If insurance were not widespread, many physicians would be reluctant to order an expensive intervention unless it offered a good chance of substantial benefit — that is, unless it was cost-effective. Indeed, without U.S.-style cost-insensitive insurance, many expensive diagnostic and therapeutic innovations would not be developed and brought to market.¹ The insured patient, on the other hand, will usually want any and all care that might possibly be of net benefit, regardless of cost. The physician may recognize that the intervention under consideration is not cost-effective but may recommend it anyway, for a variety of reasons: to keep the goodwill of the patient, to protect against a malpractice suit, or in the belief that the “primacy of patient welfare” makes the denial of such care “inappropriate” and “unethical.”

The doctor’s dilemma is the nation’s problem. Some policy experts think that if patients had “more skin in the game” — that is, had less insurance — the problem would be solved. It would not. Even the most ardent advocates of deductibles and copayments acknowledge the need for an annual cap on patients’ payments, beyond which insurance takes over completely. There is no consensus on the right level for the cap, but it is generally recognized that the average U.S. household, with large debts and minimal financial assets, could not handle much more than $5,000. But the extreme skew in annual health care expenditures, with 5% of individuals accounting for 50% of spending in any given year, means that many health care decisions, and especially those involving big-ticket interventions, will be made by and for patients whose costs have exceeded the cap.

Another popular “solution” is to eliminate care that does more harm than good — that is, “unnecessary” care. Such elimination would be desirable, but the potential savings from this source are smaller than is usually claimed. It is true that after the fact, many interventions turn out to be useless or even harmful for some patients. But the heterogeneity of patient populations and uncertainty about the response of individual patients to an intervention means that it is often difficult or impossible to determine in advance which ones will prove to help particular patients and which will turn out to have been unnecessary.

There is no escaping the fact that many interventions are valuable for some patients even if, for the population as a whole, their cost is greater than their benefit. Under what circumstances are they likely to be ordered, and when are they likely to be withheld? The context within which the physician practices, his or her assumption about the behavior of other physicians, and the economic and health consequences of ordering all the care that might do some good versus practicing cost-effective medicine will affect the physician’s choice. If the physician is paid on a fee-for-service basis and the patient has open-ended insurance, the scales are tipped in favor of doing as much as possible and against limiting interventions to those that are cost-effective. In that setting, who would benefit from the resources that are saved by practicing cost-effective medicine is not obvious to the physician.

In contrast, if the physician is practicing in a setting that has accepted responsibility for the health of a defined population and the organization receives an annual fee per enrollee, the chances of the physician’s practicing cost-effective medicine are substantially increased, even though all patients are insured. The physician’s colleagues are practicing the same way, and the resources saved can be used for the benefit of the defined population, which includes the physician’s patient. In Canada, which has universal insurance, per capita spending on health care is only 55% of the U.S. level because there is a limited overall budget, and all physicians in the system recognize the need for prudence in making decisions about care.

In short, when physicians are collectively caring for a defined population within a fixed annual budget, it is easier for the individual physician to resolve the dilemma in favor of cost-effective medicine. That becomes “appropriate” care. And it is an ethical choice, as defined by philosopher Immanuel Kant, because if all physicians act the same way, all patients benefit.²

Disclosure forms provided by the author are available with the full text of this article at NEJM.org.