|
Activity Number:
|
343
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Health Policy Statistics
|
| Abstract - #303118 |
|
Title:
|
Evaluating Composite Quality Measures
|
|
Author(s):
|
Mary Beth Landrum*+
|
|
Companies:
|
Harvard Medical School
|
|
Address:
|
180 Longwood Avenue, Boston , MA, 02446,
|
|
Keywords:
|
Provider profiling ; Latent variable models ; Composite measures
|
|
Abstract:
|
Data on the performance of specific providers are becoming increasingly important tools for consumers and purchasers of health care. As the number of measures increases there is increasing need for summary measures. Various composite measures have been proposed, however, there is no gold standard for summarizing quality and it is often difficult to determine which composite best summarizes quality. In this talk, we analyze individual-level data on 18 measures of quality provided to 2.3 million patients hospitalized in one of 4500 hospitals in 2005. We compare various composites focusing on evaluating the statistical properties of these summary scores. In order to evaluate composite measures against an objective standard, we also examine the extent to which hospital-level quality measured by various composite scores predicts risk-adjusted mortality at the individual level.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2009 program |
