Post-diagnostic smoking cessation and heart-disease mortality among adults smoking at cardiopulmonary disease diagnosis: An NHANES linked-mortality analysis.

Abstract

Introduction: Population-based analyses usually classify smoking status at survey enrollment, which combines people who quit before disease recognition with those who continued smoking through diagnosis and quit afterward. This study evaluated whether smoking cessation after a self-reported cardiopulmonary disease diagnosis was associated with subsequent all-cause and cause-specific mortality among adults reconstructed as smoking at diagnosis.

Methods: This secondary analysis pooled ten cross-sectional National Health and Nutrition Examination Survey (NHANES) cycles from 1999-2000 through 2017-2018 and linked survey records to mortality follow-up through 2019. Adults aged ≥40 years with self-reported cardiovascular or chronic lung disease were included when self-reported smoking initiation, diagnosis, cessation, and current-smoking information permitted classification as smoking at diagnosis. Survey-weighted Cox models compared post-diagnostic quitters with persistent smokers. Models were unadjusted, demographic-adjusted, and fully adjusted for prespecified demographic, disease-history, smoking-history, and survey-cycle covariates.

Results: Among 2319 participants, 975 were post-diagnostic quitters, and 1344 were persistent smokers. During a median follow-up of 6.33 years, 999 all-cause, 272 heart-disease, and 244 cancer deaths occurred. In the fully adjusted model, post-diagnostic quitting was associated with lower heart-disease mortality (adjusted hazard ratio, AHR=0.59; 95% CI: 0.42-0.82; p=0.001). Associations with all-cause mortality (AHR=0.90; 95% CI: 0.74-1.09; p=0.276) and cancer mortality (AHR=0.72; 95% CI: 0.49-1.06; p=0.100) were not statistically significant. Findings for heart-disease mortality were consistent across landmark, overlap-weighted, smoking-burden-adjusted, disease-restricted, quit-age reconstruction, and leave-one-covariate-out analyses.

Conclusions: Among adults retrospectively classified as smoking when cardiopulmonary disease was diagnosed, post-diagnostic cessation was associated with lower subsequent heart-disease mortality. The findings are observational and remain susceptible to survivor selection, reverse causation, recall and social-desirability bias, exposure changes after enrollment, mortality misclassification, and residual confounding.