MANAGED CARE May 2006. ©MediMedia USA
"How could the largely private U.S. health care system, characterized by arguably more competition than any other health care system in the world, be performing so poorly?"
Michael E. Porter, PhD, and Elizabeth Olmsted Teisberg, PhD, set out to provide a new way to look at competition in health care in their book, Redefining Health Care: Creating Value-Based Competition on Results, published this month by Harvard Business School Press.
Porter, the Bishop William Lawrence University Professor based at Harvard Business School, is a leading authority on competitive strategy and the competitiveness of nations, regions, and cities. Teisberg, an associate professor at the University of Virginia's Darden Graduate School of Business, has focused her research on strategy, innovation, and risk analysis.
Their book caps a decade of study of competition in the American health care system. In it, they argue that health care organizations define their business too broadly, providers define their role too narrowly, and both are limited by a mindset that is too local.
What's needed is competition over results in the form of risk-adjusted outcomes data. Failure to overhaul the nature of competition could result in the country turning in desperation to a single-payer approach that doesn't address the central issue of improving value.
Teisberg holds a master's degree as well as a PhD in engineering-economic systems from Stanford University's School of Engineering, a master of engineering degree from the University of Virginia, and a bachelor's degree from Washington University in St. Louis. Porter holds a PhD in business economics and an MBA from Harvard and an undergraduate degree in aerospace engineering from Princeton. Both authors spoke recently with Senior Contributing Editor Patrick Mullen.
MC: You argue that competition in health care today is too broad, too narrow, and too local. Explain what that means.
PORTER: Typically, institutions define their business too broadly, as health care or as a hospital. That's misaligned with value and can obscure value differences. A hospital may be great at treating one medical condition but substandard in another. If it competes simply as a hospital, that obscures those differences. Health plans have to see themselves as overseeing care for a series of medical conditions rather than providing health insurance, because that's not where they can add value. Competition is too narrow in that the system is organized around discrete services, interventions, or specialties, or acute care versus chronic care. All those things need to come together in a cycle of care in an integrated way. With major stroke care, 95 percent of the cost of care isn't in the hospital. It's for rehabilitation. If you try to understand the cost and value of stroke care without looking at the cost and duration of rehabilitation, you miss fundamental insights about how to drive the value of care. The third part of the current competitive mismatch is that care is too local. The system presumes that every consumer or patient should get health care from the nearest local provider. Therefore, that provider has to offer too many services. Another presumption is that patients won't, shouldn't, and don't need to travel. For many services, patients should at least be willing to seek out excellence within a metropolitan area. Better information would support their ability to do that. If services must be provided locally, they don't have to be managed locally. We need a system where excellent providers can manage networks of services across geography.
MC: Cleveland Clinic running Rochester General Hospital's cardiac program is an example you discuss in the book.
PORTER: Absolutely. Heart surgeons at Rochester General work for the Cleveland Clinic, are trained at the Cleveland Clinic, practice and measure results to the standards of the Cleveland Clinic, and enjoy all the purchasing advantages of the Cleveland Clinic. That's the way to think about how to drive value. Excellent providers should go across geography rather than be restricted to being independent standalone local providers.
TEISBERG: Competition needs to drive improvement in quality and efficiency in addressing specific medical conditions or combinations of medical conditions over the full cycle of care, rather than shifting costs or driving costs up. Competition based on things like cost shifting is a zero-sum game.
MC: Which is how managed care operated through the last decade.
PORTER: Exactly. Conceptually, managed care did not have to operate that way but that's how it got defined. The managed care moniker is pretty broad, and it could have been applied in a lot of ways. It could have meant managing medical conditions over the care cycle. Instead, it got defined as exercising bargaining power to shift costs. A tremendous irony in health care is that if you look at the world in terms of how to reduce costs, you'll usually raise costs. One principle of value-based competition is that higher quality means lower cost. You have to ask how to increase value and drive quality up. That's ultimately the only way to reduce cost.
MC: How far along are health plans in the transition from that zero-sum game — which everyone now sees didn't work — to the role you advocate?
TEISBERG: One important thing they're missing is how to reward physicians. They still think only in terms of higher pay. Health plans need to understand that it makes sense for them to steer subscribers to physicians who treat particular conditions most effectively. That will help drive quality up and costs down, and the added patient volume rewards physicians and care teams that do the best jobs for a particular condition. The point of the health care system is not to decrease cost; the point is to increase health and increase health care value for patients.
PORTER: Almost all health plans have moved away from the most egregious tactics they used for a while, but even the best still fall short of truly playing the roles that we suggest. We find a lack of realism among some health plan executives about how far they've moved. They'll say, "Oh no, we don't have restrictive networks anymore." But because of the pricing model and reimbursement structures, individuals who go out of network end up paying two or three times as much. So in effect, there still is a network. Plans are still restricting choice of providers.
MC: Where do you see improvement among plans?
PORTER: Health plans are grappling with how to measure value and results. It's tough because there's not a lot of data but some have moved further than others. Health plans are doing a better job of providing medical advisers to work with patients. We see more disease management and disease-prevention efforts. Those are positive signs.
