Using Artificial Intelligence to Optimize Delivery of Weight Loss Treatment (ReLearn)

Project ReLearn is testing the efficacy and cost-effectiveness of an Artificial Intelligence system for optimizing weight loss coaching. Participants are randomized to a 1-year weekly gold standard behavioral weight loss remote (video) group treatment or the AI-optimized treatment, which is made up of a combination of remote group treatment, short video call and automated message. In the AI-optimized condition, the system monitors outcomes (via wireless scale, mobile phone app, and wristworn tracker) and, each week, assigns each participant the treatments they have responding to the best, within certain time constraints.

Study Overview

• Status: Active, not recruiting
• Conditions: Obesity; Weight Loss
• Intervention / Treatment: Behavioral: Standard Behavioral Weight Loss Treatment; Behavioral: AI-optimized Behavioral Weight Loss Treatment

• Study Type: Interventional
• Enrollment (Estimated): 336
• Phase: Not Applicable

Contacts and Locations

Study Locations

• United States
  • Pennsylvania
    • Philadelphia, Pennsylvania, United States, 19104
      • Drexel University Center for Weight, Eating and Lifestyle Science

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years to 70 years (Adult, Older Adult)
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

• Individuals must be of overweight or obese BMI (27-50 kg/m)
• Individuals must be adults (aged 18-70)
• Able and willing to engage in the remote program
• Able to engage in physical activity (defined as walking two city blocks without stopping)
• Individuals must also provide consent for the research team to contact their personal physician if necessary, to provide clearance or to consult about rapid weight loss
• Access and willingness to use an Apple or Android smartphone
• Satisfactory completion of all enrollment procedures

Exclusion Criteria:

• Medical condition (e.g., cancer, type I diabetes, psychosis, full-threshold eating disorder) that may pose a risk to the participant during intervention or cause a change in weight
• Currently pregnant, breastfeeding, or planning to become pregnant in the next 12 months
• Recently began or changed the dosage of medication that can cause significant change in weight
• History of bariatric surgery
• Weight loss of > 5% in the previous 3 months

Study Plan

How is the study designed?

Design Details

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

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Active Comparator: BWL-S; 1 year of remote gold standard, small group-based behavioral weight loss treatment with an MS-level clinician.	Behavioral: Standard Behavioral Weight Loss Treatment; Behavioral weight loss treatment is the current gold standard treatment for obesity.
Experimental: BWL-AI; 1 year of remote weight loss treatment made up of a combination of (1) remote small group-based behavioral weight loss sessions, (2) 12-minute individual video calls, (2) automated text messages. An MS-level clinician will deliver the group treatment. Most video calls will be delivered by a paraprofessional coach, but some by an MS-level clinician. Each week the AI system will select one of the interventions for each participant based on which treatment the participant has responded to the best, within certain time constraints.	Behavioral: AI-optimized Behavioral Weight Loss Treatment; AI-optimized Behavioral Weight Loss Treatment will continuously vary intensity (automated text message, videoconference group, individual coaching call) and coach type (paraprofessional or MS-level expert) based on continuously-monitored participant digital data.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Weight change	Weight will be measured using the Fitbit Aria Air wireless scale, which is accurate to 0.2 kg. In order to maximize accuracy, we will (1) provide instructions (e.g., place scale on flat, hard surface; weigh upon waking, without clothes, after using the bathroom), (2) require participants to confirm that they are following instructions at each assessment point, (3) use an average of 5 consecutive daily weights for each timepoint, (4) remove errant weights (e.g., >1 kg change in 1 day).	Baseline, 1-month, 6-month, and 12-month assessment
Costs	All time spent training counselors, participants and supervising counselors, will be tracked by the project coordinator, as will time counselors spend delivering individual and group treatment. The web portal will track counselor time on the portal, e.g., reviewing food records and texting. While the basic time commitments are pre-set by condition, several factors will vary including participant no-shows, participant drop-outs, and counselor adherence to time limits.	Baseline, 1-month, 6-month, and 12-month assessment

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Physical Activity	Minutes of moderate to vigorous physical activity (MVPA) will be measured via Fitbit, a consumer-grade wrist-worn activity tracker which has superior compliance and close-to-equivalent accuracy as a research-grade accelerometer. Number of days meeting MVPA goal will be used by the AI system, with MVPA minutes as a secondary outcome.	Baseline, 1-month, 6-month, and 12-month assessment
Calorie intake	Seven days of calorie intake as derived from the Fitbit app's food log will be used as a secondary outcome. Participants will already be tracking their consumption using the app because data are used by both the coaches (per standard treatment) and the AI system.	Baseline, 1-month, 6-month, and 12-month assessment
Acceptability as Measured by Likert Self-report Scale	Participants will be asked to rate satisfaction with and perceived effectiveness of the program. Counselors will rate effectiveness and ease of use.	6-month and 12-month assessment

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
Self-regulation capacity	Self-regulation capacity will be measured directly using the Brief Self-Control Scale (which has strong retest reliability and internal consistency and predicts weight loss and dietary health).	Baseline and 1-month assessment
Autonomous motivation	Autonomous motivation will be measured by the internally consistent and valid Treatment Self-Regulation Questionnaire.	Baseline and 1-month assessment
Depressive symptoms	Depressive symptoms will be measured with the 20-item Center for Epidemiologic Studies Depression Scale. This self-report scale is widely used and has proven reliable across a variety of demographic groups and in both clinical and nonclinical populations.	Baseline and 1-month assessment
Binge eating	Binge eating will be measured with the Binge Eating Module of the Eating Disorders Examination (EDE), which is a well-validated, reliable measure of eating disorder symptoms that differentiates between individuals with and without Binge Eating Disorder in	Baseline and 1-month assessment

Collaborators and Investigators

• Sponsor: Drexel University
• Investigators: Principal Investigator: Evan M Forman, PhD, Drexel University Center for Weight, Eating and Lifestyle Science

Study record dates

Study Major Dates

• Study Start (Actual): March 22, 2022
• Primary Completion (Estimated): August 31, 2025
• Study Completion (Estimated): March 31, 2026

Study Registration Dates

• First Submitted: January 26, 2022
• First Submitted That Met QC Criteria: February 7, 2022
• First Posted (Actual): February 9, 2022

Study Record Updates

• Last Update Posted (Actual): August 24, 2025
• Last Update Submitted That Met QC Criteria: August 19, 2025
• Last Verified: August 1, 2025

More Information

Terms related to this study

• Keywords: Obesity; Weight Loss; Artificial intelligence; Behavioral treatment
• Additional Relevant MeSH Terms: Body Weight Changes; Weight Loss; Body Weight
• Other Study ID Numbers: R01DK12564

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No