Related Experiment Videos
Medical profiling: improving standards and risk adjustments using hierarchical models
J F Burgess1, C L Christiansen, S E Michalak
1Management Science Group, Department of Veterans Affairs, Bedford, MA 01730, USA. burgess@world.std.com
Journal of Health Economics
|September 8, 2000
Summary
Profile analysis for healthcare performance extremes requires medically meaningful standards and advanced statistical methods. Hierarchical regression offers robust risk adjustment for accurate provider profiling and quality assessment.
Area of Science:
- Health Services Research
- Biostatistics
- Medical Quality Assessment
Background:
- Performance profiling is crucial for identifying healthcare extremes.
- Current methods are often limited by statistical standards and inadequate risk adjustment.
- Provider case-mix differences can significantly impact analysis outcomes.
Purpose of the Study:
- To propose medically meaningful standards for performance profiling.
- To introduce hierarchical regression methods for improved risk adjustment.
- To enhance the accuracy and reliability of medical unit profiling.
Main Methods:
- Proposed medically meaningful standards to replace statistical ones.
- Utilized hierarchical regression to manage random variation.
- Implemented risk adjustment for provider case-mix differences.
Main Results:
- Hierarchical regression effectively handles multiple levels of variation.
- Risk adjustment using these methods accounts for provider case-mix.
- The approach supports the proposed medically meaningful standards.
Conclusions:
- Conclusions from performance profiling are sensitive to chosen standards and methods.
- Medically meaningful standards and hierarchical regression improve risk adjustment.
- This framework enables meaningful medical unit profiling based on agreed-upon standards.