Abstract
Appropriate and diverse representation of minority populations in clinical research studies improves long-term outcomes for all demographics and requires targeted solutions to recruitment challenges. During a two year project at Boston Medical Center (BMC), 191 newly diagnosed, treatment-naïve cancer patients and 76 non-cancer control donors were consented to study relationships between cancer and thrombosis biomarkers. 56% of the cancer patients self-identified as minorities (32% Black, 17% Hispanic, 3% Asian, 4% other or multiple), which reflect the overall demographics of patients undergoing surgery at BMC. However, control donor recruitment for the study was only 36% minorities during the first 6 months (n=9 of 25). These control subjects were younger and white, in part because the study was only advertised on the Boston University Medical Campus. Recognizing potential bias and lack of appropriate representation, two initiatives were developed and executed to balance the cancer and control populations with respect to ethnicity and age. First, active recruitment for older, minority-identifying control subjects expanded into the surrounding community of largely African American and Hispanic populations, by hosting donor recruitment days, with one of the days targeted to older males (≥ 50 years) to better reflect the ongoing cancer patient cohort. Second, consent forms were translated into Haitian Creole and Spanish to enable improved communications with these patients who frequent the hospital. To encourage participation from individuals with a negative perception of research, study coordinators received training in consenting and applied it to provide enhanced patient education. This resulted in a significant improvement in the control’s minority percentage to 64% over the project timeline (n=49, 44% Black, 11% Hispanic, 9% Asian). This approach also helped to balance age distributions in the cancer and control populations. The majority of cancer patients were in their 50s and older (81%, mean 59 ± 12 years; n=155). Control subjects in their fifth decade or above was only 32% in the first 6 months, but improved to 55% (mean 46+ 14 years, n=42) by the end of the project. The study was not designed to match comorbidities, but human immunodeficiency virus (HIV) infection, and hepatitis which are known to affect coagulation are highly prevalent in the BMC population. Of the consented cancer patients, 5% had hepatitis and 2% reported active HIV infection, while in the control population 6% had hepatitis and 5% had HIV. Patient demographic, clinical data, analysis results, follow-up data on treatment and thrombotic events, as well as plasma and cell pellets from this study are freely available to the research community from: cssi.cancer.gov/cancer-thrombosis. Overall, frequent data monitoring and adjusting recruitment strategies to emphasize community outreach contributed to balancing ethnic and age distributions in the study populations during this two year study.
Citation Format: Morgan P Thompson, Elizabeth R Duffy, Jasmin H Bavarva, Cheryl Spencer, DJ Stearns-Kurosawa, Rachana Agarwal, Michelle A Berny-Lang, Chris Andry. Consenting challenges toward age, ethnicity, and co-morbidity matching of cancer patient and control populations at a minority-majority safety net hospital [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 A091.