Health care system, on verge of collapse, must heal itself, reformer says

Interview with Paul Ellwood, M.D., founder and chairman of InterStudy, a non-profit health policy research organization in Excelsior, Minn., discusses the causes of the current health care crisis, the need for reform and the likely results if no major reform is undertaken in the near future.

Published April 1, 1991  |  April 1991 issue

Since 1970, Ellwood and InterStudy have been working to introduce market forces into an industry that has been relatively free of competition. Ellwood is credited with coining the terms health maintenance organization (HMO) and preferred provider organization (PPO), and his efforts with the Nixon administration were crucial to the passage of HMO-enabling legislation in the early 1970s. He and his 20 professional associates at InterStudy have also been instrumental in developing many other health care reform ideas, including the current movement to measure and manage the effect of health service on function and well-being.

fedgazette: In recent years, health care costs have been rising at about twice the rate of inflation; as measured in gross national product, health care is about 12 percent, twice that of the defense industry; and more and more people, including the fully employed, cannot afford health insurance. How did we get to this point?

Ellwood: There are many reasons, and I'll do my best to list them for you. First of all, consumers overvalue health care and have no idea what it's worth. Second, the industry structure and incentives have worked against efficiency. Third, advances in technology have been costly.

Fourth, we're having difficulty converting good science into good service. We're having difficulty being sufficiently discriminating about who will benefit from medical care. We have excess demand and an industry that isn't structured to be efficient, and then we have rapid changes in science and technology that have been difficult to keep pace with.

Finally, the health system has not done a very good job of using information. It hasn't made the same sort of progress that other industries have made in interpreting, storing and disseminating masses of data, particularly on outcomes of medical care.

And I'll add a further word about cost: the annual U.S. expenditure on health care is now so huge, in excess of $600 billion dollars, and prices have been rising so fast, that the combination of these things is simply pricing health care beyond everyone's means. There are startling comparisons that illustrate this point, for example, the rise in cost in medical care last year was greater than the cost of Desert Storm; the amount of money that the United States spends on medical care is larger than every national economy in the world except for five.

The other dilemma we have is that there is so little concentration in this largest of industries so that control remains in the hands of a powerful, autonomous profession. The result is an industry that resists change. The health industry has typically been one that U.S. presidents and business CEOs have been reluctant to challenge. They just assume that it's too complicated. And if business leaders get involved, it's often as members of boards of hospitals that perceive of reforms as detrimental to their power and financial ability.

fedgazette: Assuming no changes are made in the health care system over the next 10, 20 or more years, what would be the future of health care in this country?

Ellwood: At current rates, our doubling time for spending is about seven years, so we'll be dealing with expenditure levels in excess of a trillion dollars very soon. Let me read you this statistic: "Between 1980 and 1987, real median household income grew by 6.4 percent, or $1,560. At the same time, average real household health care costs grew by 35 percent, or $1,412." That means 91 percent of the real income gain was spent on health care.

Rising medical care expenditures are jeopardizing the economy. They are jeopardizing improvements in standards of living, reducing the amount of money we have to spend on education or housing or other worthwhile services. And, to make matters worse, what we're getting for our health care dollar is very disappointing. The infant mortality rate in Japan, which spends less than half as much for medical care as we do, is twice as good as ours. Their life expectancy is five years greater than ours. All of which says that lifestyle is probably more important than all the things we do with medicine.

So we've got an industry that's truly in crisis and an industry that people are reluctant to challenge, and yet those same people have the ultimate personal stake in the health industry's ability to maintain their health. Everybody needs a doctor from time to time, and if an executive or political leader advocates a change, chances are they'll have a doctor telling them they're dead wrong, that they're going to be killing people by undertaking the proposed reforms.

fedgazette: What type of reforms does InterStudy suggest?

Ellwood: We think that any effort to reform the system has to be comprehensive, that we've gone as far as we can with piecemeal reforms. We have three objectives: one, controlling costs; two, improving access; and three, improving quality and effectiveness.

