Blended Resources for Integrated Diabetes Guidance and Empowerment (BRIDGE)

Effectiveness and Implementation of an AI-Driven Blended Care Model for Type 2 Diabetes Management in Primary Care: A Single-Blind, Cluster Randomized Controlled Trial in Zhangjiagang, China

The goal of this cluster randomized clinical trial is to learn if an AI-enabled blended care model (the BRIDGE program) works to treat type 2 diabetes in adults. It will also learn about the cost-effectiveness and implementation feasibility of this model in primary care settings. The main questions it aims to answer are:

Does the AI-driven intervention lower HbA1c levels (blood sugar) compared to standard care?

Does this model improve participants' quality of life, self-management behaviors, and digital literacy?

Researchers will compare the "Diet-Medicine Companion" (Shi Yi Ban Lv) mini-program combined with family doctor support to standard community care to see if the blended care model works to manage diabetes.

Participants will:

Use the "Diet-Medicine Companion" mini-program to upload diet photos daily and receive AI feedback for 6 months

Receive periodic guidance and phone reminders from case administrators (family doctors)

Complete questionnaires and blood tests (HbA1c) at baseline, 3 months, and 6 months

Study Overview

• Status: Not yet recruiting
• Conditions: Type 2 Diabetes Mellitus
• Intervention / Treatment: Device: AI-driven Dietary Management and Human-in-the-loop Support

• Study Type: Interventional
• Enrollment (Estimated): 1200
• Phase: Not Applicable

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

• Aged 18 to 70 years (inclusive).
• Glycosylated hemoglobin (HbA1c) level ≥ 6.5%.
• Resided in the local area for at least 6 months.
• Capable of using a smartphone to take photos and use the WeChat mini-program.
• Willing and able to provide informed consent.

Exclusion Criteria:

• Diagnosed with Type 1 diabetes, gestational diabetes, or secondary diabetes.
• Presence of severe diabetic complications.
• Received radiotherapy or chemotherapy within the past 6 months.
• Diagnosed with severe intellectual disabilities, Alzheimer's disease, or other serious psychiatric disorders.
• Current participation in other research projects that may affect the results of this study.
• Presence of other severe disabilities or medical conditions deemed unsuitable for participation by the investigators.

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Treatment
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Single

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: Intervention; Participants in this arm will receive the "Diet-Medicine Companion" (Shi Yi Ban Lv) intervention. This involves using an AI-powered WeChat mini-program for daily dietary logging and receiving real-time feedback. Participants also receive proactive monitoring and support from family doctors (case administrators), including telephone reminders if they are inactive for more than 7 days and targeted health guidance based on their data.	Device: AI-driven Dietary Management and Human-in-the-loop Support; The intervention consists of a 6-month AI-enabled blended care program using the "Diet-Medicine Companion" (Shi Yi Ban Lv) WeChat mini-program. Patient Component (AI Support): Participants are required to upload dietary photos and blood glucose records via the mini-program at least once daily. The system, powered by a Large Language Model (LLM), provides immediate, personalized dietary feedback and answers diabetes-related queries via an AI chatbot. Provider Component (Human Support): Case administrators (family doctors) monitor patient data through a provider dashboard. The protocol involves "human-in-the-loop" support, where doctors provide telephone reminders if participants are inactive (no uploads) for more than 7 days. Doctors also intervene to provide medical guidance or health education based on specific system alerts or patient needs.
No Intervention: Control; Participants in this arm will receive standard community care (usual care) for type 2 diabetes management provided by their local community health centers. They will complete the same schedule of assessments (HbA1c tests and questionnaires) at baseline, 3 months, and 6 months but will not receive the AI mini-program intervention or the specific case administrator support provided to the experimental group.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Change in Glycosylated Hemoglobin (HbA1c) Levels	Glycosylated Hemoglobin (HbA1c) reflects the average plasma glucose concentration over the preceding 3 months. It serves as the gold standard biomarker for long-term glycemic control in diabetes management. Data will be obtained from venous blood samples collected at community health centers. A lower HbA1c percentage indicates better glycemic control.	Baseline, Month 3, and Month 6

