Hospital size and demographics associated with HCAHPS and implications for hospital value-based purchasing (HVBP).

Abstract

38 Background: Hospital Consumer Assessments of Healthcare Providers and Systems (HCAHPS) scores appear to be lower in densely populated areas and for urban academic medical centers (AMC). These patient satisfaction scores contribute to CMS reimbursement. We hypothesized that demographics and hospital factors such as size were significant determinants of HCAHPS. Methods: HCAHPS, hospital bed, and county demographics were obtained from the Hospital Compare, American Hospital Directory, and US Census Bureau websites, respectively. An aggregate positive score was created for each hospital. Multivariate regression modeling was performed for the aggregate positive score and all ten dimensions of HCAHPS. Standardized partial regression coefficients were used to assess strengths of postulated predictors (Table). A predictive formula based on the 4 strongest predictors was developed. An adjustment model corrected for demographic and hospital factors. Hospitals were rescored using this model to generate an adjusted score (WIPSAS – weighted individual patient satisfaction score). HCAHPS hospital ranking was compared to WIPSAS ranking for all NY State hospitals. HCAHPS were obtained from 3,192 hospitals and demographic data collected from 3,144 counties. Results: ‘Hospital beds’ and ‘non-English speaking’ were the strongest negative predictors while ‘% Bachelor’s’ and ‘white alone’ were the strongest positive predictors of HCAHPS. The adjusted r2 was.22 for the model and.155 for the predictive formula. WIPSAS altered the ranking in NY State. The adjusted ranking moved 3 AMCs into the top ten (prior #42, #43, #46). Conclusions: Demographic and hospital factors influence HCAHPS. Adjusted scores may be more representative of a hospital’s actual performance. CMS should utilize WIPSAS adjusted scores in order to more fairly calculate HVBP reimbursements. [Table: see text]