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.
It goes without saying that everyone's experience in The US health care system is unique. And yet for researchers, policymakers, and population health professionals, their job is at a fundamental level to walk us back from so many singular data points and reveal the big picture. It's how we can issue sweeping statements about life expectancy and mortality rates, and it's how we can track the rise and fall of something like national health expenditures over time. National health expenditures representing a universe of more than three forty million Americans, 6,000,000 employers, 6,000 hospitals. Now, these all encompassing findings are interesting and useful, but they're not very nuanced, and they tend to hide significant variation below the surface, variation that may be even more interesting and even more useful than the top line numbers, especially when thinking about the most promising pathways to improve our system.
Speaker 1:What happens when we stratify those big numbers? What can we learn when we break them down according to other factors such as, for example, income? That's our question for today's episode of A Health Podyssey. I'm here with Doctor. Betsy Q.
Speaker 1:Cliff, assistant professor at the University of Chicago Department of Public Health Sciences. Together with her coauthor, Giacomo Miele, from the American Board of Internal Medicine, she has a new article in the July issue of Health Affairs looking at, quote, trends in US health care spending by income, 02/2023. This is really fascinating stuff, and I can't wait to dive in. Doctor. Betsy Q.
Speaker 1:Cliff, welcome to A Health Podyssey.
Speaker 2:Well, thanks, Rob. Thank you so much for having me. I'm really excited to be here.
Speaker 1:Well, great to have you. And why don't we just dive right in? Your paper is about health care spending and the sort of starting point that you offer is, of course, that spending growth in The US slowed in the mid-2000s. By way of background, can you describe what we know from previous research about the factors driving that slowdown?
Speaker 2:Yeah, absolutely. So there's actually been some good research published about what's caused the slowdown, not necessarily by our team, but by some others. And it's not one thing. It's more of sort of a constellation of factors. So what people have found is that procedures that, you know, there's a few things going on.
Speaker 2:Procedures that used to be done in the hospital, for example, which is a really expensive way to do procedures, are being moved to outpatient settings. So that's lowering the cost. Another thing that's happened that's a generally positive development is that a lot of diseases are better controlled. So cardiovascular disease is a good example here. There are fewer people having heart attacks, which are really expensive things.
Speaker 2:So this is a positive reason for the spending slowdown. Some other things that have happened is within commercial health insurance, which covers the majority of people aged 65 and under, prices are still growing, but they're growing at a slower rate. And some other evidence has shown that administrative costs are growing at a slower rate than they used to be. Finally, there's some researchers who have said that sort of this explosive growth or more explosive growth in the 80s and 90s gave rise to a more sort of value mindset in the 2000, 2010s, potentially helped along by the Affordable Care Act. So they've termed this sort of the value zeitgeist where it's something that's sort of hard to prove, but definitely plausible that healthcare systems and policymakers are just more focused on containing spending growth in ways that bring more value out of the healthcare systems.
Speaker 2:There's other factors as well, but those are some of the major ones that people have identified recently.
Speaker 1:Great, I love that term, the value zeitgeist. It's sort of hard to imagine just with the extent that we hear about value so much over the last twenty years that there may have been an era where that wasn't the case.
Speaker 2:But,
Speaker 1:a lot of time has passed since then, I guess.
Speaker 2:It's not, yes. It's a great term. It's not mine, but, I enjoy it as well, yeah.
Speaker 1:Okay. Well, in addition to kind of thinking about the outside forces behind the slowdown, I know we can also examine sort of among what segments of the population that the slowdown was most pronounced by stratifying that slowdown according to other characteristics. And that's really the focus of your paper. So I'm wondering if you can say a little bit about, the factor you used in your paper, which is income, breaking it down by income. Why is that a useful lens?
Speaker 2:Yeah, that's a really great question, Rob. And so I think it really gets at what the value of spending reductions is. And are spending reductions universally considered a good thing? So spending, we could mechanically reduce. The factors I pointed to in your last question are all sort of positive reasons for we think that's either quality improving or at least not quality deteriorating reasons for spending slowdowns.
Speaker 2:But we could actually mechanically reduce spending in The United States by just cutting off access to health care, for example. The real key is not sort of some magic spending number. The real key is spending money on things that are valuable for health. We're talking about spending slowdowns, but we forget that over time, we've increased money on spending on health care, and that's actually been helpful for improving health, right? While there's a lot of talk about cost in the health care system, none of us really want to go back to a time of bloodletting and anesthesia less amputations.
