Paper of the Year 2024: An Interview With Craig Pollack
Article: “Using the Moving to Opportunity Experiment to Investigate the Long-Term Impact of Neighborhoods on Healthcare Use by Specific Clinical Conditions and Type of Service”
Authors: Craig Evan Pollack, Debra G. Bozzi, Amanda L. Blackford, Stefanie DeLuca, Rachel L. J. Thornton & Bradley Herring
Published Online: 10 Sept 2021
DOI: https://doi.org/10.1080/10511482.2021.1951804
Below is a transcript of our interview with Paper of the Year winner and research lead Craig Pollack, discussing the research process for him and his team while conducting the study documented in "Using the Moving to Opportunity Experiment to Investigate the Long-Term Impact of Neighborhoods on Healthcare Use by Specific Clinical Conditions and Type of Service”,
Claudia Aiken:
First let me say, Craig, thank you so much for joining us and congratulations again to you and your team on winning HPD’s Paper Of The Year for your article “Using the Moving to Opportunity Experiment to investigate the Long Term Impact of Neighborhoods on Healthcare Use”. To to kick things off, could you tell us a little bit about your own interest in linking
housing and health, and how you came to this kind of area of research?
Craig Pollack:
Thank you so much for talking with me about the research. On behalf of our team, we’re honored to receive the award.
As a primary care physician, I come to this from my clinical perspective working with a wide variety of patients and seeing the ways that their housing situation, their housing stability impacts their health. So much of what happens in the clinical encounter is shaped by where they live including whether they're able to come to their clinic appointments, whether they're able to afford their medicines, if they're able to take their medicines as prescribed, their ability to engage in healthy behaviors, and who they're able to live with in terms of the social support.
Julia Sands:
In your paper, you and your co-authors begin by laying out some of the existing research that looks at how our homes and neighborhoods affect our health. Could you also describe the two main challenges for researchers who want to test these connections?
Craig Pollack:
So there's a really long history of studying the connection between housing and health as exemplified by Florence Nightingale and the tenements in New York City. There’s just a huge body of research to draw upon which in recent years has really grown. I think there are several challenges in trying to study the connection between housing and health. The first main challenge is that people's decisions about where they live are not random. Where we choose to live is due to a wide variety of factors around our family structure, what we can afford, what our preferences are, and these are often shaped by historical and contemporary structural racism, discrimination. These factors that shape where we live may also be related to our underlying health status, our use of healthcare. The confounding that exists can be really hard to grapple with in studies. The second main challenge that we've seen in a lot of the studies that look at the impacts of housing and neighborhoods on health is that they often focus on shorter term outcomes. To look at health over a really long period of time, especially as, for example, children age and grow up, there's a real paucity of data.
Claudia Aiken:
Well, to overcome these two challenges that you've just told us about. you and your co-authors harness the Moving to Opportunity experiment. So for those of our readers who might not be as familiar with that, could you briefly, briefly explain what that experiment was?
Craig Pollack:
The Movement to Opportunity experiment grew out of these challenges to systemic racism, where tenants and lawyers worked together to try to counter the factors that led poor, underserved families to be constrained to living in under-resourced neighborhoods. Based on some promising findings from the results of Fair Housing lawsuits including, for example, Gautreaux in Chicago, HUD created this demonstration project called Moving to Opportunity. Moving to Opportunity essentially was a randomized control trial, which is really unique for a social experiment. About 4,600 families living in public housing in high poverty neighborhoods across 5 different cities were randomized to different study arms. The first was a control arm, the second was a traditional voucher arm, a Section 8. Voucher, which didn't have restrictions on where families could move, and then the 3rd was a voucher that needed to be used in lower poverty, more resourced neighborhoods. Those different study arms are really important in addressing some of the kind of confounding that I I talked about earlier that was such a limitation on prior studies linking housing and neighborhood environments and health.
