Last week I was walking through a local hospital when a well-meaning hospital administrator stopped me and asked me what I was doing there. After being startled a bit by the nature of the question, I responded that I was a doctor, and that this was a logical place to see patients.

The question, however, lingered in my mind and led me back to an op-ed piece published by the Bangor Daily News on August 5. In that column Dr. Erik Steele fervently defended the role of hospital administrators as cost cutters who also saved lives. Although the justification for his opinion is clear (i.e. he is a hospital administrator) the data he cites are at best incomplete, and at worst misleading. In particular, it is important to set the record straight about “for-profit” hospitals, since he used that type of institution as an example of how hospitals with more administrators can save money.

There are two recent scientific studies both authored by Dr. Gordon Guyatt from Hamilton, Ontario, both in highly respected Journals – The Canadian Medical Association Journal and The Journal of the American Medical Association – that show a troublesome trend for U.S. providers. These studies were performed by pooling data from many previous studies and from government statistics; hence the results are not ambiguous and, for the most part, unassailable. In the first paper, published last year, Guyatt and colleagues, after adjusting for income, education and other illnesses demonstrated that “for profit” hospitals in the United States had a 2 percent higher overall mortality rate (despite more administrators) than not-for profit institutions. That means there were two more deaths for every hundred people admitted to these hospitals compared to equivalent admissions in not-for profit hospitals. Moreover, “for profit” hospitals tended to hire less skilled personnel for direct care services, and still charged third party payors the same rate as not-for profits.

In the second study, published earlier this year in JAMA, “for-profit” renal dialysis units had an 8 percent greater death rate than not for profit units, again, adjusted carefully for level of income, education and other illnesses.

Simply put, the evidence does not support claims that more administration saves lives. Moreover it certainly does not save health care dollars. In this week’s edition of The New England Journal of Medicine, Drs. Woolhandler and Himmelstein examined trends in administrative costs for health care over the last decade in the United States. They found that contrary to popular belief, administrative overhead actually grew over the last ten years. In 1999, health administration costs were nearly $300 billion and accounted for 31 percent of health care expenditures in the U.S. This compares unfavorably with Canadian healthcare in which the percent of administrative costs are half (16 percent) than in the U.S. system. Moreover, between 1969 and 1999, the share of the U.S. health care labor force accounted for by administrative workers grew from 18 percent to 27.2 percent. Simply put, we are spending more money on non-patient related services.

Dirigo, the nascent Maine state health plan, recognized the issue of growing administrative costs and responded with a plan to provide near-universal coverage while offering state oversight of hospital budgets. Unfortunately, the Maine Hospital Association fought successfully to block this aspect of the legislation. But soon we will have to face that issue again. This time we should rely on an evidence based approach, not an emotional argument, to make the necessary decisions that will benefit everyone.

Clifford J. Rosen, M.D. is an adjunct staff scientist at The Jackson Laboratory and professor of nutrition at the University of Maine.