Piloting a Reinforcement Learning Tool for Individually Tailoring Just-in-time Adaptive Interventions

Obesity Clinical Trial

Official title:

Piloting a Reinforcement Learning Tool for Individually Tailoring Just-in-time Adaptive Interventions

Clinical Trial Details — Status: Not yet recruiting

Administrative data

• NCT number: NCT05751993
• Other study ID #: 22-0149
• Secondary ID: R21CA260092
• Status: Not yet recruiting
• Phase: N/A
• Start date: September 2024
• Est. completion date: February 2025

Study information

• Verified date: June 2024
• Source: UNC Lineberger Comprehensive Cancer Center
• Contact: Brooke Nezami, PhD, MA
• Phone: 919-966-5852
• Email: bnezami[@]unc.edu
• Is FDA regulated: No
• Study type: Interventional

Clinical Trial Summary

The purpose of this pilot study is to conduct a 12-week pilot feasibility study testing usability of a reinforcement learning model (AdaptRL) in a weight loss intervention (ADAPT study). Building upon a previous just-in-time adaptive intervention (JITAI), a reinforcement learning model will generate decision rules unique to each individual that are intended to improve the tailoring of brief intervention messages (e.g., what behavior to message about, what behavior change techniques to include), improve achievement of daily behavioral goals, and improve weight loss in a sample of 20 adults.

Description:

Reinforcement Learning (RL), a type of machine learning, holds promise for addressing the limitations of previous approaches to implementing JITAIs. Adaptive RL applications work by updating information about expected "rewards" (i.e., proximal outcomes) based on the results of sequentially randomized trials. To realize the potential of adaptive interventions to reduce health disparities in cancer prevention and control, mHealth interventionists first need to identify methods of using digital health participant data to continually adapt decision rules guiding highly tailored intervention delivery. This research team has developed a reinforcement learning model (AdaptRL) that reads in and analyzes user data (e.g., calories, weight, and activity data from Fitbit) in real-time, uses RL to efficiently determine which message a participant should receive up to 3 times per day, and creates a JITAI tailored to optimize daily behavioral goal achievement and weight loss for each participant. The objective of this study is to test the feasibility of using this reinforcement learning model in a pilot weight loss study.

Recruitment information / eligibility

• Status: Not yet recruiting
• Enrollment: 20
• Est. completion date: February 2025
• Est. primary completion date: February 2025
• Accepts healthy volunteers: Accepts Healthy Volunteers
• Gender: All
• Age group: 18 Years to 55 Years

Eligibility

Inclusion Criteria:

1. Age 18-55 years

2. Body Mass Index of 25-40 kg/m2

3. English-speaking and writing

4. Has a smartphone with a data and text messaging plan

Exclusion Criteria:

1. Currently participating in a weight loss, nutrition, or physical activity study or program or other study that would interfere with this study

2. Currently using prescription medications with known effects on appetite or weight (e.g., oral steroids, weight loss medications), with the exception of individuals on a stable dose of SSRIs for 3 months)

3. Previous surgical procedure for weight loss or planned weight loss surgery in the next year

4. Currently pregnant or planning pregnancy in the next 4 months

5. Lost 10 or more pounds and kept it off in the last 6 months

6. Report a heart condition, chest pain during periods of activity or rest, or loss of consciousness on the Physical Activity Readiness Questionnaire (PAR-Q; items 1-4). Individuals endorsing joint problems, prescription medication usage, or other medical conditions that could limit exercise will be required to obtain written physician consent to participate

7. Pre-existing medical condition(s) that preclude adherence to an unsupervised exercise program, diabetes treated with insulin, history of heart attack or stroke, current treatment for cancer, or inability to walk for exercise

8. Type 1 diabetes or currently receiving medical treatment for Type 2 diabetes

9. Other health problems which may influence the ability to walk for physical activity or be associated with unintentional weight change, including cancer treatment within the past 5 years or tuberculosis

10. Health or psychological diagnoses that preclude participation in a prescribed dietary and exercise program, including history of or diagnosis of schizophrenia or bipolar disorder, hospitalization for a psychiatric diagnosis in the past year, a current diagnosis of alcohol or substance abuse

11. Report a past diagnosis of or receiving treatment for a DSM-5-TR eating disorder (anorexia nervosa, bulimia nervosa, or other diagnosis)

12. Moving out of the area in the next 4 months

13. Out of town for a week or more during the study period

14. Another member of the household is a participant or staff member in this trial

15. Not willing to attend two study visits

16. Not willing to wear a Fitbit every day

17. Reason to suspect that the participant would not adhere to the study intervention

18. Have participated in another study conducted by the UNC Weight Research Program within the past 12 months

Related Conditions & MeSH terms

• Obesity
• Overweight
• Overweight and Obesity

Intervention

Behavioral:
• ADAPT
The intervention is testing the feasibility of a reinforcement learning model to pull in participants' behavioral data (calories, activity, and weight) and use this data along with participants' past behavioral goal achievements to deliver the type of message that should be most effective for a given participant at a given time. At each decision point (morning, midday, and evening on a daily basis), the system evaluates which behaviors a participant is eligible to receive a message about (eating, activity, self-weighing), which intervention options a participant is eligible to receive, and then chooses what type of behavioral message a participant should receive. Over time, the model uses participant data and response to interventions to better tailor message choice.

