Regarding “Adverse-event figures could be far higher: study”:

I'm responding to your article about the Health Affairs paper that said that one-third of all hospitalized patients at three tertiary hospitals suffered adverse events. This is consistent with two other studies that came out in November 2010, one from the inspector general of HHS and the other by Christopher Landrigan et. al. in the New England Journal of Medicine that showed that harmful adverse events had not abated in the 10 years following the publication of “To Err is Human.” This is in spite of the efforts that many, if not most, hospitals have made in the past 10 years to use systems approaches to reduce adverse events. Such approaches have had success in reducing hospital infections, but overall adverse events occur at the same or greater rate than when “To Err is Human” came out on the internet in 1999.

There may be non-compliance, but I have an alternate possibility based on my research into the origins of “To Err is Human.” The data on which it was based needs to be looked at again. A reasonable interpretation of that data, first published as part of the Harvard Medical Practice Studies, is that most adverse events are the result of individual physician error. Further, the Harvard Studies were blind by design to the occurrence of multiple errors by the same doctor.

In my e-mail exchanges with Dr. Lucian Leape who was an author of both the Harvard Studies and “To Err is Human,” I learned that the data from the former was reinterpreted to produce the latter. The onus was thereby shifted, on purely theoretical grounds, from erring doctors to faulty systems. The lack of improvement in the number of adverse events mandates that we look at the contribution of repeatedly erring doctors to the number of adverse events. One million or more lives in the next 10 years depend on this.

Dr. Philip Levitt
J.F.K. Medical Center, Atlantis, Fla., and St. Mary's Medical Center, West Palm Beach, Fla.