Geographic Variations in Lung Cancer Lobectomy Outcomes: The General Thoracic Surgery Database

Abstract

Lung cancer ranks as the top cancer killer in the United States. Using The Society of Thoracic Surgeons General Thoracic Surgery Database (GTSD), the geographic variability of lung cancer lobectomy for operative mortality and major morbidity were examined. From January 2009 to June 2015, the GTSD lung cancer lobectomy records (excluding robotic procedures) were assigned to a US Census region using hospital location. Surgeons performing fewer than seven lung cancer lobectomies per year were categorized as "low volume." The American College of Surgeons Oncology Group criteria were used to classify patients as "high risk." Applying the published GTSD risk algorithms, regional unadjusted and adjusted odds ratios were computed using univariable and multivariable generalized estimating equation logistic regression. Across geographic regions, patient risk factors and outcomes were compared using Kruskal-Wallis and χ2 tests. From 2009 to 2015, there were 39,078 lung cancer lobectomies that met study inclusion criteria (31.5% Northeast, 23.5% Midwest, 31.1% South, and 14.0% West). Fewer high-risk cases were seen in the West region (18.9% Northeast, 19.6% Midwest, 19.9% South, and 15.9% West; p< 0.001). Across geographic regions, there was no statistically significant difference in the proportion of low-volume surgeons (39.8% Northeast, 44.8% Midwest, 45.8% South, and 56.3% West; p= 0.0512). Adjusted odds ratios for operative mortality and major perioperative morbidity did not show statistically significant differences across regions (p= 0.761 and p= 0.600, respectively). Despite geographic variations in the proportion of high-risk lobectomies, the risk-adjusted mortality and morbidity outcomes did not vary by region.