Hospital Report Cards TM Maternity Care and Women's Health Methodology 2005 4
© Copyright 2005 Health Grades, Inc. All rights reserved.
May not be reprinted or reproduced without permission from Health Grades, Inc.
Newborn mortality rates were examined by birth weight and normal newborn categories. Newborn mortality
rates were compared by examining the percentage of newborns who died while in the hospital with the
national percentage of newborns who died while in the hospital for each of six different weight categories.
The weight categories were stratified as follows:
(1) 150 to 499 grams
(2) 500 to 999 grams
(3) 1000 to 1499 grams
(4) 1500 to 1999 grams
(5) 2000 to 2499 grams
(6) 2500 plus grams or normal newborns
Only single birth patients were considered. The actual mortality rate for each category was compared to a
predicted rate based on the national mortality rate for the weight category. To derive a z-score,
HealthGrades summed the actual deaths and the predicted deaths (the national average mortality rate for
each weight category) and compared the two using a combined standard deviation. (The standard deviation
used in the z-score calculation adjusted for the different mortality rates of the different weight categories.)
The z-scores were ordered and percentiles were assigned based on the z-score within the hospital group
(NICU or non-NICU hospitals). Low percentile rankings corresponded to the hospitals which had fewer
deaths than expected, and the greatest statistical significance associated with this difference.
Assignment of Star Ratings for Maternity Care
The four factors were weighted using predetermined weights based on consensus from a physician panel.
Each factor's percentile score was multiplied by its weight and then summed to create an overall score.
Based upon each hospital's overall score, HealthGrades applied the following rating system.
Best Top 15% of all hospitals within 17 states
Average Middle 70% of all hospitals within 17 states
Poor Bottom 15% of all hospitals within 17 states
