Investigating the Impact of Guaranteed Income on Mental Health and Sleep

Dr. Stanhope, Dr. Hamilton, and Dr. Roll join EPITalk host and co-author, Dr. Patrick Sullivan, to take a deep dive into their exciting recently published study regarding how guaranteed income improves financial and mental wellbeing in a randomized trial among Black women in Georgia. “Improvements in stress and sleep following 24-months of Guaranteed Income, results from a randomized trial among Black women in Georgia“ is published in the February 2026 Issue (Vol. 114) of Annals of Epidemiology. 

Read the full article here:

 https://doi.org/10.1016/j.annepidem.2025.12.010 

Episode Credits:

• Executive Producer: Sabrina Debas (Episodes 1-18) and Sofina Tran
• Technical Producer: Paula Burrows
• Annals of Epidemiology is published by Elsevier.
  
  
  

Transcript

Patrick Sullivan 0:10

Hello, you're listening to EpiTalk, Behind the Paper, a podcast from the Annals of Epidemiology. I'm Patrick Sullivan, editor-in-chief of the journal, and in this series, we'll take you behind the scenes of some of the latest epidemiologic research featured in our journal. Today we're here with Dr. Kate Stanhope and her colleagues to discuss their article, Improvements in Stress and Sleep Following 24 Months of Guaranteed Income, results from a randomized trial among black women in Georgia. You can read the full article online in the February 2026 issue of the journal Annals of Epidemiology at www.annalsof epidemiology.org. So first I want to give you a short introduction to our guests. Kaitlyn Stanhope is an assistant professor in the Department of Epidemiology at Emory University's Rollins School of Public Health. She has expertise in measuring and analyzing structural and social exposures and their association with health outcomes during pregnancy, postpartum, and as women age. Dr. Leah Hamilton is the Beaver Endowed Professor of Social Work at Appalachian State University, a senior fellow at the Jane Family Institute, and a faculty affiliate at the Center for Social Development of Washington University in St. Louis. She researches economic justice and is principal investigator of the Family Economic Policy Lab, she leads the evaluation of several cash transfer projects across the U.S. Dr. Hamilton's work has appeared in outlets such as the New York Times, the Wall Street Journal, the Washington Post, and more. Stephen Roll's research focuses on promoting asset building, debt management, and economic security in lower-income populations. Working with the Center for Social Development, where he's the director of research, he's involved in research on cash transfers, child savings, and the role of public and employer-based benefit programs in promoting or harming economic and social mobility. His recent work focuses on the role of cash transfer programs in improving household balance sheets and economic mobility outcomes, which include studies of the expanded child tax credit and guaranteed income experiments in Georgia, New York, and Missouri. After reading these introductions and knowing what your manuscript's about, I'm so happy to have the chance to talk with all of you today and thank you for making time to share some more context and background about your study. Happy to be here. Great. Kate, let's kick things off with you. Just tell us about the purpose of the study. Like how did you come into this question and what were you really trying to answer with the analyses that you did?

Kate Stanhope 2:46

I'm gonna go ahead and give more context than perhaps you're asking for here because I really like this story. But I learned about this project on WAVE. I was driving home one night and I heard Rose Scott talking to the community founder of the In Her Hands Project about the project. And I was like, this is great. This is a guaranteed income pilot among black women in Georgia. It's really rigorous. And a lot of my work is focused on racial disparities and health outcomes. But a lot of times the interventions we're doing are very downstream. They're behavioral and they're clinical. And I thought, why aren't we doing this everywhere? So I Googled it. I found Leah and I reached out and I said, I'm so excited about this project. If there's ever an opportunity to think about how this impacts health, can we brainstorm about it? And we were able to brainstorm, and one of the kind of the framework we were thinking about are what health outcomes would we anticipate guaranteed income would impact? So there has been a pretty large body of work on benefits of guaranteed income. And Stephen and Leah can talk more about what those are. In the US, there's been limited evaluation of physical health impacts. And so we're really interested in kind of trying to evaluate whether receipt of guaranteed income can actually impact people's physical health. And we also wanted to look at something, one that we thought guaranteed income would impact. So something that was stress-sensitive, that we thought a reduction in stress would improve, but also something that we really thought, yes, this would have long-term health benefits. So we talked about a number of different potential mechanisms by which we thought guaranteed income would improve health. And we ended up focusing on sleep as a primary outcome because sleep has sort of cascading consequences across the life course. It's probably beyond the scope of this interview, but even acute changes in sleep can increase your vulnerability to infectious disease. And it's something that is very sensitive to stress. And so reduction in stress can actually very quickly improve your sleep. So we thought this should work. Like guaranteed income should improve sleep, and also it could benefit health kind of across health outcomes. So that was sort of the motivation I'll let my colleagues add.