To do this, I believe that first we have to have universal health insurance coverage. We can't have 10 percent or more of the population either uncovered or inadequately covered. Incidentally, most of the people who are uncovered are employed at least part-time. Secondly, we have to confine universal health insurance to those forms of treatment that are effective and at rates that are appropriate for the treatment. Also, health care must be delivered by efficient, high-quality organizations that are publicly accountable for their impact on health.

All of these things combined will result in a situation where we can have greater competition and diversity. Buyers of health care would be better informed, and the same accounting procedure used to inform the public could be used to monitor treatment and to inform practitioners of effective procedure.

Our goal is to make health care organizations accountable for their results, and to establish potent perceptions of their function and well- being, or quality of life, as a new bottom line in medical care. Sound, scientifically defensible methods are available for comparing the effectiveness of health care. We just need to use them.

There are a fiscal set of analogies that can be made in medical care right now, in fact, we're interested in creating something analogous to the Financial Standards Accounting Board for health results. We would call this the Outcomes Management Standards Board, which would specify the types of information that health care organizations would have to collect about their performance, how they would collect the information, analyze it and make it available to the public. And then we would like to create something akin to the Federal Reserve Board—the Federal Health Board—that would register health organizations on the basis of their accountability and would be responsible for identifying the uniform set of health benefits to be covered by health insurance policies. The scope of uniform effective health benefits would be adjusted from time to time, just as the Fed adjusts the discount rate.

All of these things would be done for the same reasons the Securities Exchange Commission (SEC) was created, to improve the performance of the markets. But it's our contention that if purchasers don't have better insight into what they're buying, the health market won't work. The trend in medical care is for employers to play a greater and greater role in determining where their employees go for medical care.

What we have here is the possibility of increasingly sophisticated employer/buyers making demands for more definitive and discriminating information. Right now, the principal source of information is claims data, which tells group buyers what they are paying for but doesn't tell them if they bought more rapid return-to-work and more productive employees.

If the United States is to rely on market forces we have to have a market, we have to have multiple buyers. We can't have a single buyer—namely, the government. If we are to have multiple buyers and universal coverage, this inevitably leads to employers being required to provide health insurance or to be willing to pay into a larger pool. Free riders will literally kill the health market.

fedgazette: More specifically, how do you see competitive forces at work in the health care industry?

Ellwood: I see, first of all, health care largely delivered by organizations rather than individuals, with these organizations managing their outcomes and results and then reporting those to buyers. Buyers, that is, group purchasers and employers, are going to have to become more aggressive about what they purchase.

In the early 1970s—in the Twin Cities—a very aggressive health care market was developed. It's my view that purchasers of health care failed to take advantage of the situation. To take advantage of it, purchasers are going to have to do two things. First, they're going to have to go to defined contribution plans. In other words, they will have to set aside a certain amount of money per person for health care, and each year they might modify that amount of money, but it will be a fixed amount—unlike defined benefits plans that pay whatever it costs to cover the benefits.

One of the easiest ways to make this benefit change is to offer a cafeteria-style benefit plan; for example, an employee is allowed a certain amount of dollars for health care, but the employee can also choose to spend that money on other fringe benefits, like pensions, vacations and so forth. It's a trade-off. The other way employers can exert greater influence is to be more prescriptive about where employees get their care. The most popular form of this approach is called point-of-service health benefits: the employer identifies the place where employees can get their full range of coverage, if employees use doctors or facilities that are not part of the plan, the employer contribution will be greatly reduced.

On the matter of accountability, we have never, as a matter of standard operating procedure, followed patients' recovery precisely. Each health facility has their own method of following patients, but most have no sophisticated method for tracking the impact of medical care on the patients. Furthermore, the health system does not describe patients in a single, uniform manner to make comparisons between one doctor and another, or one hospital and another. So the health system has no scoreboard, as it were, of effectiveness. And, as a result, the likelihood of various treatments benefiting the majority of patients is relatively low. We're having a great deal of difficulty selecting which patients are going to benefit most from treatment.

