Abstract
Profiling provider performance for the assessment of quality involves a number of issues related to selection of appropriate quality measures, subsequent data collection and analysis, and selection of standards of comparison. This article emphasizes the limitations of current data systems for this purpose and discusses hierarchical modeling as the optimal analytic approach for analyzing resulting data. Mention is made of the difficulties of achieving large enough sample sizes for statistical significance at the individual provider level. Finally, the article discusses feasible options for profiling quality.
Publication types
-
Research Support, Non-U.S. Gov't
-
Research Support, U.S. Gov't, P.H.S.
MeSH terms
-
Data Collection / methods
-
Data Collection / standards*
-
Diagnosis-Related Groups
-
Health Services Research
-
Humans
-
Models, Statistical*
-
Mortality
-
Outcome Assessment, Health Care / organization & administration
-
Outcome and Process Assessment, Health Care / organization & administration
-
Practice Patterns, Physicians' / economics
-
Practice Patterns, Physicians' / standards*
-
Practice Patterns, Physicians' / statistics & numerical data
-
Quality Assurance, Health Care / organization & administration*
-
Selection Bias