Automated Health Coaching for Type 2 Diabetes

(GODART-P&F Trial)

Overseen By Tejossy John, MPH

Age: 18+

Sex: Any

Trial Phase: Academic

Sponsor: University of Alabama at Birmingham

Must not be taking: Insulin

Disqualifiers: Pregnancy, Cardiac event, Renal failure, others

No Placebo Group All trial participants will receive the active study treatment (no placebo)

Compensation: Varies

Number of Visits: 1

108 Participants Needed

Trial Summary

Do I have to stop taking my current medications for the trial?

The trial information does not specify whether you need to stop taking your current medications. However, it does exclude patients currently on insulin treatment.

What data supports the effectiveness of the treatment Automated Health Coaching for Type 2 Diabetes?

Research shows that digital health coaching, which is part of the treatment, can significantly improve diabetes management by enhancing behavior change and clinical outcomes. Virtual coaching combined with apps has been found to provide better glycemic control and higher satisfaction compared to apps alone, indicating the potential effectiveness of this approach. ^{1 2 3 4 5}

Is automated health coaching for type 2 diabetes safe for humans?

The research on digital health coaching for type 2 diabetes, including virtual coaching and gamified interventions, suggests that these approaches are generally safe for humans. They have been shown to improve diabetes management and patient satisfaction without reported safety concerns. ^{1 2 3 6 7}

How does automated health coaching differ from other treatments for type 2 diabetes?

Automated health coaching for type 2 diabetes is unique because it uses artificial intelligence to provide personalized, adaptive support, making it more accessible and cost-effective compared to traditional in-person counseling. This approach can offer real-time, individualized guidance and has shown potential to improve insulin resistance and other health outcomes, especially for under-resourced individuals. ^{1 2 3 8 9}

What is the purpose of this trial?

The purpose of this study is to pilot and assess the feasibility of implementing an artificial intelligence-assisted individualized lifestyle modification intervention for glycemic control in rural populations, which can be delivered even with regular landline phone service. This study will provide us with the knowledge to plan a well-powered optimization trial in the future to develop an optimal (low-cost) intervention package that can be delivered in a sustainable manner to the rural portions of America.

Research Team

Tapan Mehta, MD

Principal Investigator

University of Alabama at Birmingham

Mohanraj Thirumalai, PHD

Principal Investigator

University of Alabama at Birmingham

Eligibility Criteria

This trial is for adults over 18 with Type 2 Diabetes, who speak and read English, have a doctor's approval to join, and HbA1c levels between 7% to 10.5%. It's not for those pregnant or planning pregnancy, in another diabetes study, on insulin treatment, with recent renal failure or major heart events.

Inclusion Criteria

Physician consent to participate in the study

Ability to converse in and read English

Your HbA1c levels need to be between 7% and 10.5% for the first phase of the study, and between 8% and 10.5% for the second phase.

Exclusion Criteria

I have not had renal failure in the last 6 months.

Present or soon-planned pregnancy

You are currently participating in a program to manage diabetes or weight.

Timeline

Screening

Participants are screened for eligibility to participate in the trial

2-4 weeks

Feasibility Phase

Participants test the feasibility, usability, and accessibility of the GODART platform

2 weeks

1 visit (virtual)

Intervention Phase

Participants receive the intervention with different experimental conditions for 6 months

6 months

Weekly virtual coaching sessions

Follow-up

Participants are monitored for safety and effectiveness after treatment

4 weeks

Treatment Details

Interventions

• Adapted Reward Level
• Fixed Gamified Reward Level
• Weekly Automated Health Coaching
• Weekly Human Health Coaching

Trial Overview The study tests an AI-assisted lifestyle change program aimed at controlling blood sugar in rural areas via regular phone service. Participants receive either automated or human health coaching weekly with varying reward systems to see which method works best.

Participant Groups

Group I: Arm 4 Experimental Treatment 2 Interventions

Fixed Reward + Weekly human coaching

Group II: Arm 3 Experimental Treatment 2 Interventions

Fixed Reward + Weekly automated coaching

Group III: Arm 2 Experimental Treatment 2 Interventions

Adaptive Rewards + Weekly human coaching

Group IV: Arm 1 Experimental Treatment 2 Interventions

Adaptive Rewards + Weekly automated coaching

Who Is Running the Clinical Trial?

University of Alabama at Birmingham

Lead Sponsor

Trials

1,677

Recruited

2,458,000+

Findings from Research

Diabetes-related apps combined with virtual coaching can significantly improve glycemic control and promote meaningful behavior changes compared to using apps alone, addressing common treatment barriers effectively.

Users of these enhanced apps report high satisfaction and prolonged engagement, suggesting that virtual coaching may be a valuable tool in diabetes education, although further research is needed to explore its long-term effects across diverse populations.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Virtual Coaching to Enhance Diabetes Care. Ramchandani, N.[2020]

A systematic review of 42 articles identified seven core incentive mechanisms used in mobile health (mHealth) tools for diabetes management, with education being the most common, followed by reminders and feedback.

