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.
Hello, and welcome to a health podocy. I'm your host, Rob Lott. There are some things in life it seems we can always count on. The sun rising in the East and setting in the West, the steady march of time, and the fact that I always seem to misplace my keys, especially when I'm running late. Now for the longest time, at least in US health policy circles, it always seemed that the steady rate of growth for health care spending was one of those things.
Rob Lott:Starting in the sixties and extending for nearly fifty years, it seemed that despite so many policy changes, health care spending grew at a real per capita rate of about 4% annually. There were a few slowdowns and a few surges, but for the most part, the rate was pretty reliable and seemed to be unlikely to change anytime soon. But then starting in 2009, that's exactly what happened. Things changed. For the period from 2009 to 2019, inflation adjusted per capita health care spending in The US grew at just 1.7% annually.
Rob Lott:That's a pretty big drop. Many experts have proffered many explanations for the slowdown. Most people assume there's no single answer, but still, it sure would be nice to know what exactly was going on there. To what we should attribute the ten year fadeaway of our old faithful 4% growth? And that's our question for today's episode.
Rob Lott:I'm here with Doctor. Sherry Glied, a health economist and professor of public service at the NYU Wagner Graduate School of Public Service. Doctor. Glied was dean of the Wagner School for more than ten years and previously served as assistant secretary for planning and evaluation at the Department of Health and Human Services and on the president's council of economic advisers, among many other roles. Together with her coauthor, Brendan Liu, also of NYU, she has a new article in the January issue of Health Affairs titled, quote, anatomy of a slowdown, decomposing the moderation in health care cost growth twenty ten to twenty nineteen.
Rob Lott:I cannot wait to sift through that decomposition and see what we find. Doctor. Sherry Glied, welcome to A Health Odyssey.
Sherry Glied:Thanks, Rob.
Rob Lott:Now before we jump in, doctor Glied, if you don't mind, I have a quick programming note. This is our first episode back after a short hiatus at the 2025, and I am really excited for a few reasons. One, it's a new year. And although the break was nice, it's good to get back to it. And two, in addition to our usual audio recording, we're also recording video, which means that in addition to all the usual audio podcast platforms, you can also check us out on health affairs YouTube channel, where in addition to hearing our chat, you can also now see me gesticulating wildly with my hands as we talk.
Rob Lott:This is an experiment. We're gonna try it for a few months and see how it goes before, making any long term decisions about the use of video going forward. So to our listeners and our watchers, let us know what you think. All right, so let's start with a little bit of history. As I mentioned in the introduction, over the period from about 1970 to 2008, the inflation adjusted average annual rate of growth in healthcare spending was about 3.7%.
Rob Lott:And then if we look at the period just from 2000 to 2008, the rate was similar, 3.4%. When you look back at those wide swaths of time, what factors were driving growth at that point?
Sherry Glied:So there's some literature trying to figure out why healthcare cost growth has continued for such a long time, and one factor people thought about is whether it's an expansion in insurance coverage, like more people have insurance and then they spend more money, but the best estimate suggests that expansion of insurance can only explain a small share of the growth that's happened over time and that is actually true in the period that we look at as well. So people have largely concluded that it's mostly changes in technology that we keep getting more and more stuff we can do for people and particularly that that technology enters a system that doesn't really control it very effectively. So new technologies come along and then they just disseminate and diffuse and raise costs, and we're kind of stuck with that. So I think that's sort of been the story up until now.
Rob Lott:Okay. Great. And then as we, said in the introduction, there was this slowdown. This is the focus of your, paper. Between 2009 and 2019, the rate of growth was 1.7% annually.
Rob Lott:That's half of what the rate was during the previous time windows. And your research paper sought to decompose all of the factors potentially driving that slowdown. And before we talk about the findings, I wonder if you can sort of bring us back to 2016, 2017. How were people talking about this? What what were people thinking about?
Rob Lott:What was the conventional wisdom about what was going on here?
