Therapeutic Efficacy and Safety Evaluation of AI in the Management of Diabetes: A RCT Trial

Official title

Evaluation of the Therapeutic Efficacy and Safety of Artificial Intelligence-based Decision-making Technology in the Integrated Management of Diabetes Mellitus: a Longitudinal, Open-labeled, Randomized Controlled Trial

Overview

Purpose: To evaluate the efficacy of artificial intelligence (AI)-based decision-making technology in managing glycated hemoglobin (HbA1c) and blood glucose levels compared to the control group. Methods: For the AI Intervention group, the patients will be trained to independently use the diabetes telemedicine platform application. Each patient will be equipped with a glucometer and exercise bracelet, and the data will be automatically transmitted to the medical server via Bluetooth. The healthcare platform will analyze the uploaded data and provide feedback suggestions on medication, diet, and exercise automatically. The platform will also monitor the medical and lifestyle data of the patients every two weeks, offer feedback based on the analyses, and remind the patient to adhere to the self-management protocol based on the platform. The platform is a digitally integrated healthcare platform that patients can use independently without the need for monitoring and assistance by healthcare professionals. The glucometer and pedometer bracelet will automatically connect to the platform through Bluetooth. The patient lab sheet identification and structured conversion system, AI for food picture identification and calorie calculation systems, and the AI decision-making system are on the cloud server. Patients upload image information, such as lab sheets and meal pictures, through the patient's diabetes mobile health system, and the cloud platform intelligently analyzes the patient's disease, medication, and daily life status to develop personalized solutions according to individual control goals. Free outpatient visits will be provided to both the intervention and control groups every twelve weeks. For the conventional treatment group, patients will receive a free blood glucometer and will have regular outpatient appointments. There is no limit to the number of outpatient visits; however, they are required to regularly monitor and record their blood glucose, diet, and exercise data to ensure that the medical team objectively conduct their diagnosis and treatment activities. The medical team will provide free outpatient visits every 12 weeks, along with advice on medication, diet, and exercise based on the individual's blood glucose level. Expected results: A significant difference in HbA1c change from baseline to 48 weeks and improved FPG and 2-hour postprandial blood glucose levels in the AI intervention group were observed.

Detailed description

Follow-up Plan::

Visit 1(-4W\~-1W): Obtain the written informed consents of the patients, conduct the demographic survey, medical record survey, drug history investigation, subject compliance investigation, vital signs checkup, laboratory tests, imaging, and other instrument examinations, as well as evaluate the comorbidities of diabetes.

Visit 2 (D0): Educate the intervention group operating the platform system, evaluating diabetic hypoglycemia events, enhancing patients' self-management abilities, and knowledge mastery. Lab tests will be conducted at 12-week intervals, including Visit 3 (12W), Visit 4 (24W) or Visit 5 (36W), and Visit 6 (48W).

Interventions

• Other artificial intelligence
  The platform will also monitor the medical and lifestyle data of the patients every two weeks,offer feedback based on the analyses, and remind the patient to adhere to the self-management protocol based on the platform.
• Other Routine diagnosis and treatment group for diabetes
  There is no limit to the number of outpatient visits for the control group; however, they are required to regularly monitor and record their blood glucose, diet, and exercise data to ensure that the medical team (endocrinologist and nutritionist) objectively conducttheir diagnosis and treatment activities. The medical team will provide free outpatient visits every 12 weeks, along with advice on medication, diet, and exercise based on the individual's blood glucose level.

Primary outcome measures

• HbA1c [Time frame: 48weeks]
• Fasting Blood Glucose (FBG) [Time frame: 48 weeks]
• 2-hour Postprandial Blood Glucose (2hPPG) [Time frame: 48 weeks]
• Hypoglycemic events [Time frame: 48 weeks]

Secondary outcome measures (7)

• Healthcare expenses [Time frame: 48weeks]
• Insulin and oral hypoglycemic agent dosing [Time frame: 48 weeks]
• Serum lipids [Time frame: 48 weeks]
• blood pressure [Time frame: 48 weeks]
• BMI [Time frame: 48 weeks]
• body weight [Time frame: 48weeks]
• Fasting C-peptide level [Time frame: 48weeks]

Eligibility criteria

Inclusion Criteria：

• Age: ≥18 years，≤75 years;
• Diagnosed with type 2 diabetes for ≥1 year;
• 7.0% ≤HbA1c ≤11%;
• Body mass index ≥18.5 kg/m2;
• Proficient ability to use smart phones;
• Agreed to utilize a digital integrated healthcare platform for diabetes care and research;
• Informed consents are obtained from the participants.

