Surgeon Specialty and Disparities in Lung Cancer Resection Mortality

Purpose: Disparities in lung cancer surgical outcomes are well documented. However, it is unknown how surgeon specialty is related to these disparities. In this context, we evaluate how disparities in 30-day mortality are associated with surgeon specialty by evaluating outcomes for general, cardiothoracic, and non-cardiac thoracic surgeons.

Methods: Using 100% Medicare claims between 2015-2019, we identified patients undergoing resection for lung cancer using appropriate Current Procedural Terminology (CPT) codes. Operating surgeons were identified using National Provider Identifier list provided by the American Board of Thoracic Surgery (ABTS). Surgeons were designated as general surgeons if not on the ABTS list, cardiothoracic surgeons if they performed cardiac valve or coronary artery bypass graft operations during the study period, or non-cardiac thoracic surgeons otherwise. Disparities in patient populations by surgeon specialties were assessed for race (non-Hispanic Black vs. non-Hispanic White), dual-eligibility status (dually enrolled in Medicaid and Medicare vs. non-dual eligible, and geographic residence (rural vs. urban) using a chi-squared test. Black race, dual-eligible and rural residence were classified as vulnerable populations. Finally, we evaluated 30-day mortality differences between surgeon specialties using logistic regression models.

Results: Of 58,547 patients, 11% of resections for lung cancer were performed by general surgeons, 37% by cardiothoracic surgeons, and 51% by non-cardiac thoracic surgeons. There were significant differences in the demographics of patients by sub-specialty (Table 1), with non-cardiac thoracic surgeons more likely to treat Black patients and those living in urban areas. The primary outcome of 30-day mortality differed by surgeon specialty, from a low of 1.4% for non-cardiac thoracic surgeons to a high of 2.6% for general surgeons (p < 0.001). In addition, Figure 1 illustrates that disparities in mortality for vulnerable populations was associated with surgeon specialty, with large spreads for general surgeons (2.5% and 3.4% for non-dual-eligible vs dual-eligible, 2.4% and 3.0% for urban vs rural, and 2.5% and 3.8% for White vs Black patients). In contrast, disparities for non-cardiac thoracic surgeons were minimal. Notably, Black patients undergoing resection by general surgeons had the highest overall mortality at 3.8%, nearly double the rate for our entire study population (2.0%) and almost triple the rate for Black patients treated by non-cardiac thoracic surgeons (1.3%).

Conclusion: Non-cardiac thoracic surgeons had lower overall mortality rates and a reduction in disparities for vulnerable populations. Ongoing efforts to mitigate disparities in lung cancer resection quality should consider the value of regionalized care based on surgeon specialty, quality improvement in current practice settings, and barriers to access.

Identify the source of the funding for this research project: NHLBI T32HL007749- Sidra Bonner
National Clinician Scholars Program at the University of Michigan- Sidra Bonner
K08CA237638-01- Lindsey Herrel