Medicaid ‘Unwinding’ Could Lead to Eviction Crisis, New Study Suggests

As a growing number of states look to expand Medicaid to provide other health-related social needs such as housing, it stands to reason that the loss of Medicaid benefits might also have housing-related impacts. That’s the question Georgia Tech Assistant Professor Catherine Bradford and her co-authors set out to explore in a new study that looks at the housing repercussions of the 2005 disenrollment of participants in Tennessee’s Medicaid program. Bradford spoke with Spotlight recently about the study—the transcript has been lightly edited for length and clarity.

Before we talk about this specific study, tell us a bit about your work in general.

I am an assistant professor at Georgia Tech who focuses primarily on health policy. My primary research interests lie in risky behaviors and substance use and more recently I’ve started to get really interested in the intersection between health and housing and risk.

And was this the first study you’ve done that focuses on housing?

I have a few other studies that focus on eviction. One looks at the relationship between psychiatric treatment and eviction rates—when we see more psychiatric treatment access, we see evictions go down. And I have another project looking at how evictions relate to fatal overdose mortality.

And what prompted you to look at the Medicaid connection?

This is something that my co-authors, Catherine Maclean and Mir Ali, and me were really interested in thinking about in light of the current Medicaid unwinding. And there was this really interesting, somewhat exogenous shock that happened in Tennessee in 2005 that gave us some information about what we might expect now, although it’s obviously quite difficult to extrapolate.

In 2005, the Tennessee Medicaid program called TennCare disenrolled, essentially overnight, about 190,000 beneficiaries. And what we do is link that disenrollment to what happened to eviction rates and eviction filing rates in Tennessee as compared to other southern states. And we found that Tennessee experienced a 25% increase in evictions and a 28% increase in eviction filings relative to the other southern states. And so, if that disenrollment is reflective of what we could expect to happen nationally with the current Medicaid unwinding, we could see a wave of evictions, which is going to exacerbate the issues that already come along with losing your health insurance coverage.

Is there pre-existing research that shows that Medicaid coverage can lead to a decrease in evictions? Or is that something that’s not been looked at?

I think we were the only people to link disenrollment to housing outcomes. There have been other studies that look at Medicaid expansion and how that led to housing outcomes and that seems to improve housing outcomes. But to our knowledge, we’re the first to look at the impact of losing health insurance coverage on housing related outcomes.

And do you have any idea how many people have come off the roles in Tennessee as a result of the current unwinding?

I’m not sure about Tennessee in particular, but the estimates of the national numbers were something between eight to 24 million. And we’ve already seen, at least given January 2024 numbers, we’ve already seen 16 million people being disenrolled nationally.

So, what is your hope for how this study gets used?

Well, I hope that it brings attention to the broader impacts of disenrollment. Of course, it is natural for us to think of the immediate effects on health, right? When people are disenrolled, that could mean that their chronic conditions are not going to be managed effectively. It could mean that new conditions might emerge as a result of a disconnect from preventative care. And it could be that any health shock that does occur could bury a family in medical debt and really impact their financial stability.

But that’s not the only thing that’s going to happen. There’s going to be ripple effects from this dis-enrollment and one aspect is housing. Housing is a really important social determinant of health. It’s a really important aspect in family stability in general. And losing your health insurance is not just going to impact your health. It’s going to impact a lot of aspects of your life, one of which being housing.

Medicaid is being expanded to include housing in some states—I’m assuming that’s something you think is well worth looking into?

I think it’s well worth look looking into, and I think it’s appropriate. We know that there is a really strong relationship between health and housing and a strong relationship between risky behaviors and housing. If we’re worried about people’s health, we need to be worried about the determinants of health in general and protecting people’s housing is going to allow them to protect their health.

Is there any sort of follow-up planned or are there other studies that you’re interested in doing along these same lines?

Not with disenrollment in particular, but I am really interested in thinking about this relationship between health and housing. And not just evictions, but homelessness. I mean, we’re experiencing, at least since 2007 when they started monitoring the numbers, the highest numbers of people living without homes that we’ve had. It’s hard to even measure homelessness because it looks so different—they could be living in their cars, they could be living in motels, they could be living with other families, and that impacts families for generations. And having housing instability really disconnects you from the social services that are intended to help vulnerable people. So, I’m interested to dig into this in a variety of different ways to think about eviction and housing instability more broadly, to think about homelessness and the various ways that it impacts people and the policies that we could put in place to potentially help these families.

This is a little off the point of your study, but Georgia is one of the states that has work requirements for Medicaid. Do you have a view on how that potentially impacts this whole connection?

We do know that even without work requirements, I think it’s something like 80% of enrollees live in households that have someone who’s working in Georgia. I think it’s something like 60% of people on Medicaid are working. And so, it’s hard to really think about the cost and the benefits here, but we do know that a lot of people on Medicaid are already working. I understand the desire to connect people to the labor force, but I really think that we need to be protecting people, regardless.

And finally, were there particular groups who seemed to be the most impacted by the spike in evictions?

Our data came from The Eviction Lab and while they do so much wonderful work, they do not provide individual level data. So, we don’t know who is being impacted. But the work that’s already out there has shown us that the people who tend to be impacted by eviction are the most vulnerable. It disproportionately impacts people of color, disproportionately impacts women of color, and Black mothers tend to be a population that experience the brunt of the eviction epidemic that we’re seeing.

 Is there anything that you want to add?

I do want to be careful about how we extrapolate our results, because we were looking at a policy change that occurred in 2005, and many things have changed in the past 20 years. The housing market is very different. The way that we administer health is very different. With the TennCare disenrollment, it was primarily non-disabled, childless adults who were impacted. With the current Medicaid unwinding, it’s expected to disproportionately impact older Americans, which might be sheltered from these types of eviction outcomes. but also, people with limited English proficiency and disabled individuals who are at much higher risk for housing instability. We do need to be very cautious with the way that we extrapolate our results to the current times, but I think that we have good reason to think that housing instability, while perhaps not at the forefront of people’s immediate attention, will be impacting vulnerable for years to come.

One takeaway that I want people to think about from this project is that it’s really important that we expand our understanding of health. There are so many aspects that go into the health of an individual and a family and a household and we need to consider all of those potential determinants when we are crafting our policy interventions.