In November 1999 Prof. Alain Enthoven gave The Donald C. Ozmun and Donald B. Ozmun and Family Lecture in Management at the Mayo Clinic, Rochester, MN
The "managed care revolution of the 1990s" achieved an important, if temporary, success. It stabilized health insurance premiums and National Health Expenditures (NHE) as a percent of GDP for 5 or 6 years. But it now appears to be failing.
We see a great outpouring of anger, resentment, hostility and frustration expressed in state and the federal legislatures as doctors, other providers and many patients express strong dissatisfaction with managed care.
Moreover, cost containment seems to be breaking down. Average annual premium increases for large purchasers are running around 9 or 10 percent. The actuaries of the Health Care Financing Administration predict that National Health Expenditures as a share of GDP will rise from their present 13.9% to 14.6% in 2001 and 16.0% in 2007. One percent of GDP is about $90 billion, i.e. a lot of money.
High and rapidly increasing health expenditures create serious problems for our society. They strain public finances: government now pays 46% of the health care bill. A lot of this is at the expense of education and other important public goods. They reduce growth in real wages for working people. And they price health insurance out of reach for families of moderate means and taxpayers who would like to help them: 43 million people in this country are now uninsured and the number is growing. According to a recent survey of uninsured Californians, affordability is by far the reason cited most frequently for not buying health insurance. So health care expenditures must be contained somehow. Managed care was supposed to be part of the answer. What went wrong? And can it be fixed?
The Consumer-Patient Side
Let me begin with the consumer-patient side of the managed care revolution. The overwhelming majority of employed insured consumers entered the late 1980s with traditional insurance coverage that paid fee-for-service, offered "free choice of provider," go anywhere and do practically anything called medical care and expect to have it paid with no questions asked. Employers apparently paid 100% or some other high percentage of the premium. I say "apparently" because every economist will tell you that so called "employer paid" insurance really comes out of wages. Employees do not understand this, so they were left with no serious interest in the cost of care. Most had no idea what their coverage or their care costs.
There was a great variety in the details of what and how employers paid. Some paid 100% of everything. Some paid the employees' coverage in full and none or only part of dependent coverage. Others paid 75% or 80% of the premium, whatever plan the employee chose. Few gave employees a choice and the full savings if they chose a less costly plan. Employer policies were shaped by collective bargaining in some cases, and by the Internal Revenue Code in all cases. That is, so called "employer contributions" are tax free to the employee without limit, thus giving the employer and employee a greatly attenuated reason to care about health care cost increases.
This arrangement was very inflationary. When caring for insured people, providers could resolve every doubt in favor of doing more with no direct negative financial consequences for patients or themselves. When combined with rapidly expanding technology, these incentives led national health expenditures to increase rapidly through the 1980s, from 8.9% of GDP in 1980 to 13.6% in 1993. Over a 5-year period in the late 1980s and early 1990s, premiums in California doubled.
Although over the long run, employers could take health insurance premiums out of what would have been wage increases, with premiums so large and rising so fast, they couldn't take them out of wages without actually cutting wages which would have led to great conflict and dissatisfaction. So over a remarkably short period of time, many simply moved people from traditional insurance to managed care. From 1988 to 1998, the percentage of employees covered by traditional FFS fell from 71% to 14%. That is a very large change in a very short period of time.
Some employers offered their employees a wide range of choice combined with a defined contribution that left employees with economic responsibility for premium differences. The short list includes Stanford, Harvard, the University of California, and many public employee health insurance plans including those for federal employees and public employees in California. I chaired the Health Benefits Advisory Council of the California Public Employees Retirement System and the Committee on Faculty/Staff benefits at Stanford for about 6 years during those times and I do not believe we experienced a "managed care backlash" because nobody was in an HMO who didn't want to be.
