Type 2 DiabetesNCT07116902
Artificial Intelligence-based Methods to Predict Disease Progression in Youth With Type 2 Diabetes
A study testing phone application in people with Type 2 Diabetes, run by University of California, San Francisco at 2 sites in United States.
Trial summary
Currently, clinicians are unable to predict a patient's risk of long-term disease progression and development of a long-term complication based on the data that is available to them. Read more about this study
- Who is it for?
- Ages 10 to 21 years
- Conditions the study is for
- Type 2 Diabetes
- What is being tested?
- phone application
- Type of treatment
- Medical device
- Could I receive a placebo?
- Yes, some participants may receive a placebo
- How many people
- 50 participants
- Where
- 2 sitesUnited States
- Study dates
- Apr 2026to Sep 2026
- Phase
- Interventional
- Are there results?
- Not yet: the study is under way
Key requirements
Who can take part, as listed by the sponsor. Only the study team can confirm whether you are eligible, so take this list to your doctor.
- Ages 10 to 21 years
- All sexes
- No healthy volunteers
- Child, Adult
Who may be able to join
- Age 10- 21 years
- Diagnosis of T2D based on clinical diagnosis or ICD 9 and 10 codes
- Duration of T2D ≥ 3 months
- HbA1C ≥ 7% which is the target HbA1C recommended by the American Diabetes Association
- Stable medication regimen (No medication changes and no change in basal insulin dose by more than 20% in the 2 weeks prior to enrollment)
- Ability to wear CGM for a total of 6 weeks while in the study.
Show all 9Show fewer
- English or Spanish speakers.
- Willing to abide by recommendations and study procedures.
- Willing and able to sign the Informed Consent Form (ICF) and/or has a parent or guardian willing and able to sign the ICF.
Who cannot join
- Pancreatic autoantibody positivity (GAD-65, insulin, IA-2, ICA 512, ZnT8).
- Plan for undergoing bariatric surgery during the study period
- Anticipated use of systemic glucocorticoids during the study period
- Unable to stop taking more than 500mg/day of Vitamin C during the study period as this may affect the sensor readings.
- Presence of a condition or abnormality that in the opinion of the Investigator would compromise the safety of the patient or the quality of the data.
- Presence of a condition or abnormality that in the opinion of the Investigator would cause repeated hospitalizations or significant changes in medications.
Trial locations and contacts
2 sites in 1 country. Contact the study team or a site near you to ask about taking part.
United States2 sites
UCSF Benioff Children's Hospital Oakland, Pediatric Diabetes Clinic
Oakland, California, 94609
Shylaja A Srinivasan, MD, 415-353-9084, shylaja.srinivasan@ucsf.edu
Laura A Dapkus Humphries, NCPT, 628-224-8364, laura.dapkus@ucsf.edu
UCSF Benioff Children's Hospital San Francisco, Madison Clinic for Pediatric Diabetes
San Francisco, California, 94158
Shylaja A Srinivasan, MD, 415-353-9084, shylaja.srinivasan@ucsf.edu
Laura A Dapkus Humphries, NCPT, 628-224-8364, laura.dapkus@ucsf.edu
About this study
Official title: Artificial Intelligence-based Methods to Predict Disease Progression in Youth With Type 2 Diabetes: A Digital Twin Study
Currently, clinicians are unable to predict a patient's risk of long-term disease progression and development of a long-term complication based on the data that is available to them. The first aim of this is to develop and validate an Artificial Intelligence (AI) powered prediction model for Type 2 Diabetes (T2D) disease progression using existing data from previously collected studies and real-world electronic health medical data. Investigators will use clinical, pharmacologic, and genomic factors to develop the prediction model based on the most relevant clinical outcomes of change in Hemoglobin A1c (HbA1c) and the development of a microvascular complication.
Despite the availability of newer medication options, lifestyle intervention is not effective in most youth and current therapeutic options are ineffective at producing sustained glycemic control. Newer and innovative methods are needed to identify the youth at highest risk of progression in terms of increase in HbA1c and development of long-term complications and to motivate behavioral change in youth. The goal of this aim is to create an AI-powered digital twin model for 50 youth with T2D using their baseline clinical, genetic, pharmacologic and lifestyle data and utilize AI algorithms developed in Aim 1 to simulate disease progression and treatment response. Investigators will then evaluate the digital twin model in an randomized controlled trail and prospectively compare the generated digital twin data to observed values over one year. Investigators will also measure whether knowledge of the digital twin prediction with targeted healthcare recommendations influence medication and lifestyle change adherence in the digital twin arm (n= 25) compared to the control arm (n= 25).
