UCSF

Background

Surgical care comprises 30% of hospital admissions and half of overall hospital costs. Surgical complications, in particular, increase hospital costs by approximately $20,000 per admission and extend hospital stays by 9.7 days. How to improve surgical care quality and reduce costs has attracted much attention from payers and policymakers. Pay-for-performance (P4P) schemes, along with other efforts, have emerged as a key tool for improving quality at reduced cost by linking payment to performance measures.

Objectives

The overarching aim of this dissertation is to investigate the impact of P4P programs on improving the value of care for surgical patients. To address critical gaps in the literature, three primary aims are: (1) to identify the optimal payment design that maximizes the impact of P4P programs, (2) to evaluate the effectiveness of a mandated national P4P program launched by the Centers for Medicare and Medicaid Services, and (3) to assess the potential negative consequences of P4P programs on patient outcomes.

Methods

To address Aim 1, a systematic review was conducted using five databases to understand how much variation in surgical outcomes is attributable to the design of P4P programs. Studies were selected for review based on PRISMA guidelines, and the quality of individual studies was evaluated based on the STROBE checklist. To address Aim 2, an empirical study employing a propensity score weighted difference-in-differences design was performed to evaluate the impact of P4P programs on quality and cost outcomes of surgical care. To address Aim 3, another empirical analysis, building on the same statistical design as Aim 2, was performed to investigate the unintended consequences of P4P.

Results and Implications

The systematic review highlighted large variation in outcomes associated with payment designs of P4P programs. The evidence regarding preferred payment designs was inconclusive due to mixed results and studies lacking rigorous designs. Future studies are needed to draw stronger inferences about the best P4P payment design.

The first empirical study found that the incidence of surgical site infection, length of stay, and hospital costs decreased following the implementation of a mandated national P4P policy. Policymakers may therefore consider these findings when considering the continuation and expansion of this P4P program, and other payers may also consider implementing a similar policy.

Building on this analysis, the second empirical analysis found that patients treated in low-income-serving hospitals had worsened surgical quality outcomes following the P4P policy. This highlighted the importance of monitoring for potential unintended consequences of P4P programs and indicated that solutions to addressing the quality chasm in surgical outcomes need to be developed if equity in outcomes, while improving care for all surgical patients, is to be achieved.

Conclusions

P4P may be a promising strategy if designed carefully. This dissertation indicates that the increasing popularity of P4P programs could be a window of opportunity for providers, hospitals, and payers to align quality of care with expenditures through incentives. Monitoring for potential unintended consequences, especially for populations at risk, and strategies to counteract these unintended consequences are recommended components of all P4P programs.