Urbania TH, et al. Chest 2020.
BACKGROUND: Follow-up of chest computed tomography (CT) findings suspicious for lung cancer may be delayed because of inadequate documentation. Standardized reporting and follow-up may reduce time to diagnosis and care for lung cancer.
STUDY DESIGN AND METHODS: We implemented a reporting system that standardizes tagging of chest CT reports by classifying pulmonary findings. The system also automates referral of patients with findings suspicious for lung cancer to a multidisciplinary care team for rapid review and follow-up. The system was designed to reduce the time to diagnosis, particularly for early-stage lung cancer. We evaluated the effectiveness of this system using a quasi-experimental stepped wedge cluster design, examining 99,148 patients who underwent diagnostic (non-screening) chest CT from 2015 to 2017 who had not had a chest CT in the preceding 24 months. We evaluated the association of the intervention with incidence of diagnosis and surgical treatment of early-stage (I, II) and late-stage (III, IV) lung cancer within 120 days after chest CT.
RESULTS: 40% of patients received the intervention. Among 2,856 (2.9%) patients who were diagnosed with lung cancer, 28% had early-stage disease. In multivariable analyses, the intervention was associated with a 24% greater odds of early-stage diagnosis (OR = 1.24, 95% CI, 1.09-1.41) and no change in the odds of late-stage diagnosis (OR=1.04, 95% CI, 0.95-1.14). The intervention was not associated with the rate of surgical treatment within 120 days.
INTERPRETATION: In this large quasi-experimental community-based observational study, implementation of a system that combines standardized tagging of chest CT reports with clinical navigation was effective for increasing the diagnosis of early-stage lung cancer.