Examining Disparities in Breast Cancer Screening Rates in a Program Designed to Provide Free Care to Women in the State of Pennsylvania

Background. Female breast cancer is the most diagnosed cancer in women and the second leading cause of cancer death in women (behind lung cancer) (U.S. Cancer Statistics Working Group, 2013b). The purpose of this research was to examine client and provider level characteristics driving the disparities in morbidity and mortality as it relates to breast cancer screening, thus providing a clear indication of what client and provider factors are essential for successfully reducing disparities in breast cancer screening beyond removing barriers of cost. Specific aims include: (1) To characterize the client population utilizing the services of the Healthy Woman Program in Pennsylvania; (2) To Identify client and provider predictive factors that influence women with an initial screening mammogram to return for rescreening within 24 months; and (3) To use a local case study to explore service delivery factors influencing woman with an initial screening mammogram to return for rescreening. Methods. This study was a retrospective analysis of prospective data from the Healthy Woman Program (HWP), investigating women enrolled in the HWP in Pennsylvania and their rescreening rates. As the HWP dataset primarily contained client-level variables, additional structural provider-level variables were derived from Komen Foundation data, and data from the United States Department of Agriculture (USDA) and the Census, to extend the investigation beyond the individual woman to specified contextual effects. A local qualitative case study augmented the quantitative analysis to explore more process-oriented contextual effects, with an emphasis on how services are organized to support client outreach and retention, and promising practices for getting clients to return for rescreening. Results. The major predictors of an abnormal mammogram were having a diagnostic mammogram rather than a screening mammogram, being younger than 40, and living in an urban area. The major predictors of a breast cancer diagnosis were being white, being younger than 40, living in a large rural town, and being a current smoker. The major predictors of dropping out after an abnormal mammogram were being white, and having a diagnostic mammogram rather than a screening mammogram. The major predictors of delay in diagnostic work up were Black race and screening mammogram. The major predictors of returning for screening were living in a large rural town, being a non-smoker, having no symptoms prior to the initial mammogram, having a normal result from the initial mammogram, and living with others. Comprehensive breast cancer education, delivered in community settings, and tailored to cultural, social, and literacy issues, results in a robust referral network. More reminders than mandated by HWP Guidance, through multiple channels, rather than the client alone, increase returns for annual rescreening. Conclusions. Prior investigation of rescreening rates among low-income women has focused on client-level factors, yielding a narrow range of targets for systems change. This study investigated both client and provider level characteristics, analyzing them together and separately, to increase understanding of the multifactorial predictors of rescreening and widen the set of potential targets for change. Instituting a protocol of allowing community groups to organize a few women to have screenings in one day and allowing for interpreters who are family or community members would improve rescreening rates. Methods for ensuring patients return should be modified as the HWP sequence may not be enough.Dr.P.H., Public Health -- Drexel University, 201