AI-Enabled Tailored Messaging Around Cardiovascular Health Increases Engagement (Preprint)

Abstract

<sec> <title>BACKGROUND</title> Cardiovascular (CV) disease risk factors disproportionately affect underserved populations, emphasizing the need for innovative self-management support. This pilot randomized trial evaluated AI-enabled text messaging interventions based on the American Heart Association's Life’s Essential 8 (LE8) framework across three Federally Qualified Health Systems. </sec> <sec> <title>OBJECTIVE</title> To assess participant engagement with AI-enabled text messaging based on a variety of factors including language, timing of messages, and message topic. </sec> <sec> <title>METHODS</title> Participants were randomized into three groups: generic messages, AI chatbot-tailored messages with behavioral nudges, and AI chatbot messages supplemented by pharmacist engagement. </sec> <sec> <title>RESULTS</title> Preliminary findings showed higher engagement with lifestyle topics (diet, physical activity, weight management) compared to health-related content (blood pressure, blood glucose, cholesterol and medicine), with no significant differences between English- and Spanish-speaking participants or across patients with different CV diseases. A significant difference in the interaction, measured as the number of incoming text message patient responses was observed between weekdays and weekends (U = 243, p &lt; 0.01). The AI chatbot (without a pharmacist) arm demonstrated significantly higher response rates than the generic message group (p=0.0102), highlighting the potential of conversational AI to enhance patient interaction. </sec> <sec> <title>CONCLUSIONS</title> These results underscore the promise of AI-driven messaging systems in addressing CV disease risk factors in underserved populations and suggest that future research should focus on personalization, and optimizing message delivery timing. </sec> <sec> <title>CLINICALTRIAL</title> ClinicalTrials.gov ID: NCT06324981 </sec>

Ähnliche Arbeiten