But many, indeed most, employers made serious mistakes. Large numbers of consumers were converted-often involuntarily-from the freedom of FFS coverage to the limitations of HMOs, often without much explanation of the limitations or their relationship to cost-containment. Neither the employers nor the managed care organizations wanted to emphasize the limitations on choice of doctor. So people approached managed care with the expectations they had acquired under FFS. Suddenly people found themselves under limitations they had not experienced before. In many cases, people were forced to change doctors, not permitted to go to the doctor they wanted, or were denied proposed medical procedures. Because they experienced no direct visible financial benefit, these differences between FFS and managed care coverage were perceived as a pure "takeaway." Employers waited too long to respond to rising costs and then were in too much of a hurry and they were too reluctant to offer employees a choice and expose them to premium differences.
The backlash was made worse by the fact that large numbers of people were offered no choice of health insurance plan. A survey done by the Harvard School of Public Health found that 42% of Americans with employer-based health insurance had no choice of plan. Even of those with choices, 20% complained that they did not have enough variety of choice, and 31% of the total sample said their employer forced them to change health plan in the past five years. This and other surveys found that dissatisfaction levels among those without choice are typically twice as high as among those with choices.
Yet another negative consequence of the policy of little or no consumer choice was that the managed care plans, quite rationally, came to see the employer rather than the employee as the customer they must please.
So these employer policies, combined with physician anger that I will discuss later, made managed care a tinderbox in which incidents, real or imagined, could produce national conflagrations. For example, "early" (i.e. within 24 hours) hospital discharge for uncomplicated vaginal deliveries was tested in numerous studies with inconclusive medical results. Neither proponents nor opponents had good data on which to base their cases. But when HMOs and other insurers attempted to implement this as a standard for coverage-in the context of a crisis atmosphere regarding health care costs and intense pressure from government and employers to restrain premiums-in 1995 and 1996, 25 states and the federal government adopted early discharge laws generally requiring coverage of 48-hour stays for uncomplicated vaginal deliveries. This was obviously symbolic politics; people wanted to make a statement about managed care. A study reported that such legislation in Maryland, the first state to adopt it, added about $5.5 million to the annual cost for maternity stays.
Some HMOs tried to enforce the contractual exclusion of coverage for "experimental" or "investigational" procedures. For the most part, with respect to such contractual provisions, HMOs are the agents of government and employers who contract with them and determine the coverage they want to buy. Practically all health insurance policies, public or private, for-profit or non-profit have such exclusions. They are not an invention of profit-seeking HMO carriers, though the media and politicians often implied they were. Some carriers tried to enforce this provision for coverage of High Dose Chemotherapy/Autologous Bone Marrow Transplant or Stem Cell Transplant for metastasized breast cancer and wound up on the losing end of huge punitive damage judgments. This became another women's issue and the Congress passed a law that all carriers serving federal employees would cover this treatment. This seems ironic in view of the fact that recently published randomized controlled trials do not support the efficacy of this procedure. Public opinion would not allow HMOs to enforce their contracts.
I do not doubt that managed care has made errors and damaged the health of some people. But I doubt that these errors caused the backlash. After all, the medical literature is replete with studies that show that the health care system has been making many serious errors and killing large numbers of people through negligence for years-all independent of managed care-and that has produced no backlash. (I am personally disappointed at how difficult it is to get health care providers and the public interested in quality measurement and improvement.) The report of the Harvard Medical Practice Study to the State of New York "estimated that about 2,500 cases of permanent total disability resulted from medical injury in New York hospitals in 1984. Further, we found evidence that medical injury contributed at least in part to the deaths of more than 13,000 patients in that year. Negligent adverse events … were associated with 51% of all deaths from medical injury." The New York Times reported on November 9, 1999: " According to the Centers for Disease Control and Prevention, 5 percent of people admitted to hospitals, about 1.8 million patients a year, will pick up an infection there. Twenty thousand of them will die as a direct result … . hospital-acquired infections will contribute to the deaths of 70,000 more people …" If errors that injure people cause backlashes, we would have had a big one long before managed care became prevalent. I believe the fundamental problem is that doctors and patients, who were used to and expecting unlimited care, were revolting at the very idea of limits.
