America’s high spending on health care comes with tradeoffs. While negative factors exist—such as the expensive cost of drugs and medical procedures—there are positives too.
In the second part of his keynote address at the 2019 AMA State Advocacy Summit, Ashish K. Jha, MD, MPH, director of the Harvard Global Health Institute, talks about the tradeoff of positive and negative outcomes stemming from high spending in America’s health care sector.
Below is a lightly edited full transcript of his presentation. You can also listen to the full episode on Apple Podcasts, Google Play or Spotify and follow along with Dr. Jha’s presentation deck.
Dr. Jha: Administrative costs are higher here, but that can't be the entire story, right? Because that's only a small part of the health care system. Let's talk about what else is part of the story.
Remember, the only equation that we're going to talk about today in total spending is quantity times price. I think I've tried to make the argument that quantity isn't the big explanatory factor for higher spending. And so that only leaves one other factor, which is price. … Let’s think about this. Could it be price is the problem? And we'll come back to what do I mean by price. So, one issue is: Is it the prices?
There is a very famous paper … by one of my heroes, an economist at Princeton named Uwe Reinhardt. I'm sure many of you know of Uwe. He passed away about 15 months ago, but he wrote a very famous paper that came out in 2003 called, “It's the Prices, Stupid.” And it basically argued, with much less data, that a lot of what was driving American health care spending—higher health care spending in America—was prices. Prices of what?
If you go out to Washington today or even survey Americans, of course, one big thing everybody's worried about is prices of pharmaceuticals. Is that true? Are prices of pharmaceuticals higher here? And the answer it turns out is, of course they are, and we know this. Pharmaceutical spending per capita across these countries—the mean is about $750, and we're almost twice as high. And when you get to individual drugs, you start seeing this.
Actually, before I get to individual drugs, let me just make the other point that other people have looked at utilization of drugs. Do we take more medicines than other countries? It turns out we don’t; we're about average overall. The French actually tend to take more medicines than almost anybody else. But on some things we’re on the high side, on some things were on the low side. About 80% of prescriptions that are written in the U.S. are for generics.
And it turns out that we're not consuming so much more medication than everybody else, but our prices are much higher. And we see this when you get the individual drugs … almost everywhere you look, our drug prices are higher. But pharma makes up about 15% of all health care spending. So, it can't just be about pharma. …
But there is one point that is worth making, which is the American consumer, by paying higher prices for certain things, certainly is the engine for innovation in terms of new drugs that are coming out around the world. And other countries, by paying a lot less, get to free ride a little bit on the U.S. And how do you deal with that? And do you make other countries pay more? And how do you do that? And will that help? Those are different conversations. But the point is we are the innovation engine of the world when it comes to pharmaceuticals, and other things as well.
Physician salaries, I think, is a complex issue, and it's a complex issue because almost nobody—no other country—has physicians who graduate with substantial medical debt. Medical education's paid for. And if medical education is paid for, then that counts not so much in medical spending, but in education spending. That gets to somebody's earlier question of, are you really comparing apples to apples in terms of what counts? But the bottom line is medical debt is substantially higher in the U.S. than it is almost anywhere else. Length of training, especially for a lot of specialty physicians, tend to be a lot longer here.
And then here's the last point that's worth making, and that's the opportunity costs. And here's what I mean. If you're a college student in the United States, you're not asking yourself, “Do I go be a doctor in America or do I go be a doctor in France?” That's not the standard question most people ask. They're wondering, “Do I want to become a doctor? Do I want to become a lawyer? Do I want to go work on Wall Street? Do I want to become an engineer?”
These are the questions people are asking, right? Those are the tradeoffs people are thinking about. And it turns out that we pay our professionals better across almost every field. Our top lawyers get paid more than the top lawyers in France. Our engineers get paid more than the engineers of France or Germany or the UK, on average.
And so, and this is an important point, if we took physician salaries, cut it by a third because we said, “Oh, doctors are paid too much.” Two things. First of all, you might shave off 1-2% of health care spending by doing that. You could make all doctors work for free. Enslave them, and you could save 6, 7, 8% of the health care spending. Maybe 7–8%.
But the bottom line is, and Uwe, again—Uwe Reinhardt, I'm going to quote him—said, “If you cut physician salaries by a third, you can maybe shave off 2% and you're going to demoralize the health care workforce that's involved in taking care of our population.”
You're also going to get a totally different group of people going into medicine. If you paid physicians in the U.S. the way physicians are paid in France, over a generation, you're going to get a very different group of people going into medicine. And I believe that will have substantial costs.
So, I think the issue of physician salaries in this is complicated. And, so, I show you the simple data. The physicians get paid more, and that is a fact. Physicians in America get paid more. But it's very important for me to point out because a lot of people say, “Your data says doctors get paid too much.” And I'm like, at no point have I said that, at no point is that what's showing up in the data. We get paid more, but there's a lot of complexity to that.
