In the debate over health reform, nearly everyone agrees on one thing: spending more on prevention will help bring health care costs under control. From Peter Orszag, the director of the Office of Management and Budget, to American’s Health Insurance Plans, an industry association, there is consensus that the United States spends too much on treatment and not enough on prevention.

If more were spent on prevention, the reasoning goes, the savings in treatment would be large enough to reduce the relentlessly rising share of national income devoted to medical care, currently 16 cents of every dollar. In turn, controlling costs would help employers and state and federal governments in their struggle with the high costs of health insurance and health programs.

Unfortunately, as Mark Twain said: “It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so.” Voluminous evidence has accumulated over the last four decades on prevention’s potential for controlling costs. As I pointed out in recent interviews with the Wall Street Journal online and Columbia Journalism Review, it consistently points to the opposite conclusion: despite savings in treatment, prevention usually increases medical spending.

An article in the February 14, 2008 New England Journal of Medicine presented the most recent and largest compilation of evidence. Based on 599 studies published between 2000 and 2005, the authors reported that spending more on prevention increased medical spending over 80 percent of the time. The situation hasn’t changed since I concluded in my 1986 Brookings Institution book, Is Prevention Better than Cure?, that prevention usually added to medical spending.

Prevention comes in many forms: vaccines; treatment of risk factors for disease, such as elevated blood pressure or cholesterol; screening to catch disease early, when it may be more treatable; programs and advice to help people alter habits like smoking or overeating, which put them at higher risk of disease.

While a few of these approaches can save more than they cost (such as flu vaccines for the elderly, when the cost per dose of vaccine is low enough), most add to medical spending. The additional spending can be modest. Smoking cessation programs cost just a few thousand dollars for each year of life they save. But in other cases the additional spending is substantial.

Statins for people whose only risk factor for heart disease is their elevated cholesterol levels cost hundreds of thousands of dollars for each year of life saved. To read more about these and other examples, readers can consult my report for the National Coalition on Health Care here.

Why do people think more prevention will reduce medical costs when decades of studies show it won’t? Part of the answer comes from the natural tendency to focus on the individual who ultimately develops disease and to forget about the larger picture.

If you think of the man who ends up needing an expensive bypass operation after years of smoking and untreated hypertension, it seems obvious that it would have cost less to prevent his heart disease – and, of course, better for him. But it is not possible to identify that particular man ahead of time, only to identify him and people like him who are at risk of heart disease.

Prevention addresses risks of disease, not disease itself. It must be provided to all those at risk, often for many years. Yet many of them would never develop heart disease even without prevention and some will develop it in spite of prevention.

The cost per person, per year, can look low. But when those costs are summed over all the people who receive the preventive intervention, and all the years they receive it, the total cost usually exceeds the treatment savings.

How then can reputable studies claim, as some do, that prevention saves many dollars for each dollar spent? The answer is that those studies include more than just medical care spending in their tallies. They value the future wages of people whose lives and health are preserved by prevention, and even the time people would spend on treatment.

We value good health, and our time, and the estimates in those studies reflect that value. But they do not address the issue of medical costs.

A study of chicken pox vaccine provides a good example. Published in the Journal of the American Medical Association in 1994, the year the vaccine was approved, the paper stated that the vaccine “would save more than $5 for every dollar invested in vaccination.” That $5 included parents’ time, valued at their average wage rate, and children’s future earnings.

When only medical costs were considered – those of the vaccine and those of the treatment avoided because of the vaccine – the article showed that every dollar spent on the vaccine returned 90 cents in treatment savings. Not bad, but not cost-saving.

Is the vaccine worth it? Yes. It brings better health for a modest increase in medical spending. But it does not reduce medical spending.

And what about the frequent assertion that the U.S. spends only 3 percent of its health care dollars on prevention, a number so ridiculously low that it seems obvious prevention must be underfunded? Sometimes the number is put as low as 1 percent, sometimes as high as 5 percent. Researchers at the Altarum Institute have discovered that whatever the estimate, and wherever it appears, the source usually given is a report that appeared in CDC’s Morbidity and Mortality Weekly Report in 1992.

It is hard to tell from that brief report, a summary of a larger unpublished study, exactly what the numbers include, but they are based on data for 1988 and are now 20 years old. So the Altarum group went back to the beginning and developed new estimates. Their estimates show that at least 8 percent to 9 percent of national health spending goes to prevention.

To put even that larger number in perspective, recall that, while treatment takes place only in the medical sector, prevention takes place throughout the economy, as it must to keep people healthy. Examples are everywhere: improvements in highway design, safety features in cars, health inspections for restaurants, water treatment plants, sewage systems, features of buildings required by safety codes, work safety measures, limits on pollutants, and many, many more. An accurate examination of the balance of prevention and treatment needs to look beyond the medical sector.

International comparisons also suggest that spending more of our medical dollars on prevention will not reduce health care costs. In 2004 the Commonwealth Fund surveyed patients in five countries: the U.S., Canada, the United Kingdom, Australia, and New Zealand. With the exception of flu shots for the elderly, where Australia ranked first, the U.K. second, and the U.S. third, the U.S. led on the use of the preventive measures examined: blood pressure checks, screening for cervical cancer, mammography, and advice from a physician on diet and exercise.

The percentages of patients in the U.S. who received these services were higher, sometimes much higher – evidence that the U.S. is already devoting more of its medical spending to prevention than other high-income countries. Yet these other countries all have longer life expectancies and lower health care costs.

Prevention is often worth doing because it brings better health. But with prevention, as with treatment, better health comes at a higher price most of the time. The best medical care is based on the best evidence. Health reform will be most likely to succeed when it, too, is based on the best evidence. That evidence shows that spending more on prevention is not the way to control health care costs.

Louise B. Russell, PhD., is Research Professor at the Institute for Health, and Professor in the Department of Economics, Rutgers University, New Brunswick N.J. Her research focuses on the methods and application of cost-effectiveness analysis. Her most recent book, co-edited with Marthe Gold, Joanna Siegel, and Milton Weinstein, is Cost Effectiveness in Health and Medicine, the widely-cited report and recommendations of the U.S. Panel on Cost-Effectiveness in Health and Medicine. 732–932-6507; lrussell@​ifh.​rutgers.​edu