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Change in Body Mass Index (BMI)	Calculated as weight in kilograms divided by the square of height in meters (kg/m^2). BMI is used as an indicator of body fatness and weight management status. A lower BMI (within the healthy range) generally indicates better weight control outcomes.	Baseline, Month 3, and Month 6
Change in Blood Pressure	Measured as Systolic and Diastolic Blood Pressure in mmHg using a standard sphygmomanometer. Lower blood pressure values (approaching the normal range) indicate better cardiovascular risk management.	Baseline, Month 3, and Month 6
Incidence of Diabetes Complications	The occurrence of new or worsening diabetes-related complications, including but not limited to neuropathy, retinopathy, nephropathy, and cardiovascular events. Data will be collected through medical record reviews and patient self-reports. A lower incidence rate indicates better disease management	Baseline, Month 3, and Month 6
Quality of Life (EQ-5D-5L)	Assessed using the EuroQol 5-Dimension 5-Level (EQ-5D-5L) questionnaire. It covers five dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Responses are converted into a single utility index score ranging from less than 0 to 1. Higher scores indicate better health-related quality of life.	Baseline, Month 3, and Month 6
Self-Management Behaviors (SDSCA)	Assessed using the Summary of Diabetes Self-Care Activities (SDSCA) questionnaire. This instrument measures the frequency of self-care activities (including diet, exercise, blood sugar testing, and foot care) over the past 7 days. Higher scores indicate better adherence to diabetes self-management activities.	Baseline, Month 3, and Month 6
Treatment Satisfaction (DTSQs)	Assessed using the Diabetes Treatment Satisfaction Questionnaire (DTSQs). This tool evaluates the patient's satisfaction with their current treatment regimen, including perception of hyperglycemia/hypoglycemia and convenience. Higher scores indicate greater satisfaction with the treatment.	Baseline, Month 3, and Month 6
Digital Literacy (C-eHEALS)	Assessed using the Chinese eHealth Literacy Scale (C-eHEALS). This scale measures the participants' combined knowledge, comfort, and perceived skills at finding, evaluating, and applying electronic health information to health problems. Higher scores indicate higher levels of electronic health literacy.	Baseline, Month 3, and Month 6
Medical Costs	Total direct and indirect costs related to diabetes management. Direct costs include expenses for examinations, medications, outpatient/inpatient visits, transportation, and caregiver fees. Indirect costs include caregiver time costs. Data will be collected via patient questionnaires and healthcare utilization records. Lower costs indicate a reduced economic burden.	Baseline, Month 3, and Month 6

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
Frequency of Dietary Uploads (Adherence)	Measured as the total count of dietary records (photos or text entries) uploaded by each participant via the "Diet-Medicine Companion" (Shi Yi Ban Lv) mini-program. Data is automatically logged by the system's backend server. This metric serves as a direct objective indicator of participant engagement and adherence to the intervention protocol (which requires daily uploads). Higher numbers indicate better adherence.	Throughout the 6-month intervention period

Collaborators and Investigators

• Sponsor: Peking University

Study record dates

Study Major Dates

• Study Start (Estimated): March 1, 2026
• Primary Completion (Estimated): March 1, 2026
• Study Completion (Estimated): October 30, 2026

Study Registration Dates

• First Submitted: January 21, 2026
• First Submitted That Met QC Criteria: January 21, 2026
• First Posted (Actual): January 28, 2026

Study Record Updates

• Last Update Posted (Actual): February 23, 2026
• Last Update Submitted That Met QC Criteria: February 19, 2026
• Last Verified: January 1, 2026

More Information

Terms related to this study

• Keywords: digital health; AI; Type 2 Diabetes Mellitus; blended care
• Additional Relevant MeSH Terms: Endocrine System Diseases; Metabolic Diseases; Glucose Metabolism Disorders; Diabetes Mellitus; Nutritional and Metabolic Diseases; Diabetes Mellitus, Type 2
• Other Study ID Numbers: IRB00001052-25198

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES

IPD Sharing Time Frame

Start Date: Within 6 months after publication of the primary manuscript

End Date: 3 years after publication

• IPD Sharing Access Criteria: De-identified individual participant data (IPD) that underlie the results reported in the article will be available to researchers who provide a methodologically sound proposal. Data will be available beginning 6 months and ending 3 years following article publication. Proposals should be directed to the Principal Investigator, Ping He (phe@pku.edu.cn) or the study contact. To gain access, data requestors will need to sign a data access agreement.
• IPD Sharing Supporting Information Type: STUDY_PROTOCOL; SAP; ANALYTIC_CODE

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No