Surgeon-Dependent Variability in Cost of Gastrointestinal Operations in a Single Cancer Center

Abstract

IntroductionCancer operations are increasingly utilizing specialized equipment and technology. Related costs are often not known to the responsible surgeon. We seek to evaluate cost aspects of care episodes attributable to the surgeon's management decisions. MethodsFinancial cost data in a tertiary academic cancer center were queried over 3 y. Consecutive patients undergoing gastrointestinal operations followed by inpatient admission of two or more days were included, excluding patients with 40+ d admissions. Analysis of variance, Kruskal–Wallis, and multiple regression statistics were utilized. ResultsThe study population included 1540 patients: 54% men and 46% women, with a median age of 64 y (range 15-95). Eight surgeons conducted major (82%) and minor (18%) operations, with a minimally invasive surgical approach in 60.4%. Procedures included colorectal (37%), pancreatic (19%), esophagogastric (18%), hepatobiliary (18%), and small bowel resections (8%). Total direct costs differed between surgeons with an analysis of variance coefficient range between −$3265 and +$6163 (P < 0.001). Surgeons' cost differences were observed for central medical supply, operating room (OR) supply, total OR, inpatient room, laboratory, pharmacy, supportive care (P < 0.001), and radiology costs (P < 0.02). OR supply cost was the dominant consistent domain with significant differences between surgeons in all case subcategories. When controlled for case category and minimally invasive surgical approach, multiple regression showed the most significant variations between surgeons in ORs, medical supply, and nutrition costs (P < 0.001), followed by laboratory costs (P < 0.01). Top OR supply costs were staplers and energy devices. ConclusionsEven in a highly subspecialized surgical environment, surgeons’ variable utilization of ORs and medical supplies is strongly linked to variations in care-related costs. Specific queries into supply items should reduce costs and optimize value generated.