- Conditions
- Type 2 Diabetes
- Being tested
- phone application, Standard of Care (SOC)
- Keywords
- T2D, Type 2 Diabetes, AI, Artificial Intelligence, CGM
- Medical subject headings
- Diabetes Mellitus, Type 2
What participants receive
What participants receive. In a randomised study each person is assigned to one group by chance.
phone applicationDevice
Participants in the digital twin arm will receive information on their disease progression which will be based on projected change in HbA1C in alternative realities and specific recommendations on medication dosing and lifestyle changes based on this data. The digital twin information will be presented on an iPad in a game- like manner. The alternate realities will include scenarios of change in medication adherence, physical activity metrics, dietary changes etc.
Standard of Care (SOC)Other
Participants in the control arm will receive standard of care which is medication change recommendations based on HbA1C and blood glucose values every 3 months and standard lifestyle education.
| Study group | Role | What they receive |
|---|---|---|
| Digital twin arm | Experimental | Participants in the digital twin arm will receive information on their disease progression which will be based on projected change in HbA1C in alternative realities and specific recommendations on medication dosing and lifestyle changes based on this data. The digital twin information will be presented on an iPad in a game- like manner. The alternate realities will include scenarios of change in medication adherence, physical activity metrics, dietary changes etc. Receives: Device: phone application |
| Control arm | Placebo comparator | Participants in the control arm will receive standard of care which is medication change recommendations based on HbA1C and blood glucose values every 3 months and standard lifestyle education. Receives: Other: Standard of Care (SOC) |
What is being measured
The primary outcome is the main question the study is designed to answer.
Change in HbA1C
The primary outcome will be the ability of the digital twin model to accurately predict longitudinal disease progression measured as the digital twin predicted HbA1C versus measured HbA1C and the difference in HbA1C between the digital twin arm and control arms.
Measured: From enrollment to the close out visit at the 1-year mark
Secondary outcomes (10)
Sleep Quality
Change in measures related to sleep quality. (Actigraph measured duration, Insomnia Severity Index (ISI)).
Respondents rate each element of the questionnaire using Likert-type scales. Responses can range from 0 to 4, where higher scores indicate more acute symptoms of insomnia.
Scores are tallied and can be compared both to scores obtained at a different phase of treatment and to the scores of other individuals.
A total score of:
0-7 indicates "no clinically significant insomnia" 8-14 means "sub-threshold insomnia" 15-21 means "clinical insomnia (moderate severity)" 22-28 means "clinical insomnia (severe)"
Measured: From time of enrollment to the study close out visit at the 1-year mark
Psycho-Social Outcome
Change in measures related to psycho-social outcomes for Quality of Life. (PedsQL-Diabetes Module).
The Pediatric Quality of Life (PedsQL) Diabetes Module scores quality of life in children and adolescents with diabetes using a 0-100 scale, where higher scores indicate better quality of life. The PedsQL Diabetes Module uses a 5-point response scale (0-4) for each item.
It assesses various aspects of life affected by diabetes, including physical symptoms, treatment barriers, and emotional well-being.
Measured: From time of enrollment to the study close out visit at the 1-year mark
Sugar Intake
Change in measures related to sugar intake (g/day).
Measured: From time of enrollment to the study close out visit at the 1-year mark
Physical Activity
Change in measures related to physical activity (days/week)
Measured: From time of enrollment to the study close out visit at the 1-year mark
BMI
Change in BMI (BMI z-score).
Measured: From time of enrollment to the study close out visit at the 1-year mark
CGM Time In Range
Change in CGM based measures including time in the target glucose range of 70 to 180 mg/dL.
Measured: From time of enrollment to the study close out visit at the 1-year mark
CGM Time Above Range
Change in CGM based measures including time above the target glucose range of greater than 250 mg/dL for glucose and co-efficient of variation of glucose.
Measured: From time of enrollment to the study close out visit at the 1-year mark
Psycho-Social Outcome
Change in measures related to psycho-social outcomes for Diabetes Distress. (Diabetes Distress Scale).