Speaker 2:We want to spend some money in the health care system. And getting so to your question, really looking at income is looking at when the distribution of healthcare spending and whether those increases, regardless of the pace, whether those are distributed evenly across the population or how they're specifically distributed across the population. And I think the worry is that spending might be slowing down because some people are getting maybe sort of what you would call priced out of potentially valuable healthcare. And we know that income is a huge determinant of your healthcare access. It determines what type of health insurance you have.
Speaker 2:It determines your likelihood of having health insurance. It determines your ability to pay for out of pocket costs of healthcare. As spending increases and as new technologies help us improve our lives and improve our health, the question is whether income is a determinant or an increasingly important determinant of that health care access.
Speaker 1:Okay, well, let's talk about your findings in that context. You looked at MEPS data from 2005 to 2023 and divided the population by five income quintiles. And then you measured the total spending for each of those quintiles as well as, inpatient spending, outpatient, emergency department, and prescription drug spending. I know that's a lot, but maybe you can give us some of your top line findings?
Speaker 2:Yeah, absolutely. So I've been told that your listeners are quite smart, but just in case they're not familiar with exactly what MEPS is, MEPS stands for the Medical Expenditure Panel Survey, and it's a national survey of healthcare spending that's done every year by the Agency for Healthcare Research and Quality. They survey about 20,000 to 30,000 people, and it's considered sort of the most precise national health survey of health spending. And then importantly, when we talk about these results, it encompasses both the amount that people pay out of pocket, plus the amount that their insurer pays for them. So you can think of it as sort of the total cost of health care services.
Speaker 2:So what we found when we used the maps is that people with the highest income quintiles consistently have the highest levels of spending, and that between 2005 and 2023, that gap between the lowest income quintile and the highest income quintile is growing. So in that period of time, when you look at real dollars of health care spending, when we look at the lowest income quintile, is that they've had flat or no growth. But when we look at the highest income quintile, their spending has grown at 1.8% per year. And then like you said, we split it by the type of service. That spending growth in the highest income quintile is driven primarily by spending growth in outpatient services and in prescription drugs.
Speaker 2:We don't see as many disparities in emergency department or inpatient services. And that kind of makes sense to us because prescription drugs and outpatient spending tend to be a lot more discretionary than those other two services. And then we mostly see this in the 65 population, which is where you have much more variable insurance coverage than 65, where most people are, nearly all people are covered by Medicare.
Speaker 1:Wow, a lot to take away there. Really interesting findings. I'd love to kind of look under the hood with you in just a minute. But first, let's take a quick break. And we're back.
Speaker 1:I'm here with Doctor. Betsy Q. Cliff, talking about her new paper in the July issue of Health Affairs looking at trends in US health care spending by income, two thousand and five to twenty twenty three. And just a moment ago, Doctor. Cliff, you talked about some of your top line findings and basically pointed to that the growth in spending was really concentrated at that highest income level and that the lowest income quintile had almost no growth or that it was basically flat.
Speaker 1:And I'm curious if that was surprising to you or if there was anything else about those findings that were sort of particularly, surprising that you weren't expecting when you set out on the study.
Speaker 2:Yeah, great question. So when I've talked to friends about this project who are knowledgeable but not like total health policy nerds like me, they have been surprised that the lowest income individuals are not spending the most because we do know that chronic disease tends to be more prevalent in lower income populations, right? So for people like me who hew closer to health economics as sort of a theoretical discipline, we've known for a long time that healthcare is what we would call a normal good. That is people with higher incomes tend to spend more on healthcare. The levels, I have to say, were not that surprising to me.
Speaker 2:What we didn't talk about in the last answer is that there's within this 02/2023 period, there were some differences in levels of growth over time. Prior to the implementation of the Affordable Care Act in 2014, the highest income quintile has the highest level of growth. That actually reverses in the twenty fourteen to twenty seventeen period when portions of the Affordable Care Act came online and you see higher growth among lower income populations. And then when we looked at sort of the most recent years, the most recent year we had access to was 2023. So we looked at this twenty eighteen to twenty twenty three period, and you see the highest income quintile again growing much more quickly than the lower income quintiles.
Speaker 2:And that was surprising to me and something that I want to know more about. I'm not entirely sure why the trend that we saw from 2013 to 2017 reversed so dramatically in 2018 to 2023, but it's something that we certainly want to look at further.
Speaker 1:Wow, yeah, I'm already kind of imagining possible factors here, and I know your numbers can't tell us what's going on. But when you point to 2014 as sort of an implementation stage for the Affordable Care Act, are you talking specifically about the exchanges coming online and perhaps people who might not have otherwise been covered gaining access to insurance as perhaps what's going on there?
