documentation and identified conclusions that were summarized in a paper in the Journal of the American Medical Association in 1998.23 Somewhat overstated, the quality of care in the United States approximates a random walk. The quality is highly variable. There is a substantial amount of overuse, misuse, and underuse of the available science and technologies— all independent of geography, independent of payment type, and independent of when or where people were trained to practice as physicians.
It was a rather devastating conclusion. From that roundtable, the IOM launched the next series of studies, overseen by the Committee on Quality of Health Care in America, chaired by Bill Richardson, formerly the president of Johns Hopkins University and now head of the Kellogg Foundation. Chuck Buck and I were on that committee, and several others participated, including Jerry Grossman. In the second piece of work that we did, we focused on the issue of safety. It led to the To Err Is Human report published in 1999. This spring, we published the final report Crossing the Quality Chasm.
The quality and safety reports are linked together. Essentially, the examination of quality and the examination of safety were both looking at the symptoms of a system that is broken. As we examined that system, we came to more and more firm conclusions that there was a mismatch between the rate and the quantity of scientific and technological innovation occurring and the ability of the delivery system to deliver it safely and responsibly. When we published the last report, it was an attempt to understand in more detail the nature of that mismatch and to recommend interventions that could accelerate the rate of innovation on the delivery side to improve the match between the care and the caring.
What I would like to do is briefly review those symptoms for you, talk about the reasons for the conclusions we reached in the IOM report, and then describe the opportunities for innovation on the delivery side. In so doing, I will touch briefly on some of the policy levers, because there are significant policy barriers or opportunities among the various tools that we have available for intervening.
Wide variations in quality practice were documented back as far as 1975 in the small area variation analysis by John Wennberg, MD, and in a variety of studies across the country over the last 30 years. The more recent