Diabetes Distress Scale (DDS) is used to assess the level of emotional distress related to managing diabetes.
- Individual Item Score: Each item on the DDS is rated on a 6-point scale.
- Minimum: 1 (not a problem)
- Maximum: 6 (a very significant problem)
- Total Score: The DDS yields an overall distress score calculated by averaging the responses to the individual items.
- Minimum: 1
- Maximum: 6
- Interpretation of Scores:
- < 2.0: Little or no distress
- 2.0-2.9: Moderate distress
- ≥ 3.0: High distress
Measured: From time of enrollment to the study close out visit at the 1-year mark
Psycho-Social Outcome
Change in measures related to psycho-social outcomes for Overall QOL and Wellbeing using the Patient-Reported Outcomes Measurement Information System (PROMIS) Global Health questionnaire.
- PROMIS Global Health is a broader measure of overall physical and mental health.
- PROMIS Global Health scores are reported as T-scores, with a mean of 50 and a standard deviation of 10 in the US general population.
- Higher T-scores indicate better physical and mental health. A score of 60 is one standard deviation above the mean, while 40 is one standard deviation below the mean.
Measured: From time of enrollment to the study close out visit at the 1-year mark
Psycho-Social Outcome
Change in measures related to psycho-social outcomes for Overall QOL and Wellbeing using the World Health Organization- Five (WHO-5) questionnaire.
- Measures well-being, specifically positive psychological states.
- It consists of five statements relating to the past two weeks.
- Aims to assess the subjective experience of well-being, capturing feelings of positive mood, energy, and interest. A short, 5-item questionnaire.
- Typically uses a 6-point Likert scale for each item, with higher scores indicating better well-being.
Measured: From time of enrollment to the study close out visit at the 1-year mark
Timeline
From the start of the study to its completion; the mark is today. Estimated dates are the sponsor’s plan.
- First posted12 Aug 2025
- Study startApr 2026 (estimated)
- Primary completionSep 2026 (estimated)
- Study completionSep 2026 (estimated)
Resources
- Official record on ClinicalTrials.govThe registry entry this page is built from, with its full change history.
- Other trials for Type 2 DiabetesStudies of the same condition, open and under way.
- More trials run by University of California, San FranciscoThe sponsor's other registered studies.
- What the trial phases meanHow a treatment moves from first tests in people to approval.
More details
For doctors and researchers: how the study is designed, who runs it, and its registry IDs.
Study design
- Study type
- Interventional
- Phase
- Phase N/A
- Primary purpose
- Diagnosis
- Allocation
- Randomised
- Intervention model
- Parallel groups
- Masking
- None (open label)
- Enrolment
- 50 participants (estimated)
Sponsor and investigators
- Lead sponsor
- University of California, San Francisco
- Sponsor type
- University, hospital or other
- Collaborators
- Stanford University, American Diabetes Association
- Responsible party
- Sponsor
- Investigators
- Shylaja A Srinivasan, MD, Principal investigator, University of California, San Francisco
IDs, dates and publications
- ClinicalTrials.gov ID
- NCT07116902
- Sponsor's protocol ID
- 24-42259
- Other IDs
- #7-24-ICTST2DY-05
- Status last verified
- Dec 2025
- First submitted
- 13 Jun 2025
- Last updated
- 4 Dec 2025
- Results posted
- No
- Data monitoring committee
- No
- FDA-regulated drug
- No
- FDA-regulated device
- Yes
- Individual data sharing
- No
Study at a glance
From the registry record, last updated by the sponsor on 4 Dec 2025. Recruitment status is as the sponsor reported it: confirm with the study team.
- Official title
- Artificial Intelligence-based Methods to Predict Disease Progression in Youth With Type 2 Diabetes: A Digital Twin Study
- ClinicalTrials.gov ID
- NCT07116902
- Acronym / sponsor ID
- 24-42259
- Conditions
- Type 2 Diabetes
- Keywords
- T2D, Type 2 Diabetes, AI, Artificial Intelligence, CGM
- Study type
- Interventional
- Recruitment status
- Not yet recruiting
- Enrolment
- 50 participants (estimated)
- Lead sponsor
- University of California, San FranciscoUniversity, hospital or other
- Collaborators
- Stanford University, American Diabetes Association
- Study start
- Apr 2026 (estimated)
- Primary completion
- Sep 2026 (estimated)
- Study completion
- Sep 2026 (estimated)
- First posted
- 12 Aug 2025 (actual)
- Results
- Not posted in the source record
Clinical summary
The research question as the sponsor registered it. This describes what the study plans to test, not what it found.
