156
How Hospitals Respond to Value-Based Purchasing Program
The study focuses on 2,448 acute-care hospitals filing Medicare Cost Reports to CMS 2012 – 2016. Critical Access Hospitals (CAHs) are excluded from the study because of their cost-based payment model. Hospital quality information is acquired from Hospital Compare. Hospital quality is assessed by about 20 measures that cover four domains: clinical care, patient experience, safety, and efficiency. Each measure is scored on both achievement compared to peer hospitals nationwide in the performance year and improvement compared to the hospital’s baseline year. The better one of the achievement score and the improvement score is taken as the final measure score. The domain score is further calculated as the sum of these measure scores. For the overall quality, the Total Performance Score (TPS) is calculated as a weighted average of the four domains. Finally, a linear exchange function translates TPSs into HVBP adjustment factor, which determines a hospital’s value-based incentive payments. Top performers are rewarded with bonus and bottom ones get a discount of full reimbursement.
To measure quality improvement, we create the Total Improvement Score (TIS) following the TPS method, where TIS is calculated as weighted average of improvement scores on all quality measures in four domains. Moreover, weighted average of achievement scores is calculated as Total Achievement Score (TAS) to identify poor-quality hospitals. Cost-cutting as an alternative strategy is measured by change of total operating cost in percentage of revenue (ΔCOST). To capture hospital’s volume shifting, change of the percentage of private payer discharge is used (ΔVOL).
Panel regressions using the HVBP adjustment factor as the main predictor together with hospital characteristics as controls are examined. TAS, ΔCOST, and ΔVOL are used as dependent variables separately in these regressions. Evidence of both quality improvement and volume-shifting are found in poor-quality hospitals. Sub-sample studies using TAS and hospital characteristics are to be explored. And we further plan to test simultaneity of TAS, ΔCOST, and ΔVOL. Simultaneous equations modeling on quality improvement, cost-cutting, and volume shifting using two-stage least square will be applied.