One way to adress the need to improve ambulatory medical care is to ascertain what factors can predict variations in performance quality between practitioners and practice sites. Identifying predictors of quality will guide the choice of interventions to increase quality. We propose to utilize an existing data base involving over 800 practitioners in 16 practice groups to test a series of hypotheses about quality of care differences among primary care practitioners. In this data base, detailed chart review was done to measure conformity to standards of care. Two kinds of performance will be examined: performance for eight pediatric and adult medical conditions, and performance on these same conditions following a randomized quality assurance intervention. We will investigate specific predictions concerning the following relationships: (1) The effect of practitioner role (staff physician, resident, non-physician health professional) upon the performance of particular functions, and the extent to which practitioners perform better on female-related patient care tasks. (2) The extent to which practitioners' performance on other tasks predicts performance on a given task, and the extent to which improvement on a given task following quality assurance can be predicted by improvements on other tasks and by baseline performance. (3) The extent to which the group leader's performance and his/her change in performance in response to quality assurance predicts performance and change in performance by members of the group, and the extent to which this effect depends on how well a practitioner is intergrated into the group. (4) The extent to which the number of practitioners in the group, and the relative work-share of a practitioner for a given task, predicts change in performance after quality assurance. (5) The extent to which change in performance after quality assurance is predicted by the practitioner's attitudes toward quality assurance interventions.
