Independent survival benefit of early stage diagnosis in lung cancer using propensity score matching and stratified cox proportional hazards models

Lung cancer prognosis is strongly influenced by stage at diagnosis, but differences in patient characteristics, tumor features, and treatment allocation can complicate the interpretation of observational survival comparisons. This study evaluated the association between early-stage diagnosis and overall survival using propensity score matching, stratified Cox regression, and complementary sensitivity analyses. We analyzed 13,314 adults with primary lung cancer diagnosed between 2018 and 2022 using data from the Surveillance, Epidemiology, and End Results (SEER) program. Diagnostic stage was classified as early (Stages I–II) or late (Stages III–IV). One-to-one propensity score matching was performed using demographic, tumor, and treatment characteristics, yielding 4,348 patients in the matched cohort. Overall survival was assessed using Kaplan–Meier methods and Cox proportional hazards models. Variables violating the proportional hazards assumption were handled using stratified Cox regression. Robustness was further assessed using inverse probability of treatment weighting (IPTW), doubly robust estimation, histology-specific analyses, time-split models, and pre-pandemic versus COVID-era analyses. Early-stage diagnosis was consistently associated with substantially better overall survival. In the propensity score-matched cohort, early-stage disease was associated with a lower hazard of death than late-stage disease (aHR 0.41, 95% CI 0.37–0.46; P < 0.001). Increasing tumor size was associated with progressively higher mortality, with tumors > 60 mm showing more than twice the hazard of death compared with tumors < 20 mm (aHR 2.27, 95% CI 1.87–2.76; P < 0.001). The stage-survival association remained consistent across IPTW and doubly robust analyses, as well as across NSCLC and SCLC subgroups and pre-pandemic and COVID-era periods. Time-split analysis indicated that the survival advantage associated with early-stage diagnosis was strongest during the first 24 months of follow-up. Early-stage lung cancer diagnosis was strongly and consistently associated with improved overall survival across multiple analytical approaches and clinical subgroups. The findings reinforce the prognostic importance of stage and tumor burden while highlighting the potential value of timely diagnosis. Given the retrospective observational design and the possibility of residual confounding, the results should be interpreted as evidence of a robust association rather than definitive proof of causality.