Speaker 2:Yeah, I think the two, especially for those lower income quintiles, the expansion of Medicaid that began in 2014 and the health insurance exchanges coming online are the two biggest factors.
Speaker 1:Got it. And then that I'm thinking about that sort of later period, obviously COVID 2020, and then by 2022, 2023, we were to some extent coming out of the pandemic. Do you have a sense of there may have been kind of a backlash or a rebound or something like that going on?
Speaker 2:Yeah, I think that's a lot harder to pin down than the 2014 change. We just don't know. We don't know if that rebound affected people at different income levels differently. I will say we see in the data a real spike in drug spending among the higher income in this most recent period. So I don't know if that's GLP-1s or some other types of drugs.
Speaker 1:Got it. Let me ask you another question about sort of trying to parse those numbers. You talk about sort of at the lowest quintile over the sweep of from 2005 to 2023 being essentially flat. And I think when we talk about health spending, there's this sort of basic assumption that there's at least some growth, Prices are always going up to some extent. There are other factors always sort of rising.
Speaker 1:And so what does that look like when you're flat? I mean, it's sort of hard to wrap your head around. Are people accessing fewer services, or, I guess there's another world where they're healthier. But how do you think about flat spending?
Speaker 2:Yeah, it is tricky to wrap your head around. You're really asking a methodological question at that point because so these are inflation adjusted dollars, right? So if you just look at the absolute amount that people are spending without adjusting, it would be increasing. And then the other thing is we are adjusting for differences in age and differences in chronic disease over time. So our numbers essentially are trying to sort of really isolate the impacts of income and also take out the effect of inflation.
Speaker 2:So they may not perfectly reflect what people's lived experience is because they're, because we're trying to isolate the effects of income. So it's not the absolute dollars that you would see sort of in a healthcare bill, which clearly are not adjusted for all those things.
Speaker 1:Got it. Okay. Well, so kind of taking a step back here, we know, lately it's been common to describe our economy these days as a K shaped, economy, where, you know, after the initial COVID declines, there was kind of a bifurcation where one section of the economy saw rising incomes and asset gains, that's sort of the upper arm of the K, whereas another section continues to lose out with declining purchasing power and heavier cost burdens. That's the lower arm of the K, if you will. And, you know, we saw something similar after the Great Recession of the late 2000s.
Speaker 1:How do you think about the findings of this paper in the context of kind of that broader picture of our economy?
Speaker 2:I think these findings definitely do mirror what's happening in the broader economy. You know, people who have higher incomes are spending more, that's growing more quickly. People who have lower incomes, I mean, you can see if you look at the graphs in our paper, it looks a little bit like a K. They're not spending as much. And that even looks like, again, in real dollar terms, that looks like a little bit of a decline recently.
Speaker 2:So I think the big question is what that means for health. On one sense, it could mean that there's a healthier population. It could mean that prices are growing slower than inflation for at least a certain segment of the population. But it also could mean that there's issues around access. And these findings, even though they are to your introduction, they're a little bit more under the hood than like national expenditure dollars, but they're still pretty 30,000 foot, right?
Speaker 2:And I think one of the big things that we want to understand is, you know, spending encompasses both price and quantity. So it is certainly the case that people in the lowest income quintile pay lower prices per service than people in the highest income quintile. And that is because many more people in the lowest income quintile compared to the highest income quintile have Medicaid, which reimburses providers at a lower rate. But it's probably not all a price story. There's probably some difference in service utilization as well.
Speaker 2:And then the question becomes, what services are people with higher incomes utilizing more than people with lower incomes? Are they things that we think are truly discretionary, you know, like medical cosmetology comes to mind where it's not really an equity issue? Or are they services that are truly lifesaving and important for health, preventive cancer screens or medications to control chronic disease? I think that those are questions that our paper unearths, but certainly does not answer, and ones that I'm eager and I hope other researchers are eager to uncover.
Speaker 1:Great, well, that's a perfect point, I think, to wrap up on with topics for further research and continued work in this space. Doctor. Betsy QCliff, thank you so much for your work on this really interesting paper and for taking the time to chat with us here today.
Speaker 2:Thanks so much for having me. I really enjoyed my time.
Speaker 1:And to our listeners, thanks for tuning in. You describe herself as a total health policy nerd. If you see yourself, as, either total, partial, or maybe even just, thinking about stepping your toe in the policy nerd universe, you're in the right spot. Tune in next week. Leave a review.
Speaker 1:Recommend the podcast to a friend, and, check us out at healthaffairs.org. Thanks, everyone.