Emergence of a Dysfunctional Industry Structure
The next major problem area is the emergence of a dysfunctional industry structure. Back in the 1970s when we were developing the ideas that have become known as managed care and managed competition, Paul Ellwood and I envisioned that in a competitive environment of multiple and responsible consumer choice of health plan, groups of doctors, or networks of independent doctors, would find it in their interest to commit to work together to manage quality and cost of care for their voluntarily enrolled populations. They would accept total capitation payment for all the costs of care (with appropriate reinsurance for high cost cases). Among the many strategies these medical groups would pursue to improve quality and cut cost, would be:
Match the resources used to the needs of the populations they serve (thus correcting the specialty imbalances and surplus of beds and high tech facilities); Concentrate costly complex services in regional centers that would realize economies of scale and experience; Adopt what we now call Continuous Quality Improvement, including study of practice variations and adoption of best practices, process improvement or reengineering, benchmarking, analysis of errors and development of methods to prevent them, etc. Follow rational personnel substitution and referral processes among specialists, generalist physicians and other health professionals; Evidence-based medicine and outcomes measurement and management, avoiding treatments that were inappropriate or that conferred very low marginal health benefits on patients relative to cost. And development of formularies based on the judgments of the physicians treating the patients in order to achieve a high degree of compliance, and the bargaining power that comes with it.
These groups or networks would team up with one or a few carriers as partners who would market their services, including pricing, advising on plan design, contracting with employment groups, and the many other functions they perform. Examples would include Med Center's Heath Plan at the St. Louis Park Medical Center, the arrangement the Kelsey Seybold Clinic had with one carrier as its main marketing partner, or the mutually exclusive arrangements at Kaiser Permanente, Group Health Cooperative of Puget Sound and Harvard Community Health Plan. The doctors would bear the risk of costs and make all of the medical decisions, including quality and utilization management, and they would determine the premium by mutual agreement with the carrier in recognition of the demands of a competitive market. I emphasize partnership because, in the long run, health plans cannot succeed without the loyalty, commitment and responsible participation of the doctors. Also we envisioned a gradual transformation so that people would have time to develop the underlying delivery systems.
What We Got Was Very Different
Some of the largest employers offered multiple choice of plan, but most employers didn't. In the interest of lower administrative costs, they wanted to "outsource" health benefits to one carrier. And carriers promised lower premiums (in the short run) to employers who would do "single plan replacement." So employers put pressure on managed care organizations to include just about every doctor and hospital in the state in their networks so that patients would be able to go to the providers of their choice after all. That policy had three negative consequences. First, it weakened the bargaining power of the managed care organizations. Managed care can't drive very hard bargains if every provider knows that the network must include them. It was practically a reinvention of "any willing provider" which negates managed care's bargaining power. Second, it meant that managed care had to accept the inefficient overutilizer as well as the efficient, and the poor quality doctors as well as the good quality doctors. That important dimension of selectivity was impaired. And third, this policy did not lead to the formation of cohesive groups of providers that could work together to improve quality and economy of care. Worse yet, the combination of carriers seeking to meet the cost containment demands of government and employers, and doctors who wanted to maintain the status quo produced an adversarial relationship with a great deal of conflict over capitation rates, other methods of payment, and utilization decisions.
In a manner reminiscent of employer policies, the HMOs insulated the medical groups from the market in the sense that the medical group could see no benefit, in the form of more patients, or happier patients, from accepting a lower capitation rate. A lower capitation rate simply meant higher profits for the carrier.
Narrowly focused plans such as staff model HMOs and other closed panel organizations practically disappeared. The mantra became "the market wants choice." Of course people want choice if they don't have to pay for its extra costs. It is as if Stanford University said to its professors that it would pay in full the price of purchase and maintenance of any car we chose. In that case, we would be looking at new Jaguars, BMWs and Lexi (the plural of Lexus) instead of keeping old Hondas. There will be little market for plans focused on individual medical groups or care systems as long as employees are cost unconscious and see nothing to gain from accepting limited choice.
