Related Experiment Videos
Risk adjustment in claims-based research: the search for efficient approaches
L L Roos1, S M Sharp, M M Cohen
1Department of Community Health Sciences, Faculty of Medicine, University of Manitoba, Winnipeg, Canada.
Journal of Clinical Epidemiology
|January 1, 1989
Summary
Claims data effectively predict surgical readmissions and mortality. Adding the American Society of Anesthesiologists
Area of Science:
- Health Services Research
- Medical Informatics
- Biostatistics
Background:
- Routinely collected administrative data, such as claims, are increasingly used for healthcare quality assessment.
- Existing claims-based indices for comorbidity and severity require validation for predicting surgical outcomes.
- The American Society of Anesthesiologists' (ASA) Physical Status score is a well-established measure of patient health status.
Purpose of the Study:
- To develop and test claims-based indices for predicting postsurgical readmissions and mortality.
- To evaluate whether incorporating the ASA Physical Status score improves risk adjustment accuracy compared to claims data alone.
Main Methods:
- Development and testing of claims-based comorbidity and severity measures using hospital discharge abstracts and prior hospitalization data.
- Analysis of readmissions and mortality following three common surgical procedures using multiple logistic regression.
- Comparison of predictive model fit (Somers' Dyx) with and without the inclusion of the ASA Physical Status score.
Main Results:
- Claims-based comorbidity measures demonstrated reasonable predictive accuracy for postsurgical readmissions (Somers' Dyx: 0.23–0.38) and mortality (Somers' Dyx: 0.46–0.72).
- In most analyses (5 out of 6 cases), adding the ASA Physical Status score did not significantly improve the predictive performance of the claims-based models.
- The predictive power of claims data for surgical outcomes is substantial, suggesting their utility in risk adjustment.
Conclusions:
- Claims-based data provide a robust foundation for risk adjustment in predicting surgical outcomes.
- The incremental benefit of adding the ASA Physical Status score to claims-based models for risk adjustment appears limited in this context.
- Challenges in enhancing risk adjustment through more intensive data collection warrant further research into their implications for healthcare research and policy.