Supplement to Effectiveness of Digital Versus In-Person Diabetes Prevention Programs

项目摘要 我们寻求预算补充，以支持我们正在进行的随机对照试验（RCT）的完成， 在ClinicalTrials.gov上注册为NCT 05056376，由国家糖尿病和消化研究所资助 和肾脏疾病补助金R 01 DK 125780。随机对照试验研究了全自动数字 糖尿病预防计划（dDPP），Sweetch Health Ltd.，利用人工智能技术适应生活方式 指导，以及基于人类指导的糖尿病预防计划（hDPP）的护理标准。尽管 糖尿病前期对发展2型糖尿病构成重大风险，数百万美国成年人缺乏足够的 生活方式咨询，以减轻这种风险。像Sweetch这样的移动的健康（mHealth）技术提供了一种可扩展的 解决这一普遍问题。 该项目的目标是确定人工智能驱动的dDPP是否与传统的hDPP一样有效， 改善糖尿病前期成人的健康状况。这一人群特别容易受到影响， 从干预中受益匪浅。 我们的随机对照试验解决了慢性病预防和健康行为改变方面的证据缺口， 通过初步试验的短期结果本研究将完成最终随机化 到2023年底，研究参与者将招募368名年龄在18-75岁之间的超重/肥胖的糖尿病前期成年人。 参与者平均分为两组：第一组（N=184）接受全自动Sweetch数字健康检查。 试剂盒（dDPP组），而组2（N=184）通过亲自或远程参与当地CDC认可的hDPP 学习 在基线和1个月间隔时，使用体动仪对两组进行严格的身体活动测量 在整个研究中。我们假设dDPP在实现CDC方面不劣于hDPP。 12个月时2型糖尿病风险降低终点。我们还预计，dDPP将获得更高的 由于其灵活性和方便性，具有良好的参与性和可接受性。 该补充是必要的意外费用相关的活动，增加医疗保险报销， 2019冠状病毒病疫情引发的hDPP和家庭研究访问的差旅费。完成本 研究将阐明使用人工智能进行全自动数字干预的潜力，以提供有效，可扩展， 可持续和具有成本效益的促进健康的行为改变干预措施。它的成功可以深刻地 影响糖尿病预防的可扩展性。

PROJECT SUMMARY We seek a budget supplement to bolster the completion of our ongoing randomized controlled trial (RCT), registered as NCT05056376 on ClinicalTrials.gov, funded by the National Institute of Diabetes and Digestive and Kidney Diseases grant R01DK125780. The RCT studies the comparison between a fully automated digital diabetes prevention program (dDPP), Sweetch Health Ltd., which leverages AI technology for adaptive lifestyle coaching, and the standard of care human coach-based diabetes prevention programs (hDPPs). Despite prediabetes posing significant risks for developing type 2 diabetes, millions of U.S. adults lack adequate lifestyle counseling to mitigate this risk. Mobile health (mHealth) technologies, like Sweetch, provide a scalable solution to this widespread issue. This project’s objective is to determine whether the AI-driven dDPP is as effective as traditional hDPPs in improving health outcomes for prediabetic adults. This population is particularly susceptible and stands to benefit greatly from interventions. Our RCT addresses an evidence gap in chronic disease prevention and health behavior change, underpinned by promising short-term results from our preliminary trial. The study, set to complete randomization of final study participants by the end of 2023, will recruit 368 prediabetic adults aged 18-75 who are overweight/obese. Participants are split evenly into two arms: Arm 1 (N=184) receives the fully automated Sweetch digital health kit (dDPP arm) while Arm 2 (N=184) engages in a local CDC-recognized hDPP through in-person or distance learning. Both arms undergo rigorous physical activity measurements using actigraphy at baseline and 1-month intervals throughout the study. We hypothesize that the dDPP will be non-inferior to the hDPP for achieving the CDC’s type 2 diabetes risk reduction endpoint at 12 months. We also anticipate that the dDPP will garner higher engagement and acceptability due to its flexibility and convenience. The supplement is required for unexpected costs related to actigraphy, increased Medicare reimbursement for the hDPP, and travel expenses for home study visits, prompted by the COVID-19 pandemic. Completing this study will clarify the potential of fully automated digital interventions using AI for delivering effective, scalable, sustainable, and cost-effective health-promoting behavioral change interventions. Its success can profoundly impact scalability in diabetes prevention.