Initially MTO was set up to look at factors like economic self-sufficiency, and health wasn't really part of the equation. But over time the investigators realized that health was an important component of things that was potentially changing with households across different study groups, so people were able to look at the health effects in the shorter and longer term.
Julia Sands:
Why did the study team decide to keep probing beyond your previous work, which found reduced rates of hospitalization among children whose family received a voucher, but no differences in emergency room visits?
Craig Pollack:
We were really interested in trying to kind of delve deeper into those findings, to understand what exactly was changing with healthcare use. There were some really significant and striking findings with respect to children’s hospitalization, but that was overall hospitalization. Children get admitted to the hospital for a wide range of different reasons, and that some of those might be more closely tied to where they live than others. In addition to the specific conditions that people might get admitted to the hospital for, they may also receive different types of services within and outside of the hospital setting that may be tied to their neighborhood.
Claudia Aiken:
Yeah, that makes total sense. I think one of the, you know, awesome and kind of intimidating things about your study, was that you were able to link these MTOdata with 21 years worth of hospital discharge and Medicaid data. Can you tell us just a little bit more about the choice to use those particular data, and then any challenges that you might have faced in accessing them and then linking the data sets?
Craig Pollack:
This process took a long time to successfully link the data, and it required negotiating with the HUD around the MTO data, as well as with each of the states regarding the health data. Some of the health data state all payer data, which gives us information about every time somebody was admitted to the hospital or use an emergency department, regardless of the type of insurance that they have, and then some of the data that we used was Medicaid. We are really grateful for the research partnership with had with HUD and for the state agencies being willing to work with us on it.
The reason we wanted to use this data is because, first of all, it provides an objective measure of healthcare use, so we were not relying on surveys where people may misremember and where there might be non-response to the survey in ways that can bias the findings. It allows us to look over a really long period of time to do so in an efficient manner.
Julia Sands:
Our next question is about the academic and media reception around Moving To Opportunity, and how your study contributes to that. The moving to opportunity study is frequently discussed in academic circles as well in the broader media landscape related to housing. And it's at the center of a debate about whether it is better to try to invest in poor neighborhoods and communities or to help them relocate into lower poverty neighborhoods. How do you think your work fits into this debate?
Craig Pollack:
First of all, when we started this project, I think there was some skepticism that we would see a significant effect, because advocates understood that Moving to Opportunity was that it wasn't as effective as an experiment as contemporary mobility programs. A lot of households in MTO who received vouchers weren't able to use them effectively to lease up in non-poor neighborhoods, and those that were able to lease in non-poor neighborhoods didn't always remain living in those neighborhoods for a long period of time. Raj Chetty's study found that children who moved, especially those at young ages, had higher earnings as adults created a lot more interest in the study and around neighborhood factors more generally.
Claudia AIken:
That makes sense. Sso along those lines, at HPD we really think of policymakers as our audience. I think this paper is super policy relevant. Could you talk about how you think your findings inform policy? For example, how might they relate to the kind of new ability to use Medicaid waivers to address certain housing needs, for example. And then are there any takeaways for the general public as well?
Craig Pollack:
It's such an important question, how does the research we do matter? We wanted to try to get the word out to people in the housing and policy space which is one of the reasons why we're so excited to publish in Housing Policy Debate. We hope that better understanding the health impacts of Moving To Opportunity can underscore the urgency of helping increase access to opportunity neighborhoods for families with vouchers.
We also think the study has implications for people in the health sphere where there has been increasing interest in the healthcare system investing in housing in different ways. There's a lot of experimentation going on in that sphere, for example, with Medicaid 1115 waivers that can be used to pay for short term housing assistance and the like. Our hope is that this type of work can kind of help create the case that healthcare systems should be looking at these types of approaches.
I want to have a word of caution about that, though, because I think there’s a strong argument to be made as to why it matters for getting people stably housed in neighborhoods that are resourced, and sometimes in the healthcare world that can be twisted into “It's going to save healthcare dollars”. That's a very high bar for healthcare interventions. Most health care interventions don't actually end up saving money, but they still can be very valuable for individuals' well-being and is ethically the right thing to do.
