Competition the Cure for Healthcare
Michael Porter is considered by many the world's foremost authority on competition and strategy. He discusses the need for fundamental reform in the way the United States delivers healthcare. Q&A. Key concepts include:
- American healthcare is broken structurally, rewarding the wrong actions and punishing the patient.
- Competition correctly placed in healthcare can reduce cost, improve physician performance, and create better results for patients.
Editor's Note— Last month HBS Working Knowledge offered an excerpt from Redefining Health Care: Creating Value-Based Competition on Results, by Harvard Business School professor Michael E. Porter and Elizabeth Olmsted Teisberg. The U.S. healthcare system is dysfunctional, a Rube Goldberg contraption that rewards the wrong things and doesn't create value for the consumer. In this Q&A, Porter discusses his research.
Roger Thompson: What went wrong with the American model? On paper it looks ideal. It's private, it's competitive, yet it doesn't seem to work.
Michael E. Porter: The United States has a system with the wrong kind of competition, on the wrong things. Instead, we have a zero-sum competition to restrict services, assemble bargaining power, shift the cost to others, or grab more of the revenue versus other actors in the system.
Zero-sum competition does not create value; it can actually destroy value by adding administrative costs and leads to structures involving health plans and providers and other actors, which are misaligned with patient value. In a world of zero-sum competition, for example, providers will consolidate into provider groups to gain clout against insurers. But, as we point out in our book, the provider group doesn't create any value. Value is not created by breadth of services but excellence in particular medical conditions.
Zero-sum competition was a natural evolution given the historical roots of the field. In some ways, the field of healthcare is stuck in the past. Health plans think of themselves as insurance companies because that is what they started out doing. Providers are organized around the old functional structure of specialties, rather than integrated care organizations. It is a lot like the distinction between functional structures and business unit structures in management thinking, except that healthcare is still stuck in the functional model.
These forms of competition and organization have also been institutionalized in medical education and in physician certification. Nobody deliberately set out to create this mess, but unfortunately, we are left with how to deal with it.
Q: Fifteen years ago during the first Clinton Administration there was a national appetite for some kind of major healthcare reform. Do you see any such window reopening given five years of double-digit increases in health insurance cost?
A: The window is certainly opening, and it is a global window. All advanced economies, even those with universal insurance, are starting to confront the reality that the healthcare system is not aligned to drive value for the patient, or the health outcomes per dollar spent. The roles, strategies, and structures of the various actors in the system are not designed around value, nor is competition reinforcing improvements in value.
For example, let's take the prescription drug benefit, the big reform of today that everyone is talking about. This is not a real reform—it is just putting more money into the system, and the latest example of zero-sum competition. If drugs are too expensive, let's get the government to pay more. Now that healthcare costs have gotten so high, it is squeezing consumers, employers, state governments, and federal governments. The pain is so great that thinking is beginning to open up.
The United States has tried a lot of models. We tried managed care, gate keepers, co-pays and the like, and they failed. When we published our earlier article on healthcare in June 2004 in the Harvard Business Review, we were far less optimistic. But we uncovered so many interesting experiments and changes in the practitioner community that we were drawn to try to operationalize our thinking into recommendations that would allow the actors to do things differently, and dramatically increase the value we could achieve from all the money being spent.
Q: What are the key steps? Where do we begin?
A: One of the most hopeful things we discovered in the course of this research is that the revolution, if you will, has already started. The U.S. [healthcare] system can be reformed from the bottom up. Any hospital, physician practice, health plan, or employer can take positive steps in the direction of value-based competition today and be better off, even if nobody else changes. They will be more efficient, achieve better customer satisfaction, and the physicians involved will feel more pride in what they are doing. We included many examples in the book, such as the Cleveland Clinic, M. D. Anderson Cancer Center, and Dartmouth Hitchcock Medical Center, which are moving toward results measurement and integrated practice unit structures.
Reforming the U.S. system does not require a top down, big bang, government-led regulatory change. That having been said, the government can actually help a lot through modifying and extending key policies, particularly in the area of results information and removing restrictive and unnecessary impediments to competition. The corporate practice of medicine law is a perfect example. The law makes it difficult for doctors to be employed by corporations, which makes no sense today because we want doctors and hospitals to work seamlessly as an integrated practice. Another example is physician certification by state, which means that when the Cleveland Clinic wants to offer a national second-opinion service, they have to scramble around within their staff physicians to find doctors that have certifications in all the states even though the Clinic is a preeminent medical center. There are a striking number of other examples such as this, which actually work against delivering value to patients.