Other benefits of high prices—as I've already argued, I think we get very high-quality physicians and nurses. We get much faster access to diagnostics and procedures. We have nicer amenities and facilities. … Now you can ask, “Well, maybe we shouldn't spend on that and maybe we should be more frugal.” We can come back to that, but those are the realities. There are benefits of higher prices. I think both innovation and higher-quality physicians is a very important part of that.
Let's talk about health outcomes. So, I'm going to now show you the slide, just like I showed you the health care spending slide that you've seen. I'm going to show you the health care outcomes slide that every policy talk begins with, which is life expectancy, right? Japan, Switzerland, Australia, France, and here we are on our lonesome self on the right. We know our life expectancy is lower than it is for these other high-income countries.
And by the way, we like to often say this is sort of an indictment of the health care system. And we can come back to thinking about, is this an indictment of our health care system in a bit? …
And so people often say … “If we could just do what Denmark does.” Right? Here's Denmark, great health outcomes, 80.8 years. Here's Sweden at 82.3. …
Anybody know Denmark's population? Six million. 10-15% smaller than Massachusetts. Slightly bigger than Minnesota. And, so, I often say, look, if you want Sweden, we've got Sweden. It's called Minnesota. Right? About the same size. Maybe Sweden's a little bigger, but smaller than Denmark. But the point is, we have Minnesota, and we've got Sweden right here.
It's got health outcomes as good as Sweden, as good as Denmark. We have Hawaii that does even a little better. There's Connecticut edging out Germany, tied with the U.K., just below the Netherlands. The bottom line is we've got states that are the size of some of these European countries with health outcomes that are comparable. We also have states with very bad health outcomes—Mississippi and Arkansas and a bunch of other places. And so, looking at averages across these countries is a real challenge.
And whenever people say to me, “Boy, Finland can do this, America should be able to do this,” I think, Finland, 5 million—lovely place by the way, very homogenous, very high income. And the notion that our country of 320 million people with all the diversity and complexity that our country has can do what Finland can do with its 5 million people seems to me to be a challenge.
It doesn't mean we can't learn anything from Finland. I think we can. But we've got pockets of health outcome excellence that are every bit as good as any of these European countries. So, it's just worth thinking about: Are we comparing apples to apples when we compare America to Denmark? And whenever you hear politicians talk about America to Denmark, it's worth remembering, we're talking about a country of about 6 million people versus a country of 320–330 million people.
Somebody brought up neonatal mortality and I want to talk about it. So, the two numbers that every health policy expert likes to talk about whenever we want to talk about how problematic American health care is, is we begin with life expectancy and show that America is an outlier. Then we show neonatal mortality and show that America is an outlier.
And we say the American health care system is terrible. It's the standard, whenever they give you sort of the standard talking points of how you give a health policy talk, it's always in there. Begin with those two slides. I think those two slides are problematic. I've already shown you why on life expectancy, it's a little simplistic to just put America up there compared to other countries.
So, here's neonatal mortality. We are higher. We're are among the highest. We are the highest in this group. Neonatal mortality is affected by all sorts of things, and certainly social factors matter immensely and they drive low birth weight. And low birth weight is the big risk factor for neonatal mortality. So, there's been some nice work that says, what about neonatal mortality? And now I'm getting to the question that that gentleman back there asked 30 minutes ago, “What about neonatal mortality given low birth weight?”
What if you took babies that are low-birth-weight preemies? … And the two countries that do the best are the U.S. and Germany. Remember, I've been showing you all of this data that Germany and America seem to be kind of on the higher end of a lot of stuff, and then when it comes to mortality given low birth weight, we actually do pretty well.
And it starts making us ask all sorts of questions about what are the responsibilities of the health system? Is the fact that we have higher smoking among pregnant women, is that a failure of the health system? Maybe. But it's at least worth thinking through that question. Given that we have more poverty in America than there is in the Netherlands, and poverty is a pretty big predictor of low birth weight, are our higher rates of poverty an indictment of you and me?
And the health system that delivers care to people? Maybe. But at least it's worth having the conversation at that level of nuance. Not put up this slide and say our health system is underperforming. I feel like that kind of misses a lot of the subtlety. …
And people say, “Well people in the Netherlands have much lower rates of diabetes.” Everybody's biking around everywhere in Amsterdam. Right? And so, is that a failure of our health system that we don't have enough people active? Maybe. I'm not saying it's not, but I'm saying we've got to think through that a bit more carefully than looking at the obesity rate of the American population and saying the American health system is underperforming. Just seems to me like that's a leap that we should make with a lot more care than policymakers have been willing to make. …
Yes, we have a high-cost health system. It's driven primarily by administrative costs and prices. Health outcomes at the population level are worse, but if you were to get sick, this is a good place to do it. So, I like to say, yes, we have more strokes, but if you're going to have a stroke—and, please, nobody, please nobody have a stroke—but if you're going to have a stroke, this is probably the best country in the world to do it.
In the interest of time, I am going to just make a couple of quick points. …
If you think about the policies that were enacted as part of the Affordable Care Act, on the cost and quality side of things, they really on the cost side focus on quantity. How do we reduce health care services?
