Twenty Years of Research on the Quality of Medical Care

Since 1964 we have built on earlier work, with some refinements, particularly in the formulation of explicit criteria of process and outcome; the detailed standardization of case mix when outcomes signify quality; the prespecification of outcomes for follow-up, when adverse outcomes are only the occasion for later assessment of process; a greater emphasis on more subtle organizational characteristics in the study of structure; and the identification of the separate effects of structural attributes by multivariate analysis. We have also paid more systematic attention to questions of measurement, including the veracity and completeness of the record; the procedures of criteria formulation; and the reliability, validity, and screening efficiency of the criteria. A notable advance is the use of decision analysis to identify optimal strategies of care, including the introduction of patient preferences and monetary cost in the specification of such strategies, and the use of decisional algorithms to portray the criteria of quality.