Responsibility for utilization decisions was mixed. Typically in California where delegation to medical groups was most advanced, the carriers retained most of the risk of hospital costs, so they had to be involved in utilization decisions. Elsewhere they retained all of the risk. Naturally this created conflict with the doctors. It meant that treatment decisions were being made in part by persons who were not in direct contact with the patient. There were many complaints about medical decisions being made by "bean counters" and by MBAs (perish the thought!). At Stanford, we did a study on the complex question of "medical necessity", its meaning and interpretation in practice. One of the findings was that, at least in California, only doctors made determinations of medical necessity. But still, these company doctors were not the ones actually caring for the patients. In view of the extensive literature on inappropriate utilization of services developed by the UCLA-RAND researchers, and the desire of carriers to enforce authoritative practice guidelines, I have little doubt that the carriers' doctors improved decisions in many cases. This shared responsibility added to conflict and complexity. The conflict was all the more severe in cases in which the doctors bore none of the financial responsibility and were being paid fee-for-service.
The average medical group in California contracts with 15 different managed care plans, each with its own referral requirements and network, utilization management processes and formulary. This creates maddening complexity and frustration. This makes it impossible to master all the rules and procedures of the different plans. Worse yet, these rules are of the plan's making, not the doctors'.
A particularly dysfunctional area is formularies. In Kaiser Permanente, formulary decisions are made by committees of practicing doctors who treat the patients who will be taking the drugs, advised by pharmacists and other experts. They know that if they choose a less effective drug just to save money, they are more likely to have a disappointed patient back in their office complaining: "doctor, you haven't cured me." So their incentives are balanced and pointed to seeking value for money. One of the benefits of this arrangement, which includes extensive consultations with the other physicians and pharmacists in the group, is a very high degree of compliance with the recommendations of the P & T Committee. That, in turn, translates into greatly enhanced bargaining power when the Kaiser Permanente purchasers negotiate with the drug manufacturers. Moreover, the patients are always covered for the drugs their doctors prescribe, so an important source of conflict is avoided.
By contrast, in the case of the large carrier HMO, formulary decisions are made by doctors and other experts other than the ones treating the patients. Their incentives are more to hold down the drugs budget than to optimize the care process. Compliance is poor and frustration is high because the treating physicians were not consulted. And patients are not necessarily covered for the drugs their doctors prescribe. To comply, doctors would have to use different drugs for patients with the same medical condition, depending on which insurance plan covered the patient, impairing the doctors' ability to acquire experience with the drugs of their choice. And the savings from negotiated discounts appear to go to the health plans, not to the doctors or their patients.
It was or is inevitable that serious changes to contain the growth of health expenditures would be bad news for many doctors, including some loss of income and autonomy. Most American doctors have been trained in a fee-for-service cost unconscious culture. A serious expenditure containment strategy must address issues of utilization of services. The work of Wennberg and others showing very wide variations in practice patterns must call into question the appropriateness and evidence base for much of what many doctors do. The UCLA-RAND and other studies of appropriateness suggest large amounts of inappropriate hospitalization and surgery. Price controls or negotiated price reductions on doctors and hospitals alone, without addressing actual practice patterns, won't come close to solving the problem of expenditure growth.
But the way the managed care revolution played out was especially irritating for doctors for some of the reasons I have explained. Because most (not all) doctors did not want to accept responsibility for managing quality and cost of care, someone else tried to do it for them. We face a dilemma. Physicians are the best-qualified people to manage quality and cost. It would help a lot if many more were trained for this task. Neither doctors nor patients will accept anyone else doing it. But apparently most doctors don't want to do it either.
The managed care revolution did not play out in the form of rapidly growing medical groups and other physician networks, what I call "Delivery System HMOs." These organizations were overtaken by the for-profit "Carrier HMOs," really insurance companies that sold a comprehensive coverage contract to employers and then contract with all sorts of doctors. They could and did move quickly because they had access to capital and because they didn't need to build medical infrastructure. They didn't need to change doctor culture or practice settings. They just contracted with existing FFS providers. The Carrier HMOs represented a very superficial change in the delivery system, not the fundamental change that is needed to achieve economical health care.
