Comparative effectiveness research can fill part of the information void. Legislators in Washington have never been more enthusiastic about supporting comparative effectiveness research, which compares interventions with the most relevant alternatives, not just placebo, and focuses on effectiveness in real-world practice (1). Policymakers recognize that patients and physicians need up-to-date information about which treatments work best. However, as they draft legislative language, they debate such issues as the funding mechanism, the administrative structure of the responsible agency, the mechanisms for ensuring accountability, and the scope of the work (2, 3). Among these topics, the role of cost-effectiveness information may be the most controversial. The American College of Physicians, in a closely reasoned and deeply informed position paper (4), forcefully argues that this issue should be beyond debate: Cost-effectiveness analysis is essential to the work of any agency dedicated to comparative effectiveness.

As they note, objections to cost-effectiveness analysis often center on methodological concerns. Many studies, particularly old ones, have not followed recommended analytic practices (5), which is, as the American College of Physicians observes, a reason for insisting on adherence to accepted standards (6), not for rejecting the method. Cost-effectiveness studies, like conventional analyses of clinical effectiveness, can become outdated if they are not updated consistently and frequently. Committing the necessary resources to keep analyses current would solve this problem. Other objections are based on fundamental misunderstandings. For example, some critics have complained that cost-effectiveness analysis requires placing a dollar value on human lives. These critics confuse cost-effectiveness analysis with cost–benefit analysis, which sometimes does require monetizing lives. Cost-effectiveness analysis is preferred for health care evaluations precisely because it does not impose this requirement; it simply estimates how much must be spent by using a particular intervention to gain a given health effect, telling us how to achieve the greatest health effect for the dollar.

The deepest objections undoubtedly stem from anxiety about how policymakers and private payers would apply cost-effectiveness information. Whether it is used for coverage determinations, the design of copayments, or individual patient care decisions, it rewards care that produces large benefits in relation to costs and penalizes expensive care that improves health little. Not all products and services that are profitable today would be successful in a health system that rewarded value this way. However, even to those who wish to maintain the status quo, policies to promote cost-effective care may be more attractive than price controls, broad cuts in reimbursement, and other cost-control strategies, because these policies offer the prospect of improving health while reining in expenditures.

Efforts to perfect the methodology can and should continue, but they should not delay federal efforts to support the production of cost-effectiveness analyses. Congressional interest in comparative effectiveness, like the prominence of health reform in the presidential election, is a direct result of widespread public dissatisfaction with the U.S. health care system. The per-person health care expenditure of $7000 in 2006 was more than double the average of the wealthy nations of the Organization for Economic Cooperation and Development (7, 8). Health expenditures in the United States have been growing much faster than those in Canada and the largest western European countries; Organization for Economic Cooperation and Development data show, for example, that per capita U.S. health expenditures in 1970 were 17% greater than in Canada, 76% greater than in France, 33% greater than in Germany, and 113% greater than in the United Kingdom. By 2005, per capita expenditures were 92% greater than in Canada, 90% greater than in France, 95% greater than in Germany, and 135% greater than in the United Kingdom. Measures of population health have improved no more in the United States than in these countries, offering little comfort to those who presume that uniquely high expenditure is purchasing better health.

If we tolerate this level of inefficiency today, we may not be able to for long. Should current trends continue, in 50 years Medicare expenditures alone will account for 9% of the entire economic output of the United States. Only one third of the projected expenditures have a designated source of funding (9). This funding deficit is a looming catastrophe for the U.S. economy (10, 11).

The aging of the U.S. population has contributed surprisingly little to health expenditure growth in recent decades. More important have been increases in the volume of care per patient; the substitution of more costly interventions; and the introduction of expensive new diagnostic tests, procedures, and other treatments that have often added to, not replaced, older forms of care (12). The effects of these technological contributions to health expenditure growth will be greatly magnified as baby boomers age and become heavy users of a high-cost health system. Because health insurance pays for the great bulk of this care, most of the growth in expenditures is directly influenced by health insurance coverage and design.

Unlike in other wealthy nations, public and private health insurers in the United States typically do not consider cost-effectiveness when deciding which forms of medical care to reimburse (13, 14). Coverage decisions hinge instead on evidence of effectiveness. Other nations analyze evidence of effectiveness as well; however, they have concluded that health intervention evaluations are incomplete if they do not address value for money. Cost-effectiveness is seldom, if ever, the sole decision criterion—political and ethical considerations matter greatly and can trump cost-effectiveness—but it is central to decisions about whether to adopt new drugs, devices, and procedures.

In contrast, when Medicare officials have taken tentative steps toward introducing value considerations into coverage policy, they have failed. In 2000, the Health Care Financing Administration published a Notice of Intent to Publish a Rule that would have excluded those items or services from Medicare coverage that cost more than but were no more beneficial than already covered items and services (15). The agency retreated in the face of vehement public criticism, and to this day Medicare coverage deliberations exclude discussion of costs.

This reluctance to consider cost has enabled expensive drugs and procedures to receive coverage whether their health benefits are great or small. This cost-blind approach to coverage policy—and, more generally, to the provision of medical care—impedes efforts to offer low-cost health insurance in the United States. The United States' failure to develop affordable plans is responsible for the large and growing number of Americans who are either uninsured or rely on such safety net programs as Medicaid (16, 17). The high cost of insurance is also the single greatest obstacle to implementing comprehensive health reform at the national level, and it places the innovative programs of Massachusetts and a few other bold states in peril.

A cost-aware coverage policy informed by cost-effectiveness data will not necessarily result in sharp reductions in the range of covered products and services. Many countries apply cost-effectiveness analysis when they develop clinical guidelines because they find it useful for identifying the patients who will receive the greatest benefit from a health intervention (18). Because a cost-effectiveness analysis calculates the cost at which a drug or device provides good value, it can also be used to negotiate more favorable pricing, an opportunity forfeited by purchasers who ignore costs in making coverage decisions and setting copayments. In addition, refusing to consider costs does not avoid rationing of care, despite assertions to the contrary; instead, it exacerbates the problem by driving up insurance premiums. When costs rise, health insurance coverage and many forms of care grow too expensive for more Americans—not only the chronically ill and working poor, but increasingly the middle class.

A menu without prices means that the cost of the meal should be no object, not that the food is free. Even the wealthy prefer more affordable fare most of the time. In health care, affordable choices are scarce, compelling many Americans to go without timely and appropriate care. We won't have an efficient health care system until we learn the value of individual health interventions and use the knowledge effectively. When we dine where the menu has no prices, we should not be surprised by the size of the bill.