A Novel Approach for Fully Automated, Personalized Health Coaching for Adults with Prediabetes: Pilot Clinical Trial
Open accessPublished version of a preprint
Abstract
Background: Prediabetes is a high-risk state for the future development of type 2 diabetes, which may be prevented through physical activity (PA), adherence to a healthy diet, and weight loss. Mobile health (mHealth) technology is a practical and cost-effective method of delivering diabetes prevention programs in a real-world setting. Sweetch (Sweetch Health, Ltd) is a fully automated, personalized mHealth platform designed to promote adherence to PA and weight reduction in people with prediabetes. Objective: The objective of this pilot study was to calibrate the Sweetch app and determine the feasibility, acceptability, safety, and effectiveness of the Sweetch app in combination with a digital body weight scale (DBWS) in adults with prediabetes. Methods: This was a 3-month prospective, single-arm, observational study of adults with a diagnosis of prediabetes and body mass index (BMI) between 24 kg/m2 and 40 kg/m2. Feasibility was assessed by study retention. Acceptability of the mobile platform and DBWS were evaluated using validated questionnaires. Effectiveness measures included change in PA, weight, BMI, glycated hemoglobin (HbA1c), and fasting blood glucose from baseline to 3-month visit. The significance of changes in outcome measures was evaluated using paired t test or Wilcoxon matched pairs test. Results: The study retention rate was 47 out of 55 (86%) participants. There was a high degree of acceptability of the Sweetch app, with a median (interquartile range [IQR]) score of 78% (73%-80%) out of 100% on the validated System Usability Scale. Satisfaction regarding the DBWS was also high, with median (IQR) score of 93% (83%-100%). PA increased by 2.8 metabolic equivalent of task (MET)–hours per week (SD 6.8; P=.02), with mean weight loss of 1.6 kg (SD 2.5; P<.001) from baseline. The median change in A1c was −0.1% (IQR −0.2% to 0.1%; P=.04), with no significant change in fasting blood glucose (−1 mg/dL; P=.59). There were no adverse events reported. Conclusions: The Sweetch mobile intervention program is a safe and effective method of increasing PA and reducing weight and HbA1c in adults with prediabetes. If sustained over a longer period, this intervention would be expected to reduce diabetes risk in this population. Trial Registration: ClincialTrials.gov NCT02896010; https://clinicaltrials.gov/ct2/show/NCT02896010 (Archived by WebCite at http://www.webcitation.org/6xJYxrgse)
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The AI-led DPP platform was previously tested in a pilot trial in adults with prediabetes, including an initial calibration cohort to optimize its reinforcement learning algorithm.
“This platform was previously tested in a pilot trial involving adults with prediabetes, which included an initial calibration cohort to optimize its reinforcement learning algorithm.12”
From An AI-Powered Lifestyle Intervention vs Human Coaching in the Diabetes Prevention Program: A Randomized Clinical Trial (2025) · cites this paper as [12]
Methods (Calibration and Intervention Cohorts; page unavailable) describes a pilot study in adults with prediabetes that began with a calibration cohort and used its data to adjust activity metrics and refine the app. Methods (Description of Sweetch Mobile Platform; page unavailable) describes machine learning and adaptive algorithms, but our copy does not identify a reinforcement learning algorithm or say that the calibration cohort optimized one. (LLM, conf=0.85)
Experimental — accuracy not yet measuredOf 12 DCP studies assessing dysglycemia, 9 (75%) reported positive population health outcomes and 3 (25%) showed neutral effects.
“Of the 12 studies (92.3%) assessing the effects of dysglycemia, 9 (75%) reported positive population health outcomes, such as significant HbA1c reduction compared to the control group in Kim et al [45] using virtual reality (VR) technology and Everett et al [46] using AI interventions (P<.05). Furthermore, 3 (25%) studies [40,44,47] showed the effect was neutral.”
