142nd APHA Annual Meeting and Exposition

Annual Meeting Recordings are now available for purchase

298629
Predicting Fiscal Year (FY) 2015 Hospital Value-based Purchasing (VBP) redistribution slope

142nd APHA Annual Meeting and Exposition (November 15 - November 19, 2014): http://www.apha.org/events-and-meetings/annual
Tuesday, November 18, 2014

Alice Li, MS , Decision Analytic & Research, Press Ganey Associates, Inc., Chicago, IL
Jenhao (Jacob) Cheng, PhD, MS , Decision Analytic & Research, Press Ganey Associates, Inc., Elkridge, MD
Nikolas Matthes , PRESSGANEY INC, Wakefield, MA
Background: Under CMS' the Value-Based Purchasing Program (VBP), CMS withhold from participating hospitals a fixed percentage of base DRG payments. Based on their overall VBP scores hospitals can earn back part, all, or more than their withhold. Law requires that the total amount of withheld payments be redistributed and each hospital’s VBP payment will depend on its overall VBP score and the slope of redistribution function. Predicting the redistribution slope is crucial for hospitals to estimate financial impact early. We developed predictive models for the FY 2015 VBP slope.

Data: For the process and HCAHPS domains, we used hospital compare data for over 3,000 hospitals and Press Ganey’s proprietary data for over 1,500 hospitals. For the outcome and efficiency domains, we used hospital compare data. We also used financial data from the most recent MedPAR data and hospital characteristics information from AHA survey.

Methods: First, we forecast measure-level performance scores at the corresponding performance assessment periods for each hospital with exponential weighted moving average (EWMA) models. While double-EWMA was applied to proportional measures (process, HCAHPS and mortality) with a clear trend, single-EWMA was more appropriate for the other ratio measures. Then overall VBP scores were calculated by plugging the predicted measure-level scores into a series of calculation formulas. Finally, the redistribution slope was obtained with nonlinear combination of individual VBP scores and withheld dollars. Additionally we looked into the distribution of the predicted VBP scores and the variation by hospital characteristics was analyzed with descriptive analysis and multiple ANOVA model.

Results: The predicted overall VBP score centers around 40 for FY 2015 with a near-normal distribution and a slightly right skewness driven by the HCAHPS, outcome, and efficiency domains. The estimated slope is 2.65, indicating that the break-even point will be at 37.7. Higher VBP performance was predicted for hospitals in the West and Midwest region, hospitals of smaller bed size, in private ownership, and in rural setting.

Conclusion: With the introduction of the efficiency domain and two new outcome measures, the average VBP score is low by at least 25% in FY 2015 compared with FY 2014.  Consequently the payment adjustment factor is increased. Prediction on the new measures poses specific challenges. Hospitals are encouraged to pay more attention and allocate additional resources to improve performance on the spending efficiency and patient safety outcomes measures.

Learning Areas:

Provision of health care to the public
Public health or related organizational policy, standards, or other guidelines
Public health or related public policy
Public health or related research

Learning Objectives:
Describe how predictive modeling is applied to VBP measures Describe how the redistribution slope is calculated Describe how VBP scores vary by hospital characteristics

Keyword(s): Quality of Care, Quality Improvement

Presenting author's disclosure statement:

Qualified on the content I am responsible for because: I have been a senior research analyst for over 6 years in the area of hospital quality improvement research, focusing on research in clinical process, patient satisfaction, outcome and efficiency quality measures. Among my scientific interests has been analysis of identifying key drivers of higher quality of care.
Any relevant financial relationships? No

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.