Participation of under-represented communities in an online cognitive ageing platform and predictors of willingness to be contacted for future research.

Background: Black and Hispanic communities are under-represented in cognitive ageing research. MindCrowd is an online platform that offers interactive cognitive games and facilitates research. We sought to assess participation in MindCrowd and willingness to be contacted for future studies by race, ethnicity and other demographic variables.

Methods: Adults were recruited into MindCrowd through national media campaigns and enhanced local engagement (ELE) in and around four cities with robust black (Atlanta and Baltimore) and Hispanic (Miami and Tucson) populations. Recruiting in ELE regions involved direct contact with potential participants via community forums or established research cohorts. Participation was defined as completing two 5-minute cognitive games and a demographic questionnaire, and participation incidence was calculated per 100 000 adults using census data and was compared by region using incidence rate ratios (IRRs). Willingness to be contacted for future research was defined as participants submitting an email address for researcher follow-up and was modelled as a binary outcome using logistic regression to generate adjusted ORs (aORs) for age, sex, region, race and ethnicity.

Results: From 17 June 2022 to 28 December 2023, 49 934 adults participated in MindCrowd. Nationwide, the majority were female (87.2%) and >50 years old (75.2%). In ELE regions, 8.4% of participants were black versus 2.3% in non-ELE regions; 21.8% were Hispanic in ELE regions versus 16.4% in non-ELE regions. MindCrowd participation incidence was higher in ELE than non-ELE regions (32.4 vs 17.9 per 100 000 adults; IRR=1.78, 95% CI, 1.73 to 1.83). Overall, 42.1% of participants nationwide were willing to be contacted for future research, with elevated odds for all ELE regions (vs collective non-ELE regions). Compared with white participants, black participants had 23% higher odds of willingness to be contacted (aOR=1.23; 95% CI, 1.10 to 1.36); compared with non-Hispanic participants, Hispanic participants had 20% higher odds (aOR=1.20; 95% CI, 1.13 to 1.27).

Conclusions: Following nationwide and regional efforts to recruit a diverse participant pool into the online platform MindCrowd, regions with ELE had higher rates of participation than the rest of the nation across race and ethnicity categories. Controlling for region, black and Hispanic individuals showed marginal but statistically elevated willingness to be contacted for future research.