Duke University Press
Joseph White - Choice, Trust, and Two Models of Quality - Journal of Health Politics, Policy and Law 24:5 Journal of Health Politics, Policy and Law 24.5 (1999) 993-999

Choice, Trust, and Two Models of Quality

Joseph White


Evaluations of "managed care" face a basic difficulty: the term itself refers to far too many phenomena to be useful. In the previous issue of this journal, I presented a taxonomy of health care cost control methods, one purpose of which was to provide language with which to bypass discussion of "managed care" altogether (White 1999). Why, then, would I even pretend to evaluate the "managed care backlash?"

My defense is, you can evaluate the backlash without assessing the outputs of "managed care." The real issue is whether individuals could rightly be concerned about how the new world of American health insurance changes peoples' relationships to their own medical care. By reducing choice of provider, the new system tends to reduce individuals' willingness to trust the quality of the care they will receive. From that standpoint, the backlash is clearly appropriate.

"Managed Care" as Restriction

Consider how the term "managed care" becomes relevant to the lives of people who do not read this journal. Once upon a time, health insurance seemed simple. If you had it, it normally covered the services of virtually any physician or hospital. Insurance administrators interfered only rarely, if at all, with the treatment recommendations made by physicians. So an insured person with medical need would find a provider she was willing to trust, and get treated.

Today, from a patient's perspective, it's not so simple. You have to [End Page 993] worry: is your doctor in the plan? Twenty years ago, there was not as much bickering over every little thing. Now, the patient is sick and the patient is stressing over will these charges be met, will the insurance company assume these costs, what might go wrong with the insurance. 1 Whether you call these restrictions and uncertainties "managed care" in your own definitional framework or not, we should agree that (1) these are restrictions compared to the previous model and (2) these restrictions commonly are explained as being forms of "managed care," with the argument being that the health plan is managing care so as to make it more affordable without reducing, or even while increasing, quality.

Thus Karen Ignagni (1998), president of the American Association of Health Plans, asserts that "a well-administered health plan promotes optimal patient care at affordable cost by ensuring that patients get the right care, at the right time, and in the right setting."

Many other participants in this forum may focus on the empirical question of whether the plans live up to this standard or, more precisely, how well they compare to the performance of the previous system, as measured by population statistics. But the backlash is based on the inherent trust claim in Ignagni's statement: You can trust the plan to ensure that you get the right treatment.

Ordinarily, people see choice (including the choice of whom and when to trust) as a way to control their fates. In the old world, some people chose the restrictions of managed care--for instance, Kaiser Permanente's limited network--in order to have lower coinsurance charges. But the revolution of the 1990s has not been driven by individuals' own choices. The key choices have been made by employers, who decide which plans, with what restrictions, will be subsidized for their employees. It is not just "perception" that people face new restrictions. Would normal people object to such a thing, without prodding from "politics"? Of course they would. Why, then, might backlash be inappropriate?

Rationales for Restriction

One answer could be that health plans give people what they really need. If they want more, it is because they are spoiled brats who refuse to recognize [End Page 994] limits; or, like children, they simply do not know what is good for them (medicine as candy). In the most sophisticated version, the patients aren't really children, but they are adults who cannot resist consuming too much of something that is free, or are misled into consuming excessively. Therefore the benefits of managed care must be imposed on people whether they like it or not.

Many analysts may believe versions of this argument, but normal people would demand a high burden of proof for such paternalism. There would have to be clear and convincing evidence that the plans on average are not just comparable but better quality (in order to erase doubt); clear and convincing evidence that control of health care costs deserved such priority that public preferences should be overridden; and clear and convincing evidence that no alternative methods of cost control were possible. I emphasize "clear and convincing" because, unless the evidence is strong, disagreement is a matter of opinion and values, and the case for paternalism weak.

In practice, the health plan industry rarely defends itself in such stark terms. Instead, people are told that the new world gives them what they really wanted all along. Or, its advocates argue that if the new system does not give them what they want just yet, then it will do so soon. We need a little tweaking of the risk adjustment systems, competition, and so on (for a fine example, see Enthoven and Singer 1997).