From Digital Health Interventions to Prevent Type 2 Diabetes Mellitus: Systematic Review (2025) · cites this paper as [46]
The Results (Effectiveness and Safety; page unavailable) report a significant A1c reduction of 0.1% (P=.04) in the intervention cohort, and the Introduction (Study Intervention; page unavailable) identifies Sweetch as AI-based. But the Methods (Statistical Analysis; page unavailable) compare participants’ measurements before and after the intervention, not against a control group. Everett et al. also do not report the citing review’s totals of 9 positive and 3 neutral studies. (LLM, conf=0.90)
Experimental — accuracy not yet measuredKim et al using virtual reality and Everett et al using AI interventions showed significant HbA1c reduction compared to the control group (P<.05).
“such as significant HbA1c reduction compared to the control group in Kim et al [45] using virtual reality (VR) technology and Everett et al [46] using AI interventions (P<.05).”
From Digital Health Interventions to Prevent Type 2 Diabetes Mellitus: Systematic Review (2025) · cites this paper as [46]
The Results (Effectiveness and Safety) report a significant A1c reduction of 0.1% in the Sweetch intervention cohort (P=.04), and the Discussion (Significance of Findings) identifies the intervention as using machine learning. But the Methods (Statistical Analysis) describe a comparison with the cohort’s own baseline, not a control group; the paper therefore does not support the claimed between-group reduction. (LLM, conf=0.91)
Experimental — accuracy not yet measuredIn this review, AI used with wearables showed positive effects on glycemic status.
“In our review, AI used with wearables showed positive effects on glycemic status [46,56].”
From Digital Health Interventions to Prevent Type 2 Diabetes Mellitus: Systematic Review (2025) · cites this paper as [46]
The Results (Effectiveness and Safety; page unavailable) report a significant reduction in A1c, supporting a positive effect on one measure of glycemic status, but no significant change in fasting glucose. The Methods (Description of Sweetch Mobile Platform; page unavailable) describe AI using mobile-phone and digital-scale data, not a wearable device. The cited trial therefore does not support the wearable-specific part of the claim. (LLM, conf=0.88)
Experimental — accuracy not yet measuredAI-based DPP studies reported to date have been single-group or other nonrandomized designs.
“The AI-based DPP studies reported to date have been single-group or other nonrandomized designs.12,20,21”
From An AI-Powered Lifestyle Intervention vs Human Coaching in the Diabetes Prevention Program: A Randomized Clinical Trial (2025) · cites this paper as [12]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin
(2002) · New England Journal of Medicine · doi:10.1056/nejmoa012512 · Cited by 19,011 · cited 2× · see citing works
“Regular physical activity (PA; 150 min per week), weight loss, and healthy diet are highly effective strategies for preventing or delaying the onset of type 2 diabetes [8-10]; however, randomized controlled trials (RCTs) of these interventions have required intensive one-on-one or group lifestyle coaching, which raise…”
BackgroundTuomilehto, J., Lindström, J., Eriksson, J. G., Valle, T. T., Hämäläinen, H., Ilanne-Parikka, P., Keinänen-Kiukaanniemi, S., Laakso, M., Louheranta, A., Rastas, M., et al. (2001) · New England Journal of Medicine · doi:10.1056/nejm200105033441801 · Cited by 10,873 · see citing works
Prediabetes: a high-risk state for diabetes development
Tabák, A. G., Herder, C., Rathmann, W., Brunner, E. J., & Kivimäki, M. (2012) · The Lancet · doi:10.1016/s0140-6736(12)60283-9 · Cited by 2,958 · see citing works