So government does have a key role to play, but luckily, we do not have to wait for government.
In other countries such as the United Kingdom or Germany, government is far more directly involved in delivering healthcare. In these countries, change will be highly political. But in the United States, and this is one of our great strengths as a country, we tend to be more fluid and flexible. Once Americans understand the problem, we can make considerable headway.
Q: Would you be surprised to see some of your ideas surface in the next presidential campaign?
A: I would be surprised if healthcare was not a major issue. I would hope that our ideas would surface, because previous reforms have missed the mark. There is no plan for American healthcare right now; nobody has one. The only plan is to put more money into the system.
Q: You must have seen the recent report about the massive differences in Medicare costs from state to state, and even within states, with no apparent difference in care. How do you explain those differences?
A: The recent findings are just the latest in a body of important work by John Wennberg and his colleagues at Dartmouth. Those findings were an inspiration for our work. Indeed, they were the last piece of the puzzle for us in truly understanding the problem. We have known for a long time that U.S. healthcare is high cost. What everyone assumed was that U.S. healthcare was very high quality, which at least made the high cost understandable.
What Wennberg did, as did others, was to show that U.S. healthcare is not high quality. Indeed, there is every possible quality problem you can imagine: incorrect diagnoses, drug errors, unnecessary complications, and failed treatments. Also, there are huge differences in quality across providers and across geography.
For us, this framed the problem. How could we have a high-cost system that also has poor quality, where providers differ so much in value but these differences persisted? In a world of normal competition, that cannot happen. The high quality players grow and thrive, and the poor quality players go out of business or fix their problems. So finally, the right question became, why is competition failing? And that led us to the distinction between zero-sum competition and positive-sum competition, and the central importance of value.
Q: Why isn't high quality, value-driven healthcare more expensive?
A: Healthcare is not like buying a car. If you want leather seats in a car, this costs more because leather costs more than plastic. If you want a TV set with a bigger screen and more features, that is more expensive; it takes more circuits, more material, and so on. Healthcare is very different, especially today when we already treat virtually every medical condition in some way. Most of the time, the best quality healthcare is also the lowest cost care. The reason is that the lowest costs arise when the patient stays healthy, or gets healthy faster.
If you get the diagnosis right, you save a lot of wasted and unnecessary treatment, and costs go down. If you avoid making mistakes, costs go down. If excellent surgery allows the patient to go home sooner, costs go down. If you actually cure the disease, the patient does not need to have any more office visits or drugs. And so on, as we discuss extensively in the book. Of all the fields we have worked in, this is the field where the notion that quality is free is the most powerful.
One of the central themes of our book is that the way to drive down costs in healthcare is to drive up quality. That is the dynamic we have to harness. And in order to drive up quality, there is only one way that will work: We have to measure results. And in order to use results to drive quality, we have to create competition on results at the medical-condition level.
Q: But you also note in your book that physicians have been resistant to the idea of measuring results because they fear the information can be used against them in malpractice lawsuits. Wouldn't you expect that to be the case?
A: We believe that the more you measure results, the fewer lawsuits you will have, in part because many lawsuits grow out of ignorance. Today, if there is a bad outcome, the assumption is that the doctor did a bad job. But with evidence on outcomes and true risks, the burden of proof to sue a doctor will rise. Doctors will learn to be happier in a world where they are measuring results more than they are today.
We take the difficulty of results measurement head-on in our book. The book documents a number of medical conditions in which results are measured today, including complex areas such as organ transplants and cardiovascular surgery. The book shows that it is possible to make sophisticated adjustments for patient risk. We need to get on with measuring results, and the ability to do so will certainly improve. Gaming of the system will be corrected. The whole process of working hard to measure results better and better is extremely good for patients.
If doctors can demonstrate results, they will not have health plans looking over their shoulders and trying to second-guess their practices. Today's quality movement in healthcare is a welcomed step in the right direction, but it is not really focused on results; it is focused on methods, or processes of care.
Q: What about pay for performance?
A: Most pay for performance programs are actually pay for process compliance, not pay for results. While many in the field understand that such programs must move toward outcomes, there is a deeper problem. Pay for performance assumes you should have to pay more for high quality.