Advocates of the new system recognize public concerns about choice and quality. When the public relations folk at Aetna felt a need to defend their acquisition of Prudential HealthCare, I (presumably like many health policy authors) received a letter from Aetna's chairman. Its first sentence proclaimed that the purchase "represents an important advance in our ability to offer our members larger geographic coverage and greater choice and to work with participating physicians and other health care providers to improve the quality of care that our members receive" (Huber 1999). The difficulty is, advocacy for "managed care" assumes a specific relationship between choice and quality. The backlash is based, fundamentally, on a different sense of the relationship.

Trusting Plans

In the old system people had to trust providers, especially doctors. In the new system, they are told to put their trust in health plans.

There is a lot wrong with putting trust in physicians. Physicians do have economic incentives to overserve. Physicians can make mistakes. [End Page 995] Physicians can have personal problems. Unfortunately, you have to trust a physician even if you first go through a health plan. And there are some serious problems with the idea that people should choose to trust a plan, rather than a doctor:

Trust about What?

Choice among plans must involve hundreds of quality dimensions. The usual version of this point is that any representative report card would be too complex to be useful. The counterpoint is, choosing doctors, especially specialists, is simpler. You don't have to investigate how your cardiologist would do on skin cancer. Assessing the quality of a plan is not only more difficult but less relevant. The performance that gives your plan high ratings for average quality may not be quality in what you need.

Advocates of restrictive networks will argue that you can trust the plan to choose your doctor. But if there were an objectively correct method, every plan should include (or exclude) the same providers. Obviously the managers of different plans disagree, and none will admit being wrong. A patient might infer the managers change their minds frequently, because it can be difficult to figure out what doctors and other providers are even in a single plan. In any event, it is harder in the new system to pursue quality in your own individual circumstance by seeking out, through your own information, the most appropriate provider of care. A person might judge based on reports from people they know, or credentials, or even report cards about provider performance. Yet, in a restricted network, there is a reasonable chance they will discover that the provider they desire is not in the plan.

When Trust Must Be Given

The choice might be simplified if people could choose the plan that fits their particular health need. Yet in the ideal model, people would choose a health plan at a given time each year. So they may choose when they do not know what they need. If one plan has a better record on cancer, and the other is better on heart disease, you might want to wait and see which you contracted. Choosing the better plan for cancer and then needing cardiac treatment would be, in a much milder way, like going to an oncologist for heart trouble. [End Page 996]

When Trust Is Lost

Most important, what happens when something goes wrong? The theoretical issue here is how well errors can get corrected. In the current debate it is usually posed in terms of procedures for appeal of a plan's decision to refuse some service. Yet an even more fundamental issue is the effect of restricted networks.

What happens when you believe your care is not going well? In a world of flawed competition, such as Medicare+Choice as I write, people can drop a plan that they do not like. But for good reasons involving risk selection, that is definitely not the ideal model. Under the ideal competitive model, people must stay in the plan they believe is serving them poorly until the open enrollment period. In the real-life employment-based insurance world, of course, many people cannot switch to any alternative plan at any time.

They might still try to switch providers within their plan. In the old system, if you didn't like your doctor's care, you could find a different one. Yet that has to be harder in a restrictive network. There may be few alternatives. Your gatekeeper might not allow it (and may have financial incentives not to). In short, when people choose a plan, they must significantly limit their future ability to respond to problems that they perceive in their own care.

Two Models of Quality

In essence, the old system and the new one involve two different models of how choice can lead to quality.

In the first model, people pursue quality in advance, through their choice of plan. Then measuring plan performance becomes, as in the title of a recent special issue of Health Affairs, the definition of "quality in an information age" (1998). If quality as measured is unacceptable, the plan's managers will fix the problem. There is much informed commentary about the design of report cards and their usefulness to plan enrollees. There is little questioning of the model itself.

Yet if we were discussing "health planning," rather than "health plans," there would surely be claims that planning cannot provide quality because it calls for one-size-fits-all policies, is inflexible, presumes that extremely complex problems beyond the scope of human cogitation nevertheless can be solved by analysis, and so on. We would be told it is hardly self-evident that the best way to provide quality is to centralize (sorry, "integrate") [End Page 997] , rely on aggregate data, and control. We might worry that deciding in advance discourages learning from mistakes.

