What Causes High Health Care Costs in the United States?

Sep 8, 2011, 11:45 AM, Posted by

Miriam J. Laugesen, Ph.D., is an assistant professor of Health Policy and Management at the Mailman School of Public Health at Columbia University. She is a 2009 recipient of a Robert Wood Johnson Foundation Investigator Award in Health Policy Research.


In a Health Affairs paper published in its September issue, Sherry Glied, Ph.D., and I find that Medicare fees paid for office visits are 27 percent higher than public fees in other countries, and fees for office visits paid by private insurers are 70 percent higher, on average, in the United States than in other countries. Fees for hip replacements paid by U.S. Medicare are 70 percent higher in the US and private insurers pay 120 percent more.

We show how higher physician fees, particularly among orthopedic specialists—rather than physicians providing more services or having higher training costs—explain higher incomes of U.S. physicians compared to physicians in other countries.

For some time, there’s been some awareness that prices are higher, on average, in U.S. health care. For example, people say that we pay more for pharmaceuticals than people in other countries. However, understanding the details of the cross-national differences is difficult, especially in the area of physician services. The category of physician services spending varies substantially, even within countries. To try to make this comparison cleaner we focus on just two areas of medicine and two specific physician services: basic office visits provided by primary care physicians, and hip replacements provided by orthopedic surgeons in Australia, Canada, France, Germany, the United Kingdom (England), and the United States.

One important difference between this study and others is that we take into account differences in utilization, practice expense costs, education costs, and lost earnings. That’s important, because the usual explanations for that difference might be related to those differences. The paper shows these differences don't account for higher earnings, and that is a new interpretation. Another difference is that we tease out the public-private differences in payment rates that are typically not compared.

Let me explain a little more about how the comparison of the cross-national educational cost comparison works. This requires taking account of the length spent in training and education, which is time physicians could have spent earning money doing a job for pay (lost earnings), and tuition costs borne by physicians themselves (direct educational costs).

The first issue is how long people spend in medical school and training. In general it’s actually more similar than we’d first think, with a slightly longer training for specialists compared to primary care on average. So we’d expect that physicians everywhere factor this into their fees. If it’s going to be a similar proportion of fees across all countries, it’s a wash. The second is the issue of medical school cost. In the U.S. the fees medical students pay to attend medical school are higher than in other countries where it is usually more subsidized. As a result physicians do leave medical school with debt in the United States.

Assuming most physicians try to recoup that cost of lost earnings and direct costs, we calculated the additional 'premium' we would expect physicians would need to earn to pay for the costs of their training. Surprisingly, it’s not as large as you would think.

To calculate that, we use a formula that discounts that expense over a lifetime of earnings. In terms of the gap within the U.S. between orthopedics and primary care, there is only a $3,100 difference. That’s because, on average, a primary care physician needs to recoup $21,300 per year compared to their colleagues in other countries, while orthopedists would need to earn an additional $24,400 per year more than their subsidized colleagues in other countries.

In the U.S., the actual difference in earnings is much greater—primary care physicians earn $186,582 while orthopedists earn $442,450 here. That translates into a difference of $255,568 per year, which is more than the difference in education costs.

Some people have asked me if this is an argument for doing away with fee-for-service reimbursement methods and encouraging greater pay for performance. Not necessarily. We should definitely try to get more value in health care, and we should consider different methods of reimbursement, such as bundling certain types of services. Physician reimbursement is one part of my Robert Wood Johnson Foundation Investigator Award project called “The Politics of Relative Values.” In that project I’m looking at the way Medicare develops payment levels for physicians under the Resource-Based Relative Value Scale.

However, looking at what other countries are paying leads me to conclude that there may not necessarily be a relationship between higher fees or higher physician incomes and fee-for-service methods of payment. Many of these countries use fee-for-service compensation methods. The difference is they are paying less per service. While we do like to think fee for service methods are the cause of our woes, it’s important to consider that many health care systems pay for some or all of their services this way. That issue definitely warrants further consideration.

Read more about Laugesen’s study.

Read a New York Times story on the study.

Read the study.

This commentary originally appeared on the RWJF Human Capital Blog. The views and opinions expressed here are those of the authors.