RESPONSE TO THE CRITICISM BY HENRY AARON OF THE NEJM ARTICLE ON ADMINISTRATIVE WASTE IN US HEALTH CARE

The article we took from medscape for our news summary on Administrative waste CLICK HERE included almost a critique of the Woolhandler article by a Dr. Henry Aaron. 

The Woolhandler article is of great importance to the UHC movement since some 25% of our health care budget would be freed for patient care by scaling our administrative costs more towards similarity with the Canadian system. In his NEJM editorial, Dr. Aaron of the Brookings Institution said the administrative costs in the United States may be 24% lower than the estimates of Woolhandler et al. He said the excess spending on healthcare administration in 1999 is probably closer to $159 billion, not $209 billion cited in the study. Aaron added that it is still not clear that the United States would save money if it converted to the Canadian system. While Dr. Aaron characterizes the U.S. healthcare system as "an administrative monstrosity," he said the latest comparisons "clearly exaggerate" the differences between the North American neighbors. All these assertions were repeated in the medscape article.

The response from co-author Himmelstein follows.

**

In commenting on our article in today's New England Journal of Medicine comparing health care administrative costs in the U.S. and Canada, Henry Aaron opines that such comparisons merely "titillate policymakers and others but provide them with little useful guidance" (1,2). We will not, here, argue with his political judgement on the irrelevance of national health insurance other than to point out that last week, 7,784 physicians, including some of the most prominent leaders of American medicine, called in the Journal of the American Medical Association for the very reform that Aaron dismisses (3).

Our purpose here is to refute Aaron's methodologic critique of our analyses. His claim that we overstate the potential administrative savings of national health insurance is based on demonstrably false assumptions about comparative health care wages and prices in the U.S. and Canada.

First, Aaron claims that in calculating potential administrative savings we should have used percentage differences in administration's share in the U.S. and Canada, rather than absolute dollar amount differences.

A concrete example may illustrate this apparently arcane point. Seniors' drug coverage is far more expensive in the U.S. than in Canada. This U.S. excess is attributable to two factors. First, for most drugs Americans pay about twice as much as the Canadian price for the identical product. Second, insurance overhead for administering drug coverage is far higher in the U.S. than overhead in Canada because of Canada's far simpler insurance system. Hence, the overall cost to society of a two-month supply of a drug in Canada might amount to $100 dollars ($2 for the overhead and $98 for the medication). The comparable cost in the U.S. would be about $209 ($13 - about 6% - for insurance overhead and $196 for the medication). Aaron would claim that the potential administrative savings in this example are $8 (6% - 2% x $209). Our estimate would be $11 ($13 - $2).

In essence, Aaron assumes that the administrative effort (and cost) of paying a $209 drug claim must be more than twice that entailed in paying a $100 claim for the same bill. We assume that the paperwork costs of processing these two claims would be identical, if the insurance systems were the same.

Aaron introduces his point with a hypothetical example of a nation whose health care costs are one-tenth that of the U.S. If that were the case in Canada, it would indeed be true that our absolute dollar method would overstate the potential administrative savings. But his example is a false one. In fact, our analysis of data from the U.S. and Canadian Census Bureaus shows that health administrative workers in Canada make somewhat higher wages, on average, than their American counterparts (while the few insurance executives in Canada make far lower incomes that U.S. CEO's, the clerks and secretaries in Canada make more). Aaron's reasoning, properly applied to the real data rather than his misleading hypothetical example, would indicate that we underestimate the potential administrative savings of national health insurance.

Aaron goes on to suggest that we also overestimate potential administrative savings because of differences in relative wages - i.e. the somewhat lower incomes of Canadian physicians relative to their U.S. counterparts and the relatively higher wages of Canadian secretaries and accountants. This argument ignores the fact that physicians' net incomes account for only 10% of health spending, and that lower-status clerical workers - who are far more numerous than physicians - are generally paid more in Canada. Hence, the real data refutes his implication that taking into account cross-national wage differentials between clinical and administrative personnel would lower the estimated administrative savings.

Our paper underwent an extraordinarily long and detailed review process at The New England Journal of Medicine, during which several reviewers vetted our analysis in great detail. These reviewers made useful suggestions for technical revisions, which we incorporated in our analysis. These revisions - notably valuing physician time spent on administration based on physicians' net rather than gross earnings, which decreased our estimate of potential administrative savings by about $20 billion - adopted conservative assumptions, to assure that our projection of potential administrative savings stood on firm ground. In contrast, Aaron's critique of our methods is without technical merit and ignores the actual data on several theoretical points that he raises.

Even if all of his critiques were true, and the potential administrative savings of adopting national health insurance "only" amounted to $159 billion in 1999 (equivalent to $218 billion in 2003) our paper's principal conclusion is clearly correct "A large sum might be saved in the United States if administrative costs could be trimmed by implementing a Canadian-style health care system." --------------------------------------------------

1) Woolhandler, Campbell, and Himmelstein. "Costs of Health Administration in the United States and Canada" NEJM August 21, 2003

2) Aaron, Henry "The costs of health care administration in the United States and Canada - Questionable answers to a questionable question" NEJM August 21, 2003

3) "Proposal of the Physicians' Working Group for Single-Payer National Health Insurance." JAMA August 13, 2003. Woolhandler, Himmelstein, Angell, and Young. Endorsed by 7,784 physicians and medical students.

Full article and the editorial are in the same issue of NEJM accessible through your nearest medical library. N Engl J Med 2003:348:768-775,801-803
The medscape article may still be available CLICK HERE
 The Project EINO news summary is available after (Oct 1 2003) CLICK HERE , before Oct 1 2003 CLICK HERE.