The alternative to a planning model would emphasize decentralization so people with the most information could make decisions, flexibility so people could respond to new information and pursue their own interests based on their own intelligence, and innovation over precommitment of resources--all on the theory that maximizing choices improves results. Most important, the alternate model would say that, rather than trying to get everything right in advance, we need to improve our ability to correct errors.

Readers should recognize the irony: if this were the old debate about "health planning," the second vision of quality would be associated with advocates of "markets," and the first with advocates of politics. In the current debate about the role of health plans, positions are reversed. Fans of the health plans' restrictions on individual choice claim that theirs is a market-oriented solution, because the plans must pass muster in the marketplace. But, aside from the fact that it doesn't exist for most individuals, a market for plans does not and cannot substitute for individual choice of provider. The problems that I have identified here are built into the plan model. Choice of plans cannot be choice of doctor. It must be at a much higher level of aggregation, at a time when people have less idea what they really need, and must involve a commitment that, if seemingly mistaken, is not as easily corrected.

The language used to discuss these two models has varied with the context, and so have the value ladings of the distinction. Most generally, the distinction might be posed as "social interaction versus intellectual cogitation." When the subject is whether government should act or leave choice to the market, it may be "politics versus markets." When the subject is whether analysis should dominate democratic procedure, it is "planning versus politics"--and politics has moved from the cogitation to the interaction side of the line! In practice, both models of a good system have their points. Planning can be useful, and so can a system of much more decentralized transactions. The merits will vary with the situation. Careful analysts of organizations recognize that the two modes of choice are not totally separate: "Plainly cogitation is surrounded by interaction and interaction contains cogitation" (Wildavsky 1979: 113). For the purposes of this forum, however, the key point is that there are two models, and the new American insurance system only allows for one.

Public distrust of "managed care" is really distrust of the new role of the health plan. This distrust should not be dismissed as a result of simplistic [End Page 998] notions about the way the world should work. If individuals' wanting choice is simplistic, then so is the advocacy that presumes plans can guarantee quality. If action based on central analysis makes more sense for individuals' health treatments than for investment in medical capital goods, that needs to be shown by extensive argument, not simply assumed.

Conclusion

The "managed care backlash" involves normal peoples' resentment of a new set of restrictions on an important part of their lives: their ability to influence their own medical care. You can call that illegitimate because peoples' notions of the importance of quality and choice are unrealistic; yet to do so you have to assume that only the planning model of how to pursue quality is legitimate. That makes sense neither from the perspective of ordinary people nor from that of much of our social science.

If public concerns about choice and quality are not irrelevant in principle, then the case that objection to the new insurance world is "inappropriate" must fall back on the belief that the public's wishes, being undisciplined by reality, must be overridden. Put differently, choice is not important because individuals are not competent to make good choices. There are some sophisticated variations on that theme. You may believe them if you wish. Do not expect the public to agree.

Tulane University Medical Center

Joseph White is associate professor of health systems management at the Tulane University School of Public Health and Tropical Medicine. A scholar of federal budget policy and politics before he began work on health care policy, he is the author of Competing Solutions: American Health Care Proposals and International Experience (1999), numerous other articles on health care system reform and cost control, and a forthcoming book on Social Security and Medicare reform.

Notes

1. I thank Sydelle Zinn for this formulation, and much else that is good in my life. I should add my appreciation to Curtis Florence, Dahlia Remler, and Ken Thorpe for comments on the first draft of this essay. No one other than myself bears any blame for failings real or perceived.

References

Enthoven, Alvin C., and Sara J. Singer. 1997. Perspective: Markets and Collective Action in Regulating Managed Care. Health Affairs 16(6):26-32.

Huber, Richard L. 1999. Letter to author and, presumably, many other people. 20 January.

Ignagni, Karen. 1998. Covering a Breaking Revolution: The Media and Managed Care. Health Affairs 17(1):26-34.

Quality in an Information Age. 1998. Special issue of Health Affairs 17(4).

White, Joseph. 1999. Targets and Systems of Health Care Cost Control. Journal of Health Politics, Policy and Law 24(4): 653-696.

Wildavsky, Aaron. 1979. Speaking Truth to Power: The Art and Craft of Policy Analysis. Boston: Little, Brown.

Share