Screening
Transporting the effects of lung cancer screening from the National Lung Cancer Screening Trial to a US target population screened within the past year Xiaomeng Chen* Xiaomeng Chen Chen Chen Chen Chen Chen Chen Chen Chen Chen Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina at Chapel Hill
Background: The National Lung Screening Trial (NLST) showed that low-dose computed tomography (LDCT) versus chest radiography (CXR) reduces lung cancer mortality. However, the broader population eligible for and undergoing lung cancer screening differs from trial participants; these differences, along with others, may impact real-world population effectiveness. We aimed to estimate the effects of screening with LDCT versus CXR on lung cancer mortality, had the NLST been conducted in a target population undergoing screening in 2024.
Methods: We used individual-level data from the NLST and 2024 Behavioral Risk Factor Surveillance System (BRFSS). BRFSS respondents meeting NLST eligibility criteria and reporting lung cancer screening in the past year were included. We applied inverse odds weights for trial participation, combined with BRFSS survey weights, and Poisson regression to estimate the average effects of LDCT versus CXR on lung cancer mortality in the screened population (i.e., average treatment effect in the treated). We estimated 95% confidence intervals (CIs) using a robust variance estimator.
Results: This analysis included 53,452 NLST participants and 4,259 BRFSS respondents (weighted n=1,683,411) recently screened with LDCT. Compared with the NLST, the target population was older, more racially and ethnically diverse, and had more comorbidities. In the unweighted NLST population, the overall lung cancer mortality rate was 280 deaths per 100,000 person-years (PYs); the incidence rate ratio (IRR) comparing LDCT versus CXR on lung cancer mortality was 0.80 (95% CI 0.69-0.92). In the weighted NLST population, the lung cancer mortality rate was 453 deaths per 100,000 PYs; the IRR of lung cancer mortality for LDCT versus CXR was 0.88 (95% CI 0.69-1.13).
Conclusion: Compared to the unweighted NLST population, the weighted NLST population had substantially higher lung cancer mortality rates but a slightly attenuated relative mortality benefit of LDCT versus CXR.