Search

One barrier to searching for novel mutations in African American families with breast cancer is the challenge of effectively recruiting families—affected and non-affected relatives—into genetic research studies. Using a community-based participatory research (CBPR) orientation, we incorporated several evidence-based approaches through an iterative fashion to recruit for a breast cancer genetic epidemiology study in African Americans. Our combined methods allowed us to successfully recruit 341 African American women (247 with breast cancer and 94 relatives without breast cancer) from 127 families. Twenty-nine percent of participants were recruited through National Witness Project (NWP) sites, 11 % came from in-person encounters by NWP members, 34 % from the Love Army of Women, 24 % from previous epidemiologic studies, and 2 % from a support group. In terms of demographics, our varied recruitment methods/sources yielded samples of African American women that differ in terms of several sociodemographic factors such as education, smoking, and BMI, as well as family size. To successfully recruit African American families into epidemiological research, investigators should include community members in the recruitment processes, be flexible in the adoption of multipronged, iterative methods, and provide clear communication strategies about the underlying benefit to potential participants. Our results enhance our understanding of potential benefits and challenges associated with various recruitment methods. We offer a template for the design of future studies and suggest that generalizability may be better achieved by using multipronged approaches.

The purpose of this study was to investigate the effectiveness of replicating an evidence-based model, the Witness Project, for increasing breast and cervical cancer screening with African American women in a variety of locations and organizations in the United States. The quantitative and qualitative methods included a cadre of process and outcome measures to evaluate the effectiveness of the four-phase replication process. The intervention was replicated in 25 sites with 401 volunteers, delivering cancer education and screening programs to over 10,000 women. Key components and criteria of successful replication were established, and preliminary screening outcomes demonstrated a 43.4% increase in mammography in women aged 40 and older. This study demonstrates that the Witness Project model can be effectively replicated, that the replication process can be standardized, and that the replication sites were able to obtain positive screening results comparable to the original intervention outcomes. The model was not able to be effectively replicated with just the "turnkey" toolbox approach, but required additional technical assistance.

Background. Few studies have investigated the contribution of inherited genetic variation to breast cancer disparities and fewer have examined whether risk alleles shared in families may contribute to higher risk of aggressive tumors. One barrier is the challenge of effectively recruiting African American families for research. Methods. Using a community-based participatory research (CBPR) orientation, we utilized multi-level methods in an iterative fashion to recruit for a breast cancer genetic linkage study. We recruited affected relative pairs as well as discordant sibling pairs where possible. We analyzed our recruitment yield by method and calculated descriptive statistics for several demographic and clinical characteristics of families selected for genotyping, which includes 281 individuals from 106 pedigrees (197 affecteds, 84 unaffecteds). Results. We recruited 341 African American women (247 with breast cancer and 94 unaffecteds) from 127 families. The top three yields were obtained via collaboration with the Love Army of Women (34% of sample), The National Witness Project (NWP, 29% of sample), and previous epidemiologic studies (24% of sample). Within the set of 106 pedigrees selected for genotyping, 27 pedigrees have one affected relative, 68 pedigrees have 2 affecteds, 10 pedigrees have 3 affecteds, and one pedigree has 6 affecteds. Where adjudicated (n=141, 72%), the mean and median age at diagnosis was 51.6 years (SD 12.0) and 51.6 years, respectively. Affected women were younger at menarche (p=0.002) and more likely to be hysterectomized (p=0.018) compared to unaffected women. According to tumor receptor status, approximately 26% of cases are ER negative, 40% are PR negative, 49% are HER2 negative, and overall 16% are ER/PR/HER2 negative. Conclusions. Approaching recruitment of African American pedigrees with use of a multi-pronged approach that includes collaboration from the community can greatly facilitate success. While our sample has a slightly younger age of onset compared to African American cases in the general population (SEER median age for African Americans is 58 yrs), the sample approximates the US proportion of all invasive breast cancers in the US with regard to triple negative status (∼15%). Citation Format: Youjin Wang, Deborah O. Erwin, Lina Jandorf, Detric Johnson, Veronica Meadows-Ray, Mattye Willis, Heather Ochs-Balcom. Recruitment of African American breast cancer pedigrees for The Jewels in Our Genes Study: Recruitment outcomes and sample characteristics. [abstract]. In: Proceedings of the Sixth AACR Conference: The Science of Cancer Health Disparities; Dec 6–9, 2013; Atlanta, GA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2014;23(11 Suppl):Abstract nr B12. doi:10.1158/1538-7755.DISP13-B12