January 3, 1996

Systems Analysis of Adverse Drug Events

Author Affiliations

American Medical Association Chicago, Ill

JAMA. 1996;275(1):33. doi:10.1001/jama.1996.03530250037017

To the Editor.  —There is no question that the contributions made by Dr Bates and colleagues1 and Dr Leape and colleagues2 are important to our knowledge and understanding of adverse drug events. Their research could make a real difference in the health care community's ability to deliver better and safer medical care.However, it is unfortunate that the authors needlessly chose to dramatize their work by making sweeping, scientifically invalid estimates regarding annual rates of iatrogenic injuries and deaths numbering in the millions and hundreds of thousands, respectively. It is simply not scientifically possible to estimate current national iatrogenic injury and death rates based solely on the Harvard Medical Practice Study's (MPS's) decade-old sample of patient discharge records, which included only one state (New York) during just 1 year (1984).3We have no desire to minimize the extent or nature of medical injuries, but we are concerned