DESCRIPTION (provided by applicant): For breast cancer (BC), the most common cancer of US women, the 5-year relative survival after diagnosis is now nearing 90%, resulting in an estimated 2.3 million US BC survivors in 2005. Yet, because BC is the 2nd leading cause of cancer death, and survivors have increased risks of recurrence, second cancers, adverse outcomes, and premature death, it is critical to identify modifiable factors that can reduce the risk for these conditions and improve survival, particularly in racial/ethnic minority populations who have less favorable outcomes. Some epidemiologic evidence suggests that body size and physical activity may be such modifiable factors and, moreover, that they are associated with community "built environment" characteristics such as the number of walkable destinations and the availability of recreation facilities and supermarkets. Given the limited success of public health efforts to reduce overweight/obesity and increase physical activity, two prominent goals of the NIH, it is important to consider alternative influences on physical activity behavior. This goal is particularly important for BC survivors, who are at greater risk of weight gain and decreased physical activity. To date, no studies have directly assessed the relationship between the built environment and BC outcomes; therefore, this study aims to consider the combined effects of physical activity, body size, and a comprehensive set of measures of the built environment on overall survival, BC-specific survival and second primary cancer after BC diagnosis. In doing so, this study will assess the independent associations between each of these variables and outcomes, and the relationships of these variables to each other. Utilizing Geographic Information Systems technology, this study efficiently combines data that objectively characterizes the built environments of the Greater San Francisco Bay Area of northern California, data on cancer diagnosis and follow-up from the population-based Greater Bay Area Cancer Registry, and interview data on individual level risk factors from over 4,000 multiethnic (African-American, Hispanic, Asian-American and non-Hispanic white) BC patients who participated in two population-based studies. Regression models will be used to assess the association between variables of interest and multivariate adjusted Cox proportional hazards modeling will be used in the survival analyses. PUBLIC HEALTH RELEVANCE: This study has the potential to identify how community environments may be changed to improve physical activity that, in turn, may improve outcomes in BC survivors. If the built environment proves to influence behaviors and outcomes, then this knowledge can be used by urban and public health planners to advocate changes to the built environment, and by health care professionals to design post-diagnosis treatment programs and/or recommendations for taking greater advantage of the existing built environment to improve health behaviors among BC survivors.

This study has the potential to identify how community environments may be changed to improve physical activity that, in turn, may improve outcomes in BC survivors. If the built environment proves to influence behaviors and outcomes, then this knowledge can be used by urban and public health planners to advocate changes to the built environment, and by health care professionals to design post-diagnosis treatment programs and/or recommendations for taking greater advantage of the existing built environment to improve health behaviors among BC survivors.