Preparing for Participation in the Centers for Medicare and Medicaid Services’ Bundle Care Payment Initiative – Advanced (BPCI-A) for Major Bowel Surgery

Abstract

\n BackgroundAs healthcare costs rise, there is increasing emphasis on alternative payment models to improve care efficiency. Bundled payment represents an alternative reimbursement model gaining popularity within the surgical sphere. We aimed to assess where the largest opportunities for care improvement lay and how best to identify patients at high-risk of suffering costly complications.MethodsWe utilized itemized CMS claims data for a historical group of 252 bundled payment patients and performed a cost analysis to identify opportunities for improved care efficiency. Based on the results of this cost analysis, we identified readmissions as a target for improvement. We then assessed whether the ACS NSQIP surgical risk calculator could accurately identify patients within our bundled payment population who were at high-risk of readmission using a logistic regression model.ResultsReadmissions accounted for 12.8% of the average total care episode cost with a coefficient of variation of 2.72, thereby representing the most substantial, non-fixed cost for our bundled payment patients. Patients readmitted within their 90-day care episode were 2.53 times more likely to be high-cost (>$60,000) than patients not readmitted. However, the ACS NSQIP surgical risk calculator did not accurately predict patients at high-risk of readmission within the first 30-days with an AUROC of 0.58.ConclusionsOur study highlights the importance of reducing readmissions as a central component to improving care for bowel surgery bundled payment patients. Preventing such readmissions requires accurate identification of patients at high-risk of readmission; however, current risk prediction models lack the adaptability necessary for this task.