And then third, you asked about the general public, and we hope that this adds to the growing incentive for the general public to promote having affordable housing in high opportunity, well-resourced neighborhoods.
Claudia Aiken:
Great. Such a good point, too, about how often we frame these questions around health care savings. But that's really not always the best metric, I'm sure.
Craig Pollack:
Yes, there are instances where there can be real benefits to healthcare’s bottom line and these cases tend to focus on the high use, high cost individuals. These are incredibly important populations to focus on. But we also need to make sure we’re remembering other populations, including children where there can be real compounding effects on health over the long term.
Julia Sands:
How can other researchers expand upon your findings? What particular challenges are there to research at the intersection of health and housing that you hope will be overcome? I know that you talked a little bit about the challenges of linking the data between Medicare costs and the Moving to Opportunity study. Do you have any ideas on how that could be improved for researchers in the future?
Craig Pollack:
I think that we've been excited that there's been a move towards a greater linkage of data. So, for example, we've been working with HUD to link their data to cancer registry and Medicare claims called the Seer Medicare data. That data is now available for researchers to use. I think that's an example of a place where it takes a lot of upfront costs and time to create this research platform that, we hope, can yield a tremendous amount of insight as to the connection between housing and health at a broad scale. I think that type of data linkage can be incredibly important for helping us get the contours and understand the landscape.
I also think there's a real need to understand some of the mechanisms that are underlying these associations. For example, in the case of asthma, we've been doing a lot of work in Baltimore with the housing mobility program, trying to understand what are the factors, whether those are home exposures, the amount of stress, outdoor air pollution, you name it that may be kind of underlying some of the changes that that we're observing. The goal is that by understanding those pathways better, we’ll be able to help serve these families and reduce the burden of asthma.
Article: “Using the Moving to Opportunity Experiment to Investigate the Long-Term Impact of Neighborhoods on Healthcare Use by Specific Clinical Conditions and Type of Service”
Authors: Craig Evan Pollack, Debra G. Bozzi, Amanda L. Blackford, Stefanie DeLuca, Rachel L. J. Thornton & Bradley Herring
Published Online: 10 Sept 2021
DOI: https://doi.org/10.1080/10511482.2021.1951804
Below is a transcript of our interview with Paper of the Year winner and research lead Craig Pollack, discussing the research process for him and his team while conducting the study documented in "Using the Moving to Opportunity Experiment to Investigate the Long-Term Impact of Neighborhoods on Healthcare Use by Specific Clinical Conditions and Type of Service”,
Claudia Aiken:
First let me say, Craig, thank you so much for joining us and congratulations again to you and your team on winning HPD’s Paper Of The Year for your article “Using the Moving to Opportunity Experiment to investigate the Long Term Impact of Neighborhoods on Healthcare Use”. To to kick things off, could you tell us a little bit about your own interest in linking
housing and health, and how you came to this kind of area of research?
Craig Pollack:
Thank you so much for talking with me about the research. On behalf of our team, we’re honored to receive the award.
As a primary care physician, I come to this from my clinical perspective working with a wide variety of patients and seeing the ways that their housing situation, their housing stability impacts their health. So much of what happens in the clinical encounter is shaped by where they live including whether they're able to come to their clinic appointments, whether they're able to afford their medicines, if they're able to take their medicines as prescribed, their ability to engage in healthy behaviors, and who they're able to live with in terms of the social support.
Julia Sands:
In your paper, you and your co-authors begin by laying out some of the existing research that looks at how our homes and neighborhoods affect our health. Could you also describe the two main challenges for researchers who want to test these connections?