Exclusion criteria

• Presence of other types of diabetes, such as type 1 diabetes and gestational diabetes;
• Severe diabetic complications;
• Medical history of chronic liver diseases, including hemochromatosis, hepatocellular carcinoma, autoimmune liver disease, cirrhosis, viral hepatitis (including hepatitis A, B, and C), or hepatolenticular degeneration;
• Kidney injury (serum creatinine ≥1.5 times the upper limit of the reference) ; Serum ALT and AST levels elevated >2-fold;
• Medical history of mental disorders, such asschizophrenia, depression, or bipolar affective disorder;
• Excessive alcohol intake or drug abuse in the past 3 months;
• Use of medications affecting glucose metabolism, such as corticosteroids or ·consumption of immunosuppressive and anti-obesity medications in the past 3 months;
• Pregnancy, planning for pregnancy, or lactation; or any other conditions unsuitable for trial participation;
• Participatingor plan to participate in other clinical trials; and other cases that are inappropriate to participate.

Criteria are shown verbatim from the registry (in English). Final eligibility is always assessed by the study center.

Healthy volunteers: No

Study design

Allocation

Randomized

Model

Parallel assignment

Masking

Open label

Primary purpose

Treatment

Study locations

China · 1 center

• The First Hospital of Jilin University — Changchun

Publications

• Lee YB, Kim G, Jun JE, Park H, Lee WJ, Hwang YC, Kim JH. An Integrated Digital Health Care Platform for Diabetes Management With AI-Based Dietary Management: 48-Week Results From a Randomized Controlled Trial. Diabetes Care. 2023 May 1;46(5):959-966. doi: 10.2337/dc22-1929. PMID 36821833
• Kim EK, Kwak SH, Jung HS, Koo BK, Moon MK, Lim S, Jang HC, Park KS, Cho YM. The Effect of a Smartphone-Based, Patient-Centered Diabetes Care System in Patients With Type 2 Diabetes: A Randomized, Controlled Trial for 24 Weeks. Diabetes Care. 2019 Jan;42(1):3-9. doi: 10.2337/dc17-2197. Epub 2018 Oct 30. PMID 30377185
• Dobson R, Whittaker R, Jiang Y, Maddison R, Shepherd M, McNamara C, Cutfield R, Khanolkar M, Murphy R. Effectiveness of text message based, diabetes self management support programme (SMS4BG): two arm, parallel randomised controlled trial. BMJ. 2018 May 17;361:k1959. doi: 10.1136/bmj.k1959. PMID 29773539
• Agarwal P, Mukerji G, Desveaux L, Ivers NM, Bhattacharyya O, Hensel JM, Shaw J, Bouck Z, Jamieson T, Onabajo N, Cooper M, Marani H, Jeffs L, Bhatia RS. Mobile App for Improved Self-Management of Type 2 Diabetes: Multicenter Pragmatic Randomized Controlled Trial. JMIR Mhealth Uhealth. 2019 Jan 10;7(1):e10321. doi: 10.2196/10321. PMID 30632972
• Doupis J, Festas G, Tsilivigos C, Efthymiou V, Kokkinos A. Smartphone-Based Technology in Diabetes Management. Diabetes Ther. 2020 Mar;11(3):607-619. doi: 10.1007/s13300-020-00768-3. Epub 2020 Jan 25. PMID 31983028
• Sun C, Sun L, Xi S, Zhang H, Wang H, Feng Y, Deng Y, Wang H, Xiao X, Wang G, Gao Y, Wang G. Mobile Phone-Based Telemedicine Practice in Older Chinese Patients with Type 2 Diabetes Mellitus: Randomized Controlled Trial. JMIR Mhealth Uhealth. 2019 Jan 4;7(1):e10664. doi: 10.2196/10664. PMID 30609983
• Wang H, Yuan X, Wang J, Sun C, Wang G. Telemedicine maybe an effective solution for management of chronic disease during the COVID-19 epidemic. Prim Health Care Res Dev. 2021 Sep 29;22:e48. doi: 10.1017/S1463423621000517. PMID 34583801
• Bulut C, Kato Y. Epidemiology of COVID-19. Turk J Med Sci. 2020 Apr 21;50(SI-1):563-570. doi: 10.3906/sag-2004-172. PMID 32299206

Identifiers

NCT: NCT06957093 · K2024283

Primary sources (government registries)

• ClinicalTrials.gov (US NIH)

View this study on ClinicalTrials.gov ↗