Statistical Collaboration for Competing Risk Analysis: Smoking Cessation and the Risk of Second Primary Lung Cancer Among Lung Cancer Survivors

• Activity Number: 286 - Missing Data Methods
• Type: Contributed
• Date/Time: Wednesday, August 11, 2021 : 1:30 PM to 3:20 PM
• Sponsor: Section on Statistical Consulting
• Author(s): Sophia Luo* and Eunji Choi and Summer S. Han
• Companies: Stanford University School of Medicine and Stanford University School of Medicine and Stanford University School of Medicine
• Keywords: competing risk analysis; lung cancer; time-to-event; survival analysis; cumulative incidence; right-censored data

Abstract

In biomedical research, study outcomes are often time-to-event data under competing risks. For example, one may be interested in cancer recurrence as the study outcome, but this outcome may be precluded by competing events, such as death. In estimating cumulative incidence, taking into account competing risks is important but often neglected. In this study, we compare the cumulative incidence estimates of second primary lung cancer (SPLC) among lung cancer survivors in the Multiethnic Cohort Study using the Aalen-Johansen estimator that takes into account competing risks versus the naïve Kaplan-Meier estimator. We apply a cause-specific proportional hazards model and a Fine-Gray subdistribution hazard model to evaluate the association between SPLC risk and the effect of smoking cessation following initial primary lung cancer (IPLC) diagnosis. We find that competing risk analysis using the Aalen-Johansen estimator gives a more accurate estimation of the cumulative incidence of SPLC compared to the Kaplan-Meier method that provided an overestimated risk, and that smoking cessation after IPLC diagnosis is significantly associated with reduced SPLC risk among lung cancer survivors.