Measuring and predicting the performances of practicing physicians: an overview of two decades of research at the University of Utah.
This paper presents the major findings of research carried out by a team of investigators at the University of Utah School of Medicine in the 1960s and 1970s (1) to better understand the multidimensional and complex aspects of practicing physicians' performances, and (2) to investigate relationships between assessments of students in medical school and physicians' performances. The major research began in the 1960s, and data were assembled from a variety of sources to identify factors of physicians' performances. The original study sampled 507 practicing physicians, classified into four highly diverse and representative groups. The team also investigated to what degree these physicians' medical school performances (measured by grades) were predictors of their practice performances. Factor analysis of the data on the physicians' performances resulted in the identification of 25 to 29 factors within each of the four groups. A representative profile of the performances of each physician studied showed how well the physician performed on each of the factors appropriate to his or her specialty and background. Correlation analysis demonstrated that these physicians' performances in medical school, as measured by grade-point averages (GPAs), were almost completely independent of their later performances in their practices, a disturbing finding and one that the team maintained cannot be dismissed by arguments of "restriction of range." In the teams' later studies, 61 measures of physicians' performances were correlated to medical school GPAs in the two basic science years and the two clinical science years. No association beyond chance was observed, and nonsignificant relationships held up regardless of the physicians' specialties or years of experience.(ABSTRACT TRUNCATED AT 250 WORDS)