Sherry Glied:So there's sort of different conventional wisdoms happening at the same time, but I think it's important to remember that that decade between 2009 and 2019 is one with a lot of policy change happening, right? That's the ACA, and so there are big expansions of coverage, but also value based purchasing and, you know, readmission reduction, payment reductions and all that stuff happening in Medicare, lots and lots of sort of ferment in the health environment, and you see analysts kind of responding to that. So first of all, the beginning of this period, we're coming out of the great recession. So healthcare costs slowed in 2008 because people's incomes fell, but the CMS actuaries and lots of other people thought, well, that's gonna end, and in fact, it did end, and then healthcare costs are gonna accelerate again because that's what typically happened with previous recessions. And then the ACA happened and people had kind of different views of what that might do.
Sherry Glied:Some people thought covering all these new people, that's just gonna drive up costs a lot, and other people were more sanguine about that possibility, and then sort of other themes come in. Some people thought technological innovation is slowing down. We've seen a decline in the rate of innovation in pharmaceuticals, so maybe the change is gonna come through slower technological innovation, but on the other hand, at this period that you're sort of talking about, that's when all those hep c drugs came on, Civaldi, you know, all of that, and people were very nervous about it. There were new cancer drugs, there were other innovations that some people thought, no, no, no, we're going back to a rapid rate of growth because we have all this innovation that's happening. And then some people were fans of value based payment and all the Medicare innovations, and they thought that's gonna slow things down, it will surely slow things down in Medicare, maybe it will also have follow on effects in the other, in the private sector especially.
Sherry Glied:The CMS actuaries were not very keen on that, neither CMS nor CBO gave a lot of credit for all those innovations, but some analysts said, well, they're wrong, we're actually gonna see a lot more benefit than we thought. And then one last factor people keep pointing to is that this is also a period of expansions of HSAs, so increased cost sharing, maybe that's what's doing the trick. One of the things I think you should note as you think about all these different explanations is that they actually have pretty specific targets, so if you think it's higher cost sharing, you should expect that any slowdown is gonna happen in the commercial insurance market, And if you think it's Medicare payment reforms, then you ought to think that any slowdown that's gonna happen is gonna happen mostly in Medicare, right? So one of the sort of animating impulses in our paper is to say, well, if we actually look at this by payer and by service, maybe we could sort this out a little bit better than these kind of global explanations do.
Rob Lott:Great. Well, that's a perfect segue for us to dig into your paper. You identified four major contributors once you kinda dug in and started to decompose the the actual contributions to the slowdown. And I'm hoping maybe we can sort of go through those four patterns one at a time. You can say a little more about what's going on under under the hood there.
Rob Lott:Especially because I think while the what is interesting, what was the pattern, I think what we can really learn from is the sort of why why we saw those patterns sort of surfacing during this time period. So let's go through it. The first one was, and correct me if I misrepresent these here, hospital and pharmaceutical spending across payers.
Sherry Glied:So I want to actually, let me step back for one second. I just want to note one thing that what we did in this paper is we looked at what the CMS actuaries predicted was gonna happen in the next decade, and it's important to keep that in mind because they're already looking at things like the aging of the population, changes in payment systems within Medicare, so that's already baked into their forecast and we've pulled that out. We also pull out the changes in policy that happened over this decade. So we're saying kind of net of all of the things that people have already sort of baked in, what happened? So what is it that the CMS actuaries didn't expect to happen that actually happened afterwards?
Sherry Glied:And I think that's a distinction that's important here because for example, expansions and coverage might've had an effect on spending, but people expected that, right? So what we see that isn't expected is that there's a pretty big shift in hospital spend from inpatient to outpatient care that's happening in Medicare and Medicaid, much more in Medicare and Medicaid than in commercial. In commercial, much of that had already happened, so you sort of see Medicare and Medicaid shifting into outpatient care. Kind of makes sense as we think about providers becoming more comfortable in providing certain kinds of hospital services in outpatient settings over this period, and that's a contribution on the hospital side. On the pharmaceutical side where there's a lot of savings relative to forecast, one of the things that happened is that there's a big shift from branded drugs to generic drugs.