Sepah, S. C., Jiang, L., & Peters, A. L. (2014) · The Diabetes Educator · doi:10.1177/0145721714531339 · Cited by 139 · see citing works
Michaelides, A., Raby, C., Wood, M., Farr, K., & Toro-Ramos, T. (2016) · BMJ Open Diabetes Research & Care · doi:10.1136/bmjdrc-2016-000264 · Cited by 128 · see citing works
Flagged first, then by citations
Research works printed but not linked: 19
- [1]Edwards CM Cusi K Prediabetes: a worldwide epidemic Endocrinol Metab Clin North Am 2016 12 45 4 751 64 10.1016/j.ecl.2016.06.007 27823603 27823603 · DOI
- [3]Mainous AG Tanner RJ Baker R Zayas CE Harle CA Prevalence of prediabetes in England from 2003 to 2011: population-based, cross-sectional study BMJ Open 2014 6 9 4 6 e005002 10.1136/bmjopen-2014-005002 24913327 PMC4054625 24913327 · DOI
- [4]Wang L Gao P Zhang M Huang Z Zhang D Deng Q Li Y Zhao Z Qin X Jin D Zhou M Tang X Hu Y Wang L Prevalence and ethnic pattern of diabetes and prediabetes in China in 2013 J Am Med Assoc 2017 6 27 317 24 2515 23 10.1001/jama.2017.7596 28655017 PMC5815077 28655017 · DOI
- [5]Anjana RM Deepa M Pradeepa R Mahanta J Narain K Das HK Adhikari P Rao PV Saboo B Kumar A Bhansali A John M Luaia R Reang T Ningombam S Jampa L Budnah RO Elangovan N Subashini R Venkatesan U Unnikrishnan R Das AK Madhu SV Ali MK Pandey A Dhaliwal RS Kaur T Swaminathan S Mohan V ICMR–INDIAB Collaborative Study Group Prevalence of diabetes and prediabetes in 15 states of India: results from the ICMR-INDIAB population-based cross-sectional study Lancet Diabetes Endocrinol 2017 8 5 8 585 96 10.1016/S2213-8587(17)30174-2 28601585 28601585 · DOI
- [6]Brannick B Wynn A Dagogo-Jack S Prediabetes as a toxic environment for the initiation of microvascular and macrovascular complications Exp Biol Med (Maywood) 2016 6 241 12 1323 31 10.1177/1535370216654227 27302176 27302176 PMC4950274 · DOI
- [8]Pan XR Li GW Hu YH Wang JX Yang WY An ZX Hu ZX Lin J Xiao JZ Cao HB Liu PA Jiang XG Jiang YY Wang JP Zheng H Zhang H Bennett PH Howard BV Effects of diet and exercise in preventing NIDDM in people with impaired glucose tolerance. The Da Qing IGT and Diabetes Study Diabetes Care 1997 4 20 4 537 44 9096977 9096977
- [11]Herman WH The economics of diabetes prevention Med Clin North Am 2011 3 95 2 373 84, viii 10.1016/j.mcna.2010.11.010 21281839 21281839 PMC3061284 · DOI
- [12]Grock S Ku JH Kim J Moin T A review of technology-assisted interventions for diabetes prevention Curr Diab Rep 2017 9 23 17 11 107 10.1007/s11892-017-0948-2 28942537 28942537 · DOI · Paper page
- [14]Fukuoka Y Gay CL Joiner KL Vittinghoff E Novel diabetes prevention intervention using a mobile app: a randomized controlled trial with overweight adults at risk Am J Prev Med 2015 8 49 2 223 37 10.1016/j.amepre.2015.01.003 26033349 26033349 PMC4509889 · DOI · Paper page
- [16]Castro Sweet CM Chiguluri V Gumpina R Abbott P Madero EN Payne M Happe L Matanich R Renda A Prewitt T Outcomes of a digital health program with human coaching for diabetes risk reduction in a medicare population J Aging Health 2017 1 1 898264316688791 10.1177/0898264316688791 28553807 PMC5944079 28553807 · DOI · Paper page
- [17]Hamman RF Wing RR Edelstein SL Lachin JM Bray GA Delahanty L Hoskin M Kriska AM Mayer-Davis EJ Pi-Sunyer X Regensteiner J Venditti B Wylie-Rosett J Effect of weight loss with lifestyle intervention on risk of diabetes Diabetes Care 2006 9 29 9 2102 7 10.2337/dc06-0560 16936160 16936160 PMC1762038 · DOI · Paper page
- [18]Ilanne-Parikka P Laaksonen DE Eriksson JG Lakka TA Lindstr J Peltonen M Aunola S Keinánen-Kiukaanniemi S Uusitupa M Tuomilehto J Finnish Diabetes Prevention Study Group Leisure-time physical activity and the metabolic syndrome in the Finnish diabetes prevention study Diabetes Care 2010 7 33 7 1610 7 10.2337/dc09-2155 20413514 20413514 PMC2890369 · DOI
