of the first to help answer those outstanding questions, using completely new data from an original multi-state survey.”

Adrianna McIntyre, assistant professor of health policy and politics, lead author 

The study will be published June 29 in JAMA Health Forum.

The researchers surveyed 2,210 adults ages 19 to 64 in those four states whose 2022 income was at or less than 138{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497} of the federal poverty line. Participants were asked whether they and/or their dependents had been enrolled in Medicaid at any point since March 2020, when states paused Medicaid disenrollment as part of the COVID-19 federal public health emergency. They were also asked about their current health insurance and ability to access care, as well as their demographic information.

Most survey respondents (71{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497}) respondents reported that they and/or a dependent had been enrolled in Medicaid at some point since March 2020. Among adult respondents who had Medicaid, 12.5{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497} reported that they had been disenrolled by fall 2023. State-specific adult disenrollment levels varied: 16{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497} in Arkansas, 15{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497} in Texas, 8{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497} in Louisiana, and 7{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497} in Kentucky. Of those disenrolled, 52{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497} had found another source of insurance, though a meaningful share of these respondents reported experiencing coverage gaps. The remaining 48{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497} of those disenrolled reported being uninsured. Disenrollment was significantly higher among those who were younger, rural, and/or employed, and women were likelier than men to become uninsured. The reported disenrollment rate was much lower, on average, for dependent children (5.4{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497}).

The study also found that those who had been disenrolled had significantly worse access to health care compared to those who did not lose their Medicaid coverage. Those disenrolled reported more cost-related delays in care (51{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497} versus 27{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497}) and delays or skipped doses of medications (45{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497} versus 27{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497}); said that care was less affordable than the previous year (47{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497} versus 22{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497}); and did not go for an annual check-up during the prior year (57{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497} versus 34{e60f258f32f4d0090826105a8a8e4487cca35cebb3251bd7e4de0ff6f7e40497}).

“In prior research, even brief coverage gaps have been associated with care disruptions and negative health outcomes,” said McIntyre. “Our findings suggest that state and federal policymakers should pursue policies to mitigate adverse outcomes associated with coverage disruptions-;not just during the Medicaid unwinding, but in the years to come, as issues related to eligibility redeterminations and continuity of coverage will remain relevant in Medicaid.”

The researchers noted that the study’s findings may be limited in their generalizability for a number of reasons, including that the survey sample was limited to residents of four states who reported lower incomes and that there were considerable variations in states’ approaches to unwinding.

Benjamin Sommers, Gabriella Aboulafia, Jessica Phelan, E. John Orav, Arnold Epstein, and Jose Figueroa, all of Harvard Chan’s Department of Health Policy and Management, were co-authors.

The study received funding from the Commonwealth Fund and the Episcopal Health Foundation.