Leemore Dafny, PhD, interviews Michael Chernew, PhD, the Leonard D. Schaeffer Professor of Health Care Policy and Director of Healthcare Markets and Regulation Lab at Harvard Medical School.
Leemore Dafny: The system is broken. We’re trying new options. Can you share with us why you are committed to these new options and how we should think about evaluating them?
Michael Chernew: I think it’s important to understand that some method to control health care spending growth is really imperative. Historically, health care spending has grown at about two percentage points faster than national income. That’s simply not sustainable. For public systems, such as Medicare, this drives a huge concern about taxes. [Dartmouth College economist] John Skinner did some work suggesting that if the gap between health care spending growth and national income growth fell by half, to one percentage point, by 2060 (which I understand is a long way off), taxes would need to rise to 70 percent for the highest earners and proportionally for everybody else. There’s a lot of deleterious consequences to high tax rates, and that simply can’t happen.
In the private sector, there’s concern that health care spending growth puts pressure on wages. So one way or another, we have to make a change. All of the solutions that we might talk about — new payment models, a whole bunch of things on the patient demand side in terms of benefit design — all of those things have flaws, and we essentially have to weigh the extent to which we want to keep the existing system of payment at much, much lower, much more slowly growing fees; or move to a model that allows providers that become efficient at translating those efficiencies into profits; or put the entire burden on patients. There are flaws with all of that. There’s not going to be a perfect solution.
At the end of the day, the delivery system has to become efficient. We spend a lot of time talking about financing, but the financing is really just a mechanism to allow the delivery system to succeed if they become efficient, and I think these types of models allow them to do that. Given the pressures we face with an aging population, and a desire to encourage medical innovation, the only way we’re going to be able to finance that is if we can get efficiencies. And the only way we’re going to be able to get efficiencies is if the delivery system has a financing apparatus that allows [health care organizations] to succeed when they become more efficient. In the existing fee-for service-models, if you become more efficient, you tend to lose. In these new payment models, if you can find ways to become more efficient on a population basis, you can actually win. I don’t know the right structure for the models. I certainly see challenges to organizations to succeed under these models. We have a lot to learn in moving along this journey, but we’re running along this path, not because the place we’re going is so wonderful, but because the alternative seems so difficult.
Dafny: Thank you for that summary about the imperative to get there. I especially think that will be meaningful to all those who are listening to hear that coinsurance and copayments and narrow networks are not going to do it. That having the health care system, the delivery system, reinvent itself is an essential component of getting there.
Chernew: What sometimes people don’t realize is there is a huge amount of innovation going on in these models. Just an enormous amount of venture capital, an enormous amount of payer activity to try and figure out what works. It’s unlikely there’s going to be one correct model that will eventually will go everywhere. We’re going to see different things in different parts of the country.
Dafny: I think at one point, when we were first hearing about these models, there was grave concern that certain sets of providers would stint on care, and would reap enormous benefits, and leave [the rest of] us kind of holding the bag. The converse appears to have happened, with many organizations saying there’s not enough of a payback to get involved in a Medicare ACO. Can you tell me why that happened?
Chernew: My personal opinion is that the payers have been trying to pull the savings out of the system very quickly. So that means that the shared savings percentages are lower than I would like, and they tend to rebase the payment benchmarks down after a certain amount of time in the original models, so they pull the savings out of the system. Everybody, I think, would agree that you’re going to have to get the savings to flow outside of the health care system. And remember, we’re talking about changing the rate of growth, not pulling out absolute savings — but if you do that too quickly, your desire to capture the savings may reduce the amount of savings that you have to capture. In the original regulatory models, I think they went too far along the spectrum of hating it when the provider system profits.
You need to allow there to be profits in the provider system in order to induce these efficiencies. And you need to set the system overall on a spending trajectory that we can afford. And if you’re constantly trying to pull money out of the system every time you see an efficiency, the provider system doesn’t nearly have the incentives to participate in creating those efficiencies — and that’s where the balance is. Because we know we have to control spending, and we’re going to have to pull money out of the system relative to what otherwise would have happened, but if we’re too aggressive in the way in which we do that, we won’t get participation.
Dafny: One of the advantages that the commercial sector has over the public sector, with regard to ACOs, is that the commercial policies can restrict provider choice. Is that a tool that you think would be valuable for the Medicare providers, Medicare ACOs, to access?
Chernew: It would. There are mechanisms [with] which they might work around that problem. There’s the Medigap Select programs, and of course we have the Medicare Advantage program, which providers could potentially participate in. So there are a number of ways to get around that particular issue, but I do agree that it’s an issue. I think the bigger advantage that the private sector has is that it’s easier to work collaboratively with the providers than it is for Medicare, because Medicare works for the entire country with a very fixed, assigned set of rules, whereas the commercial sector can learn over time, can modify as needed. It can make exceptions for particular providers doing various things. That level of flexibility, I think, helps the commercial side a lot.
On the opposite end of the spectrum, of course, the market power of the government to say, “This is what the rates are,” and the vast number of people that it controls, enables them to deal with pricing concerns. Where even if you get this exactly right on the private side in terms of the form of payment, the level of payment is still going to be subject to all the concerns we have about provider market power. So we need to figure out how to manage that aspect of this, because it’s certainly true that if organizations grow to the point where they have the scale to succeed under these models, they may not be willing on the commercial side to pass those savings along as quickly as, eventually, we would like. So there’s a balance that goes on there.
Dafny: A topic of great interest to the health care marketplace. Thank you so much, Dr. Chernew, for your time.
This interview originally appeared in NEJM Catalyst on February 26, 2016. Listen to or read Part 1 of the interview.