Chairman’s Letter to JAMA Editor

Letter to the editor of JAMA written by Dr. Richard E. Anderson, chairman of The Doctors Company Board of Governors, praising the critique of the IOM report on medical errors in the JAMA article by Clement J. McDonald, MD, et al., “Deaths Due to Medical Errors Are Exaggerated in IOM Report.”

To the Editor:
Dr. McDonald and colleagues provide an important critique of the IOM report on medical errors and of the Harvard Medical Practice Study (MPS) that is integral to it. McDonald et al. correctly argue that the basis of estimating death rates due to medical adverse events was inappropriate because a high-severity group was chosen for analysis without a control group to provide context and because a causal relationship was not established between the existence of adverse events and subsequent death.

In fact, the headline number of 98,000 deaths annually due to medical error does not represent actual deaths but is conflated from a flawed analysis of fewer than 200 actual deaths in the index 1984 study. (The lower number of 44,000 deaths was derived in the same manner from a 1992 study of data from Colorado and Utah.) The original MPS authors noted that a blinded analysis by a second team of their own reviewers failed to identify the same set of adverse events as the first team, but they did find the same incidence of adverse and negligent adverse events. Nonetheless, the authors declared their data reliable. This is roughly equivalent to saying it does not matter whether we incarcerate the innocent or the guilty as long as the overall number of convictions matches the crime rate. Even more remarkable, the MPS reviewers agreed only 10% of the time on the simple presence or absence of medical negligence. The study methods were sufficiently idiosyncratic that the authors themselves found no correlation whatsoever between their determinations of medical negligence and the outcome of malpractice verdicts.

It is interesting that the IOM report calls for a national goal of a 50% reduction in medical error. Although this is indisputably a worthy target, if we were to take the MPS data at face value, this has already been achieved between 1984 and 1992 (55% decline in deaths due to medical error from 98,000 to 44,000).

It is unfortunate that the authors of the IOM report chose to use the headline-grabbing death numbers from 2 flawed studies. Use of the death numbers not only undermined the integrity of the IOM’s otherwise strong report but has led healthcare policymakers to declare solutions based on faulty data.