Carrier HMOs rounded up many FFS doctors who resented managed care and really didn't want to be there. These doctors not only lacked any incentive to cooperate with managed care, but they were motivated by resentment and fee-for-service to fight it. And they had a powerful weapon readily at hand: they could recommend to patients procedures, medications, and lengths of hospital stay that the HMOs would challenge or deny. Recalling the studies documenting large amounts of inappropriate surgery, hospitalizations and prescribing, and wide variations in medical practice not explained by differences in medical need, an HMO doing its job properly would have to question many doctor recommendations. But non-medical people in general have no idea at all about inappropriate procedures or "unnecessary surgery," and generally assume their doctor is better than average anyway. In addition, in many areas there is no professional consensus that could serve as the basis for a determination of medical necessity. So it is entirely understandable that patients would take the side of the doctor they knew and trusted and be furious at the insurer who was trying to deny payment for what the doctor recommended presumably for the unworthy goal of saving money.
In July 1999, the Kaiser Family Foundation published a survey that found that 87% of doctors said that their patients have experienced health plan denials of coverage for health services over the last two years. In all the public discussion, it was assumed that of course the insurance company was in the wrong and the doctor was right. All the literature about inappropriate hospitalizations and surgery was forgotten. It is an unequal contest. Managed care can only work with the willing cooperation of doctors. And if managed care cannot work for this reason, America will have to turn for a solution to its cost problem more along the lines of what the Canadians have done, which American doctors are likely to like even less.
Can It Be Fixed?
I am not optimistic. First, it is very hard to get employers and employees to understand the potential benefits of competition and the meaning and value of rational economic incentives in health care. It is hard to get voters and their representatives in legislatures to support the changes needed for responsible multiple choice arrangements (e.g. a limit on the tax break). Health care is a particularly difficult area in which to make and sell rational public policy because the issues are so complex.
Beyond that, I believe that advancing medical technology is facing us with a set of choices that will become increasingly painful. My Stanford colleague Victor Fuchs recently reported "Health care expenditures on the elderly have outpaced the gross domestic product by 3.5 to 4.0 percent per year in recent decades … partly attributable to demographic change … by far the more important factor, however, is the rapid growth of age/sex specific consumption of health care by the elderly. If the trends of the past decade or two continue until 2020, the elderly's health care consumption in that year will be approximately $25,000 per person (in 1995 dollars), compared with $9,200 in 1995." Fuchs' analyses show, for example, that between 1987 and 1995, the per person consumption of angioplasty by males 75-79 years old has increased 20% per year; for CABG, 9%; for Cardiac Catheterization, 10%; for Hip Replacement, 18%; for Knee Replacement, 12%; and more. And costly new procedures are being developed all the time. I believe this is going to make it all the more important that we find ways of developing and motivating delivery systems that improve efficiency and provide costly services only when they are likely to provide good value for money. In the face of Fuchs' data, I do not believe efficiency improvement alone can come close to solving the problem, though it could help and ought to be pursued.
The anti-managed care legislation working its way through federal and state legislatures is not likely to do much to ameliorate the backlash because it does not get at the fundamental causes. The causes of the backlash are much deeper than the specific irritations or grievances we hear about. I think they are first, that the large insured employed American middle class rejects the very idea of limits on health care because they don't see themselves as paying for the cost. We in America have created an entitlements mentality. And second, that while doctors (in partnership with managers) are the best qualified people to manage quality and cost of care, and the only people that doctors and patients will accept in that role, most doctors don't want to do it and they feel outraged if someone else tries to do it.
Here are the lines along which the managed care backlash might be ameliorated.
First, consumers must be required to face responsible choices (with some protection for the poor). That is, a consumer who wants the freedom of a fee-for-service wide-access plan should have to pay the full difference in premium with net after tax dollars. That means we need a limit on tax free employer contributions set at the level of a very efficient health plan, and employers must make defined contributions that do not subsidize more costly plans if they want to take advantage of the tax break for their employees. Better still, as Nancy Dickey has recently written in a Sounding Board in the NEJM, the AMA proposes that the present tax exclusion be replaced by an income-related refundable tax credit usable only for health insurance. This policy would give consumers a personal and serious reason to be interested in an economical health plan.
I appreciate that this would be extremely difficult to sell politically, but the logic in support of it is overwhelming.
