The Cost of Health Care — Highlights from a Discussion about Economics and Reform

Atul A. Gawande, M.D., M.P.H., Elliott S. Fisher, M.D., M.P.H., Jonathan Gruber, Ph.D., and Meredith B. Rosenthal, Ph.D.

N Engl J Med 2009; 361:1421-1423October 8, 2009

Video

Adobe Flash Player is required to view this feature. Download the latest version here. If you are using an operating system that does not support Flash, we are working hard to bring you alternative formats.

Download a Transcript of Video (PDF)

Poll Results

Contributors

President Barack Obama has argued that health care reform is essential to the future economic health of the United States. But the economics of both health care and reform are daunting. Is it possible to pay for health care coverage for all Americans without moderating our rapidly rising health care costs? And what do we know about the best ways of achieving cost savings? On September 14, 2009, the Journal hosted a roundtable discussion about the most promising ways of slowing the growth of health care costs, their potential effects on medical practice, and the likelihood that the current health care reform effort will be a step in the right direction. Moderator Atul Gawande was joined by health economics experts Elliott Fisher, Jonathan Gruber, and Meredith Rosenthal. What follows are highlights of their discussion. The video, along with a related reader forum and polls, can be found at NEJM.org.

The Payment System

Dr. Meredith Rosenthal: I think looking at reforming the health care payment system would be number one on my list for ways to start approaching cost control. We have very good reason to believe that because we pay for units of service, for visits, for surgeries, for days in the hospital, that we get more of these things. And there's good evidence from experiments and natural experiments that show very significant changes in patterns of health care when payments change.

Dr. Elliott Fisher: I think it's a subconscious influence on the way many of us practice. And a lot of it is driven by the environment within which we practice. Hospitals are paid the same way — and often are recruiting more physicians who provide services that have higher margins that are more profitable, so that the hospitals can balance their books.

Fee for service does not pay for us to have long conversations with our patients. When we're feeling constrained, it's much harder for us to have that long conversation with a patient with heart failure to see if we can safely manage them at home. The default position in many communities becomes, “Gosh, I'm too busy. I better send them to the emergency room.” The emergency room physician recognizes that they don't have time to manage the patient with heart failure in the emergency room, so admits them to the hospital. To the extent that those resources are available for us and we're not paid to do the things that we really would like to do or know we should do, we see huge differences in the likelihood of really unnecessary hospitalizations in different communities.

Dr. Atul Gawande: So if payment systems are one of the core things to address, how do you address it?

Dr. Jonathan Gruber: The right answer is managed care 2.0 done smart. And by “done smart,” what we mean is smarter on the input side, which is, don't just pay a flat fee, pay in a risk-adjusted way based on what's wrong with the patients; and, on the output side, don't just reimburse [providers] flat out, but reimburse them based on the quality and the value of what they deliver. And put [them] together in what we call an accountable care organization.

The problem is we're pretty far from legislative language. My feeling is [that] the science of coverage has gotten so far past the science of cost control that I feel like the President did a bit of a disservice holding up cost control as such a key feature of this round of reform.

Dr. Gawande: So what is a possible next step?

Dr. Rosenthal: There needs to be a transitional payment model that would be paired with a transition in the delivery system. Most of the proposals along those lines involve bundled payment of one form or another. Bundling payment for readmissions with the primary admission is the smallest piece of that. But there are payment models like Prometheus payment that are looking to put together all the components of care for a clinically defined episode and try to get various clusters of providers maybe acting as medical groups, maybe working together collaboratively to accept risk for those episodes.

A Public Insurance Option

Dr. Gawande: Then there [are] all these other possible approaches — one is the idea of a public option for insurance coverage.

Dr. Gruber: The important aspect of a public option is, will they regulate rates that are paid to providers within the public option? And that's really what the battle should be about. The less important part is who owns the plan. The truth is, if they're paying competitive private rates, it doesn't really matter that much.

Dr. Rosenthal: If the public option could come to the table with a reformed payment system, a radically different payment system, and perhaps not contract with every provider, allow certain accountable care organizations and others to come forward, it could perhaps innovate in a way that's just not possible based on the way health plans compete in our system now.