Patrick Sullivan 4:53

Great. Thank you so much. So, Leah, could you walk us through the methods that were used to carry out the research and just talk a little bit about your approach and what did you love about it? Where do you think some of the limitations, if any, are of the research approach?

Leah Hamilton 5:07

Yeah, thanks so much. This project has been just, you know, a highlight of my career to get the privilege to work on because it came directly out of community listening sessions. There was a task force in the old fourth ward of Atlanta, which is Dr. Martin Luther King's home neighborhood or Ebenezer Baptist Churches, and asked people in the neighborhood in community listening sessions, what do you need to achieve greater economic stability? And they said, cash. We need cash. And that's really where this pilot grew from, is exactly what community members said they needed. And then the program partnered with community members to design the intervention. And the community members helped determine what's uh an amount of money that feels meaningful in your life. And they circled around $850 a month. But then some folks said, you know, $850 is great, but what I really need is a car or a down payment on an apartment or something. So they decided to test that and they randomized participants into two treatment groups, receiving $850 a month for 24 months, and then the other half received 4,300 in the first month, and then 700 a month for the following 23 months. So the total transfer is the same. So that was a really interesting mechanism, but it's both rigorously designed, but also driven by community. And I got ahead of myself a little bit. The selection of participants was also completely randomized. The organization did a really great job of getting out to community organizations in the old fourth ward of Atlanta. They expanded to a rural county cluster in southwest Georgia and then a suburban college park. All of those are predominantly black and historically black neighborhoods. The participants did not have to be black, but the program did a really great job of targeting these neighborhoods that are predominantly black. So we have a sample that's about 97% black women who are making less than the area median income. The eligibility criteria was less than 200% of the federal poverty line. And our team came in and we wanted to do one, a rigorous study, but more importantly, a study that upheld the same values as the program design. So my team used a community-based participatory research design. We met with community advisory councils multiple times before the research began and throughout the evaluation to say what are the research questions that are most important to you. And this is what we're finding. This is how do we interpret this data? I partnered with some fantastic students at Clark Atlanta University, doctoral students, to conduct qualitative interviews. I partnered with Stephen's team at Wash U to do, you know, state-of-the-art, rigorous evaluation on our surveys. And we also have administrative data we can talk about and took every opportunity to have brilliant researchers like Kate who are interested in doing add-on studies to demonstrate all the different ways that guaranteed income could impact someone's life.

Patrick Sullivan 8:12

Thanks. And I think we'll try and touch back on this idea about the different disciplines and the different dimensions of this research question, maybe a little bit in the next section. I wanted to think about, Stephen, particularly what kinds of sources of bias you were concerned about when comparing the guaranteed income recipients to non-recipients and how that played into the design and the analysis of the study data.

