Subscribe
Share
Share
Embed
Each week, Health Affairs' Rob Lott brings you in-depth conversations with leading researchers and influencers shaping the big ideas in health policy and the health care industry.
A Health Podyssey goes beyond the pages of the health policy journal Health Affairs to tell stories behind the research and share policy implications. Learn how academics and economists frame their research questions and journey to the intersection of health, health care, and policy. Health policy nerds rejoice! This podcast is for you.
00;00;00;00 - 00;00;37;21
Alan Weil
Hello and welcome to “A Health Podyssey”. I'm your host, Alan Weil. Heart failure is a leading cause of hospital admissions and there are striking racial disparities in its incidents. Black men and black women die of heart failure at rates about 50% higher than white men and white women, respectively. The American Heart Association notes that the ability to properly manage heart failure is significantly affected by the social determinants of health and the costs of treatment are often a barrier to receiving proper care.
00;00;38;18 - 00;01;13;08
Alan Weil
What is the cost of racial and ethnic disparities in heart failure? And what can we do about those disparities? That's the topic of today's episode of “A Health Podyssey”. I'm here with Andrew Anderson, assistant professor in the Department of Health Policy and Management at the Tulane School of Public Health and Tropical Medicine. Dr. Anderson and colleagues published a paper in the May 2023 issue of Health Affairs assessing financial costs associated with disparities in preventable heart failure hospitalizations among Medicare beneficiaries in the US South.
00;01;14;00 - 00;01;36;28
Alan Weil
They found significant disparities in preventable heart failure hospitalizations between black, Hispanic, American Indian, Alaskan Native beneficiaries and white beneficiaries with tens of millions of dollars of costs each year associated with these disparities just in the South. We'll discuss these findings in today's episode. Dr. Anderson, welcome to the program.
00;01;37;17 - 00;01;40;09
Andrew Anderson
Thanks for inviting me to do this. I'm excited for the conversation.
00;01;41;03 - 00;01;55;15
Alan Weil
I'm looking forward to learning more about this subject with you. Let's start with just sort of the motivation for the study. You focused on racial and ethnic disparities related to heart failure. Why did you select this condition as the focus of your work?
00;01;56;06 - 00;02;19;13
Andrew Anderson
Yeah. So first I would just say this work is kind of couched in a broader literature and cost of illness. But there's been far fewer studies that have been focused on cost of illness related to disparities. And we know that the United States spends tons of money on health services. But the overarching question here is how much would we save if we could use this for other purposes if we eliminated racial and ethnic disparities?
00;02;20;13 - 00;02;47;28
Andrew Anderson
So there's been lots of studies before this that have looked at disparities that have cost billions of dollars, lost productivity, lost lives, direct medical costs. And so the reason why we specifically focused on heart failure admissions and preventable admissions is because of the disproportionate mortality among black Americans. In 2017, I think it was around one and a half times in the South and throughout the United States.
00;02;47;28 - 00;03;18;12
Andrew Anderson
It's also a leading cause of admissions and readmissions, which is also extremely costly. Nearly two thirds of the annual per patient cost for heart failure is actually driven by hospital admissions. And so the main aim of this paper was to try to describe the costs associated with disparities between black and white Medicare beneficiaries. But we also looked at the differences for Hispanic Americans, Asian American and Pacific Islander, as well as American Indian, Alaska Native populations.
00;03;18;28 - 00;03;43;05
Alan Weil
Yeah, we'll get into those findings as we go along. So this is a high cost, high burden and high disparity illness. You looked at preventable heart failure hospitalizations. Now, that's a concept that maybe isn't immediately intuitive to people. When is a hospitalization preventable? And how do we think about that in terms of interpreting your results?
00;03;43;22 - 00;04;08;06
Andrew Anderson
Yeah. So the short answer to that question is that these are unscheduled admissions that are not due to a transfer from another facility or for a cardiac procedure, which would be probably the most of the admissions. A lot of them are actually planned. We defined preventable using the consensus definition developed by the Agency for Health Care Research and Quality, AHRQ, from their prevention quality indicators.
00;04;08;17 - 00;04;34;06
Andrew Anderson
And so preventable admission rates is more so measuring differences in access to care or quality of outpatient care in the community. It's a measure of population health. So these indicators are used for county health rankings and a bunch of other population health programs. It's also a really key tool for community health needs assessments. So we use the measure that's been used repeatedly in public reporting and payment programs.
00;04;35;16 - 00;04;56;12
Alan Weil
Makes a lot of sense. Thank you for that clarification. Now, as I was looking at the paper, your focus is on the south, the southern part of the United States. But as you note in the paper, we're missing a lot of the data for states, for some critical elements. And, you know, we don't typically talk a lot about data and methods on “A Health Podyssey”.
