Optimal screening and treatment strategies for breast cancer in older women are uncertain, particularly for those aged e75. Early detection with screening mammography reduces breast cancer mortality in women aged 50-69 and possibly up to age 74, but the benefits of screening are uncertain for women aged e75. The United States Preventive Service Task Force (USPSTF) has concluded there is insufficient evidence to recommend for or against breast cancer screening in women aged e75. Although some women aged e75 may benefit from breast cancer screening, due to exceptionally good health, many have underlying chronic illnesses and may not benefit, due to limited life expectancy resulting from competing mortality risks. The current statistical models used to inform the USPSTF recommendations do not account for competing mortality risks adequately. The statistical research proposed in this application addresses important methodological issues related to competing mortality risks for breast cancer. We will investigate the impact of competing risks on breast cancer mortality using a stochastic model of the natural history of breast cancer. We will integrate Fine and Gray's competing mortality risks directly into equations of the Cancer Intervention and Surveillance Network model developed by Zelen and Lee at Dana-Farber Cancer Institute. The CISNET-DFCI model is one of 7 CISNET breast cancer models used to inform USPSTF breast cancer screening guidelines and the only purely analytic model. We will then examine the impact of competing mortality risks on age-specific breast cancer mortality by comparing mortality estimates derived from the new (FG-CISNET-DFCI) model with the original model and a model using the less robust life table method to account for competing risks. We will then use the FG-CISNET-DFCI model to explore the benefits and harms of breast cancer screening under different scenarios (e.g., stopping ages, screening intervals). These analyses will offer insight into ongoing debate about breast cancer screening in older women by providing data on the mortality benefit and over-diagnosis in populations with different age and clinical (comorbidity) profiles. The new model will also provide a methodological prototype for integrating competing risks into other CISNET models. Our project will not only help inform recommendations for breast cancer screening in specific elderly groups, but also serve as a model to address the appropriateness of cancer screening other preventive health measures in older populations.