Where’s the Value in Health Care?

As US health-care expenditures continue to spiral upward, the value of this spending is increasingly questioned. General Motors (GM), once the leading manufacturer and the largest private employer in the United States, is nearly bankrupt, in large part because it spends $1500 in health-care costs per vehicle, whereas Toyota, which will shortly overtake GM in global auto production, pays only $201 per vehicle for health care in North America and $97 in Japan [1]. Although health-care costs are easy to measure as tangible taxes or paycheck deductions, health-care benefits are much harder to evaluate and quantify. The drivers of health-care cost increases are complex [2,3], but to develop appropriate policies to constrain or reorganize health-care spending, it is critically important to properly compare health-care costs with health-care value.

In this issue of Value in Health, Luce et al. [4] provide a multifaceted assessment of the US return on investment for health-care expenditures, using three independent methodological approaches. Using (appropriately updated) societal life-year values of $99,000 to $173,000, and conservatively considering only the health care-related improvements in survival over past two decades, they find that a dollar of US health-care spending generates between $1.55 and $1.94 in overall health-care gains. Had they included the reductions in morbidity and improvements in employee productivity resulting from these expenditures, the returns on investment would have been even higher, but such gains are much more difficult to quantify than the survival improvements. When Luce et al. look at certain disease-specific results in the Medicare program and at the broader cost-effectiveness literature, they corroborate their overall average estimates of return on investment with an impressive and independent diversity of additional positive findings.

The Luce et al. study [4] is merely the latest in a series of consistent findings by several health economists demonstrating the substantial positive social value for aggregate US health-care spending [5–8]. The logic behind these robust findings is easily sketched out. Between 1980 and 2000, average US life expectancy at birth increased by 3.2 years. A substantial portion of this increased life expectancy should be attributed to increased health-care spending (Luce et al. [4], estimate 66.5% × 3.2 years = 2.13 years average survival gain) because medical care was the main factor that changed during this time period. Multiplying the 2.13 years medical care survival gain times the $100,000 to $173,000 life-year value times the number of people surviving in 2000 who would have died if 1980s medical care were still being used (312,720) generates a societal savings calculation in excess of $1 trillion because of increased medical care spending.

Murphy and Topel [6] have used similar calculations to show that a medical technology, innovation, preventive service, or treatment that could reduce cancer death rates by 1% would save the United States approximately $400 to $500 billion, with similar estimates for heart disease mortality reductions. Because it is quite plausible that US pharmaceutical industry R&D or National Institutes of Health (NIH) biomedical research budget, each of which is roughly $40 billion annually could feasibly reduce either heart disease or cancer death rates by 1% through a focused research effort costing less than a single year’s R&D expenditures, the societal returns to such biomedical research are potentially greater than 10 to 1.

The importance of this finding cannot be overstated as we grapple with the tax and payroll burdens of increasing medical care costs. Many countries and health-care systems, particularly outside the United States, have successfully cut their growth rates for health-care spending and biomedical research. But health care is a productive sector of the economy, not just a cost to employers and governments. Certainly, increasing health-care costs make some employment sectors and manufactured products, like automobiles, less competitive, but the health-care sector itself is one of the most robust areas of the economy for generating new employment. The US Bureau of Labor Statistics projects 4.4 million new health care-related jobs by 2012 [9].

The US NIH is unmatched anywhere else in the world in terms of governmental commitment to biomedical R&D (see Fig. 1), and in the tangible benefits produced by this research, including new understanding and treatments for HIV/AIDS, cancer, heart disease, and nearly every other human malady. Similarly, the United States spends more per capita on pharmaceutical R&D than Europe or Japan, reversing patterns of the early 1990s (see Fig. 2). US pharmaceutical and biotech exports also lead the world [10]. It is often suggested that European and other economies that free-ride on US biomedical R&D obtain nearly all of the benefits of US medical technology at much lower health-care costs per capita. But being the first to adopt any new technology will always be more expensive than waiting to adopt mature technologies. Europe currently spends less than 60% of the United States on pharmaceutical R&D per capita and invents only about half as many new drugs.

Figure 1. Health-care R&D in government budgets as a percentage of gross domestic product, 2002. n.a., not available. Source: Organization for Economic Cooperation and Development (OECD) R&D Database, 2003.

Figure 2. Pharmaceutical R&D expenditure in Europe, USA, and Japan from 1990 to 2003. n.a, not available. Source: EFPIA member associations, PhRMA, JPMA, Data 2003: estimate EFPIA & PhRMA (Mio euros [at 2002 exchange rates]).

According to a Bain Consultants study [11], Germany saved $19 billion because it spent much less per head than the United States on drugs in 2002. Nevertheless, Germany lost out on $4 billion from R&D, patents and related benefits that went elsewhere, $8 billion because high-value jobs went elsewhere, and $3 billion in profit had German pharmaceutical manufacturers kept pace with rivals elsewhere. A further $2 billion was lost as the country shed corporate headquarter positions. The cost of poorer-than-necessary German health was $5 billion. In sum, Germany’s $19 billion saving was in fact a $3 billion net economic loss. According to the study coauthor “When you add up all of the costs, the free rider model is actually quite expensive.”

A strong argument can be made that the major global health-care financing problem is not that the United States spends too much on health care, but rather that every other country in the world spends too little, particularly on biomedical R&D. Unquestionably, there is waste, inefficiency and harmful medical interventions within the US health-care sector [12]. A primary focus of health economics and outcomes research is to identify and reduce these negative aspects as far as possible. But as employers and politicians increasingly rush to the health-care cost cutting bandwagons it is crucial that we recognize that these costs are associated with enormous economic and societal benefits. The benefits of biomedical R&D accrue not only to this generation but to all future generations. Once the magnitude of these benefits is understood, it is only logical that we should work to protect them from blunt and simplistic cost-cutting measures.