Hospital Variation in Invasive Mediastinal Staging for Patients With Stage I Lung Cancer Treated With Radiation.
[BACKGROUND] The use of invasive mediastinal staging (IMS) before radiation therapy in patients with early stage lung cancer is variable.
- p-value P < .001
APA
Odeh B, Pohlman A, et al. (2026). Hospital Variation in Invasive Mediastinal Staging for Patients With Stage I Lung Cancer Treated With Radiation.. Chest. https://doi.org/10.1016/j.chest.2025.12.038
MLA
Odeh B, et al.. "Hospital Variation in Invasive Mediastinal Staging for Patients With Stage I Lung Cancer Treated With Radiation.." Chest, 2026.
PMID
41525890
Abstract
[BACKGROUND] The use of invasive mediastinal staging (IMS) before radiation therapy in patients with early stage lung cancer is variable. The reasons for this variation and whether it affects outcomes is unknown. In this context, we conducted a patient- and hospital-level analysis to better understand the impact of this variation.
[RESEARCH QUESTION] How does the rate of IMS before radiation therapy for patients with early stage lung cancer vary across hospitals nationally and does it impact survival?
[STUDY DESIGN AND METHODS] We used the National Cancer Database to identify patients with stage I non-small cell lung cancer treated with radiation between 2004 and 2022. The use of IMS was the main exposure variable. A hierarchical regression model was used to calculate risk- and reliability-adjusted rates of IMS using patient, tumor, and hospital characteristics. Hospitals were distributed into quintiles based on adjusted rates. Kaplan-Meier survival analyses were used to estimate survival.
[RESULTS] A total of 34,879 patients with stage I lung cancer were treated with stereotactic body radiation therapy at 1,210 hospitals. Of those, 4,016 (11.5%) underwent IMS; of which, 3,060 (76.2%) were endobronchial ultrasound. The rates of IMS before radiation varied across hospitals from 0% to 100%. When stratified into risk- and reliability-adjusted quintiles, the variation persisted with the lowest and highest quintiles having IMS rates of 13.2% and 18.5%, respectively, and an overall rate of 15.5% (P < .001). Patient and tumor characteristics were similar between quintiles. Hospitals with the lowest rates had lower survival (median survival, 46 vs 54 months for the highest; P < .001). Of the variation, 31.4% is explained at the hospital level.
[INTERPRETATION] There is wide variation in the use of IMS before radiation therapy for stage I lung cancer, which is not explained by patient or tumor characteristics. Hospitals with higher IMS rates have better overall survival.
[RESEARCH QUESTION] How does the rate of IMS before radiation therapy for patients with early stage lung cancer vary across hospitals nationally and does it impact survival?
[STUDY DESIGN AND METHODS] We used the National Cancer Database to identify patients with stage I non-small cell lung cancer treated with radiation between 2004 and 2022. The use of IMS was the main exposure variable. A hierarchical regression model was used to calculate risk- and reliability-adjusted rates of IMS using patient, tumor, and hospital characteristics. Hospitals were distributed into quintiles based on adjusted rates. Kaplan-Meier survival analyses were used to estimate survival.
[RESULTS] A total of 34,879 patients with stage I lung cancer were treated with stereotactic body radiation therapy at 1,210 hospitals. Of those, 4,016 (11.5%) underwent IMS; of which, 3,060 (76.2%) were endobronchial ultrasound. The rates of IMS before radiation varied across hospitals from 0% to 100%. When stratified into risk- and reliability-adjusted quintiles, the variation persisted with the lowest and highest quintiles having IMS rates of 13.2% and 18.5%, respectively, and an overall rate of 15.5% (P < .001). Patient and tumor characteristics were similar between quintiles. Hospitals with the lowest rates had lower survival (median survival, 46 vs 54 months for the highest; P < .001). Of the variation, 31.4% is explained at the hospital level.
[INTERPRETATION] There is wide variation in the use of IMS before radiation therapy for stage I lung cancer, which is not explained by patient or tumor characteristics. Hospitals with higher IMS rates have better overall survival.