Craig Pollack:
So there's a really long history of studying the connection between housing and health as exemplified by Florence Nightingale and the tenements in New York City. There’s just a huge body of research to draw upon which in recent years has really grown. I think there are several challenges in trying to study the connection between housing and health. The first main challenge is that people's decisions about where they live are not random. Where we choose to live is due to a wide variety of factors around our family structure, what we can afford, what our preferences are, and these are often shaped by historical and contemporary structural racism, discrimination. These factors that shape where we live may also be related to our underlying health status, our use of healthcare. The confounding that exists can be really hard to grapple with in studies. The second main challenge that we've seen in a lot of the studies that look at the impacts of housing and neighborhoods on health is that they often focus on shorter term outcomes. To look at health over a really long period of time, especially as, for example, children age and grow up, there's a real paucity of data.
Claudia Aiken:
Well, to overcome these two challenges that you've just told us about. you and your co-authors harness the Moving to Opportunity experiment. So for those of our readers who might not be as familiar with that, could you briefly, briefly explain what that experiment was?
Craig Pollack:
The Movement to Opportunity experiment grew out of these challenges to systemic racism, where tenants and lawyers worked together to try to counter the factors that led poor, underserved families to be constrained to living in under-resourced neighborhoods. Based on some promising findings from the results of Fair Housing lawsuits including, for example, Gautreaux in Chicago, HUD created this demonstration project called Moving to Opportunity. Moving to Opportunity essentially was a randomized control trial, which is really unique for a social experiment. About 4,600 families living in public housing in high poverty neighborhoods across 5 different cities were randomized to different study arms. The first was a control arm, the second was a traditional voucher arm, a Section 8. Voucher, which didn't have restrictions on where families could move, and then the 3rd was a voucher that needed to be used in lower poverty, more resourced neighborhoods. Those different study arms are really important in addressing some of the kind of confounding that I I talked about earlier that was such a limitation on prior studies linking housing and neighborhood environments and health.
Initially MTO was set up to look at factors like economic self-sufficiency, and health wasn't really part of the equation. But over time the investigators realized that health was an important component of things that was potentially changing with households across different study groups, so people were able to look at the health effects in the shorter and longer term.
Julia Sands:
Why did the study team decide to keep probing beyond your previous work, which found reduced rates of hospitalization among children whose family received a voucher, but no differences in emergency room visits?
Craig Pollack:
We were really interested in trying to kind of delve deeper into those findings, to understand what exactly was changing with healthcare use. There were some really significant and striking findings with respect to children’s hospitalization, but that was overall hospitalization. Children get admitted to the hospital for a wide range of different reasons, and that some of those might be more closely tied to where they live than others. In addition to the specific conditions that people might get admitted to the hospital for, they may also receive different types of services within and outside of the hospital setting that may be tied to their neighborhood.
Claudia Aiken:
Yeah, that makes total sense. I think one of the, you know, awesome and kind of intimidating things about your study, was that you were able to link these MTOdata with 21 years worth of hospital discharge and Medicaid data. Can you tell us just a little bit more about the choice to use those particular data, and then any challenges that you might have faced in accessing them and then linking the data sets?
Craig Pollack:
This process took a long time to successfully link the data, and it required negotiating with the HUD around the MTO data, as well as with each of the states regarding the health data. Some of the health data state all payer data, which gives us information about every time somebody was admitted to the hospital or use an emergency department, regardless of the type of insurance that they have, and then some of the data that we used was Medicaid. We are really grateful for the research partnership with had with HUD and for the state agencies being willing to work with us on it.
The reason we wanted to use this data is because, first of all, it provides an objective measure of healthcare use, so we were not relying on surveys where people may misremember and where there might be non-response to the survey in ways that can bias the findings. It allows us to look over a really long period of time to do so in an efficient manner.
Julia Sands:
Our next question is about the academic and media reception around Moving To Opportunity, and how your study contributes to that. The moving to opportunity study is frequently discussed in academic circles as well in the broader media landscape related to housing. And it's at the center of a debate about whether it is better to try to invest in poor neighborhoods and communities or to help them relocate into lower poverty neighborhoods. How do you think your work fits into this debate?