Locations

Country	Name	City	State
United States	University of North Carolina at Chapel Hill	Chapel Hill	North Carolina

Sponsors (4)

Lead Sponsor	Collaborator
UNC Lineberger Comprehensive Cancer Center	Duke University, National Cancer Institute (NCI), RTI International

Country where clinical trial is conducted

United States, 

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	Feasibility (success of using the AdaptRL model)	Feasibility as the success of using the AdaptRL model will be defined as the mean number of messages delivered per participant per day.	up to 12 weeks
Primary	Study engagement	Study engagement will be defined as the percent of person-days in which participants accessed the web app.	up to 12 weeks
Primary	Self-monitoring adherence	Self-monitoring adherence will be defined as the percent of person-days in which participants tracked at least one weight loss behavior (tracked calories, wore tracker, or self-weighed).	up to 12 weeks
Secondary	Message satisfaction	Message satisfaction will be defined as the percent of delivered messages that were rated as "liked" (compared to dislike or not rated).	up to 12 weeks
Secondary	Percent weight loss	Percent weight loss will be defined as weight change from baseline to 12 weeks calculated as a percent from baseline weight.	12 weeks
Secondary	Moderate-to-vigorous physical activity	Moderate-to-vigorous physical activity will be defined as the change in self-reported weekly minutes of moderate-to-vigorous physical activity as measured by the Paffenbarger Activity Questionnaire from baseline to 12 weeks. The minimum is 0, no maximum. Higher numbers represent higher minutes of weekly moderate-to-vigorous physical activity.	Baseline, 12 weeks
Secondary	Dietary intake	Dietary intake will be defined as the change in average daily calorie intake as measured by the Automated Self-Administered 24-hour (ASA 24-hour) dietary recalls from baseline to 12 weeks. Daily caloric intake is measured in kcals, with higher numbers indicating higher caloric intake.	Baseline, 12 weeks
Secondary	Adherence to calorie goal	Adherence to the calorie goal as the percent of person-days in which participants tracked their calories and stayed at or under their calorie goal will be measured by dietary self-monitoring data tracked in the Fitbit app and transmitted via Application Programming Interface (API) to study servers.	up to 12 weeks
Secondary	Adherence to daily active minutes goal	Adherence to daily active minutes goal, the percent of person-days in which participants met their daily active minute goal, will be measured by activity tracker data tracked in the Fitbit app and transmitted via Application Programming Interface (API) to study servers.	up to 12 weeks
Secondary	Adherence to daily self-weighing	Adherence to daily self-weighing, the percent of person-days in which participants self-weighed will be measured by Fitbit smart scales and transmitted via Application Programming Interface (API) to study servers.	up to 12 weeks
Secondary	Adherence to daily self-weighing at the participant-day level	Adherence to daily self-weighing at the participant-day level, the percent of person-days weighed after the message randomization time until the end of the day will be measured by Fitbit smart scales and transmitted via Application Programming Interface (API) to study servers.	up to 12 weeks
Secondary	Adherence to the daily self-weighing percent of person-days weighed	Adherence to the daily self-weighing percent of person-days weighed the day after the message randomization will be measured by Fitbit smart scales and transmitted via Application Programming Interface (API) to study servers.	up to 12 weeks
Secondary	Achievement of active minutes goal	Achievement of active minutes goal, percent of person-days met active minutes goal after the message randomization time until the end of the day will be measured by activity tracker data tracked in the Fitbit app and transmitted via Application Programming Interface (API) to study servers.	up to 12 weeks
Secondary	Achievement of active minutes goal percent of person-days	Achievement of active minutes goal percent of person-days met active minutes goal the day after the message randomization will be measured by activity tracker data tracked in the Fitbit app and transmitted via Application Programming Interface (API) to study servers.	up to 12 weeks
Secondary	Achievement of calorie goal (at or under goal)	Achievement of calorie goal (at or under goal) percent of person-days met calorie goal after the message randomization time until the end of the day will be measured by dietary self-monitoring data tracked in the Fitbit app and transmitted via Application Programming Interface (API) to study servers.	up to 12 weeks
Secondary	Achievement of calorie goal (at or under goal) percent of person-days	Achievement of calorie goal (at or under goal) percent of person-days met calorie goal the day after the message randomization will be measured by dietary self-monitoring data tracked in the Fitbit app and transmitted via Application Programming Interface (API) to study servers.	up to 12 weeks

External Links

•   University of North Carolina Lineberger Comprehensive Cancer Center Clinical Trials