Combining Surgical Outcomes and Patient Experiences to Evaluate Hospital Gastrointestinal Cancer Surgery Quality

Background Assessments of surgical quality should consider both surgeon and patient perspectives simultaneously. Focusing on patients undergoing major gastrointestinal cancer surgery, we sought to characterize hospitals, and their patients, on both these axes of quality. Methods Using the American College of Surgeons' National Surgical Quality Improvement Program registry, hospitals were profiled on a risk-adjusted composite measure of death or serious morbidity (DSM) generated from patients who underwent colectomy, esophagectomy, hepatectomy, pancreatectomy, or proctectomy for cancer between January 1, 2015 and December 31, 2016. These hospitals were also profiled using the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey. Highest-performing hospitals on both quality axes, and their respective patients, were compared to the lowest-performing hospitals. Results Overall, 60,526 patients underwent their cancer operation at 530 hospitals. There were 38 highest- and 48 lowest-performing hospitals. The correlation between quality axes was poor (rho = 0.10). Compared to the lowest-performing hospitals, the highest-performing hospitals were more often NCI-designated cancer centers (29.0% vs. 4.2%, p = 0.002) and cared for a lower proportion of Medicaid patients (0.14 vs. 0.23, p < 0.001). Patients who had their operations at the lowest- versus highest-performing hospitals were more often black (17.2% vs. 8.4%, p < 0.001), Hispanic (8.3% vs. 3.5%, p < 0.001), functionally dependent (3.8% vs. 0.9%, p < 0.001), and not admitted from home (4.4% vs. 2.4%, p < 0.001). Conclusions Hospital performance varied when assessed by both risk-adjusted surgical outcomes and patient experiences. In this study, poor-performing hospitals appeared to be disproportionately serving disadvantaged and minority cancer patients.