Stephen Roll 8:37

Of course. So one of the advantages of our approach is that the offer of the guaranteed income was made randomly. So the treatment group is randomly selected. So this is the initial structure of this, is just like a field experiment. So there is no bias between the treatment group and the control group generally. However, where bias does get introduced in the study is in terms of actual survey response. So the treatment group, likely because they're receiving, you know, a substantial amount of money every month, is more likely to respond to our follow-up surveys than the control group. This is a perennial problem in these types of studies. It's just much easier to engage the group who is receiving services or receiving payments or receiving whatever intervention than it is to engage the control group. And so we do have some differential response rates between the treatment and the control group. And that's the primary source of that's the primary concern that we have as far as the possibility of bias. However, when we actually look at how this is affecting, say, baseline indicators between the treatment and control groups, we don't actually see much evidence that there is strong bias emerging from this differential response. However, we also engaged in a process called generalized boosted modeling, which basically uses a machine learning algorithm to generate a set of weights that actually balances the treatment and control groups based on their baseline administrative indicators, just to further address the possibility that differential response is somehow biasing our estimates. And when we generate those weights and apply them to the analysis, we actually don't see much difference between the weighted and unweighted analyses, which provides further evidence that differential responses is not a substantial or possibly even notable driver in the outcomes we see.

Patrick Sullivan 10:37

Thanks. That's such a well-conceived and then well-executed plan to really address this aspect of bias. So thanks so much for talking about the background and the methods for the manuscript, but I want to be sure that we get to an explanation of what you found. So, what were some of the key findings from the paper?

Behind the Paper

Kate Stanhope 10:55

So we looked at two aspects of sleep and then a continuous measure of mental distress. All the outcomes were continuous. And what we found is no difference in sleep duration, which is to say the amount of time people reported sleeping between those who did and didn't receive guaranteed income. We did observe higher sleep quality among those who received guaranteed income versus not receiving it. And we also observed lower mental distress scores on this continuous measure of general mental distress, the Kessler 10. Then we also did an exploratory mediation analysis to see whether we could attribute this to financial strain because we have the advantage of having two survey time points at 12 and 24 months of follow-up. And those findings suggest that a small but meaningful proportion, perhaps 12 or 15% of the changes in sleep quality and mental distress, may be attributed to differences in financial strain at 12 months, which is, I think it's reassuring because that's kind of our hypothesis about how we think guaranteed income may impact sleep and mental distress is by reducing financial strain. Of course, there's other mechanisms. People could use increased cash resources to get better housing, to potentially have greater flexibility to get a better job, easier commute. There's a lot of mechanisms, but it is reassuring to see one of the primary mechanisms supported by our exploratory mediation analysis.

Patrick Sullivan 12:18

Thanks. So we're gonna move now to a section called Behind the Paper. And this is just trying to share some of what we don't always see when we read the paper, which is, and I think a lot of, I was gonna say earlier career, but earlier in later career, colleagues, you know, wonder like, how do we come up with these ideas? How do you like move them forward? So maybe just starting with what was the inspiration behind this analysis? Like, how did you land on researching this particular intervention of guaranteed income?

Leah Hamilton 12:47

I'll start there, but I think that we probably all have our own story. So I am a social worker who worked in foster care for many years, and I saw all the ways that social policy could do better for families and was making it more difficult for families to get ahead. I started researching our social assistance policies, TANF, SNAP, and trying to determine if those were not helping families get on their feet, do what they needed to do for their families. And the short answer is that they don't. And so that really brought me to this idea of what if we just gave people cash and trusted people to do what they needed to do with it. And at about the same time, the field of guaranteed income was exploding. This was in the wake of first the Great Recession of 2008 and then later COVID when the federal government was, you know, pushing money out the door as fast as possible. And a lot of people were beginning to say that we, well, have been saying for decades, but more people are realizing that the restrictions we put on public assistance are used to are based in the assumptions that poor people will misuse public funds. And we have decades of research saying that that's not the case, but those restrictions both increase bureaucracy and make it harder for families to get ahead. Just one really quick example is something called the Cliff effect. So for a lot of programs, if you make, you know, 10 cents over the income limit, you're gonna lose all benefits, you're gonna lose child care assistance, SNAP, Medicaid. So a lot of times families have to make really difficult decisions. Am I gonna feed my children or am I gonna take this 10 cent raise at work, right? We shouldn't be putting families in those positions. And guaranteed income kind of grew out of this idea that we need to trust people to make the best decisions for their own families. And then about that time, I received some incredible opportunities to be able to evaluate this huge growth of pilots. In the last few years, there's been about 150 guaranteed income pilots across the country, fueled both privately and publicly by cities and states. And so this has been a really great opportunity to look at what actually happens when you trust people. But I'd love to hear my colleagues kind of what brought them to this work.