As we already discussed, in healthcare excellent providers are often also the most efficient. Patient flow to excellent providers in addressing a medical condition is a much more powerful incentive than a small bonus because it feeds the virtuous circle of value improvement from volume and experience. As patients seek out the excellent providers who demonstrate good results, there will be rapid improvement of other providers, and a reallocation of capacity. Physicians practicing at institutions with little volume and substandard results will shift to other fields, or move to organizations and management structures where they can succeed.
Doctors will come to understand that there is no inherent right to practice medicine unless the organization can demonstrate good results. This is a much better way to be held accountable than today's flawed malpractices system.
Q: How would health plans change? What new roles would they take on?
A: Health plans have eroded the trust of many of their subscribers during the era of gate keeping, denial of claims, and network restrictions. As a result, some believe we ought to get rid of health plans and move to a single-payer system where the government is the payer. We disagree, because health plans have important value-adding roles in the system.
Why do we need health plans? We need them to integrate across all of an individual's health needs. Some make the case that an integrated provider system can play the role of integrator. We reject that idea because it is crucial to have competition at the provider and medical condition levels.
So a value-based system keeps the health plan separate from the provider, and providers compete at the medical condition level. The health plan is also indispensable in aggregating information. We think the health plan is the logical place in the system at which to aggregate medical records. Right now, the medical record resides with providers, and one provider can request the record from another. That is a very cumbersome and inefficient system, which creates delay and duplication.
Health plans should also inform and advise members about where to seek care, and help to navigate the care cycle. Many health plans, among them Harvard Pilgrim and Aetna, are starting to move in these directions. Health plans must become health organizations, rather than see themselves as payers or insurance companies.
Q: Could health plans help push along the movement toward electronic patient records?
A: Health plans have an important role in encouraging medical records because it will help them drive higher value in the system. Health plans can create incentives and standards to rapidly disseminate electronic medical records to all the actors in the system. There is a critical need for standardization of data and protocols so that records can be aggregated. In Massachusetts, for example, Blue Cross Blue Shield has invested heavily in a pilot program to develop a medical records platform that could be shared.
But as we emphasize in our book, IT is not the solution to the problem of healthcare delivery, but an enabler. Automating current care delivery processes will have limited benefits. Conversely, there is much that can be done to radically improve value without fully electronic records.
Q: What do we do about 45 million uninsured people?
A: We have to get them insured. Universal insurance is not simply fair; it is also the only way to truly achieve a high value system. The United States already provides emergency and acute care for the uninsured, but we go about it in the worst way imaginable. We avoid the most cost effective spending, which is in primary care and preventative care. We treat the uninsured only after they get sick, and in expensive settings, and we already pay for this. This is paid for with cross subsidies and charity. The cost of moving to universal insurance will be less than sometimes supposed because the system is already bearing much of the cost and there will be savings from providing preventive care and disease management.
To get to universal insurance, however, the problem is not just the poor. In Massachusetts, state government discovered that a significant proportion of the uninsured had incomes of $90,000 a year or higher. To get to universal insurance, not only must the poor get help, but health insurance must become mandatory.
Q: I was startled reading that in your book.
A: Individuals who are young and healthy may not think insurance is necessary. But the whole logic of insurance is that everybody needs to pay every year so that they contribute enough to pay their costs if they get sick. It is the same principle as in auto insurance. You do not buy car insurance the day before you are going to have an accident. Universal and mandatory insurance should go together, with subsidies for those who need them.
We also need to get away from the idea that either you have enough money to buy insurance, or it should be free. Individuals should contribute to the cost of their insurance to the extent that they can. This will also reduce the cost of moving to universal coverage.
I want to emphasize, however, that universal insurance is not a solution. Universal insurance enables the solution. The only solution to the healthcare problems in America is to dramatically improve the value we are getting for the money we are spending.
Q: People who have bad health habits, who smoke, drink excessively, don't exercise, they're overweight, how do you factor that in what they should be paying for insurance?
A: This is a controversial issue, but a crucial one. Historically, individuals have not been responsible for behavior that affects their health and healthcare. But healthcare is a co-produced product. The doctor and the patient produce healthcare together. If the patient does not participate, it limits what the doctor can accomplish.
We are moving to a system where the individual must be engaged, or face the consequences. More employers are asking employees to participate in health risk screening and disease management, or bear higher premiums for their health insurance.