|
Activity Number:
|
29
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 3, 2008 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Health Policy Statistics
|
| Abstract - #302284 |
|
Title:
|
Simpson's Paradox in Real Life: The Center for Medicare Services Composite Quality Score
|
|
Author(s):
|
Carlos F. Alzola*+ and Sam Ogunbo and Nikolas Matthes
|
|
Companies:
|
Data Insights and Maryland Hospital Association and Maryland Hospital Association
|
|
Address:
|
401 Glyndon St SE, Vienna, VA, 22180,
|
|
Keywords:
|
Simpson's paradox ; CMS ; Pay for performance ; Scores ; Medicare
|
|
Abstract:
|
The National Hospital Quality measures (NHQM) assess different dimensions of clinical conditions. Quality of care for AMI patients is assessed with nine different indicators along a continuum of care. The indicators, taken individually, do not provide an overall picture of the care provided by a given facility. CMS's composite score looks at each patient and each dimension of care as an opportunity to provide eligible care. All these opportunities are aggregated over all patients at a facility to form a denominator. The opportunities where patients received eligible care are also aggregated to form a numerator. The ratio of these two quantities form the Composite Quality Score. The methodology is susceptible to instances of Simpson's Paradox. We explore how often hospitals that perform better than another in each individual measure are ranked lower when compared on their CQS scores.
|
