AI Voice Agent Improves Blood Pressure Monitoring Accuracy in Older Adults

Artificial intelligence voice agents helped older adults with high blood pressure accurately report their blood pressure readings and improved blood pressure management, according to preliminary research presented at the American Heart Association's Hypertension Scientific Sessions 2025. The study involved 2,000 adults, primarily ages 65 and older, and evaluated the effectiveness of voice-enabled AI agents in engaging patients to self-report accurate blood pressure readings instead of traditional phone calls with healthcare professionals.

The AI voice agent calls were made using commercially available AI in multiple languages, including English and Spanish. When blood pressure readings fell outside threshold ranges or patients reported symptoms such as dizziness, blurred vision, or chest pain, calls were immediately escalated to licensed nurses or medical assistants. This integration of AI into clinical workflows resulted in an 88.7% lower cost-per-reading compared to using human nurses for similar tasks, significantly reducing manual workload for clinicians.

During the study period, 85% of patients were successfully reached by the voice-based AI agent, with 67% completing the call and 60% taking compliant blood pressure readings during the call. Among these patients, 68% met controlling blood pressure Stars compliance thresholds. The Medicare Advantage and Healthcare Effectiveness Data and Information Set controlling blood pressure measure increased from a previously reported 1-star rating to a 4-star rating, representing a 17% improvement. Overall, 1,939 controlling blood pressure gaps were closed.

Patient satisfaction was notably high, with average satisfaction rates exceeding 9 out of 10 on a scale where 10 represented 100% satisfaction. Lead study author Tina-Ann Kerr Thompson, M.D., expressed surprise at the high patient satisfaction scores, noting the critical importance of patient engagement and satisfaction for healthcare outcomes. The study's findings demonstrate how integrating AI into care can improve home blood pressure monitoring and completion rates, leading to enhanced quality outcomes for patients.

Eugene Yang, M.D., M.S., FACC, an American Heart Association volunteer expert who was not involved in the study, described the research as potentially game-changing. He emphasized that accurate blood pressure readings are essential for improving control, and breakthrough AI technologies like this could transform blood pressure management by reaching patients wherever they are and addressing critical barriers such as limited access to care. The American Heart Association's Target:BP initiative, which helps healthcare organizations improve blood pressure control rates through evidence-based programs, emphasizes the importance of self-measured blood pressure monitoring for all adults with hypertension.

The study was conducted with patients at Emory Healthcare in Atlanta during a 10-week period. Participants included 2,000 adults with an average age of 72 years, 61% of whom were women, all receiving care for high blood pressure. Electronic health records identified patients missing blood pressure data or whose most recent readings were not within the normal range of <120/80 mm Hg. These patients were tagged to receive calls from the AI voice agent, with multiple contact attempts made if patients did not answer initially.

While the study shows promising results, it has several limitations. The research was observational without a control group, and AI calls were not directly compared to human calls. The retrospective nature of the study meant evaluation occurred after clinically identified calls were already made. The findings are considered preliminary until published as a full manuscript in a peer-reviewed scientific journal, as abstracts presented at American Heart Association scientific meetings are not peer-reviewed.