1956-LB: Artificial Intelligence vs. Human Coaching for Diabetes Prevention—Results from a 12-Month, Multicenter, Pragmatic Randomized Controlled Trial

IF 7.5 1区医学 Q1 ENDOCRINOLOGY & METABOLISM

Diabetes Pub Date : 2025-06-20 DOI:10.2337/db25-1956-lb

NESTORAS N. MATHIOUDAKIS, MOHAMMED S. ABUSAMAAN, MARY E. ALDERFER, DEFNE ALVER, ADRIAN S. DOBS, BRIAN KANE, BENJAMIN LALANI, JOHN MCGREADY, KRISTIN RIEKERT, BENJAMIN RINGHAM, FATMATA VANDI, AMAL A. WANIGATUNGA, DANIEL ZADE, NISA M. MARUTHUR

{"title":"1956-LB: Artificial Intelligence vs. Human Coaching for Diabetes Prevention—Results from a 12-Month, Multicenter, Pragmatic Randomized Controlled Trial","authors":"NESTORAS N. MATHIOUDAKIS, MOHAMMED S. ABUSAMAAN, MARY E. ALDERFER, DEFNE ALVER, ADRIAN S. DOBS, BRIAN KANE, BENJAMIN LALANI, JOHN MCGREADY, KRISTIN RIEKERT, BENJAMIN RINGHAM, FATMATA VANDI, AMAL A. WANIGATUNGA, DANIEL ZADE, NISA M. MARUTHUR","doi":"10.2337/db25-1956-lb","DOIUrl":null,"url":null,"abstract":"Introduction and Objective: Prediabetes is highly prevalent, yet few patients receive evidence-based behavioral lifestyle support. Artificial intelligence (AI) may offer a scalable approach to diabetes prevention. This study evaluated whether a fully automated AI-based diabetes prevention program (ai-DPP), consisting of a mobile app and digital body weight scale, is non-inferior to a traditional human coach-based DPP (h-DPP) in adults with prediabetes and overweight or obesity. Methods: We conducted a two-site, pragmatic, RCT involving adults with prediabetes and overweight or obesity (NCT05056376). Participants were randomly assigned (1:1) to either an ai-DPP (Sweetch Health, Ltd) or a CDC-recognized h-DPP for a 12-month intervention. Physical activity was objectively measured using actigraphy. The primary endpoint, assessed at 12 months, was the CDC-defined composite diabetes risk reduction outcome, including achieving 5% weight loss, 4% weight loss plus 150 minutes of weekly physical activity, or a 0.2 reduction in A1C. The pre-specified non-inferiority margin was 15 percentage points. The primary outcome was analyzed using a modified intention-to-treat (mITT) approach, including participants with available 12-month data who did not use prohibited medications. Results: Of 427 screened, 368 were enrolled (183 ai-DPP, 185 h-DPP). Trial completion (85.1%) and prohibited medication use (3.5%) were similar between arms, leaving 300 (151 ai-DPP, 149 h-DPP) in the mITT analysis. Achievement of the primary outcome was similar between groups (ai-DPP: 35.8%, h-DPP: 35.6%; p = 0.97). The age - and sex-adjusted risk difference was -2.6% (lower 95% CI: -11.6%), demonstrating non-inferiority. Individual endpoints in the composite outcome also showed non-inferiority. Conclusion: A fully autonomous AI-based DPP requiring no human coaching is non-inferior to the traditional human-coach based DPP, presenting a promising, scalable alternative for adults with prediabetes. Disclosure N.N. Mathioudakis: None. M.S. Abusamaan: None. M.E. Alderfer: None. D. Alver: None. A.S. Dobs: None. B. Kane: None. B. Lalani: None. J. McGready: None. K. Riekert: None. B. Ringham: None. F. Vandi: None. A.A. Wanigatunga: None. D. Zade: None. N.M. Maruthur: None. Funding The National Institute of Diabetes and Digestive and Kidney Diseases (R01DK125780).","PeriodicalId":11376,"journal":{"name":"Diabetes","volume":"45 1","pages":""},"PeriodicalIF":7.5000,"publicationDate":"2025-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Diabetes","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.2337/db25-1956-lb","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENDOCRINOLOGY & METABOLISM","Score":null,"Total":0}