Stephen Roll 15:00

I'm happy to jump in next. So I've been on this journey for a while, largely with Leah. I actually came to this work through work I was doing around the COVID-19 pandemic. I had developed this longitudinal survey looking at or examining these two tensions, right? On the one hand, the pandemic was disproportionately harming low-income folks, frontline workers, black and brown folks economically, socially, and in terms of their health, on the one hand. On the other hand, the federal government in some ways had finally started getting serious about building a functioning social safety net. Cash was going out quickly to families. They had radically expanded the eligibility criteria for a wide variety of programs, unemployment insurance, SNAP, and so on. And so I was using the survey to explore these tensions. And then, you know, Leah sort of in parallel was working on these issues around cash transfers. And a colleague of ours actually introduced us to see if I would be interested in incorporating just a couple questions about cash transfers, particularly the expanded child tax credit, into the survey. And so that's how Leah and I got connected. And then we built that into this broader research agenda around the child tax credit expansion in 2021, which was the closest the US has ever come to providing unconditional cash at scale for families. And that was a very successful collaboration that then I think evolved into this broader collaboration around a variety of guaranteed income programs all around the country, in Georgia, in New York, here where I'm based, in St. Louis, and so on.

Kate Stanhope 16:41

Well, I'm I'm trained in in public health and epidemiology. And so I was aware of guaranteed income and I'd seen it in global settings a lot, kind of unconditional cash transfers or conditional cash transfers, which are used in maternal health in a lot of global settings. I hadn't seen a lot in the US. And then in my postdoc, I was I mostly work in maternal health, and I was part of evaluating many interventions that were designed to improve the lives of low-income pregnant or postpartum women, but actually asked them to do a lot more while they were experiencing all these social and economic barriers. So we were asking them to come back to the hospital more with their new baby, even though they didn't have a car. We were asking them to come to more appointments, even though they had to return to work maybe, maybe a week after they gave birth. And so I was really frustrated with this disconnect between we're trying to address well-known, I mean, just income and race-based disparities in maternal health by asking, like by putting the burden on the marginalized people who are experiencing the burden of outcomes. And it, I'm continuing to be frustrated with that. Anyway, so I heard about guaranteed income and I was like, well, why aren't we testing this? Like, why aren't we doing this where we're saying meet your needs in the way you think you need to meet them? And then we can talk about bringing you in for extra care or an extra program. So that's kind of where I came to it, is kind of this this desire to think more about how we intervene upstream for maternal health disparities.

Patrick Sullivan 18:05

Great. I'm trying to think about where we, I mean, we've sort of like gotten through a lot of the components of this. So just in terms of what else you feel like you want to do and is important to convey, I think you've really gotten to the key findings already. There's this question about whether any of them surprised you, or were they largely what you expected going into the analysis?

Stephen Roll 18:29

In effect, the findings were largely what we expected. I mean, it's worth noting that for the majority of US households, finances are the single biggest source of stress and anxiety they have. And that's true across U.S. households generally, so we would expect that to be even more true for households in poverty, which the majority of our participants were. And so by providing these cash transfers, right, we're directly addressing what is likely the largest source of stress and anxiety for these households. So I think that the fact that it then translated into improvements in sleep quality in particular is very encouraging and not surprising. But what I also wanted to know is this documenting this is very important. To my knowledge, there has been no other study, at least in the US, on the impacts of guaranteed income and unconditional cash transfers on sleep, right? On sleep quality, on sleep length. So documenting this is very important, especially given all the other downstream positive health effects that are associated with improved sleep.

Patrick Sullivan 19:42

Great. So I'll go around for one last round of any final or concluding thoughts that you have. And Dr. Stanhope, can we start with you?

