Preoperative acute care spending predicts costs in bundled payment programs

Surgical alternative payment models (APMs), such as the Bundled Payments for Care Improvement (BPCI) initiative, aim to reduce health care costs by issuing lump-sum payments for surgical episodes. However, these programs have not achieved net savings. Because preoperative care may influence postoperative expenditures, incorporating this phase into cost-containment strategies may improve APM performance. This study aimed to (1) evaluate the association between preoperative and postoperative acute care costs for 4 BPCI-targeted surgical procedures-percutaneous coronary intervention, cardiac defibrillator placement, lower extremity joint replacement, and spinal surgery-and (2) identify clinical and nonclinical drivers of perioperative costs.

We conducted a retrospective cohort study using 2018 Healthcare Cost and Utilization Project data from Florida, Maryland, New York, and Wisconsin. The sample included patients with both a preoperative and a postoperative acute care encounter within 90 days of surgery.

Multivariable ordered logistic regression models assessed the associations among pre-, intra-, and postoperative cost quartiles.

Episode-related costs ranged from $12,808 (bottom quartile) to $70,286 (top quartile). Patients in the highest preoperative cost quartile were significantly more likely to remain in the highest operative and postoperative quartiles. Adjusted models showed that being in the highest preoperative cost quartile was associated with a 13.7-percentage point (95% CI, 12.2-15.2; P  < .001) higher probability of being in the top postoperative cost quartile. Facility-level BPCI participation was associated with higher preoperative costs.

Preoperative spending is a strong predictor of postoperative expenditures. Accounting for this relationship may enhance the design and performance of future surgical APMs.