Presentation Authors: Meera Chappidi*, Anobel Odisho, San Francisco, CA
Introduction: Previous studies in non-urologic populations have demonstrated readmission to non-index hospitals can result in increased costs of care, and are a source of variation that can be targeted to decrease costs. However, this has not been well studied in urologic surgery populations. Therefore, our objective was to compare the cost of index vs. non-index readmissions following major urologic oncology surgeries.
Methods: Patients undergoing radical cystectomy (RC), prostatectomy (RP), nephrectomy (RN), and partial nephrectomy (PN) for urologic cancers were identified between 2010-2014 in the Nationwide Readmissions Database. Among patients who experienced readmission within the 90-day post-operative period, the cost of first readmission was compared between readmissions to index vs. non-index hospitals. Multivariable models controlling for patient, surgery, and hospital-level characteristics were used to determine if non-index readmission was associated with increased readmission costs.
Results: Following major urologic oncology surgeries, 90-day readmission rates were 40.3% (RC, n=9,964), 5.8% (RP, n=12,341), 14.9% (RN, n=11,543), 14.0% (PN, n=6,419), and 17.6% (NU, n=2,431). The percentages of readmitted patients with readmission to non-index hospitals were 26.6% (RC), 30.7% (RP), 28.7%(RN), 25.6% (PN), and 30.3% (NU). On multivariable modeling, non-index hospital readmissions were more expensive following RP ($10,826 vs. $9,585, p < 0.001), but less expensive following radical cystectomy ($12,275 vs. $14,543, p=0.001). The cost of index vs. non-index hospital readmissions was comparable following RN, PN, and NU (Figure).
Conclusions: Hospital readmission to non-index hospitals is a source of significant cost variability following RC and RP. Non-index readmissions were more expensive than index readmissions following RP, but the opposite was true for RC. This suggests a unique approach must be used for each surgery to understand how to improve readmission costs. As there is increased focus on cost containment with the implementation of bundled payment models, there is a need to better understand the factors associated with cost variability based on hospital readmission location following RC and RP.