Sherry Glied:It's a huge shift. It actually overpowers just about everything else that's happening over this period, and on top of the shift from branded to generics, there were also a bunch of prescription drugs that actually go to over the counter use, like the proton pump inhibitors. And when things move to over the counter, their prices come down. Also, that reduces the need for physician visits because people can just get their drugs whenever they want, so that shift to over the counter is also in there. We see fewer people seeking care for things that maybe they need prescriptions for, fewer people with loads of physician visits, and this happens pretty much across the board across payers.
Rob Lott:So does that suggest that that that slowdown in pharmaceutical spending was because there weren't any big blockbuster drugs or not as many, or was it because of policies that may have influenced how people were purchasing drugs?
Sherry Glied:I think it's a combination of things. So certainly there were blockbuster drugs like Sovaldi that happened in this period, so there were some really important innovations, but a couple of things. One is that shift to generics, a lot of drugs did go off patent, a lot of really important drugs went off patent over this period, and so there was this shift to generics for those drugs. And you always have this kind of play in the pharmaceutical market between new drugs that are introduced and older drugs that go off patent. And so it's that it's that dynamic that's driving lower than expected pharmaceutical costs in this period.
Rob Lott:Got it. Okay. Let's move on to the second pattern, which you identified as slower price growth, especially among private insurance.
Sherry Glied:Price growth in Medicare and to a lesser extent in Medicaid is pretty easy for the CMS actuaries to forecast. Where they might get price growth wrong would be in private coverage because private insurances in a market. If you had asked people even during this decade, what do you think is happening to price growth in private insurance? I think people would have said, Oh, it must be going up really fast because the thing that we've been worried about in price growth is really concentrated healthcare markets, and this is a period where hospital concentration had been a big concern. We actually found that the cost per inpatient stay was growing more slowly over this decade than it had in the past, and that's particularly interesting because of that shift from inpatient to outpatient care that we described earlier.
Sherry Glied:That would tend to make you think that hospital stays are gonna get more expensive. The people who are left in hospital are sicker, and that's happening, but overall, we're seeing slower growth in inpatient stays, and so we actually we were very surprised by this, and we thought we'd look at the consolidation effects and it's sort of an interesting story and maybe a little bit of an argument for doing more of this full market kind of analysis rather than, we as economists, I'm an economist by training, we do a lot of these focused, well established causal analyses but we do a lot less of this sort of macro level accounting and maybe we should be doing both of them because what we found is that yes, concentration of hospitals consolidation can have very big effects in some consolidating markets, but overall it didn't affect prices across The US very much. That's because a lot of the consolidation happened in markets prices were not driven up very much by the consolidation. We also see some moderation of physician salaries in some of the high priced specialties, so a little bit sort of more pressure on prices in the private market than people had anticipated.
Rob Lott:Got it. Okay. Third pattern you identified was Medicaid spending on long term services and supports. Say a little more about that.
Sherry Glied:Yeah. So this is one, again, where people had not really focused. Honestly, I think nobody thinks about long term services and supports, and if they do, they just think it's going to get more and more expensive over time. And this is a period in which there was a lot of effort to move people from nursing homes into home based care, and that did happen, but we actually saw declines in the use of Medicaid home based care, not people substituting back into nursing homes, but actually changes in the demographics of older people on Medicaid. More of them are married, fewer of them are smokers.
Sherry Glied:They're just doing a little bit better, and maybe that's why we just see declines overall in the use of Medicaid home based care.
Rob Lott:Gotcha. Correct me if I'm wrong. I think you were at the Department of Health and Human Services after the the Affordable Care Act included the CLASS Act.
Sherry Glied:The demise of the class act. We're gonna write a book about that someday. Yeah.
Rob Lott:Alright. I I look forward to reading it. Do you think had you known what the numbers would be, you know, looking back, would that have changed the decision in
Sherry Glied:On the class act?