As I have listened to the debate over managed care, I have often thought, "we are hearing from people who do not believe they are going to have to pay for what they are demanding." I think that is very clear in demands for expanded and unlimited rights to sue health plans and laws prohibiting 24-hour coverage for hospital stays for normal uncomplicated deliveries and many of the other demands that will raise costs. If we want public support for economizing policies, we must have citizens who believe they have a real personal interest in economical health care.
In a market of truly cost conscious consumers, I would expect some health plans would include numerous economizing measures such as 24-hour stays for uncomplicated deliveries, maximum use of outpatient surgery, acceptance of mandatory arbitration instead of the tort system for disputes, maximum use of generic drugs even when new brand name drugs offer a genuine if marginal improvement in outcome, etc. It would be important that their members be well informed before joining. Such plans are needed if we are to make health insurance affordable to the millions who can't afford it now. I think the worst thing about much of the anti-managed care legislation now passed or in process is that, acceding to the demands of middle and upper income insured people who do not realize they are in fact paying for their coverage, legislatures are outlawing the kind of truly cost-effective health care badly needed by people of modest means.
Second, there needs to be a great deal more genuine consumer choice. The choice menu ought to include wide-access "triple option" or "point of service" plans and preferred provider insurance as well as closed panel plans with very limited networks seeking to offer good quality but very economical care. We have a triple option plan at Stanford and it has proved to be very popular. It reduces conflict because the terms of exit are pre-negotiated. But exit from the HMO network or the Preferred Provider network does cost the consumer substantially more money. On the other side, the economical closed panel plan can say to enrolled members "you have chosen us because we offer an economical plan so we should be able to expect you to cooperate with our cost containment programs." The troubled Carrier HMOs tried to be all things to all people.
Individual choice of plan is important. It allows each person to choose the degree and method of cost containment that he or she finds least objectionable. And if things go badly, reasonable people will recognize that, having been given a choice, they are at least partly responsible for the outcome.
Third, I think there needs to be consumer cost consciousness both at the choice of plan level and at the point of service. People must have some reason for forgoing the diagnostic test they might want but which the doctor feels is of doubtful value.
Fourth, we need to move from all-inclusive networks to including some very selective networks in the menu. I am thinking of Med Center's Health Plan or Kelsey Seybold and its main insurance partner, in which people can sign up for care with a specific medical group and personally benefit from the ability of that group to manage cost. Put alternatively, we need partnerships between medical groups and other delivery systems on the one hand and insurance carriers on the other where the medical groups make all the important decisions including pricing, and bear the consequences in a cost-conscious competitive marketplace. The focus and market power would move from the carriers to the medical groups.
Fifth, to make this possible, we need administrative arrangements that are capable of offering multiple choice to a defined population at a low cost. I know this is possible because CalPERS has offered public employees in California a choice of more than 20 plans at an administrative cost of less than one half of one percent of premium. Economies of scale have something to do with this. The bi-partisan Managed Competition Act of 1993 provided for creation of a Health Plan Purchasing Cooperative for small employment groups and individuals in every state. In her recent Sounding Board in the NEJM, Nancy Dickey describes, as a part of the AMA proposal Voluntary Choice Cooperatives (VCCs) to enhance patients' choices. (Voluntary cooperatives raise serious problems of adverse risk selection. I hope some workable compromise can be found.)
Sixth, what about the doctors? As I noted earlier, no plan can succeed without their support. So I am encouraged that the thoughts I am expressing here are along the same lines as those of Dr. Dickey and the AMA. Evidently, the cost unconscious demands of consumers are becoming a problem for many doctors too. In the world created by the reforms I have just outlined, some doctors would choose to work in prepaid group practices or similar arrangements, while others would choose to practice with no economic restraint at all in traditional fee-for-service. But if most people with responsible choices choose various forms of managed care, including the most economical, as they typically do in employment groups that offer responsible multiple choice of plan, then increasing numbers of doctors would have to choose between a livelihood and the restraints of one or another form of managed care. That would put doctors in an economic position similar to that of the rest of us. But I hope they would see that as their choice, their responsible choice. And in the cost conscious environment I have described, they would see that economy in health care is in their patients' own direct interest too.