Drivers of Cost for Pancreatic Surgery: It’s Not About Hospital Volume
Outcomes for pancreatic resection have been studied extensively due to the high morbidity and mortality rates, with high-volume centers achieving superior outcomes. Ongoing investigations include healthcare costs, given the national focus on reducing expenditures. Therefore, we sought to evaluate the relationships between pancreatic surgery costs with perioperative outcomes and volume status.
We performed a retrospective analysis of 27,653 patients who underwent elective pancreatic resections from October 2013 to June 2017 using the Vizient database. Costs were calculated from charges using cost–charge ratios and adjusted for geographic variation. Generalized linear modeling adjusting for demographic, clinical, and operation characteristics was performed to assess the relationships between cost and length of stay, complications, in-hospital mortality, readmissions, and hospital volume. High-volume centers were defined as hospitals performing ≥ 19 operations annually.
The unadjusted mean cost for pancreatic resection and corresponding hospitalization was $20,352. There were no differences in mean costs for pancreatectomies performed at high- and low-volume centers [− $1175, 95% confidence interval (CI) − $3254 to $904, p = 0.27]. In subgroup analysis comparing adjusted mean costs at high- and low-volume centers, there was no difference among patients without an adverse outcome (− $99, 95% CI − $1612 to 1414, p = 0.90), one or more adverse outcomes (− $1586, 95% CI − $4771 to 1599, p = 0.33), or one or more complications (− $2835, 95% CI − $7588 to 1919, p = 0.24).
While high-volume hospitals have fewer adverse outcomes, there is no relationship between surgical volume and costs, which suggests that, in itself, surgical volume is not an indicator of improved healthcare efficiency reflected by lower costs. Patient referral to high-volume centers may not reduce overall healthcare expenditures for pancreatic operations.
This project was supported by the National Center for Advancing Translational Sciences, National Institutes of Health (NIH; grant number UL1TR001860), and the Agency for Health Care Research and Quality (Grant Number T32HS 022236). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.