00;04;56;20 - 00;05;16;29
Alan Weil
But here, if we're going to try to understand this phenomenon, just having kind of what I would think of as fairly basic data would be useful. But here it's missing. Can you say a little bit about where you didn't have the data that might have been the right kind of data or the best data to answer the question you had in front of you?
00;05;17;02 - 00;05;17;15
Alan Weil
Yeah.
00;05;17;16 - 00;05;42;04
Andrew Anderson
So surprisingly, we weren't able to use every state in the South. So the US census, is about, depending on how you define it, 15 to 17 states. We looked at 15 states that met the broad definition of the South based on that definition. Six of these states, which included Alabama, Louisiana, where I am, Oklahoma, Tennessee, Texas and Virginia,
00;05;42;12 - 00;06;10;21
Andrew Anderson
they didn't consistently report their hospital discharge data, which is what we used, which was the state and patient databases. And we looked at between 2015 and 2017. Of the nine remaining states, four of them, which included South Carolina, West Virginia, Maryland and Delaware; they either didn't report data on ethnicity, the unique hospital identifiers that we actually needed to link up these databases and actually the point of origin of a hospitalization or zip codes.
00;06;10;24 - 00;06;36;10
Andrew Anderson
So we needed all of these data estimates to actually do the calculations. So once we actually did all of the merging, we were only able to really focus on six states, which included Kentucky, Arkansas, Florida, Georgia, Mississippi and North Carolina. And you can immediately tell these are states that are demographically different in some cases. Right? So Mississippi and Florida definitely have different histories and also different populations.
00;06;36;15 - 00;06;44;09
Andrew Anderson
And so what we did was we tried to adjust for some of the differences in those populations when we calculated these heart failure admission rates.
00;06;44;20 - 00;07;02;00
Alan Weil
Yeah. So, I mean, you used, I guess, what we'd call sort of standard techniques to account for the fact that you didn't have everything you needed. But I do think it's just an important story, even separate from your findings about how many of the states just didn't have the information you might have wanted. Well, let's move to the findings.
00;07;02;01 - 00;07;15;14
Alan Weil
You focused on the cost of preventable admissions and the particularly the cost associated with disparities in those. Can you just give us sort of a top level of the findings from the research?
00;07;15;18 - 00;07;42;13
Andrew Anderson
Yeah. So we found over $600 million in annual excess spending across all of the groups that we looked at. And these are specifically attributed to the Medicare program. So we could have included Medicaid, commercial payors, but we really focused on Medicare. And then we, the headline is we found nearly half of the excess admissions within this cost for black beneficiaries were attributable to the disparity.
00;07;42;18 - 00;08;18;00
Andrew Anderson
Nearly 15% was for Hispanic beneficiaries and nearly 51% for American Indian and Alaska Native. And we used the reference group for white as white beneficiaries rates, which had similar demographic characteristics. And so I guess the main takeaway is that this equates to millions of dollars in excess spending for the Medicare program due to disparities in outcomes. The one other thing that I would want to mention is that we made some measurement choices here which actually apply to all disparities research where we're choosing a reference group, in this case Asian and Pacific Islander.
00;08;18;09 - 00;08;43;15
Andrew Anderson
We weren't able to actually disaggregate these distinct groups, unfortunately, but they actually had the lowest heart failure admission rate and we could have chosen them as a benchmark. But we decided to choose the white beneficiaries as a reference group because we wanted our estimates to be more comparable to previous estimates and because of the historical and present day contrast and opportunities and resources due to racism in the South as well as other parts of the United States.
00;08;44;14 - 00;09;07;29
Andrew Anderson
And then another, we could have also used the average rate across all groups. So there are multiple ways of doing this. But the takeaway here, and we did some sensitivity analysis, is that we would still find disparities regardless of the, in some cases, the disparities would be larger. In some cases it might be slightly smaller. But either way, the differences are large enough that it would lead to excess cost.
00;09;07;29 - 00;09;12;03
Andrew Anderson
And as a reminder, all of these are preventable under our definition.
00;09;12;15 - 00;09;30;22
Alan Weil
Yeah. So first of all, I just want to thank you for highlighting the thinking that went into the reference group. As you, I'm sure know, often people sort of default to white as the comparison group, the reference group, and don't give it much thought and it doesn't mean it's the wrong answer, but it's not the right answer if you haven't thought about why.
00;09;30;22 - 00;09;53;12
Alan Weil
And so I appreciate you going through the variety of options that you had there and explaining your choice. Before we take a break and then we'll come back and talk a little bit about what to do about this, I do want to just because it is a somewhat complex concept, I do want to spend a moment with you on this notion of the share of preventable admissions due to disparities.