Craig Pollack:
First of all, when we started this project, I think there was some skepticism that we would see a significant effect, because advocates understood that Moving to Opportunity was that it wasn't as effective as an experiment as contemporary mobility programs. A lot of households in MTO who received vouchers weren't able to use them effectively to lease up in non-poor neighborhoods, and those that were able to lease in non-poor neighborhoods didn't always remain living in those neighborhoods for a long period of time. Raj Chetty's study found that children who moved, especially those at young ages, had higher earnings as adults created a lot more interest in the study and around neighborhood factors more generally.
Claudia AIken:
That makes sense. Sso along those lines, at HPD we really think of policymakers as our audience. I think this paper is super policy relevant. Could you talk about how you think your findings inform policy? For example, how might they relate to the kind of new ability to use Medicaid waivers to address certain housing needs, for example. And then are there any takeaways for the general public as well?
Craig Pollack:
It's such an important question, how does the research we do matter? We wanted to try to get the word out to people in the housing and policy space which is one of the reasons why we're so excited to publish in Housing Policy Debate. We hope that better understanding the health impacts of Moving To Opportunity can underscore the urgency of helping increase access to opportunity neighborhoods for families with vouchers.
We also think the study has implications for people in the health sphere where there has been increasing interest in the healthcare system investing in housing in different ways. There's a lot of experimentation going on in that sphere, for example, with Medicaid 1115 waivers that can be used to pay for short term housing assistance and the like. Our hope is that this type of work can kind of help create the case that healthcare systems should be looking at these types of approaches.
I want to have a word of caution about that, though, because I think there’s a strong argument to be made as to why it matters for getting people stably housed in neighborhoods that are resourced, and sometimes in the healthcare world that can be twisted into “It's going to save healthcare dollars”. That's a very high bar for healthcare interventions. Most health care interventions don't actually end up saving money, but they still can be very valuable for individuals' well-being and is ethically the right thing to do.
And then third, you asked about the general public, and we hope that this adds to the growing incentive for the general public to promote having affordable housing in high opportunity, well-resourced neighborhoods.
Claudia Aiken:
Great. Such a good point, too, about how often we frame these questions around health care savings. But that's really not always the best metric, I'm sure.
Craig Pollack:
Yes, there are instances where there can be real benefits to healthcare’s bottom line and these cases tend to focus on the high use, high cost individuals. These are incredibly important populations to focus on. But we also need to make sure we’re remembering other populations, including children where there can be real compounding effects on health over the long term.
Julia Sands:
How can other researchers expand upon your findings? What particular challenges are there to research at the intersection of health and housing that you hope will be overcome? I know that you talked a little bit about the challenges of linking the data between Medicare costs and the Moving to Opportunity study. Do you have any ideas on how that could be improved for researchers in the future?
Craig Pollack:
I think that we've been excited that there's been a move towards a greater linkage of data. So, for example, we've been working with HUD to link their data to cancer registry and Medicare claims called the Seer Medicare data. That data is now available for researchers to use. I think that's an example of a place where it takes a lot of upfront costs and time to create this research platform that, we hope, can yield a tremendous amount of insight as to the connection between housing and health at a broad scale. I think that type of data linkage can be incredibly important for helping us get the contours and understand the landscape.
I also think there's a real need to understand some of the mechanisms that are underlying these associations. For example, in the case of asthma, we've been doing a lot of work in Baltimore with the housing mobility program, trying to understand what are the factors, whether those are home exposures, the amount of stress, outdoor air pollution, you name it that may be kind of underlying some of the changes that that we're observing. The goal is that by understanding those pathways better, we’ll be able to help serve these families and reduce the burden of asthma.
|
Whether you're an author, a practitioner, a student, or just someone who cares about housing - we would love to hear from you.
Email us at [email protected]. |
FOLLOW HPD:
|