Value-Based Purchasing (VBP): Identifying improvement opportunities by studying the scoring pattern and variation among individual measures

Background: CMS just released the proposed methodology for VBP for fiscal year 2013. Hospitals will be scored on improvement and achievement on 17 clinical and 8 satisfaction measures.

Data: Measure-level scores were calculated for 3211 hospitals from most recent public data from 2Q09 to 1Q10 for the assessment period and prior year for the baseline period.

Methods: Coefficient of variation (CV) was used for measure-level scores to identify measures with low mean scores and high variance for improvement opportunities. Each measure was categorized into one of the following four groups: scored by achievement, scored by improvement, tied and zero scores. Distribution of 4 score groups and rank correlation were performed to explain scoring patterns.

Results: Antibiotic selection in pneumonia and beta blocker in surgeries are the two measures with highest CV (72 and 68) presenting the greatest opportunity for improvement. All 8 satisfaction measures have higher CV and present opportunities with physician communication and staff responsiveness having the highest CV (>95). Low-performing measures were those with a high ratio of zero group to achievement group. The achievement is always the dominant over the improvement group (about 50% or more) and explaining a high correlation (>0.93) with the individual measure score. Satisfaction measures have a higher proportion in zero group (25%-31%) explaining lower satisfaction performance.

Conclusion: Hospitals are found to be lagging on two clinical measures and all satisfaction measures. Efforts to improve performance in these areas may lead to significant national improvement on both clinical process and patient satisfaction.