引用次数: 0

Abstract

Introduction and Objective: Prediabetes is highly prevalent, yet few patients receive evidence-based behavioral lifestyle support. Artificial intelligence (AI) may offer a scalable approach to diabetes prevention. This study evaluated whether a fully automated AI-based diabetes prevention program (ai-DPP), consisting of a mobile app and digital body weight scale, is non-inferior to a traditional human coach-based DPP (h-DPP) in adults with prediabetes and overweight or obesity. Methods: We conducted a two-site, pragmatic, RCT involving adults with prediabetes and overweight or obesity (NCT05056376). Participants were randomly assigned (1:1) to either an ai-DPP (Sweetch Health, Ltd) or a CDC-recognized h-DPP for a 12-month intervention. Physical activity was objectively measured using actigraphy. The primary endpoint, assessed at 12 months, was the CDC-defined composite diabetes risk reduction outcome, including achieving 5% weight loss, 4% weight loss plus 150 minutes of weekly physical activity, or a 0.2 reduction in A1C. The pre-specified non-inferiority margin was 15 percentage points. The primary outcome was analyzed using a modified intention-to-treat (mITT) approach, including participants with available 12-month data who did not use prohibited medications. Results: Of 427 screened, 368 were enrolled (183 ai-DPP, 185 h-DPP). Trial completion (85.1%) and prohibited medication use (3.5%) were similar between arms, leaving 300 (151 ai-DPP, 149 h-DPP) in the mITT analysis. Achievement of the primary outcome was similar between groups (ai-DPP: 35.8%, h-DPP: 35.6%; p = 0.97). The age - and sex-adjusted risk difference was -2.6% (lower 95% CI: -11.6%), demonstrating non-inferiority. Individual endpoints in the composite outcome also showed non-inferiority. Conclusion: A fully autonomous AI-based DPP requiring no human coaching is non-inferior to the traditional human-coach based DPP, presenting a promising, scalable alternative for adults with prediabetes. Disclosure N.N. Mathioudakis: None. M.S. Abusamaan: None. M.E. Alderfer: None. D. Alver: None. A.S. Dobs: None. B. Kane: None. B. Lalani: None. J. McGready: None. K. Riekert: None. B. Ringham: None. F. Vandi: None. A.A. Wanigatunga: None. D. Zade: None. N.M. Maruthur: None. Funding The National Institute of Diabetes and Digestive and Kidney Diseases (R01DK125780).

查看原文本刊更多论文

1956-LB：人工智能与人类指导预防糖尿病——来自12个月、多中心、实用的随机对照试验的结果

前言和目的：前驱糖尿病非常普遍，但很少有患者接受循证行为生活方式支持。人工智能（AI）可能为糖尿病预防提供一种可扩展的方法。本研究评估了由移动应用程序和数字体重秤组成的全自动人工智能糖尿病预防计划（ai-DPP）在糖尿病前期和超重或肥胖的成年人中是否优于传统的基于人类教练的糖尿病预防计划（h-DPP）。方法：我们进行了一项双中心、实用的随机对照试验，涉及患有前驱糖尿病和超重或肥胖的成年人（NCT05056376）。参与者被随机分配（1:1）到ai-DPP （Sweetch Health, Ltd）或cdc认可的h-DPP进行为期12个月的干预。使用活动记录仪客观测量身体活动。12个月时评估的主要终点是cdc定义的复合糖尿病风险降低结果，包括体重减轻5%，体重减轻4%加上每周150分钟的体育活动，或A1C降低0.2。预先规定的非劣效性差为15个百分点。主要结局采用改良的意向治疗（mITT）方法进行分析，包括有12个月可用数据且未使用违禁药物的参与者。结果：在筛选的427例中，368例入组（183例ai-DPP， 185例h-DPP）。试验完成（85.1%）和禁用药物使用（3.5%）在两组之间相似，mITT分析中有300例（151例ai-DPP， 149例h-DPP）。两组间主要结局的实现情况相似(ai-DPP: 35.8%, h-DPP: 35.6%；P = 0.97)。年龄和性别调整后的风险差异为-2.6% (95% CI: -11.6%)，显示非劣效性。综合结果中的个体终点也显示出非劣效性。结论：一个完全自主的基于人工智能的不需要人工指导的DPP不逊于传统的基于人类教练的DPP，为成人前驱糖尿病患者提供了一个有前途的、可扩展的替代方案。N.N. Mathioudakis：没有。ms . abusaman：没有。M.E.奥尔德弗：没有。阿尔弗：没有。多布斯：没有。B.凯恩：没有。B.拉拉尼：没有。J.麦格雷迪：没有。里克特：没有。林厄姆：没有。凡迪：没有。A.A. Wanigatunga：没有。D. Zade：没有。N.M.马鲁瑟：没有。国家糖尿病、消化和肾脏疾病研究所（R01DK125780）。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Diabetes 医学-内分泌学与代谢

CiteScore

12.50

自引率

2.60%

发文量

1968

审稿时长

1 months

期刊介绍： Diabetes is a scientific journal that publishes original research exploring the physiological and pathophysiological aspects of diabetes mellitus. We encourage submissions of manuscripts pertaining to laboratory, animal, or human research, covering a wide range of topics. Our primary focus is on investigative reports investigating various aspects such as the development and progression of diabetes, along with its associated complications. We also welcome studies delving into normal and pathological pancreatic islet function and intermediary metabolism, as well as exploring the mechanisms of drug and hormone action from a pharmacological perspective. Additionally, we encourage submissions that delve into the biochemical and molecular aspects of both normal and abnormal biological processes. However, it is important to note that we do not publish studies relating to diabetes education or the application of accepted therapeutic and diagnostic approaches to patients with diabetes mellitus. Our aim is to provide a platform for research that contributes to advancing our understanding of the underlying mechanisms and processes of diabetes.