© Copyright 2007 Health Grades, Inc. All rights reserved.
May not be reprinted or reproduced without permission from Health Grades, Inc.
America's 50 Best Hospitals Award
2007 Methodology
HealthGrades' America's 50 Best Hospitals Award recognizes hospitals for consistent excellence by
identifying those hospitals that have received HealthGrades' Distinguished Hospital Award for Clinical
Excellence
TM
(DHA-CE) designation for the most consecutive years.
1
Hospitals who were DHA-CE recipients for all of the five years that HealthGrades has awarded the
DHA-CE were identified.
2
Hospitals who were DHA-CE recipients for all of the last four years that HealthGrades has awarded
DHA-CE were identified.
·
The four-year recipients were sorted by z-score, using the average z-score from the most
recent DHP-CE analysis.
·
The top hospitals from this list were then added to the list from step 1 to create a list of
America's 50 Best Hospitals.
Summary of DHA-CE Award Process
To help consumers evaluate and compare hospital performance, HealthGrades has analyzed patient
outcome data for virtually every hospital in the country across 28 different patient cohorts. HealthGrades'
risk-adjustment methodology considers important differences in patient demographic (e.g., age, gender) and
clinical characteristics (e.g., underlying medical conditions) that could increase the patient's risk of mortality
or complications. The final result is expressed in the form of star ratings.
To be considered for the Distinguished Hospital Award for Clinical Excellence (DHA-CE), a hospital had to
have received high star ratings based on HealthGrades' risk-adjustment methodology and also must have
sufficient patient volume to be rated in a wide variety of cohorts (e.g., 22 cohorts were required for the most
recent DHA-CE award).
DHA-CE average z-score is the straight average of all z-scores associated with star ratings, with equal
weights given to each z-score, whether the z-score is associated with complications, in-house mortality,
30-day mortality, or 180-day mortality. To calculate a z-score we take the difference between the observed
number and predicted number of events (mortalities or complications), and divide it by the standard
deviation associated with the observed number of events.