Despite advancements in technology, the fundamental incentive-driven mechanisms have remained consistent, evolving mainly in terms of content delivery and personalization to enhance user engagement and retention.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

A systematic review on incentive-driven mobile health technology: As used in diabetes management. de Ridder, M., Kim, J., Jing, Y., et al.[2022]

The 12-week Healthy at Home digital health coaching program significantly improved insulin resistance in individuals with type 2 diabetes, showing a reduction in HOMA2-IR scores compared to usual care (p = 0.029).

This pilot study suggests that digital health coaching can be an effective and accessible intervention for under-resourced patients with type 2 diabetes, potentially leading to better diabetes management and clinical outcomes.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Digital Health Coaching for Type 2 Diabetes: Randomized Controlled Trial of Healthy at Home. Azelton, KR., Crowley, AP., Vence, N., et al.[2021]

Chronic diseases affect about half of the US population and are responsible for 70% of deaths and 80% of healthcare costs, highlighting the need for effective, individualized interventions beyond traditional healthcare settings.

The proposed digital approach utilizes real-time sensing technology and machine learning to deliver personalized behavior change interventions, aiming to empower individuals to manage and prevent chronic diseases more effectively.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

An Individualized, Data-Driven Digital Approach for Precision Behavior Change. Wongvibulsin, S., Martin, SS., Saria, S., et al.[2023]

Improving self-management skills is crucial for individuals with chronic conditions like diabetes, and having high self-efficacy plays a significant role in this process.

The digital health intervention 'Better Choices, Better Health' effectively enhances self-efficacy and supports self-management in adults with chronic conditions, demonstrating a scalable and affordable approach to improving health outcomes.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Digital Therapeutics: Leading the Way to Improved Outcomes for People With Diabetes. Kaufman, N.[2020]

The use of digital self-management solutions for Type 2 diabetes, such as DM4All and DiaWatch, led to significant improvements in metabolic outcomes, including reductions in HbA1c, body weight, blood pressure, and cholesterol levels over an average follow-up of 8 months in 100 patients.

Compared to a control group, the ProEmpower participants showed significantly better improvements in cardiovascular risk factors, highlighting the efficacy of digital technologies in enhancing diabetes management and promoting healthier lifestyles.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Clinical outcomes of a digitally supported approach for self-management of type 2 diabetes mellitus. De Luca, V., Bozzetto, L., Giglio, C., et al.[2023]

Digital games can effectively support diabetes management for both children and adults by using gamification and social components to motivate behavior change and improve lifestyle habits.

While the paper highlights examples of mobile and desktop games designed for type 1 and type 2 diabetes, it emphasizes the need for more comprehensive research to evaluate their effectiveness across different age groups and patient needs.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Digital games for type 1 and type 2 diabetes: underpinning theory with three illustrative examples. Kamel Boulos, MN., Gammon, S., Dixon, MC., et al.[2022]

A pilot study involving 177 patients demonstrated that a reinforcement learning model could effectively recommend personalized diabetes health coaching interventions, potentially improving clinical outcomes like hemoglobin A1c reduction and quality of life.

The model's recommendations led to better cumulative health outcomes compared to traditional coaching methods, suggesting that AI-driven approaches could enhance the effectiveness of diabetes management interventions.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Optimizing Health Coaching for Patients With Type 2 Diabetes Using Machine Learning: Model Development and Validation Study. Di, S., Petch, J., Gerstein, HC., et al.[2022]

The study involving 70 overweight and obese participants demonstrated that using a conversational AI health coach led to an average weight loss of 2.38% over approximately 15 weeks, indicating its effectiveness in promoting weight loss similar to traditional in-person interventions.

Participants reported a high level of satisfaction with the AI coach, scoring 87 out of 100 on usability, and showed a significant increase in healthy meal choices by 31%, suggesting that AI can effectively support behavior change in diabetes prevention.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

A Fully Automated Conversational Artificial Intelligence for Weight Loss: Longitudinal Observational Study Among Overweight and Obese Adults. Stein, N., Brooks, K.[2020]

References

1. United States pubmed.ncbi.nlm.nih.gov

Virtual Coaching to Enhance Diabetes Care. [2020]

2. England pubmed.ncbi.nlm.nih.gov

A systematic review on incentive-driven mobile health technology: As used in diabetes management. [2022]

3. Switzerland pubmed.ncbi.nlm.nih.gov

Digital Health Coaching for Type 2 Diabetes: Randomized Controlled Trial of Healthy at Home. [2021]

4. United States pubmed.ncbi.nlm.nih.gov

An Individualized, Data-Driven Digital Approach for Precision Behavior Change. [2023]

5. United States pubmed.ncbi.nlm.nih.gov

Digital Therapeutics: Leading the Way to Improved Outcomes for People With Diabetes. [2020]

6. Switzerland pubmed.ncbi.nlm.nih.gov

Clinical outcomes of a digitally supported approach for self-management of type 2 diabetes mellitus. [2023]

7. Canada pubmed.ncbi.nlm.nih.gov

Digital games for type 1 and type 2 diabetes: underpinning theory with three illustrative examples. [2022]

8. Canada pubmed.ncbi.nlm.nih.gov

Optimizing Health Coaching for Patients With Type 2 Diabetes Using Machine Learning: Model Development and Validation Study. [2022]

9. Canada pubmed.ncbi.nlm.nih.gov

A Fully Automated Conversational Artificial Intelligence for Weight Loss: Longitudinal Observational Study Among Overweight and Obese Adults. [2020]