This comprehensive guide presents a data science approach to healthcare quality measurement and provider profiling for policymakers, regulators, hospital quality leaders, clinicians, and researchers. Two volumes encompass basic and advanced statistical techniques and diverse practical applications. Volume 1 begins with a historical review followed by core concepts including measure types and attributes (bias, validity, reliability, power, sample size); data sources; target conditions and procedures; patient and provider observation periods; attribution level; risk modeling; social risk factors; outlier classification; data presentation; public reporting; and graphical approaches. Volume 2 introduces causal inference for provider profiling, focusing on hierarchical regression models. These models appropriately partition systematic and random variation in observations, accounting for within-provider clustering. Item Response Theory models are introduced for linking multiple categorical quality metrics to underlying quality constructs. Computational strategies are discussed, followed by various approaches to inference. Finally, methods to assess and compare model fit are presented.
Preface and Acknowledgments; 18. Preliminaries; 19. A causal inference framework; 20. Introduction to hierarchical linear models; 21. Hierarchical generalized linear models; 22. Composite measures and multiple outcomes; 23. Computational approaches; 24. Inference for hierarchical generalized linear models; 25. Model assessment; 26. Future directions; Bibliography; Index.
Delivery policy
Security policy
Return policy