TEISBERG: Health plans and employers need to get bold about letting go of discounts in the interest of getting rid of sky-high list prices. The whole dynamic of the system will continue to drive in a bad direction until that vicious cycle is stopped.
PORTER: The list price discount approach creates tremendous inefficiencies. Value discounts are a legitimate way to increase efficiency and value in the rest of the economy, but they don't work in health care. Health care is not a commodity, and treating a patient is specific to that patient, not the health plan of which he's a member.
TEISBERG: Discounts tend to be given over a very wide variety of medical conditions, as if you were going to treat the patient the same no matter what he does. Discounts end up driving more people into the uninsured pool and driving list prices up by more than the discount they're getting. Plans would benefit very quickly by letting go of those discounts.
MC: You contrast raising standards for all physicians by using such tools as practice guidelines with rewarding excellence. How does the system move toward identifying, recognizing, and driving business to the best providers?
TEISBERG: We need to reward results, rather than compliance with process standards. A lot of effort has gone into micromanaging providers, but we still have huge problems with variable quality and frequent errors. Today, rewarding results tends to mean pay for performance, and the assumption is that you must pay more for better performance. You can reward results by allowing more and more patients to go to doctors who get the best results. That drives a virtuous cycle of care. It enables physicians to become more efficient, to gain expertise, to develop better techniques, and to benefit from economies of scale.
MC: You argue that there will be a difficult transition in health care over the next several years, but that absent the kind of changes you advocate, that change could come in the form of a single-payer, top-down system.
TEISBERG: Doctors will be far better off competing to improve results than if they're managed from the outside and told what to do. If we have outside specifications dictating exactly how doctors are supposed to treat patients, doctors can't keep up. We would make some general improvement, thanks to the kind of work that the Institute for Healthcare Improvement is doing in preventing medical errors. We need to set clear goals by publishing risk-adjusted outcome data and then let providers achieve excellence by improving means or processes themselves. It's not realistic to specify everything about a patient that a doctor needs to consider, because patients have individual circumstances. You have to allow for judgment. When you measure results, the best providers continue to improve. It's not just that the lower part of the curve rises toward the middle; the whole curve moves up.
MC: You talk about the need for physicians to accept joint responsibility for a care cycle as part of what you're calling integrated practice units. How will physicians change their thinking?
PORTER: Physicians think about their particular role in the care cycle and try to do a good job, but they don't tend to think very hard about what happens before they play their role or what happens afterwards. But we know what happens before and after a doctor provides care has a fundamental impact on the ultimate value created. I have a cranky hip so I go to a physical therapist. I talk to him as he's contorting my hip to try to loosen it up. He says he can tell within a day or two after joint surgery whether a patient will have a problem just by how the patient is responding. But there's no feedback cycle between the physical therapist and the surgeon. The surgeon doesn't know what happened to the patient after surgery. If you want to deliver value, you have to accept some responsibility for what happens before and after. There's no way that outsiders can address which physicians are good and which are not, which costs are necessary and which are not. Physicians have to work that out amongst themselves by holding themselves jointly accountable. If they're held jointly accountable, there will be some very interesting discussions about how they're working together and whether each is carrying their weight.
MC: One key to measuring results is having risk-adjusted outcome data. You point out that a number of institutions gather such data, and that there are no huge technical obstacles to collecting it. Yet some doctors remain wary. Why is that?
PORTER: We have gold-standard outcome information for some of the most complicated medical conditions, such as heart transplants. We have valid, accepted risk adjustments in those areas. These are dealing with very sick patients. There's no reason why we can't do this for every medical condition.
TEISBERG: We've had good, accepted, peer-reviewed risk-adjusted outcome measures in intensive care for 25 years. Those are complicated, diverse, life-and-death cases, so measuring outcomes can be done. Even when very good outcome measures are available, there's resistance to using them for things like accreditation. How could it not be in physicians' interest to embrace that kind of measurement? Anyone doing any job wants to do it well and wants to know that they're doing it well.
PORTER: This issue has not been framed appropriately. We applaud efforts to improve quality, but most of the quality movement is really process improvement. We have not yet encountered a physician who doesn't want to improve the value he delivers to patients. The only way to do that convincingly is to measure results.
TEISBERG: It's important that a doctor who's getting sicker patients isn't penalized.
MC: What's your opinion of consumer-directed health plans?
PORTER: Supporting and motivating patients is critical, but we're not great fans of the consumer-driven model. The idea that a consumer has a health savings account and goes out and shops for care absent good information is not in and of itself the solution. If you're confronting a system that's fragmented, too broad, too narrow, and too local, it's like going to a crummy store. You don't have very good choices.
MC: If the changes you advocate don't happen, how much time do we have before pressures force an ill-conceived reform?
PORTER: We have two to five years, but we don't have a decade. The discussion about paying for health care would be totally different if we had results information. The changes we talk about will require a phase-in period. I'm confident that our book will have an impact. Our traditional constituency in the business community is going to start putting more and more pressure on this issue because its time has come.
TEISBERG: Just a couple of years ago, leaders throughout the health sector were starting to suggest that perhaps the only solution was to go to a single-payer system. The tide is turning as people start to understand there is an alternative.
MC: Thank you.