If you look at the value-based payments for hospitals—VBP, HRRP, some of the stuff that's been focused on physicians, efficiency metrics—they're really focused on the model that we're doing too much. We're readmitting too much. It shows up in the VDP. … I think the data are pretty clear, that most of these value-based payments just don't do much. They've had little to no impact.
Because I think our mental model has been that doctors would just do the right thing if you gave them an extra 1% on their compensation. I think that's a crazy mental model, but that's been the mental model. Let's put in 1–2% adjustment on their salaries, and all of the sudden, they'll start doing things that they already knew were really important and good for their patients, but for whatever reason, weren't doing. And it has turned out that that makes a whole lot of simple economic theory but just doesn't really work in real life.
The models that are having some impact are what I sort of talk about as a kind of accountability and changing the episode. ACOs and bundle payments—here, there's a bit more reason for optimism. There was a paper in The New England Journal two weeks ago by Michael Barnett and folks, showing about $600–800 of savings per episode for knee and hip replacement. ACOs, on average, are sort of saving 2–4%.
These are some of the big national programs. The point in my mind is, these value-based payment adjusters don't do very much in terms of outcomes and impact. Some of these other ACO-like things are having modest benefits. And there's some issues on scalability. I've started my talk with this, and I'm going to finish with this, which is, where is the action going to be? All the action, I think, all the interesting intellectual action is going to be in the states.
And I'm going to just spend a minute and just fly through. Again, a lot of you have a lot more expertise on this than I do. And this is not meant to be a comprehensive coverage of all the things that are happening in states. I just want to talk about a few where I see a lot of interesting stuff.
There's the Maryland All-Payer Model, where you have hospitals on a global budget, and you have targets set, and you're looking at per capita hospital revenue growth. There’s mixed data on whether it's having an impact. Massachusetts has a different approach. They created a health policy commission that sets targets for total health care spending growth, encourages movements away from fee-for-service models towards ACOs and alternative payment.
Let me just make a couple more points, and then I'm going to make a big one. We see this across other states. Vermont is doing this. And Arkansas. Oregon has an alternative payment advanced care model. An advanced care model.
The big picture point here is if you look at the landscape and the country, you see lots of states trying lots of different things—40 states pursuing some sort of value-based payment models. Many of them are multi-peer initiatives. Lots of states are trying ACOs or other episode programs. There's the state innovation models, et cetera.
Here's the big-picture point. I think states are trying all these things. I think it's great because they're all doing it a little differently, and we're going to learn a lot about this kind of stuff. But it's still primarily focused on quantity, on trying to reduce quantity of health care services. And I've told you that the data on ACOs and bundle payments is you can make a little bit of an impact. You can shave off 2, 3, 4% of health care spending over five years.
And that's not nothing, but what has become clear to me—and if I were asked to give this talk a year ago, I would not have said this, but I will say it now—is that states are starting to realize that a lot of the action is not just in quantity, but it's in prices. And where states are going is tiptoeing in—some people say it's more than tiptoeing, up to you—into price regulation. Because if prices are a problem, there are two ways that we know how to deal with prices: competition and regulation.
And so it starts … in easier areas like out-of-network provider charges and maxing them out… where you get into Rhode Island's hospital rate setting and ACO growth caps—but they're considering a cost target, and that will eventually get into prices. I'd argue that Massachusetts’ health policy commission is really thinking about price regulation quite a bit. You get stuff into Vermont, West Virginia, Pennsylvania.
The point is, people are trying this, are thinking about it, and again, I think a lot of you probably have a lot more on-the-ground experience on these things than I do. …
I think what we're going to see is: People always talk about states as the laboratory of democracy in America—I think it's going to be the laboratory of innovation, new care models, new efforts to do things. Blue states are going to do things very differently than red states, certainly around price regulation. I think we're going to see a lot more stuff coming out of blue states, in terms of their willingness to say, "Actually, we're going to set all payer prices at X."
So that, I think, is going to be interesting. I think states are going to be even more important in the era of divided government. I think Medicare is going to continue doing alternative payment models and ACOs. But you're not going to see big, new, bold things coming out of CMS, or certainly no big health reform coming out of Congress in the next two years. And I think what's interesting about states is what works for one state may not work for others.
I've already shown you data that Minnesota. … On almost every metric, Minnesota and Mississippi look very different. And so there's no reason to believe that the policies that will make Minnesota's health system better are exactly the policies that will help the Mississippi health system. There may be some overlap, but you know that there are going to be differences in policy.
And my belief is that this is the approach that we're going to need to take to create a uniquely American solution. We're going to have to figure this out at the state level, and we're not going to have a single health system that looks exactly the same in Boston as it does in Oregon, as it does in Miami. That's not what we're going to, I think, ever get to.
We're going to have a patchwork of places that look different from each other because the underlying populations are so different. The needs are different—and that, to me, is going to be an important feature of what I sort of think of as a uniquely American solution, with states leading the way.
—
You can listen to this episode and all “Moving Medicine” podcasts on Apple Podcasts, Google Play or Spotify.
