Improving Medicare's Hospital Compare Mortality Model

Abstract

Objective: To improve the predictions provided by Medicare's Hospital Compare (HC) to facilitate better informed decisions regarding hospital choice by the public.

Data sources/setting: Medicare claims on all patients admitted for Acute Myocardial Infarction between 2009 through 2011.

Study design: Cohort analysis using a Bayesian approach, comparing the present assumptions of HC (using a constant mean and constant variance for all hospital random effects), versus an expanded model that allows for the inclusion of hospital characteristics to permit the data to determine whether they vary with attributes of hospitals, such as volume, capabilities, and staffing. Hospital predictions are then created using directly standardized estimates to facilitate comparisons between hospitals.

Data collection/extraction methods: Medicare fee-for-service claims.

Principal findings: Our model that included hospital characteristics produces very different predictions from the current HC model, with higher predicted mortality rates at hospitals with lower volume and worse characteristics. Using Chicago as an example, the expanded model would advise patients against seeking treatment at the smallest hospitals with worse technology and staffing.

Conclusion: To aid patients when selecting between hospitals, the Centers for Medicare and Medicaid Services (CMS) should improve the HC model by permitting its predictions to vary systematically with hospital attributes such as volume, capabilities, and staffing.

Keywords: Bayesian statistics; Medicare quality of care; acute myocardial infarction; hospital compare.

Figure 1

Comparing a Low‐ and High‐Volume Hospital's Posterior Probability Density Function for Mortality under Two Models Notes. (A) for Hospital N = 17 displays the posterior probability density function for 30‐day mortality using direct standardization comparing the Hospital Compare model (without hospital characteristics) (solid) versus the expanded model (dashed) including hospital characteristics. Note the mortality in the expanded model is shifted far higher as compared to the Hospital Compare model. (B) displays Hospital N = 448. Here, both the Hospital Compare and expanded model provide almost the same predicted probabilities for mortality

Figure 2

Plots Comparing Five Chicago Hospitals under the Hospital Compare Model (A above) and under the Expanded Model That Includes Hospital Characteristics (B below) Using the Directly Standardized Approach