Hospital Report Cards TM Bariatric Surgery Methodology 2007-2008 2
2
© Copyright 2007 Health Grades, Inc. All rights reserved.
May not be reprinted or reproduced without permission from Health Grades, Inc.
Methodology for Rating Hospitals
Fair and valid comparisons between hospital providers can be made only to the extent that the risk-adjustment
methodology considers important differences in patient demographic and clinical characteristics. The risk-adjustment
methodology used by HealthGrades defines risk factors as those clinical and demographic variables that influence
patient outcomes in significant and systematic ways. Risk factors may include age, gender, specific procedure
performed, and comorbid conditions such as hypertension, chronic renal failure, heart failure, and diabetes. The
methodology is disease-specific and outcome-specific. This means that individual risk models are constructed and
tailored for each clinical condition or procedure using multivariate logistic regression.
For multivariate logistic regression-based ratings (see below), HealthGrades conducted a series of data quality
checks to preserve the integrity of the ratings. Based on the results of these checks, we excluded a limited number of
cases because they were inappropriate for inclusion in the database or miscoded.
Examples of excluded patient records were:
·
Patients who left the hospital against medical advice or who were transferred to another acute care hospital.
·
Patients who were still in the hospital when the claim was filed.
·
Patients with an invalid gender.
Multivariate Logistic Regression-Based Ratings
The initial analysis of the data utilized 19 states of all-payer data from 2003 through 2005. Bariatric surgery patients
were identified by their ICD-9 principal procedure of a bariatric surgical procedure and a principal diagnosis of
obesity/morbid obesity (see Appendix A)--a definition previously described by Santry et al.
2
Patients under the age
of 18 were excluded.
For this population, potential risk factors and the outcome measure (complications) were then defined.
1
Potential risk factors were defined as all clinically relevant diagnoses occurring in more than 0.5 percent of the
patients. In addition, patient demographic factors such as age and gender and the specific procedure performed
on the patient were also considered. Some diagnosis codes were merged together (e.g., primary and secondary
pulmonary hypertension) to minimize the impact of coding variations.
2
Complications were identified using previous peer-reviewed research
2,3
and through input from clinical and
coding experts.
In some cases an ICD-9 code can be either a risk or a complication. In these cases, a code is differentiated by the
presence or absence of a 900 post-operative complication code. For example, in the case where a patient record
contains "427.31 Atrial Fibrillation," that code is considered a risk if it occurs by itself and a complication if there is a
corresponding "997.1 Cardiac Complications NEC" code also present in the patient record. Outcomes were binary,
with documented major complications either present or not. Mortality is considered a major complication. Appendix B
lists the major complications for bariatric surgery.