Rob Lott:When it was 2011, 2012 to
Sherry Glied:So I don't think it would have changed the decision on the class act because and this is gonna get really in the weeds. The issues with the CLASS Act actually had to do with younger people, not with seniors. The budget busting concerns around the CLASS Act were really around the treatment of people 65 with disabilities, it was a very complicated problem. I will say more broadly though, that if the architects of the Affordable Care Act had seen our paper back in 2010 when they were writing it, we wouldn't be having the arguments we are having now about the enhanced premium subsidies, they would have been baked into the act because it came out costing so much less than people had anticipated in the first place. Got
Rob Lott:it. Okay. Let's hit the fourth and final sort of major contributor that you identified, which was declining administrative costs among private insurance?
Sherry Glied:Yeah. Again, some of this may be due to the MLR rule, the spending ratios that private insurers are supposed to adhere to under the Affordable Care Act, although that's been very contentious. Some of it may also have to do with increasing automation of some of the administrative work in health insurance, but we actually see this slowing down of administrative cost growth in private insurance.
Rob Lott:Great. Well, thanks for kind of walking us through all the nitty gritty there. In just a minute, I want to ask you about some of the context and maybe what we can do with these findings, but first, let's take a quick break. And we're back. I'm here talking with doctor Sherry Glied, about her paper in the January issue of Health Affairs about the moderation in health care cost growth during the period twenty ten to 2019 and looking at exactly what was behind that moderation, the different factors contributing to it.
Rob Lott:And you hopefully just sort of walked us through the four major patterns that you sort of uncovered during that time period. And earlier you provided some helpful context that the sort of baseline you started with was the CMS actuaries projections. And then what you identified were these things that were sort of unexpected or not projected. And I'm curious if you can say a little bit about sort of if we saw these sort of major digressions from what was initially projected, what does that tell you about the usefulness of the projections in general and especially over that ten year period?
Sherry Glied:So I want to actually distinguish between two aspects of projection. One is what the underlying trend looks like and the other is what is the effect of a particular policy change. Our paper really focuses more on the underlying trend piece of this, so we look at the estimates of the effects of particular policy changes, but in this paper, we're not kind of assessing whether they're right or not. In prior work I've done, we found that the policy projections are not too bad. I mean, they're off by 10 or 20%, but it's within the ballpark and they were sometimes too high and sometimes too low, so there's not some systematic problem with the projections.
Sherry Glied:The question here is how do you project the underlying trend in healthcare cost growth by sector and payer over time? And that's the really tough problem. That's really the Cassandra problem because it depends on all kinds of things like innovation and other factors that are really hard to assess, and we could talk about this later, but my intuition is that being very transparent and simple about the projections is probably the way to address this.
Rob Lott:What what can we say about the relationship between, the slowdown in spending that you zeroed in on and the kind of health care and health that we're getting for our spending, as a nation. Are we becoming better purchasers of health, or are we maybe losing our ability to shop for value? What do we know about the relationship between spending and outcomes, or is there a disconnect that sort of makes all this just an exercise?
Sherry Glied:That's a really tough question, and I think one thing to really think about and especially as we go forward is whether in fact, as earlier, I the main impetus for healthcare cost growth over long periods of time has been innovation, technological innovation that improves health or improves our ability to produce health in ways that are less unpleasant for people and our willingness to pay for those new innovations. So when you see a slowdown in healthcare cost growth, you could sort of say, well, that's great, or you might say, well, maybe that's because innovation slowed down and we're not actually seeing gains in health that we might be from innovation, and this period had been a period, especially post COVID, where we've seen real declines in health outcomes, we've seen real problems in health outcomes, so you could reasonably say that's a worry here. On the other hand, this was a period with some pretty impressive clinical innovation. Sovaldi just basically erased hep C for people who took it. We saw big improvements in cancer survival for several different cancers over this period.
