App-based Motivational Interviewing and Artificial Intelligence in Diabetes Management

Not Applicable

Recruiting

• Conditions: Diabetes Mellitus Type 2
• Interventions: Behavioral: Empower+ App & Smartwatch wearable tracker; Behavioral: App-based Motivational Interviewing & Human health coaching

• Registration Number: NCT06214520
• Lead Sponsor: Singapore General Hospital

Brief Summary

There is an urgent need for better control and prevention of complications in type 2 diabetes mellitus (T2DM). Behavioural change is critical, and while literature suggests that motivational interviewing (MI) may be effective in improving glycemic control, none has explored app-based MI designed specifically for T2DM. The overall objective of this project is to determine the effectiveness of primary care model combining app-based MI and AI-powered personalised nudges delivered through a mobile application (app) for diabetes management (EMPOWER-PLUS). The project aims to evaluate the effectiveness and implementation of MI and nudges through EMPOWERPLUS to deliver diabetes management through a randomised controlled trial (RCT). This will be a 3-arm RCT with primary outcome measure being the difference in HbA1c level at week 36 between the intervention and control arms. Secondary outcome measures include cost-effectiveness, quality of life, medication adherence, diet, and physical activity. Eligible poorly controlled T2DM patients with T2DM in polyclinics will be randomized to intervention arm who will receive EMPOWER-PLUS and smartwatch wearable on top of their usual clinical care. The first control group will have access to nudges delivered through app and smartwatch wearable in addition to usual clinical care but will not receive MI. The second control group will receive usual care (no access to MI, nudges and smartwatch wearable). This study is important to improve T2DM outcomes and reduce healthcare utilization by providing scientifically evaluated and transformative primary care model. Leveraging on digital technology and artificial intelligence to drive personalised care, behavioural change and empowerment has huge potential for scale up.

Detailed Description

Not available

Study Timeline

• Study First Submitted: 2023-12-19
• Study First Submitted QC: 2024-01-10
• Study First Posted: 2024-01-19
• Study Start: 2024-02-23
• Status Verified: 2024-11
• Last Update Submitted: 2024-11-12
• Last Update Posted: 2024-11-14
• Primary Completion: 2025-10-30
• Study Completion: 2026-06-30

Recruitment & Eligibility

• Status: RECRUITING
• Sex: All
• Target Recruitment: 525

Inclusion Criteria

• Aged 21 years and above
• Have been diagnosed with diabetes (T2DM)
• Had HbA1c result of ≥ 7.5% within past 3 months
• Physically able to exercise
• Able to read and converse in English
• Able to download the Empower+ app, use the smartphone wearable tracker, and conform to the minimum smartwatch and app monitoring schedule

Exclusion Criteria

• On bolus insulin treatment
• Require assistance with basic activities of daily living (BADL)
• Have planned major operation or surgical procedure within 9 months from the time of recruitment
• Cognitively impaired (scored < 6 on the Abbreviated Mental Test)
• Currently pregnant or lactating
• Current participants of ongoing clinical trials involving the usage of a smartphone wearable tracker or mobile health app to assist diabetes management
• Past Empower study participants who refused to participate in future Empower studies

Study & Design

• Study Type: INTERVENTIONAL
• Study Design: PARALLEL

Arm && Interventions

Group	Intervention	Description
Control 1	Empower+ App & Smartwatch wearable tracker	Control group 1
Intervention	App-based Motivational Interviewing & Human health coaching	Intervention Group
Intervention	Empower+ App & Smartwatch wearable tracker	Intervention Group

Primary Outcome Measures

Name	Time	Method
HbA1c	9 months	Difference between HbA1c measured at the in-house laboratory at polyclinics, at baseline and the 9-month mark.

Secondary Outcome Measures

Name	Time	Method
Physical activity - active minutes	9 months	Measured by the smartphone wearable tracker over 9 months.
Medication adherence	9 months	Measured by Voils, at baseline and the 9-month mark. For each item in Voils, the raw score ranges from 1 (missed medication none of the time) to 5 (missed medication every time), with a higher score indicating a poor level of medication adherence (or equivalently a higher level of medication non-adherence).
Physical activity - steps	9 months	Measured by the smartphone wearable tracker over 9 months.
HbA1c	18 months	Difference between HbA1c measured at the in-house laboratory at polyclinics, at baseline and the 6-month mark. HbA1c will be also retrieved from the Electronic Medical Record over the follow-up period (from month 10 to 18, if any).
Patient activation score	9 months	Measured by patient activation measure (PAM), at baseline and the 9-month mark. For each item in PAM, the raw score ranges from 1 (Disagree strongly) to 4 (Agree strongly), and 0 is given to option "Not Applicable", with a higher score indicating a higher level of patient activation.
Sleep pattern	9 months	Measured by the smartphone wearable tracker over 9 months.
Heart rate	9 months	Measured by the smartphone wearable tracker over 9 months.
Sleep quality	9 months	Measured by the smartphone wearable tracker over 9 months.
Direct healthcare cost	9 months	Retrieved from the electronic medical records and recorded in survey form: Costs of consultations, lab tests, medications, admissions, at baseline and the 9-month mark.
Indirect healthcare cost	9 months	Measured by survey - Work Productivity and Activity Impairment (WPAI), at baseline and the 9-month mark. The WPAI-GH consists of six questions: 1 = currently employed; 2 = hours missed due to health problems; 3 = hours missed other reasons; 4 = hours actually worked; 5 = degree health affected productivity while working (using a 0 to 10 Visual Analogue Scale (VAS)); 6 = degree health affected productivity in regular unpaid activities (VAS). For the last two items, the raw scores range from 0 to 10, with a higher score indicating the patient being more affected by his/her health problems.
Dietary changes	9 months	Measured with the Empower+ app (calorie intake, carbohydrate and sugar intake) over 9 months.
Health-related Quality of life	9 months	Measured by EuroQol-5D-5L (EQ-5D-5L) combined with global health rating on a visual analogue scale (EQ-VAS) at baseline and the 9-month mark. Each dimension in the EQ-5D-5L has five response levels ranging from no problems (Level 1) to extreme problems (Level 5). There are 3,125 possible health states defined by combining one level from each dimension, ranging from 11111 (full health) to 55555 (worst health). For EQ-VAS, the raw score ranges from 0 to 100, with a higher score indicating a better health that the participant can imagine. The converted index value will be ranging from -0.5 to 1.0, where 0 is the value of a health state equivalent to death, negative values representing values as worse than dead, and 1 is the value of full health.
Usability of the Empower+ app and Motivational Interviewing module	1 month	Measured by the System Usability Scale (SUS), at the 1-month mark (for intervention group and control group 1). For each item in SUS, the raw score ranges from 1 to 5, with a higher score for positive statements and a lower score for negative statements indicating a better usability.

Trial Locations

Locations (3)

SingHealth Polyclinic (Tampines); 🇸🇬 Singapore, Singapore

SingHealth Polyclinic (Sengkang); 🇸🇬 Singapore, Singapore

SingHealth Polyclinic (Punggol); 🇸🇬 Singapore, Singapore