- [19]Jeon CY Lokken RP Hu FB van Dam RM Physical activity of moderate intensity and risk of type 2 diabetes: a systematic review Diabetes Care 2007 3 30 3 744 52 10.2337/dc06-1842 17327354 17327354 · DOI · Paper page
- [20]Rosen LD Whaling K Carrier LM Cheever NA Rokkum J The media and technology usage and attitudes scale: an empirical investigation Comput Human Behav 2013 11 1 29 6 2501 11 25722534 25722534 PMC4338964
- [22]Klein S Allison DB Heymsfield SB Kelley DE Leibel RL Nonas C Kahn R Waist circumference and cardiometabolic risk: a consensus statement from shaping America's health: Association for Weight Management and Obesity Prevention; NAASO, the Obesity Society; the American Society for Nutrition; and the American Diabetes Association Obesity (Silver Spring) 2007 5 15 5 1061 7 10.1038/oby.2007.632 10.1038/oby.2007.632 17495180 17495180 · DOI
- [23]Pickering TG Hall JE Appel LJ Falkner BE Graves JW Hill MN Jones DH Kurtz T Sheps SG Roccella EJ Council on High Blood Pressure Research Professional and Public Education Subcommittee, American Heart Association Recommendations for blood pressure measurement in humans: an AHA scientific statement from the Council on High Blood Pressure Research Professional and Public Education Subcommittee J Clin Hypertens (Greenwich) 2005 2 7 2 102 9 15722655 15722655 PMC8109470
- [26]Oguma Y Sesso HD Paffenbarger Jr RS Lee IM Weight change and risk of developing type 2 diabetes Obes Res 2005 5 13 5 945 51 10.1038/oby.2005.109 10.1038/oby.2005.109 15919849 15919849 · DOI
- [27]Block G Azar KM Block TJ Romanelli RJ Carpenter H Hopkins D Palaniappan L Block CH A fully automated diabetes prevention program, Alive-PD: program design and randomized controlled trial protocol JMIR Res Protoc 2015 4 1 e3 10.2196/resprot.4046 25608692 25608692 PMC4319077 · DOI
- [30]Sepah SC Jiang L Peters AL Long-term outcomes of a Web-based diabetes prevention program: 2-year results of a single-arm longitudinal study J Med Internet Res 2015 17 4 e92 10.2196/jmir.4052 25863515 25863515 PMC4409647 · DOI · Paper page
Cited by
Sorted by citations
Artificial intelligence for diabetes management and decision support: Literature review
Contreras, I., & Vehi, J. (2018) · Journal of Medical Internet Research · doi:10.2196/10775 · Cited by 560 · see citing works
Challenges in Participant Engagement and Retention Using Mobile Health Apps: Literature Review
Amagai, S., Pila, S., Kaat, A. J., Nowinski, C. J., & Gershon, R. C. (2022) · Journal of Medical Internet Research · doi:10.2196/35120 · Cited by 343 · see citing works
Artificial intelligence in diabetes management: Advancements, opportunities, and challenges
Guan, Z., Li, H., Liu, R., Cai, C., Liu, Y., Li, J., Wang, X., Huang, S., Wu, L., Liu, D., et al. (2023) · Cell Reports Medicine · doi:10.1016/j.xcrm.2023.101213 · Cited by 287 · see citing works
“68 Sun et al. investigated whether a year-long cloud platform-based and intelligent, personalized exercise prescription intervention could improve the health outcomes of Chinese middle-aged and older adult community dwellers.”
BackgroundVan Rhoon, L., Byrne, M., Morrissey, E., Murphy, J., & McSharry, J. (2020) · DIGITAL HEALTH · doi:10.1177/2055207620914427 · Cited by 182 · see citing works
Digital coaching strategies to facilitate behavioral change in type 2 diabetes: a systematic review
Gershkowitz, B. D., Hillert, C. J., & Crotty, B. H. (2021) · The Journal of Clinical Endocrinology & Metabolism · doi:10.1210/clinem/dgaa850 · Cited by 67 · see citing works
Weight loss in a digital app-based diabetes prevention program powered by artificial intelligence.
