If the "Business Model" of medicine is sick, what's the diagnosis, and what's the cure?

• Author: Ohsfeldt, Robert L.
• Position: Controversy

In his thoughtful essay "The Business Model of Medicine: Modern Health Care's Awkward Flirtation with the Marketplace" in this issue, Dr. James P. Whalen highlights several shortcomings of the U.S. health care system. In his criticism, he is hardly alone. Indeed, the production of tomes cataloging the system's numerous ills has become something of a growth industry. Recent examples include As Sick as It Gets (Mueller 2001) and Oxymorons: The Myth of a U.S. Health Care System (Kleinke 2001). Authoritative international bodies such as the World Health Organization (WHO) and the Organization for Economic Cooperation and Development (OECD) have issued less-than-favorable assessments of the U.S. health care system (Docteur, Suppanz, and Woo 2003). Even in more sympathetic assessments, such as From Marcus Welby to Managed Care: The Economic Evolution of American Health Care (Dranove 2000) and The Corporate Practice of Medicine: Competition and Innovation in Health Care (Robinson 1999), the authors find that the system has ample room for improvement.

Many of the specific points Whalen raises relate to three common themes in the literature: the U.S. health care system is too expensive, produces mediocre health outcomes, and provides inadequate and inequitable access to care for the population. In some respects, Whalen's concerns about the "business model" of health care echo those voiced almost twenty years ago by Dr. Arnold Relman, then editor of the New England Journal of Medicine, in an exchange of letters with Princeton health economist Uwe Reinhardt (Relman and Reinhardt 1986). However, Whalen's principal concern is that the profit-seeking motive among entrepreneurial providers of prescription drugs, devices, or procedures drives them to devote excessive resources to promotional activity directed toward both consumers and physicians. Further, he concludes that this promotion often leads patients and their physicians to select inappropriate, excessively expensive, and even potentially dangerous treatments.

In this essay, I place Whalen's concerns in the context of the three general themes common in critiques of the U.S. system, and then I comment on the apparent negative effects of promotional activities and on his proposed solution.

Too Expensive?

Although definitional complexities abound in any cross-national comparisons of "health" spending, let us take it as a given that per capita spending on health care is greater in the United States than in any other OECD country. Let us also take it as a given that per capita gross domestic product (GDP) is greater in the United States than in any other OECD country (excluding tiny Luxemburg). In these circumstances, we would expect the higher GDP to translate into higher health care spending because health care is a "normal" good. However, in a simple linear regression of health spending per capita on GDP per capita in 1999 across OECD countries, Docteur, Suppanz, and Woo (2003) find that the United States is an "outlier," with actual health spending per capita exceeding "expected" spending by more than 30 percent.

A rationalization of this finding might rest on econometric nitpicking--using the same OECD data in a semilog rather than in a linear model reduces the prediction error for the United States to less than 1 percent (my estimate using SAS Version 8.0). Or a more fundamental explanation might relate to differences in the availability of discretionary health care services across countries: residents of other OECD countries might spend more on health care if their governments allowed them to do so. Perhaps the critics are wrong, and Americans really don't spend "too much" on health care after all.

Let us assume, however, that the critics are correct. Another potential explanation for "excess" spending in the United States pertains to differences in the prices of inputs used to produce health care. The popular press has directed considerable attention to the higher prices of certain brand-name prescription drugs in the United States than in Canada. The usual interpretation is that U.S. prescription-drug prices are "too high," but, as one might have suspected, the popular press has focused on drugs with the largest price differentials. After adjusting for purchasing-power parity, some drug prices (for the most part, generic-drug prices) are in fact lower in the United States than in Canada. Whether drug prices "on average" are higher or lower depends on the choice of consumption weights used to define the average--a classic index-number problem (Berndt 2000).

Still, let us assume that drug prices "on average" really are higher in the United States than in Canada. Other inputs used to produce health care services also have higher prices in the United States than in Canada. For example, U.S. physicians, nurses, and other skilled hospital-based employees earn more than their Canadian counterparts, after adjustments for purchasing-power parity (Fuchs and Hahn 1990; Redelmeier and Fuchs 1993; Bell et al. 1999). The difference is particularly striking for U.S. physicians, who earn up to 80 percent more than Canadian physicians.

Simply cutting the earnings of physicians and other skilled health-sector workers in the United States to Canadian levels would (in a static model) eliminate a substantial portion of the alleged 30 percent "excess" in health care expenditures. But are U.S. physicians' incomes really "too high"? Except possibly in some subspecialties, the return on investment in medical education resembles the rates of return on education in other professional occupations, such as law and business (Nicholson 2002; Weeks and Wallace 2002). The health sector must compete with other sectors for labor, so it is doubtful that U.S. physicians' incomes could be reduced substantially without adversely affecting the supply of physician services (except possibly in certain subspecialties).

Mediocre Outcomes?

If Americans spend the most on health care, we might reasonably expect the U.S. system to produce health outcomes superior to those of health systems spending less. However, the "usual metrics"--for example, healthy (quality-adjusted) life expectancy at birth, or the child mortality rate--do not reveal such superiority (see table 1). Clearly, in a nation as wealthy as the United Stares, the abnormally high child mortality rate, especially among blacks, is tragic. Characteristics of the health system affect child mortality by affecting the care that children and pregnant women receive, but a complex combination of other social, economic, and cultural conditions, through some elusive causal pathways, also contributes to this tragedy. These other conditions lie beyond the purview of the health care system. Therefore, child mortality may not be very "sensitive" to differences in health system characteristics and hence may not be especially informative about system performance.

The abnormally low estimated life expectancy at birth in the United States obviously is affected by the abnormally high child mortality rate, but death rates among adolescents and youths also can have a dramatic impact on estimated life expectancy at birth. Therefore, some specific cultural aspects of American society outside the purview of the health care system contribute to the "underperformance" of that system as measured by life expectancy at birth. Two examples are deaths associated with motor vehicle accidents and homicides.

According to WHO data, the U.S. death rate from motor vehicle accidents is approximately three times higher than the rate in Sweden or the United Kingdom, and one and one-half to two times higher than the rate in Australia, Canada, Denmark, Germany, and Japan. In the United States, unintentional injury...