Predicting Risk-Adjusted Mortality for CABG Surgery
- 1 July 2005
- journal article
- Published by Wolters Kluwer Health in Medical Care
- Vol. 43 (7) , 726-735
- https://doi.org/10.1097/01.mlr.0000167802.27044.44
Abstract
In recent years, several studies in the medical and health service research literature have advocated the use of hierarchical statistical models (multilevel models or random-effects models) to analyze data that are nested (eg, patients nested within hospitals). However, these models are computer-intensive and complicated to perform. There is virtually nothing in the literature that compares the results of standard logistic regression to those of hierarchical logistic models in predicting future provider performance. We sought to compare the ability of standard logistic regression relative to hierarchical modeling in predicting risk-adjusted hospital mortality rates for coronary artery bypass graft (CABG) surgery in New York State. New York State CABG Registry data from 1994 to 1999 were used to relate statistical predictions from a given year to hospital performance 2 years hence. Predicted and observed hospital mortality rates 2 years hence were compared using root mean square errors, the mean absolute difference, and the number of hospitals whose predicted mortality rate data was within a 95% confidence interval around the observed mortality rate. In these data, standard logistic regression performed similarly to hierarchical models, both with and without a second level covariate. Differences in the criteria used for comparison were minimal, and when the differences could be statistically tested no significant differences were identified. It is instructive to compare the predictive abilities of alternative statistical models in the process of assessing their relative performance on a specific database and application.Keywords
This publication has 17 references indexed in Scilit:
- Cardiac surgery report cards: comprehensive review and statistical critiqueThe Annals of Thoracic Surgery, 2001
- An introduction to hierarchical linear modellingStatistics in Medicine, 1999
- Comparing risk-adjustment methods for provider profilingStatistics in Medicine, 1998
- Improving the Statistical Approach to Health Care Provider ProfilingAnnals of Internal Medicine, 1997
- Statistical Methods for Profiling Providers of Medical Care: Issues and ApplicationsJournal of the American Statistical Association, 1997
- League Tables and Their Limitations: Statistical Issues in Comparisons of Institutional PerformanceJournal of the Royal Statistical Society Series A: Statistics in Society, 1996
- The Estimation of School EffectsJournal of Educational and Behavioral Statistics, 1995
- Cleveland Health Quality Choice: A Model for Collaborative Community-Based Outcomes AssessmentThe Joint Commission Journal on Quality Improvement, 1994
- Empirical Bayes methods for estimating hospital‐specific mortality ratesStatistics in Medicine, 1994
- Stein's Paradox in StatisticsScientific American, 1977