Currently, clinicians are unable to predict a patient's risk of long-term disease progression and development of a long-term complication based on the data that is available to them. The first aim of this is to develop and validate an Artificial Intelligence (AI) powered prediction model for Type 2 Diabetes (T2D) disease progression using existing data from previously collected studies and real-world electronic health medical data. Investigators will use clinical, pharmacologic, and genomic factors to develop the prediction model based on the most relevant clinical outcomes of change in Hemoglobin A1c (HbA1c) and the development of a microvascular complication.
Despite the availability of newer medication options, lifestyle intervention is not effective in most youth and current therapeutic options are ineffective at producing sustained glycemic control. Newer and innovative methods are needed to identify the youth at highest risk of progression in terms of increase in HbA1c and development of long-term complications and to motivate behavioral change in youth. The goal of this aim is to create an AI-powered digital twin model for 50 youth with T2D using their baseline clinical, genetic, pharmacologic and lifestyle data and utilize AI algorithms developed in Aim 1 to simulate disease progression and treatment response. Investigators will then evaluate the digital twin model in an randomized controlled trail and prospectively compare the generated digital twin data to observed values over one year. Investigators will also measure whether knowledge of the digital twin prediction with targeted healthcare recommendations influence medication and lifestyle change adherence in the digital twin arm (n= 25) compared to the control arm (n= 25).
Study design
- Primary purpose
- Diagnosis
- Allocation
- Randomised
- Intervention model
- Parallel groups
- Masking
- None (open label)
- Number of arms
- 2
| Arm | Type | Receives |
|---|---|---|
| Digital twin arm | Experimental | Device: phone applicationParticipants in the digital twin arm will receive information on their disease progression which will be based on projected change in HbA1C in alternative realities and specific recommendations on medication dosing and lifestyle changes based on this data. The digital twin information will be presented on an iPad in a game- like manner. The alternate realities will include scenarios of change in medication adherence, physical activity metrics, dietary changes etc. |
| Control arm | Placebo comparator | Other: Standard of Care (SOC)Participants in the control arm will receive standard of care which is medication change recommendations based on HbA1C and blood glucose values every 3 months and standard lifestyle education. |
Who may qualify
- Ages
- 10 to 21 years
- Sex
- All
- Healthy volunteers
- Not accepted
- Age groups
- Child, Adult
Key inclusion criteria (9 listed)
- Age 10- 21 years
- Diagnosis of T2D based on clinical diagnosis or ICD 9 and 10 codes
- Duration of T2D ≥ 3 months
- HbA1C ≥ 7% which is the target HbA1C recommended by the American Diabetes Association
- Stable medication regimen (No medication changes and no change in basal insulin dose by more than 20% in the 2 weeks prior to enrollment)
Key exclusion criteria (6 listed)
- Pancreatic autoantibody positivity (GAD-65, insulin, IA-2, ICA 512, ZnT8).
- Plan for undergoing bariatric surgery during the study period
- Anticipated use of systemic glucocorticoids during the study period
- Unable to stop taking more than 500mg/day of Vitamin C during the study period as this may affect the sensor readings.
- Presence of a condition or abnormality that in the opinion of the Investigator would compromise the safety of the patient or the quality of the data.
Eligibility is decided by the study team after screening; this overview does not establish that a patient qualifies. Full inclusion and exclusion criteria.
Interventions
| Intervention | Type | Arms |
|---|---|---|
| phone applicationParticipants in the digital twin arm will receive information on their disease progression which will be based on projected change in HbA1C in alternative realities and specific recommendations on medication dosing and lifestyle changes based on this data. The digital twin information will be presented on an iPad in a game- like manner. The alternate realities will include scenarios of change in medication adherence, physical activity metrics, dietary changes etc. | Device | Digital twin arm |
| Standard of Care (SOC)Participants in the control arm will receive standard of care which is medication change recommendations based on HbA1C and blood glucose values every 3 months and standard lifestyle education. | Other | Control arm |
Primary outcome
Change in HbA1C
Time frame: From enrollment to the close out visit at the 1-year mark
The primary outcome will be the ability of the digital twin model to accurately predict longitudinal disease progression measured as the digital twin predicted HbA1C versus measured HbA1C and the difference in HbA1C between the digital twin arm and control arms.
