Cancer is a leading cause of death. Cigarette smoking is the most important preventable cause of cancer-related death. The impact of smoking on cancer survivors in the post-diagnosis setting is not well studied. In this dissertation, the association between cigarette smoking after cancer diagnosis and risk of all-cause death was examined among male cancer patients of the Shanghai Cohort Study and female cancer patients of the Iowa Women's Health Study. Cox proportional hazard regression models and Kaplan-Meier method were used to compare mortality risk and survival in association with post-diagnosis smoking. Following the two cohort analyses, estimates of the proportion of death that is due to cancer, the total remaining life expectancy for patients who quit smoking and patients who continue to smoke after cancer diagnosis, as well as gains in life expectancy due to post-diagnosis smoking cessation were provided by performing a decision analysis. Findings from this dissertation work suggest that smoking cessation even after cancer diagnosis may reduce the risk of death and extend remaining life expectancy for both male and female cancer patients. The magnitude of the effect of post-diagnosis smoking cessation varies by cancer type, gender, age and stage at diagnosis. The encouragement of cancer patients to quit smoking during clinic visits at or after cancer diagnosis could be an effective strategy to improve the prognosis of cancer patients. Findings of this dissertation fill a gap in existing knowledge base regarding impact of smoking among cancer survivors and have important public health implications to patients, healthcare providers, policy makers, health insurance and pharmaceutical identities, as well as the general public.