Examining virtual research recruitment and participant diversity in a multi-center birth cohort, Childhood Allergy and the NeOnatal Environment" (CANOE)
Afshan T, Kulkarni A, Smith J, Blackshere T, Tesson E, Hartert T, Rivera-Spoljaric K, Zoratti E, Joseph C, Gangnon R, Gern J, and Singh AM. Examining virtual research recruitment and participant diversity in a multi-center birth cohort, Childhood Allergy and the NeOnatal Environment" (CANOE). J Allergy Clin Immunol 2023; 151(2):AB183.
J Allergy Clin Immunol
Rationale: Recruitment for a NIH/ECHO-supported multi-center birth cohort, “Childhood Allergy and the NeOnatal Environment” (CANOE) stopped due to the COVID-19 pandemic. Redesign of study procedures emphasized virtual and socially distanced activities. We hypothesized that “virtual” recruitment methods (social media, websites, email) would surpass “traditional” methods (in-clinic, telephone, flyers/print materials) and increase enrollment of families from diverse backgrounds and communities.
Methods: Pregnant women (n=439, target 500) were recruited from four academic medical centers in Detroit MI, Madison WI, Nashville TN, and St. Louis MO. We collected demographic and social information by questionnaires and examined race, ethnicity, age, parity, and employment status in relation to recruitment method using chi-square tests.
Results: In-clinic and telephone recruitment comprised 55% of enrollment, followed by print materials (17%), and social media and email (15%). The cohort includes families self-identifying as Caucasian/White (63%), African American/Black (27%), Hispanic/Latino (3.3%), Asian (3.5%), and mixed races (1.2%). This reflects site demographics for White and Black patients, while other populations are not as well recruited into this cohort. Recruitment method success did not vary by race, ethnicity, maternal age, or employment status (p=ns for each comparison). Most (63%) multigravida mothers (9.1% of participants) were recruited in clinic, while primigravida participants were recruited more evenly via all methods.
Conclusions: “Virtual” recruitment methods comprised a smaller proportion of cohort enrollment than hypothesized and study recruitment method did not vary by race/ethnicity; however, consideration of combined, varied, and novel recruitment methods may add to the development of best practices for more representative research study recruitment.