Graham, S. A., Pitter, V., Hori, J. H., Stein, N., & Branch, O. H. (2022) · DIGITAL HEALTH · doi:10.1177/20552076221130619 · Cited by 43 · see citing works
Engagement and weight loss: results from the Mobile Health and Diabetes Trial
Muralidharan, S., Ranjani, H., Mohan Anjana, R., Jena, S., Tandon, N., Gupta, Y., Ambekar, S., Koppikar, V., Jagannathan, N., Allender, S., et al. (2019) · Diabetes Technology & Therapeutics · doi:10.1089/dia.2019.0134 · Cited by 42 · see citing works
Mathioudakis, N., Lalani, B., Abusamaan, M. S., Alderfer, M., Alver, D., Dobs, A., Kane, B., McGready, J., Riekert, K., Ringham, B., et al. (2025) · JAMA · doi:10.1001/jama.2025.19563 · Cited by 41 · see citing works
Self-citationHealth Coaching Strategies for Weight Loss: A Systematic Review and Meta-Analysis
Sieczkowska, S. M., de Lima, A. P., Swinton, P. A., Dolan, E., Roschel, H., & Gualano, B. (2021) · Advances in Nutrition · doi:10.1093/advances/nmaa159 · Cited by 29 · see citing works
Abusamaan, M. S., Ballreich, J., Dobs, A., Kane, B., Maruthur, N., McGready, J., Riekert, K., Wanigatunga, A. A., Alderfer, M., Alver, D., et al. (2024) · Trials · doi:10.1186/s13063-024-08177-8 · Cited by 16 · see citing works
Self-citationDigital Health Interventions to Prevent Type 2 Diabetes Mellitus: Systematic Review
Duong, T., Olsen, Q., Menon, A., Woods, L., Wang, W., Varnfield, M., Jiang, L., & Sullivan, C. (2025) · Journal of Medical Internet Research · doi:10.2196/67507 · Cited by 14 · see citing works
Applications of artificial intelligence and machine learning in prediabetes: a scoping review.
Lalani, B., Herur, R., Zade, D., Collins, G., Dishong, D. M., Mehta, S., Shim, J., Valdez, Y., & Mathioudakis, N. (2025) · Journal of Diabetes Science and Technology · doi:10.1177/19322968251351995 · Cited by 11 · see citing works
Self-citationLim, P. C., Lim, Y. L., Rajah, R., & Zainal, H. (2025) · BMC Digital Health · doi:10.1186/s44247-025-00150-y · Cited by 8 · see citing works
- Authors
- Everett, E., Kane, B., Yoo, A., Dobs, A., & Mathioudakis, N.
- Affiliations
- Johns Hopkins University (US)
- Corresponding author
- Mathioudakis, Nestoras
- Venue
- Journal of Medical Internet Research
- Publisher
- JMIR Publications Inc.
- Published
- 2018-02-27
- Volume
- 20
- Issue
- 2
- Pages
- e72
- ISSN
- 1438-8871 (ISSN-L)
- PMC
- PMC5849796
- OpenAlex
- W2788565034
- Cited by
- 101
- References
- 5
- Citations by year
- 2026: 10 · 2025: 14 · 2024: 20 · 2023: 10 · 2022: 14 · 2021: 11 · 2020: 14 · 2019: 5 · 2018: 3
- FWCI
- 12.60
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- Locations
- Journal of Medical Internet Research (publishedVersion, cc-by)
- Europe PMC (PubMed Central) (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Funders
- Johns Hopkins University (UL1TR001079)
- National Institutes of Health (UL1TR001079, UL1TR003098)
- National Center for Advancing Translational Sciences
- Topics
- Health Sciences › Health Professions › General Health Professions › Mobile Health and mHealth Applications
- Health Sciences › Medicine › Endocrinology, Diabetes and Metabolism › Diabetes Management and Education
- Social Sciences › Psychology › Applied Psychology › Digital Mental Health Interventions
- Keywords
- prediabetes, type 2 diabetes prevention, mobile health, health coaching, physical activity, weight loss, HbA1c, personalized interventions, feasibility and acceptability, body mass index, glycemic control, fasting blood glucose, student retention, usability evaluation
- MeSH terms
- Adult, Female, Humans, Male, Pilot Projects, Prediabetic State, Prospective Studies, Surveys and Questionnaires, Weight Loss, Telemedicine, Mobile Applications
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
