[Skip to Content]
[Skip to Content Landing]
Citations 0
January 5, 2000

Unreliability of Physician "Report Cards" to Assess Cost and Quality of Care

Author Affiliations

Phil B.FontanarosaMD, Deputy EditorIndividualAuthorMargaret A.WinkerMD, Deputy EditorIndividualAuthorStephen J.LurieMD, PhD, Fishbein FellowIndividualAuthor

JAMA. 2000;283(1):51-54. doi:10.1001/jama.283.1.51

To the Editor: Had Dr Hofer and colleagues1 or Dr Bindman2 acknowledged any of the limitations in the study of physician profiling, their presentations would have been more balanced, and thereby encouraged greater sympathy for physicians at risk of deselection due to inept profiling. Only 1 of 4 profiling measures, glycemic control, is a quality indicator; the others are utilization measures. In the study by Hofer et al, only 1 of 3 groups of physicians, those from a West Coast health maintenance organization, had data representing that measure. This makes generalizations about weak physician-patient outcome explanatory links to all possible quality profiling measures or physician organizations rather tenuous. However, if physician impact on any measure was only 4% and all explained variation was 20%, then physician impact as one fifth of all explained variation may be more important than the authors led readers to believe.