10 secondary outcomes
Sleep Quality
Time frame: From time of enrollment to the study close out visit at the 1-year mark
Psycho-Social Outcome
Time frame: From time of enrollment to the study close out visit at the 1-year mark
Sugar Intake
Time frame: From time of enrollment to the study close out visit at the 1-year mark
Physical Activity
Time frame: From time of enrollment to the study close out visit at the 1-year mark
BMI
Time frame: From time of enrollment to the study close out visit at the 1-year mark
CGM Time In Range
Time frame: From time of enrollment to the study close out visit at the 1-year mark
CGM Time Above Range
Time frame: From time of enrollment to the study close out visit at the 1-year mark
Psycho-Social Outcome
Time frame: From time of enrollment to the study close out visit at the 1-year mark
Psycho-Social Outcome
Time frame: From time of enrollment to the study close out visit at the 1-year mark
Psycho-Social Outcome
Time frame: From time of enrollment to the study close out visit at the 1-year mark
Results and safety
This study is still under way. Results are not available in the source record yet.
Research links
- Official record on ClinicalTrials.govLast updated by the sponsor 4 Dec 2025.
- Individual participant dataSharing planned: No
Related trials for Type 2 Diabetes
Not yet recruiting
SWitching to Appropriate Prescription in Cardiovascular-Kidney-Metabolic Syndrome
Cardiovascular-kidney-metabolic Syndrome, Diabete Type 2
Active, not recruiting
Pilot of Continuous Glucose Monitor-Augmented Food Is Medicine
Nutrition, Type 2 Diabetes
Recruiting
Whole Food for Families: A Pilot RCT of a Dietary Guidelines-Based Intervention to Prevent Type 2 Diabetes
PreDiabetes, Diet, Healthy
Not yet recruiting
Safety and Efficacy of the Fully Closed Loop iLet Bionic Pancreas in Adults With Type 1 and Type 2 Diabetes
Diabetes (Insulin-requiring, Type 1 or Type 2)
Recruiting
6-Weeks of Blood Flow Restriction Training on Immune Cell Function and Bioenergetics
Type 2 Diabetes
Referral and contact
AzTrials does not enrol participants or take referrals. Contact the study team or a site listed below; they decide who can take part.
Shylaja A Srinivasan, MD principal investigator
University of California, San Francisco
2 sites in 1 country: see every site and its status.
Questions about this trial
What is this clinical trial studying?
A study testing phone application in people with Type 2 Diabetes, run by University of California, San Francisco at 2 sites in United States.
The full registered description is in the About this study section and the official record.
Who may be eligible to take part?
The sponsor lists 9 inclusion and 6 exclusion criteria. Eligibility is decided by the study team after screening, so read the full criteria and contact a listed site.
Is this clinical trial currently recruiting?
The record says: opening soon, as of the sponsor's last update on 4 Dec 2025. Status can change between updates, so confirm with the study contact before making plans.
Where is the study taking place?
The record lists 2 sites in United States. Whether a site is taking part right now should be confirmed with that site. See every site.
What are the possible benefits and risks?
They depend on the study and the treatment, and no study can promise a benefit. Some participants in this study may receive a placebo. The study team explains the known risks during informed consent, before anyone joins; the record itself does not list them.
Are results available for this study?
This study is still under way. Results are not available in the source record yet.
How can a doctor refer a patient or contact the research team?
Contact Avani A Narayan, MS (415-530-8047, avani.narayan@ucsf.edu), or a site near the patient. AzTrials does not enrol participants or take referrals.
What does the primary outcome measure?
Change in HbA1C (time frame: From enrollment to the close out visit at the 1-year mark). It is what the study was designed to measure, not a finding.
Where can I read the original study information?
Does a listing mean the treatment is approved or proven to work?
No. A listing means the study is registered. It does not show that the treatment is safe, works, or is approved by a regulator.
How often is this information updated?
AzTrials copies changed records from ClinicalTrials.gov every night and reloads the whole registry every week. The sponsor last updated this record on 4 Dec 2025.