Using Estimated True Safety Event Rates vs. Flagged Safety Event Rates: Does It Change Hospital Profiling and Payment?

• Activity Number: 328
• Type: Topic Contributed
• Date/Time: Tuesday, August 5, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Health Policy Statistics Section
• Author(s): Amy Rosen*+
• Companies: VA Boston Healthcare System
• Keywords: patient safety, hospital performance, pay-for-performance

Abstract

Hospital profiling based on administrative claims data suffers the statistical critique that those data do not accurately count true patient outcomes, such as safety events. We assess the incorporation of clinical chart abstractions, containing true counts, into statistical adjustment of widely used hospital patient safety composite measures. In our retrospective analysis of 2010 VA hospital discharge data (n=517,814 hospitalizations in 151 hospitals), we estimate the AHRQ Patient Safety Indicator (PSI) composite measure and compare it to two modified composite measures, which adjust for true counts. We find that a small number of hospitals had substantial differences in their PSI composite measures and estimate that approximately one-third of hospitals would see a change of at least 10% in performance payments if those statistical adjustments were made. In summary, we suggest that changes in hospital profiles and payments would be substantial for some hospitals if the PSI composite score used weights reflecting the relative prevalence of true versus flagged events, which is an important statistical consideration in pay-for-performance programs.