Association of Perioperative Outcomes and Utilization of Services in Early-Stage Lung Cancer


Media is loading

Document Type


Presentation Date



Presentation: 6:18

Mentor: Tyler Grenda, MD, MS, Thoracic Surgery, Thomas Jefferson University Hospital

Awards: 2021 IASLC Staging Project 9th Edition


Background: There is known variation in perioperative outcomes and utilization of services at the hospital level related to early-stage lung cancer. However, it remains unknown if there are associations between perioperative outcomes and utilization of services that may profile centers across a broader context of care. In this context, we evaluated associations between perioperative outcomes (e.g. mortality) and utilization of services (e.g. bronchoscopy, hospitalization) around the time of lung cancer resection.

Methods: We performed a retrospective cohort study of patients undergoing lung cancer resection in 2017- 2019 in the State Inpatient and Ambulatory Databases (New Jersey, Pennsylvania, Florida, and Maryland). We used hierarchical logistic regression models to determine risk- and reliability-adjusted mortality at the hospital level. Utilizing the same independent variables, a logistic regression model was used to calculate risk-adjusted rates of utilization of services within 6 months prior to the index admission, and hospitalizations within 6 months around the index hospitalization for surgical resection. Hospitals were divided into tertiles based on their risk- and reliability-adjusted mortality. We then compared associations between hospital perioperative outcomes and utilization of services.

Results:A total of 15,172 patients across 298 hospitals underwent lung cancer resection. Mean risk- and reliability-adjusted mortality varied between 1.1% and 1.5% (p<0.01) in the best and lowest performing, respectively. Risk-adjusted utilization of ambulatory procedures across the perioperative outcome tertiles were 35%, 22%, and 32% (p<0.001; lowest to highest mortality tertile). Additionally, risk-adjusted hospitalization in 6 months prior to surgery was 23%, 32%, and 26% (p<0.001), while following surgery was 38%, 55%, and 42% (p<0.001), respectively. Risk-adjusted morbidity was 38%, 48%, and 43% (p=0.02) across tertiles. There were no significant differences between hospital tertiles of perioperative outcomes and risk-adjusted utilization.

Conclusions: Overall, we observed that there is variation in perioperative outcomes for lung cancer resection across hospitals. Additionally, there was variation in utilization at the hospital level, while, however, there was no significant difference between tertiles of mortality and utilization. Further evaluation to better profile centers across the continuum of early-stage lung cancer care is necessary to target improvement, lower cost, and improve care.



This document is currently not available here.