The public reporting of comparative provider performance is widespread among public and private purchasers of health care. A key goal of these reporting efforts is to provide consumers with comparative information that can help them choose high-value providers (i.e., high-quality, low-cost). However, the evidence to date shows challenges in consumer use of these reports, reflecting problems in their design, including the underlying analytic methods, the content and display of information, and dissemination to consumers at the point when health care decisions occur. While there is great interest in and need to communicate the differential """"""""value"""""""" of providers to consumers given efforts to increase consumer cost sharing and selection of high value services and providers, there has been little progress in developing a value metric that combines cost and quality information. Challenges exist in combining measures to create a measure of value where cost and quality measures have weak associations. Current attempts at showing """"""""value"""""""" have focused on side-by-side displays of cost and quality measures. Moreover, as measurement pushes further into areas where the number of events per provider is low, problems arise in obtaining a strong signal on a provider's true performance. We propose to develop an innovative framework that conceptualizes health care value as a two-dimensional construct. Our approach will estimate optimal provider performance across the cost-quality continuum, allowing for fuller consideration of cost-quality tradeoffs in public reports.
Aim 1 of the project is to improve the precision and substantive relevance of the information provided to consumers in public reports of health care provider quality. We will do this by (a) assessing the risk of misclassifying physicians into performance tiers and associate reliabilities of physician profiles with misclassification risk and (b) examining whether performance summaries that are tailored to specific reporting goals (e.g., ranking, establishing percentiles or tiers of performance) result in more accurate and precise public reports of small providers'performance as compared to the standard approach of using provider-specific means.
Aim 2 is to develop a rigorous cost-quality modeling framework to inform the public reporting of health care provider value. We will employ bivariate hierarchical modeling to obtain measures of the cost-quality tradeoff at each level of the provider hierarchy. Public Health Relevance of the Project: Our findings will help the sponsors of public reports minimize the frequency and severity of provider performance misclassification. The ultimate benefit from our project will be to strengthen the credibility of public reports, leading to increaed use by consumers, ultimately improving the value of health care delivered in the U.S.
This project aims to improve the precision and substantive relevance of the information provided to consumers in public reports of health care provider quality and cost. Public Health Relevance of the Project: Our findings will help the sponsors of public reports minimize the frequency and severity of provider performance misclassification. The ultimate benefit from our project will be to strengthen the credibility of public reports, leadng to increased use by consumers, ultimately improving the value of health care delivered in the U.S.
|Paddock, Susan M; Damberg, Cheryl L; Yanagihara, Dolores et al. (2017) What Role Does Efficiency Play in Understanding the Relationship Between Cost and Quality in Physician Organizations? Med Care 55:1039-1045|
|Paddock, Susan M; Adams, John L; Hoces de la Guardia, Fernando (2015) Better-than-average and worse-than-average hospitals may not significantly differ from average hospitals: an analysis of Medicare Hospital Compare ratings. BMJ Qual Saf 24:128-34|
|Paddock, Susan M (2014) Statistical benchmarks for health care provider performance assessment: a comparison of standard approaches to a hierarchical Bayesian histogram-based method. Health Serv Res 49:1056-73|