Association between volume and cost in low-resection volume regions: a population-level study on pancreatoduodenectomy for pancreatic cancer patients.

Background: Pancreatoduodenectomy (PD) is a highly complex, invasive, and costly surgical procedure. Limited evidence on the PD volume-cost relationship in countries with a low population density exists. This study aimed to investigate this issue in Australia.

Methods: This retrospective cohort study included pancreatic cancer patients who had a PD at any public hospital in New South Wales, Australia between 2016 and 2019. The primary outcome was the total hospital cost during PD admission (not including patient financial burden). Study hospitals were grouped into low-volume hospitals (LVHs; <10 PDs per annum) or high-volume hospitals (HVHs). Multivariable modelling was applied to examine the association between volume and cost.

Results: Of 443 PDs, the median total hospital cost per patient at HVHs was AU$55398; significantly lower than that at LVHs (AU$62859; P = 0.001). After adjusting for available patient and clinical factors, the total cost per patient at LVHs was 22% higher than that of HVHs (adjusted estimate: 1.22, 95% CI: 1.08-1.37; P = 0.002). Similar patterns were found in three main cost components: 24% higher employee cost at LVHs than at HVHs (1.24, 95% CI: 1.10-1.41; P < 0.001), 15% higher operating cost (1.15, 95% CI: 1.00-1.31; P = 0.047), and 31% higher other costs (1.31, 95% CI: 1.12-1.53; P < 0.001).

Conclusion: Performance of PDs at HVHs was associated with substantially lower hospital costs. Our findings demonstrate the likely economic benefit of centralizing PDs in countries with a relatively low population density. Future studies should investigate related patient financial burdens.