A recent study by researchers at Johns Hopkins Medicine reveals that an AI-driven app for prediabetes effectively reduces diabetes risk, comparable to traditional human-led programs. Funded by the National Institutes of Health and published in JAMA, the study is the first phase III randomized controlled trial proving an AI Diabetes Prevention Program (DPP) can meet CDC benchmarks for diabetes risk reduction. With approximately 97.6 million U.S. adults facing prediabetes, access to human-led interventions often encounters logistical barriers.
The trial engaged 368 participants, comparing AI-driven DPP with human-led programs over 12 months. Notably, engagement was higher in the AI group, with 93.4% initiating the program compared to 82.7% in traditional setups. Both programs achieved similar outcomes, highlighting AI’s potential to overcome accessibility issues in lifestyle interventions. As AI DPPs can be automated and more accessible, they may serve as a viable alternative for patients facing challenges with conventional programs, suggesting a paradigm shift in diabetes risk management.
Source link