00;09;53;13 - 00;10;05;14
Alan Weil
So you have the full universe of admissions that were preventable. And how do you say, and these are the share that are due to disparities.
00;10;05;25 - 00;10;32;23
Andrew Anderson
Good question. So this goes back to the reference group. If we saw that for certain groups, for black beneficiaries, for Hispanic and American Indian, Alaska Native beneficiaries, their adjusted heart failure admission rate was higher than for white beneficiaries. So we calculated the difference based on the assumption if they had the same rate of as white beneficiaries and if they had the same rate, what would be the savings?
00;10;33;16 - 00;11;08;17
Alan Weil
Well, I really want to talk about sort of what we know about where these disparities come from and what we can do about them. We'll have that conversation after we take a short break. And we're back. I'm speaking with Dr. Andrew Anderson about the cost of disparities in preventable heart failure hospitalizations in the southern region of the United States.
00;11;09;08 - 00;11;35;24
Alan Weil
Before the break, we got the top line findings. Tens of millions of dollars in excess spending associated with disparities around preventable admissions. At the introduction of the episode, I mentioned some information from the American Hospital Association that talks about some of the structural factors in the health system and outside of health care that can lead to the disparities that you measure in your paper.
00;11;36;11 - 00;11;53;20
Alan Weil
What can you tell us about how health and health care exist and function in the southern region of the United States? That, first of all, led to your inquiry in the paper? And second of all, that might contribute to the disparities that you documented.
00;11;54;01 - 00;12;22;11
Andrew Anderson
One of the reasons we focused on the South is, many people don't know some of the adjoining states in the South have been called the heart failure belt, which also overlaps with a region known as the Black Belt, which is known for its fertile soil and was also used for cotton and tobacco production in forced labor plantations. And so we were really interested in how some of that very present structural racism in the United States may actually extend out into the future
00;12;22;11 - 00;12;43;22
Andrew Anderson
and to today. And we know that the South is racially and ethnically diverse and home to a large share of the nation's people of color. Black Americans in the South account for a greater share of the southern population compared to other parts of the United States. And we also know that the rates of things like diabetes and cancer are also worse in the southern region.
00;12;44;09 - 00;13;07;23
Andrew Anderson
Southerners are also more likely to be uninsured compared to individuals living in other parts of the country. We all know that the South has been slower to expand Medicaid. Most states in the South have not, except Louisiana and Arkansas, and I think more recently, North Carolina is joining the group. We know that Medicaid and SCHIP (State Children’s Health Insurance Program) eligibility levels are more limited in the South compared to other regions.
00;13;07;23 - 00;13;32;28
Andrew Anderson
And among those states that I just mentioned that have expanded Medicaid, generally, they had lower uninsurance rates to begin with compared to pre 2014. So there's lots of chronic illness in the South and there's a lot of other demographic data that we could talk about when it comes to the outcomes that would contribute to this type of difference in terms of differences and opportunities and resources.
00;13;33;09 - 00;13;35;19
Andrew Anderson
So that's why we chose to focus on the South.
00;13;36;02 - 00;13;58;06
Alan Weil
So you find these high costs, of course, they're the health costs. You're measuring the financial cost of delivering services. The health costs to the population are profound. What's your sense of what could be done, particularly if we're looking at what many would call the social drivers? But you referenced, of course, the role of health insurance as well.
00;13;58;06 - 00;14;06;24
Alan Weil
When you think about the information you've gathered in the findings of this study, where does it lead you?
00;14;07;08 - 00;14;31;21
Andrew Anderson
So as I said in the beginning, the preventable hospital admissions are really signals of serious access or performance problems in health systems and communities. And you know, heart failure has some really well known screening, diagnostic and treatment procedures. The problem is that it's inconsistently implemented or there's lower uptake among black, Hispanic, American Indian and other groups that we looked at in the study.
00;14;32;06 - 00;14;56;21
Andrew Anderson
What it takes to close those gaps really is about getting more of the right care at the right time to the right people. And so there's an under utilization of guideline directed care among these populations. So in the paper, we talked about a study that was cited using registry data where less than a quarter of, they found less than a quarter of black patients with heart failure were actually receiving pharmacotherapies that they needed to manage their heart failure.
00;14;57;11 - 00;15;18;06
Andrew Anderson
So we know a lot about what interventions work to prevent and treat heart failure. And we know that these are the same interventions that work for everybody. The problem is that these strategies that are used might look slightly different to increase the uptake of those interventions. And those differences are based on the unique barriers that certain populations face on average.