Kate Stanhope 19:51

Yeah. Um, so I think one of the things I teach social epidemiology trainees, and one of the traps I think trainees and others can sometimes get stuck in with social epidemiology is kind of being stuck almost just describing differences or disparities across axes and marginalization. And, you know, we feel like things aren't well suited to causal inference or we can't do randomized trials. And I think I would contend that changes to the social environment happen constantly. And some are like this, they're a randomized experiment that we can evaluate. So we have to seek out kind of these sorts of partnerships where we can contribute our expertise because I think this is how we move the needle on social epidemiology, but more over on health and health disparities, not just continuing to document disparities.

Leah Hamilton 20:35

Thank you so much, Kate. From someone who cares about social policy and social inequity, what's really striking to me about these results is, you know, sleep is has been well documented to be incredibly expensive to the American economy through increased cardiovascular disease, healthcare costs, lost productivity. And this is just one finding among many. We have another paper out just recently in health affairs on improvements in nutrition, which we know have huge impacts on health outcomes. And we have lots of other outcomes that have been documented to be incredibly huge cost savers in the long run to folks' both health and their long-term economic trajectory. So I really think it's important to continue documenting and thinking about how it, how much money we're saving by intervening early instead of after, you know, health problems and lost productivity have arisen and all the downstream effects to children. And then the final thing I'll say is that I know that a lot of listeners are probably thinking, guaranteed income, that's pie in the sky. That's never, you know, that's never going to happen. But there are really promising movements policy wise that are happening in 2021. The federal government experimented with monthly child transfers to almost every family in the country through the expanded child tax credit. There's less movement now at the federal level, but states are doing incredible things. Things right now to expand their family benefits, child tax credits, earned income tax credits. So, you know, if you're interested in learning more about this, there are you know policy mechanisms that are real-

Stephen Roll 22:12

One element that Leah mentioned, which is around the costs and the returns of these programs. So one comment I often give, especially in talking to the media about these types of programs, is okay, well, you found these effects. But these are very expensive programs, right? And it is true. Providing cash at scale to a lot of people is a pretty big line item in any budget. But the cost savings of doing so likely dwarf the actual costs of these programs. And you can think about this just in terms of health costs alone. Think about the health savings, are that the healthcare cost savings of expanded healthcare access, being able to afford the prescriptions you need, being able to afford the doctor's visit, the healthcare benefits of an avoided eviction, an avoided spell of child homelessness, and all the associated downstream costs. Yes, to that child's development, yes to that child's long-term economic trajectories, which are very important, but also to that child's health. Think about the costs of being forced to live in an area with a wide variety of environmental pollutants or other risks to families, and about the benefits of being able to afford to move out of those environments to someplace that is safer and more free of these environmental harms, right? So all of these things carry a huge amount of benefits that manifest over the course of people's entire lives, from childhood to adulthood to late in life. And so one challenge I often put to policymakers or to the media or other folks that raise this is yes, fighting poverty is expensive, but the costs of not fighting poverty and not fighting poverty through direct and effective tools like guaranteed income is actually an order of magnitude greater than the cost of actually fighting it.

Patrick Sullivan 24:04

Thank you for that. I just want to thank you all again for this kind of multi-sectoral collaboration that really provides the evidence, Stephen, of what you were just talking about. And I think we have to get to that point of you know, summarizing this in scientifically compelling ways for it to have policy leverage. So thanks to each of you for your role in this work, for bringing this work to Annals of Epidemiology, and for taking time today to give us a little bit more perspective and insight into how you did this work and what the implications of it are. So that brings us to the end of this episode. Thanks everyone here in this discussion for joining us today. It was just such a pleasure to be able to talk with you about your work. Thank you. I'm your host, Patrick Sullivan. Thanks for tuning in to this episode and see you next time on EPItalk, brought to you by Annals of Epidemiology, the official journal of the American College of Epidemiology. For a transcript of this podcast or to read the article featured on this episode and more from the journal, you can visit us online at www.annalsof epidemiology.org.