Participants self-reported demographic data and health history including prior history of SARS CoV-2 infection at enrollment. Daily electronic surveys developed and administered by Oracle Corporation (Redwood, CA, USA) solicited symptoms of COVID-19-like illness, infection with or exposure to SARS- CoV-2, masking habits, and any receipt of COVID-19 vaccines. Participants consenting to serological surveillance were mailed at least 4 serology kits for at-home testing of finger-prick capillary blood; written instructions were provided in English and Spanish.

We performed a descriptive analysis of the racial and ethnic distribution of children enrolled in our CCRP study. We used chi-square tests to compare the racial and ethnic distribution of enrolled children to that of all children receiving care at the site during the same timeframe, as well as among participants who completed serological testing to those who consented to testing. We also calculated propensity for completing an at-home test versus consenting and not completing any test based on race and ethnicity via logistic regression. Results were reported as adjusted odds ratios (aOR) and 95% confidence intervals (CI). All statistical tests were two-tailed, with p < 0.05 considered statistically significant. Analyses were performed using SPSS v. 27.

Results

A total of 1,501 children were enrolled during the recruitment period. Most children were older than 5 years, 51.2% were female, and 88.3% were from mostly urban counties (Table 1). Of enrolled children, 6.9% were Hispanic, 8.5% were Black, and 80.4% were White (Figure 1A). Among 190,009 unique children who received care at the institution during the same time, 12.5% were Hispanic, 23.0% were Black, and 63.0% were White (Figure 1A). Of 1,412 participants who consented to serologic testing, 7.1% were Hispanic, 8.6% were Black, and 80.0% were White (Figure 1B). Only745 of the 1,412 participants consenting to testing actually completed a serology test; of these, 6.6% were Hispanic, 7.9% were Black, and 82.0% were White (Figure 1B). The aOR of Black and Hispanic children who consented and completed testing compared to White children who consented and completed testing was 0.79 (95% CI 0.66 – 1.03) and 0.91 (95% CI 0.73 – 1.05), respectively.

Table 1: Characteristics of enrolled children in COVID-19 Community Research Partnership study.

Figure 1: Racial and ethnic distribution of children enrolled in the study compared to that of target population (A) and racial and ethnic distribution of children who consented to study compared those to completed serology testing (B). In Panel A, all proportional differences are statistically significant at p < 0.05 except for Asian race group (p = 0.50). In Panel B, none of the proportional differences are statistically significant.

Discussion

For our CCRP study, we aimed to recruit a diverse group of children by disseminating both English and Spanish advertisements, emails, flyers, and videos. We performed in-person recruitment activities at local ethnic grocery stores, churches and a large academic pediatric clinic serving a high proportion of minority patients. Despite these efforts, the proportion of minority children enrolled in the study underrepresents the targeted population, thus affecting the generalizability of any findings from the syndromic and serological surveillance. The percentage of White and Non- Hispanic children enrolled was higher than that of the targeted population, whereas the percentage of Black and Hispanic children enrolled was lower than that of the targeted population. However, among children who enrolled in the study and consented to serologic testing, less racial and ethnic disparity was noted among those who completed at-home testing. Our findings indicate that once enrolled, participants of all races and ethnicities are likely to complete study procedures. Therefore, diverse representation in research may be achieved by strengthening efforts to recruit underrepresented minorities early in study planning.

A focus group of clinical trial recruiters who work with minority populations found that taking time to build relationships with potential participants and offering flexibility to facilitate their enrollment are vital for recruitment and retention [9]. Several descriptive studies have also highlighted the importance of partnering with diverse community stakeholders in the recruitment process to engage minority populations in research [10]. A more recent study compared demographics of participants enrolled in COVID-19 clinical trials by in-person recruitment versus those enrolled by online recruitment, including social media advertising; online advertising with inclusive social media recruitment led to increase in racial, ethnic, and geographic diversity among study participants [11].

Efforts to improve diverse representation in clinical research should focus on recruitment early in the study period. Future comparative studies are required to identify optimal methods to enhance pediatric minority participation in research of disease processes like COVID-19, which disproportionally affect Black and Hispanic individuals.