00;15;18;17 - 00;15;44;17
Andrew Anderson
And it's that we're not overcoming those barriers and many of those barriers we already know when it comes to transportation in certain areas. One of the ones that we talked about in the paper was pharmacy deserts, which is, I think a paper that was published in Health Affairs, where in cities like Charlotte, North Carolina, Jacksonville, Memphis, they're disproportionately categorized as pharmacy deserts.
00;15;44;17 - 00;16;15;22
Andrew Anderson
So when it comes to low availability of pharmacies to begin with, the actual geographic accessibility of those pharmacies, that is going to further exacerbate the disparities and other barriers that might exist. And so because the United States has a history of racism and, you know, codified racial hierarchy, I mean, it's been repeatedly mentioned about racial segregation and neighborhood disinvestment and exclusionary zoning laws, especially in southern urban areas.
00;16;16;14 - 00;16;57;14
Andrew Anderson
You know, resources like good schools, clean water, safe neighborhoods, healthy foods, all of these things that are going to improve health and lead to thriving. You know, it's unequally distributed across racial lines. And so until that situation is fixed, right, that we're going to continue to see these types of disparities. But what we can do in the meantime in the health care system is to try to stop you know, actually contributing to the problem with discrimination and actual treatment and actually trying to you know, there is more of a movement now to address health related social needs within the health care system.
00;16;57;22 - 00;17;36;23
Andrew Anderson
That is a stopgap in some ways, because we're not addressing the upstream social determinants of health that I just mentioned. But it is something that can be done to try to stop the bleeding in the meantime. That's where we concluded on how to try to address some of the disparities is really improving the uptake of guideline directed care, addressing the social determinants of health, which will likely happen through a public policy in the long term, and then also training providers to try to address some of the mistreatment or lack of treatment that may be given to certain communities.
00;17;37;12 - 00;18;01;13
Alan Weil
Yeah, that seems really important. I mean, even though the origins of these problems are deep and longstanding and they arise outside of the health care system, part of what you're saying is even when you're just operating within the health care system, there are things you can do. They aren't going to solve everything, but they can mitigate some of the damage and they can address some of the disparities.
00;18;01;13 - 00;18;26;14
Alan Weil
And it's not going to solve the whole problem. But it's better than not solving at all or pretending it's not there and just leaving alone and saying, well, the origins of those problems aren't our doing, so we don't have any part to play in addressing them. I don't think anyone wants to take that approach. So I do think quantifying the burden is helpful in motivating change.
00;18;26;14 - 00;18;54;18
Alan Weil
It's only one aspect. It certainly doesn't lead to change immediately and doesn't always lead to change. But it is part of the arsenal of arguments people can use for why it's important to do something about the problems that led us to this state. Well, as we come to a close, I do want to note you were a fellow in our Health Equity Fellowship for Trainees, the HEFT program here at Health Affairs.
00;18;56;04 - 00;19;11;09
Alan Weil
I doubt most of our listeners are familiar with it. I'm not going to ask you to give a total overview of our program, but I wonder if you could just say a word about the role that that fellowship played in preparing this manuscript and getting you to a place where you're publishing it here in Health Affairs.
00;19;12;07 - 00;19;39;05
Andrew Anderson
The program had a great impact on getting this paper to where it is now. First, it was work that was unfunded, and I did this with a group of doctoral students in my department, fantastic group of students, and I was matched through this program with a cardiologist. Right? Which really helped me understand the clinical aspects of a heart failure and the treatment.
00;19;39;11 - 00;20;07;23
Andrew Anderson
And so that was really invaluable. And I was lucky to get that matching. And it led to a much more improved paper for me to have that clinical analysis. And we also got some really great tips through the matching with the associate editor, one of the associate editors at Health Affairs, to try to craft the narrative here in a way that policymakers could potentially actually want to use or understand how this information is usable.
00;20;08;04 - 00;20;24;15
Andrew Anderson
So it was a really great experience. I think the paper would not have maybe gotten published or wouldn't have gotten to this stage because there has been so much refinement over the last two and a half years actually, that we've done this work. So it's been a really great experience.
00;20;24;26 - 00;20;52;00
Alan Weil
Well, I'm happy to hear that and happy that it led to the publication of this manuscript and the little small part that we played in that. Well. Dr. Anderson, thank you so much for focusing on this critical issue, for sticking with the topic, even without funding, to find your way through the complex data limitations and challenges for really an exceptional explanation of why this is important work.
00;20;52;00 - 00;21;09;03
Alan Weil
And the results themselves really do speak for themselves about the scale of these disparities and the importance financially of addressing them, much less the humanitarian reasons for doing so. Dr. Anderson, thank you for being my guest today on “A Health Podyssey”.
00;21;09;22 - 00;21;10;17
Andrew Anderson
Thanks for having me.