- Miller, Melanie;
- Diaz, Adam;
- Conti, Catherine;
- Albala, Bruce;
- Flenniken, Derek;
- Fockler, Juliet;
- Kwang, Winnie;
- Sacrey, Diana;
- Ashford, Miriam;
- Skirrow, Caroline;
- Weston, Jack;
- Fristed, Emil;
- Farias, Sarah;
- Korecka, Magda;
- Wan, Yang;
- Aisen, Paul;
- Beckett, Laurel;
- Harvey, Danielle;
- Lee, Edward;
- Petersen, Ronald;
- Shaw, Leslie;
- Okonkwo, Ozioma;
- Mindt, Monica;
- Weiner, Michael;
- Nosheny, Rachel
INTRODUCTION: We evaluated preliminary feasibility of a digital, culturally-informed approach to recruit and screen participants for the Alzheimers Disease Neuroimaging Initiative (ADNI4). METHODS: Participants were recruited using digital advertising and completed digital surveys (e.g., demographics, medical exclusion criteria, 12-item Everyday Cognition Scale [ECog-12]), Novoic Storyteller speech-based cognitive test). Completion rates and assessment performance were compared between underrepresented populations (URPs: individuals from ethnoculturally minoritized or low education backgrounds) and non-URPs. RESULTS: Of 3099 participants who provided contact information, 654 enrolled in the cohort, and 595 completed at least one assessment. Two hundred forty-seven participants were from URPs. Of those enrolled, 465 met ADNI4 inclusion criteria and 237 evidenced possible cognitive impairment from ECog-12 or Storyteller performance. URPs had lower ECog and Storyteller completion rates. Scores varied by ethnocultural group and educational level. DISCUSSION: Preliminary results demonstrate digital recruitment and screening assessment of an older diverse cohort, including those with possible cognitive impairment, are feasible. Improving engagement and achieving educational diversity are key challenges. HIGHLIGHTS: A total of 654 participants enrolled in a digital cohort to facilitate ADNI4 recruitment. Culturally-informed digital ads aided enrollment of underrepresented populations. From those enrolled, 42% were from underrepresented ethnocultural and educational groups. Digital screening tools indicate > 50% of participants likely cognitively impaired. Completion rates and assessment performance vary by ethnocultural group and education.