Dissertations, Theses, and Capstone Projects

Date of Degree

9-2023

Document Type

Dissertation

Degree Name

Ph.D.

Program

Economics

Advisor

Partha Deb

Committee Members

Michael Grossman

Jessica N. Van Parys

Subject Categories

Health Economics

Keywords

Comprehensive Care for Joint Replacement, Provider practice style, Bundled payments

Abstract

This dissertation consists of three chapters that investigate impact of incentive changes on hospital and physician practice styles, as well as patient outcomes.

In an effort to curb the escalating healthcare expenditure, the Center for Medicare \& Medicaid Innovation has introduced several alternative payment models, which have garnered significant attention in numerous studies with varied findings. This first chapter introduces the focal point of this thesis, namely the Comprehensive Care for Joint Replacement (CJR) model, and provides a review of pertinent studies that have examined this model and related areas. Specifically, this chapter reviews and summarizes the findings from studies that have investigated the Bundled Payments for Care Improvement (BPCI) Initiative, the CJR Model, and other alternative payment models. In the second section of this chapter, a conceptual model is developed that emulates the CJR Model and provides predictions of potential model effects.

The second chapter explores and examines the mechanisms through which hospitals, as healthcare providers, respond to the incentive changes brought about by a Medicare payment innovation. The estimation results, derived from an extended difference-in-differences (DiD) framework that includes chronic or geographical controls, reveal the general and spillover effects of the Medicare policy on the targeted patients, non-targeted privately insured patients, and non-targeted patients undergoing related procedures. To examine the policy mechanisms, an econometric specification by extending the DiD method with Bartik-like instruments is developed, enabling the exploration of policy effect heterogeneity across various dimensions according to the conceptual model. The findings indicate that hospitals' responses to the policy are contingent upon their capacity in terms of discharge supplies and business patterns: greater accessibility and experience in specific discharges are associated with increased utilization, whereas the opposite leads to inconclusive model outcomes.

Building upon the previous chapter, which examined a mandatory bundled payment model, the third chapter examines the policy effects on physician practice changes when treating non-targeted patients. Employing a DiD framework, the analysis demonstrates that physicians experience effects when treating non-targeted patients, irrespective of the payment type. Effects of Practice-style changes are more pronounced when focusing solely on Model physicians, who are fully exposed to the Model. Effects are less pronounced for patients treated by a Model physician in collaboration with a peripheral physician, who hasn't treated LEJR patients but co-treats others with the Model physician. Effects are negligible when considering only the peripheral physician. The findings indicate that physicians directly exposed to the CJR model exhibit persistent practice-style changes, but these changes are contingent on exposure. In the second part of this study, Model heterogeneity is examined by employing a Bartik instrument framework. The findings indicate that surgeons with a higher proportion of Medicare patients and HHA discharge patients are more susceptible to the Model's effects and exhibit more substantial changes in their practice style.

Download

Share

COinS