Abstract A046: Collecting nonclinical data to address disparities in cancer prevention: Lessons from the field

Abstract

Abstract Background and Purpose: Research on how women at high-risk of breast or ovarian cancer make risk-reduction choices is still developing, and rarely addresses minority and underserved populations. Early findings indicate that African American women face additional burdens at every stage of the decision-making process. Because most high-risk African American women do not receive clinical care related to risk, further research requires recruitment methods that do not rely on the clinical populations commonly used in risk-reduction studies. Procedures: The Daughter, Sister, Mother Project has recruited high-risk African American and White women from non-clinical environments to collect both qualitative (semi-structured interview) and quantitative (survey) data. Non-clinical recruitment methods include contacting women through social media, online volunteer databases, and community organizations. They present unique challenges for which we have developed specific solutions. Results: Our experience recruiting non-clinical but high-risk populations has yielded solutions that may be useful to other researchers. (1) Risk prediction modeling. Because the individual risk level of women recruited through non-clinical methods is usually unknown, risk prediction modeling must be built into the data collection process. Telephone screening by trained research staff allows risk prediction modeling before study enrollment, and replaces the risk-level information that would pre-exist participant recruitment in a clinical setting. Collecting risk-related information within a survey instrument makes it possible for participants to complete a survey in one interaction, but requires risk-prediction modeling and sample trimming after the data have been collected. (2) Hacking and fraudulent accounts. Combining online data collection with social media recruitment facilitates the involvement of participants not commonly drawn into biomedical research, but also exposes studies to hackers and fraudulent participants. To avoid (a) losing incentive funds to hackers and (b) incorporating fraudulent information into study datasets, we have developed methods to distinguish “real” from “fake” participants. These include survey programming methods that detect or weed out fraudulent accounts, semi-automated methods to locate impossible data combinations, and direct phone contact with participants after data collection is complete. (3) Ongoing connections. Some African American women not already involved in high-risk clinical care are also hesitant to become involved in health-oriented research. Trust can be built through connections with community organizations, fostering ongoing two-way contact with our research team, and returning useful information to the communities where we learn it. Conclusion: Recruitment of non-clinical populations is a useful tool for cancer health disparities research, and requires creative recruitment, data collection, and data cleaning methods. Citation Format: Tasleem J Padamsee. Collecting nonclinical data to address disparities in cancer prevention: Lessons from the field [abstract]. In: Proceedings of the Twelfth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2019 Sep 20-23; San Francisco, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl_2):Abstract nr A046.