Modelling Community Disparity in Breast Cancer Screening

点击翻译按钮获取中文摘要

DESCRIPTION (Provided by Applicant): Although breast cancer screening has been proven to reduce the risk of mortality, low-income, disenfranchised communities have lower screening utilization. Much of the work examining these disparities has focused on understanding the association of individual and social characteristics with screening behavior (e.g., knowledge, social support). However, characteristics of the community in which women reside have been ignored despite the fact that such characteristics influence health and related behaviors. In order to address community disparities, it is critical to press beyond the current boundaries of our perspectives and develop a new, multilevel model. Multilevel approaches include characteristics of individuals as well as broader community level factors on individual-level outcomes. This new paradigm has the potential to increase our understanding of disparities in breast cancer screening among communities, point to new areas for interventions to increase screening, and thus have the potential to reduce the risk of mortality. Despite its great potential, there are currently no studies that have developed a multilevel model aimed at examining community variation of breast cancer screening. The proposed project will construct a conceptual model by testing the different pathways (medical infrastructure, general community infrastructure, social capital) by which women age 40 or older from metropolitan areas with adverse material and social conditions are less likely to be screened for breast cancer after controlling for individual-level barriers. Examination of these three potential pathways will lead to the development of more specific hypotheses about the mechanisms by which area adverse material and social conditions influence screening for breast cancer. We will conduct analyses of existing public use data (1999-2000 National Health Interview Survey) of more 10,000 women from 58 metropolitan areas throughout the United States, merged with community data from various existing sources to construct this model (e.g. US Census, Area Resource File). By focusing on specific pathways of the constructed multilevel model, the findings of this novel study will serve as the basis for future cancer control research grant applications. Such applications will focus on the evaluation of specific hypotheses using population-based surveys and in-depth data collection methods not available by using existing data sources.