Breast cancer remains the second leading cause of cancer morbidity and mortality among women in the US. New discoveries have resulted in the widely accepted view that breast cancer is a heterogeneous disease with molecularly distinguishable morphological subtypes. This awareness is driving the development of new paradigms for the prevention, early detection and clinical management of breast cancer. However, there are very limited data on the population effects of these novel cancer control approaches. Population modeling is a unique comparative effectiveness paradigm to fill this gap by translating advances from the laboratory and clinical trials to understanding their net effects on US breast cancer mortality. The CISNET Breast Working Group has collaborated over the past nine years to apply independent population models to evaluate cancer control practices and use results to inform clinical and public health guidelines. This proposal leverages the investment in these models and provides the continuity and cohesion of this highly productive group. The modeling groups include Dana Farber (D). Erasmus MC (E), Georgetown-Einstein (G), MD Anderson (M), Stanford (S) and Wisconsin-Harvard (W). For this application, we will extend our work by modeling populations of women with varying risk factors (e.g., breast density, HRT) for the development of specific molecular subtypes of breast cancer (based on ER and HER2). Our specific aims are to use these adapted models to: 1) compare the impact of observed practice patterns to the benefits and harms of targeting new screening and adjuvant therapy modalities based on risk factors and molecular subtypes; 2) explore the impact of improving access to new services; 3) conduct value-of information-like analyses to evaluate the relationship between performance characteristics of a new screening test (e.g. blood based biomarker) and its impact on breast cancer mortality, utilization of treatments and over-diagnosis; and 4) communicate results to end-users using a web-based platform. This work will advance the field of modeling by explicitly capturing molecular attributes of breast cancer, and in so doing, build a robust capacity to inform debates about "best practices" for cancer control interventions.