Sherry Glied:So while the overall outlook for health was not great, if you look at sort of targeted innovations, maybe it's not so bad. I think one possibility is that we're actually doing better at managing the pricing of some of these new pharmaceutical technologies, partly because of regulation but partly because, I mean, this is me speculating, don't want to put too much on this, but one thing that we're seeing, and we saw it in the case of Civaldi, we see it in other cases, is that we're seeing more sort of competing drugs come in shortly after the first innovator, and we also see a lot more insured bargaining with the pharmaceutical companies about which drug is going to be on the formulary and so on. So even sort of looking forward, if we look at the rollout of GLP-1s, what you see is that the spend actually doesn't hit the targets that people thought it would because new drugs come on so quickly now and maybe that's actually bringing us a better innovation value framework, at least in that sector of the market. It's so scary for me to sound like an optimist, but it may be that we are doing a little bit better in that part of the market.
Sherry Glied:It might have to do with how drug discovery is happening, but I don't know. I think the key question is whether we're actually seeing innovation within healthcare and whether we're getting the diffusion of that innovation.
Rob Lott:Okay. Well, you know, in the introduction, I talked a little bit about sort of old faithful, that 4% rate that we saw or 3.7% rate over so many years, and I think that theory was driven in part by you. You talk about this in the paper that even the expectation was that even if there's sort of short term containment of price growth in one area, it would be offset by increases in utilization. But obviously, the last ten years suggest that maybe those assumptions need to be revisited. And so I guess my question for you as we sort of wrap up is, okay, let's start with that point.
Rob Lott:So now what? Obviously, don't think the answer is to just stop worrying about spending growth, but it's also probably not to just sort of stick with business as usual. And so I guess how does all of this context shape how you think we should address healthcare spending and think about the future? And if the current leaders of our healthcare system were to ask you, where do we start? Where would you point them?
Sherry Glied:So I think the right thing to do is not to worry too much about the longer range forecast, but to really think about what are we doing now and are we getting value for the money we put in at a moment in time? Recognizing that in the future, the array of technologies that will be available to us both on the clinical side and on the cost containment side, frankly, will be different. So rather than trying to look into a crystal ball and imagine what things are gonna look like ten years from now, what we should be saying is if we do this thing right now, what we know right now, what is it gonna do for people? Is it actually providing a commensurate increase in health for the money that we're spending? So we can get very fixated on the long term forecasts and I think we should be spending more time thinking about the here and now, the next two or three years, what will this do, and recognizing that we can change course later.
Sherry Glied:So one of the things we say in the paper is maybe we should just make these forecasts really almost trivial. So for example, what if we said our forecast is just the five year, three year, whatever moving average of healthcare spending growth in this sector by this payer in the last three years. So there's no magic to it, there's no kind of figuring out what might happen in the world, and it's self correct because if you were wrong, then the number's gonna be high next time and then it'll get back to where it is, but I think more than whether that's the right thing to do and how close you'd get, I think it would avoid having policymakers focus so much on this sort of magical forecast number, which comes from the mouths of, you know, the seer sitting in the and temple on instead really focus on like, okay, if we do this, what is gonna happen to people right now? Because that's what we can control. And a little bit of this whole effort honestly comes from the fact that back when I was at ASPE, we ran a committee on the long term Medicare forecast, and I realized that the Medicare actuaries, they have to forecast healthcare costs seventy five years from now.
Sherry Glied:That is a ludicrous exercise. We have absolutely no idea what the world is gonna look like seventy five years from now. Right? And yet policymakers put a lot of faith in those numbers, and and I think that's a mistake.
Rob Lott:Great. Well, a great sort of sober minded, clear headed way to to look at our work and our task going forward. Doctor. Sherry Glied, thank you so much for taking the time to join us today.
Sherry Glied:Thank you very much for inviting me.
Rob Lott:To our listeners, thanks for tuning in. If you enjoyed this episode, subscribe, leave a review, recommend it to a friend. If you're watching on YouTube, smash that subscribe button. And, of course, for everyone, please tune in next week. Thank you.
Rob Lott:Thanks for listening. If you enjoyed today's episode, I hope you'll tell a friend about a health policy.