In tracking health outcomes in the past, we have tended to focus on death as an outcome. We have fairly good systems for tracking cancer patients, for instance, because we have such a clean end-point: death. On the other hand, the treatments for cancer produce all sorts of disturbances in people's quality of life—pain, hair loss, speaking loss, impotence, scarring—consequences for some that are worse than a prolongation of life. What we now must do is move to a system that allows us to compare the results of one medical firm against another and one treatment against another in terms that are meaningful to the patient.

To that end we have very good systems now that are evolving to do that. InterStudy has devised an outcomes management system that is being tested by 28 large American corporations and health insurers. It will accrue health outcomes information, analyze it and provide it to purchasers, patients and providers. The primary idea is to strengthen the market.

fedgazette: What has been the medical community's reaction to outcomes management proposals?

Ellwood: Very favorable. If I were to compare my experience in promoting HMOs and PPOs and managed care, to moving outcomes management, doctors have reacted more favorably to the latter. Doctors, of course, like the idea of knowing how their patients are doing.

Our failure to track patients well is a technical problem more than anything. It requires huge amounts of data, collected over long periods of time, and it has to be collected by more than one practitioner. The greater degree of organization of the health system and cheaper computing power, make outcomes management feasible.

The only thing doctors are worried about with tracking results is premature exposure of differences that are not scientifically justified. For example, even though patients may have the same diagnosis, there are great differences in severity of their conditions, and therefore differences in their prognosis even though they both received the same treatment. Adjustments must be made for the severity of the disease, the presence of other diseases, education, income, housing, support systems—all contribute to how well patients cope with a disease and whether a treatment works.

The outcomes management plan is an attempt to treat health care as if it's a continuous clinical trial, as if it's a continuous experiment, where every patient is compared with every patient like them across the country. The end result would be an improved ability to select treatments so patients can have a better idea of what the impact of treatment is likely to be on their lives.

On the matter of information, medicine has really lagged behind banking and other industries in its use of computers. We don't have, for instance, computerized medical records systems at a time when it's quite possible for an individual to call in and be queried by computer. During the phone call the computer could compile a medical record on the caller, give the caller suggestions on how to treat the condition or tell them that their doctor will be calling them shortly. All of this is now feasible. Smart cards, whereby a patient's medical record would be stored on a card, could be invaluable in medicine.

fedgazette: If universal health coverage is a foundation of health care reform, how would we attain such coverage?

Ellwood: The ideas I prefer are those of the economist Alain Enthoven of Stanford University. Alain would start with small employers, pooling their buying power, and then move toward a requirement over perhaps 10 years for all employers to provide health benefits in what we call a pay-or-play plan. You either purchase health care for your employees or you contribute a certain percentage of their wages to a pool to purchase health care.

I would also advocate capping the tax exclusion for health benefits. There are few fringe benefits as open-ended as this exclusion, and the tax losses for health benefits for the upcoming year will exceed $60 billion. If we could cap that tax exclusion at some reasonable level, we could then take some of those savings to use for health insurance for the poor.

Getting universal coverage is a staged process, beginning with more economical medical care for small employers and other general reforms that will help get costs in control, followed by a cap on the tax exclusion to help fund the coverage.

There's just no way to get at universal coverage in a market-oriented health system without the employer being required to cover their employees up to a certain level.

fedgazette: Some would suggest that Americans have a "cure me at any cost, I don't want to die" attitude that goes a long way toward fueling our rising health care costs. Some other countries, for example, impose age restrictions on certain types of treatment. Do Americans expect too much from their health care system?

Ellwood: Yes, I think they do, but I think that's driven as much by the health delivery and health research sector as it is by Americans' desire for eternal life. We've led people to think that medical care can do more for them than it will. Every Thursday in USA Today there is a major health story because The New England Journal of Medicine came out the day before with the latest breakthrough in medical research.