Dr. Fisher: Although we don't necessarily need a [new] public option to do it. We have, for the Medicare population, a large [provider] population that is very interested in some of these new ways of paying.

Dr. Gruber: The problem is, politically, we've tried Medicare pilots and they get killed. But if this legislation could bring the hammer down, say, “We are going to have these pilots even if local politicians don't want them,” then that would be a very significant step forward. We have a goal we'd like to achieve, which is to bring our health care costs under control in a sensible way. That's a goal that we don't know how to achieve yet.

Dr. Gawande: What about the argument, though, that we do know how: single payer?

Dr. Rosenthal: If you just look at Medicare today and all you did was expand it to the rest of the citizenry, then we'd still be facing the same problems that we're facing now. So I think you need a shift in social decision making around health care. There's been some conversation about creating an entity that would be somewhat politically insulated that could make and really implement decisions about reform — an entity with the legitimacy to guide the country toward making more reasonable tradeoffs in terms of how much and what we want to spend our health care dollars on.

Consumer-Based Approaches

Dr. Gawande: The other point of view is that we should be heading in the direction of creating medical savings accounts where people have high-deductible health care coverage and they have some skin in the game.

Dr. Rosenthal: We have lots of evidence that when people pay out of pocket, they don't make terrific decisions about what to seek and what not to seek. And because of the nature of insurance, we need to protect people against catastrophic risks, against high costs. This cost sharing would be limited to the end of the distribution of health care spending that matters least. What is driving high and growing health care spending in the U.S. is in the tail of the distribution, very high cost interventions and very sick people. And cost sharing is totally inappropriate there.

Dr. Gruber: There's no one who thinks seriously about this that thinks that a consumer-based approach can control costs in a fundamental way. Ultimately, the supply side is where it has to happen. That said, I think it's very important to learn the lesson of the managed-care backlash: people didn't see the savings. We had 3 years where employer-sponsored premiums grew by 2% or less a year, but people don't know that. And unless you give people a financial stake in what's happening, there'll be a backlash against accountable care organizations.

Comparative-Effectiveness Research

Dr. Gawande: Exactly how does someone doing research in Washington on whether one statin or another statin works, and having a result 5 years from now, translate into savings?

Dr. Rosenthal: We have a grave lack of head-to-head evidence on the effectiveness of clinical interventions. And without that information, commercial payers in particular, but also Medicare really lacks the ability to try to impose rules that might improve the value of health care spending.

Dr. Fisher: I think, as a clinician, it would be wonderful for us to understand better what the real risks and benefits of these treatments are so that we can help our patients make wiser choices. And that's the fundamental issue.

The other thing that's missing from the comparative-effectiveness discussion is that we really need to be understanding the effectiveness of different delivery systems and different approaches to delivery. Because that's where the real opportunities for both improving care and reducing spending are.

Dr. Gruber: It's impossible to argue that more information here wouldn't be good. And I think the notion that somehow making the information available will turn the government into Big Brother is just crazy within the context of our political system. Right now we are so far away from Big Brother, we are so hands-off, we do so little to really talk about what's effective, that we can only move in the right direction. It's never going to take out the huge share of decisions that are gray zone. But even a tiny effect on the growth rate is billions and billions of dollars over time.

Dr. Fisher: We see huge differences in growth rates across communities. You have San Francisco, San Diego, central Pennsylvania, many other communities where cost growth has been at least a percentage point below the national average. And no one in those communities is aware of rationing. None of the physicians in those communities would say, “We aren't providing the beneficial treatments.” So the kinds of deflections that we need to achieve trillions of dollars of savings are relatively modest. And they're mostly going to come, I think, from reorganizing care delivery, not from eliminating beneficial and wanted care.

Source Information

From Brigham and Women's Hospital (A.A.G.), Harvard Medical School (A.A.G.), and the Harvard School of Public Health (A.A.G., M.B.R.) — all in Boston; Dartmouth Medical School and the Dartmouth Institute for Health Policy and Clinical Practice, Lebanon, NH (E.S.F.); and the Massachusetts Institute of Technology, Cambridge, MA (J.G.).