This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.
—Drs McDonald and Fitzgerald correctly note that published rates of mortality from regional studies and those from selected centers may vary widely. Because of a number of biases, these comparisons may not be appropriate; however, they may also represent actual differences in currently unmeasured aspects of medical or surgical care. There are, of course, also substantial variations in outcomes within a geographic region. We believe that there are two important consequences of these variations.First, careful study of the fine structure of the processes of care across institutions may provide clues to the origin of this variability. These factors must be elucidated if we are to know enough to improve the outcomes of medical and surgical care.Second, decisions to initiate medical or surgical treatment, or analyses that model these decisions, must be context sensitive. They must take into account patient, disease, and treatment variables. These treatment variables
O'Connor GT, Plume SK, Olmstead EM. CABG Surgical Mortality in Different Centers-Reply. JAMA. 1992;267(7):933. doi:10.1001/jama.1992.03480070048023