I think we've acculturated people to assume that the health system can do more than it actually can. Even procedures that are very, very good help only a limited number of people. Take coronary bypass surgery—this is a common operation that prolongs life for perhaps 10 percent of the cases. So, if we could find that 10 percent among those coronary bypass candidates, we could save 90 percent of the expenditure. And coronary bypasses account for $8 billion per year in health costs.

It's not a matter of rationing, rather, it's a matter of rationality, of identifying the individuals who can really benefit, and that brings us back to effective outcomes management.

I might add something about methods of reform here. I believe that most of the reform in health care is going to have to come from within the industry itself. Purchasers and political leaders lack the interest, the guts, the sophistication to deal with the issue. The technique that I found best for reforming the system is to get health organizations to experiment with new methods of improving performance and then have those leaders promote those ideas.

One of the roles that InterStudy has come to play has been to persuade leaders of individual health organizations that they have a broader responsibility for the health system as a whole. There are a number of people in business, politics, labor and the health industry who are working with these broader responsibilities in mind.

fedgazette: You say that reforms are going to have to come from within the health care industry. Why should the health industry reform itself? It seems to be doing quite well with the status quo.

Ellwood: That's it, that's what makes it so hard. What's unique about reformers, though, is that they can see beyond their own short-term advantage. People must be persuaded that the larger context is failing. None of us is sure what form the collapse of the health industry might take, or when, but we think it's inevitable. The poor quality performance of the health care industry and the extraordinarily high levels of expenditure are not justifiable.

As for the politicians, the reason most have avoided addressing health care issues is because they are too complex and subtle. It's too uncertain, too difficult, too thankless an undertaking. And furthermore, the politicians seem to want one-shot quick cures and we're dealing with a problem that is going to take years to resolve. The most aggressive, precise, decisive plan today to reform the health system will take us 10 to 15 years to install.

We will need a Federal Health Board that would have responsibility to carry out these reforms, for defining the tax deductibility of health benefits and to require health care organizations to be accountable for their results.

fedgazette: Aside from the problems facing the rest of the health care system, the rural health care network faces somewhat unique problems, among them are long distances between medical facilities, a dwindling customer base and the inability to find and retain medical personnel. Do your recommendations for reform also apply to the rural health care system?

Ellwood: Yes, I think having health care delivered by organizations that would have branches in rural areas would work just fine. We have some beautiful examples of that within the Upper Midwest—based on various medical group practices.

Also, there's very little evidence that the lesser amounts of health resources in Upper Midwest rural areas are having an adverse effect on people's health perhaps because they live healthy lives. They're not exposed to the detrimental effects of the city, like pollution, for example, and they generally get more exercise. So, I'm not persuaded that the lesser intensity of health care that's available in rural areas is necessarily bad for rural residents' health. It's inconvenient to have to go significant distances for health care, but we recently completed a study for the Northwest Area Foundation which found that people in rural areas who had to travel out of town for health care were often more satisfied than those that were getting it in town. And we weren't able to establish measurable differences in health status that were attributable to the intensity of health services in the community.

fedgazette: We've talked about what may happen in the future of the health care industry, what is your estimation of what will actually happen?

Ellwood: If we can't achieve what I've described to you, the government will have to take over this industry and will have to control both supply and demand. It will set prices of services and control availability of health care through queuing and rationing.

And the big problem—when you attempt to control prices in health care—is not setting the price, it's controlling volume. If you restrict a hospital or doctor to a certain amount per procedure, then the volume of procedures goes up. Also, if you inspire incentives by introducing competitive measures, it won't do any good if nobody knows which procedures really work. All in all, the pieces of this issue are so interlocked—as I stated earlier—that financing, structure, accountability, regulation and tax policy must be reformed simultaneously.


Related Links