Sharing Digital Self-Monitoring Data With Others to Enhance Long-Term Weight Loss (FitLink)

Sharing Digital Self-Monitoring Data With Others to Enhance Long-Term Weight Loss: A Randomized Trial Using a Factorial Design

Adults attempting weight loss through lifestyle modification (LM) typically find maintenance of behavior change difficult. Outcomes might be improved if participants are provided with sustained sources of accountability and support and ongoing opportunities to reflect with others on goal progress. This study proposes that sharing digital data with other parties has the potential to improve long-term weight loss. The proposed study will enroll adults ("index participants") (N = 320) with overweight/obesity in a 24-month LM program and instruct them to use digital tools for self-monitoring of weight, physical activity, and eating on a daily basis.

Adults attempting weight loss through lifestyle modification (LM) typically find maintenance of behavior change difficult. Outcomes might be improved if participants are provided with sustained sources of accountability and support and ongoing opportunities to reflect with others on goal progress. This study proposes that sharing digital data with other parties has the potential to improve long-term weight loss. The proposed study will enroll adults ("index participants") (N = 320) with overweight/obesity in a 24-month LM program and instruct them to use digital tools for self-monitoring of weight, physical activity, and eating on a daily basis. Groups will meet remotely each week in months 1-3 to initiate weight loss. In months 4-24, intervention contact will be remote and will include the following: quarterly group meetings held via videoconference; brief phone calls with the coach held twice per quarter; and monthly text messages with the coach, with a small group of fellow group participants, and with a friend or family member outside of the program. Text message content will vary according to treatment condition, i.e. three types of data sharing partnerships: Coach Share, Group Share, and Friend/Family Share. Outcomes will be measured at months 0, 3, 6, 12, and 24. Primary aims include determining if Coach Share, Group Share, and Friend/ Family Share each improve long-term weight loss, PA, and calorie intake, and examining if effects are additive when participants are assigned to engage in more than one type of data sharing partnership. The secondary aim is to explore mediators and moderators of intervention effects. The tertiary aim is to determine how the quality of relationship with coach, group members, and friends/family member over time may differentially moderate treatment effects and predict data sharing efficacy. As digital technology makes data sharing increasingly feasible, it is critical to determine how to optimize these partnerships to improve long-term outcomes in LM.

If an index participant is randomized to have Coach Share ON, the coach will have continuous access to objective, real-time, detailed device data, and address data directly in all individual interactions with index participants (groups, individual weigh-ins, phone calls, texts), conveying that they are closely monitoring progress.

If Group Share is ON, participants in a given LM group will view and respond to aspects of each other's self-monitoring device data in their small-group text messages. Each small group participant's data will be highlighted one day of the month, and content from the Group Share Off message library also will be sent.

Index participants who are randomized to have Friend/Family Share ON will give their friend or family member access to their self-monitoring data. Text messages sent by the program bot, which includes the index participant and friend/family member, will use a similar format as described in Group Share ON. The friend or family member will be trained to respond to data in ways that enhance the participant's motivation and self-regulation.

Among index participants who are randomized to have Coach Share OFF, the coach will have no access to their device data. Index participants will be explicitly told their coach does not have data access. During groups and phone calls, the participant will self-report progress, and goal setting will be based on that self-report. Text messages from the coach will be personalized in the OFF condition, as they are when coach data sharing is ON, though these messages will not be informed by device data. Instead, text messages will provide a reminder of key goals and strategies discussed in the previous phone call and provide encouragement for participant efforts.

Among index participants randomized to have Group Share OFF, supportive accountability will not be the target of participant interactions and no data will be shared with LM program peers. Instead, once weekly prompts for small group text exchanges will encourage index participants to provide social support (primarily of the informational type) to each other. Index participants will be explicitly told that they should not include information on goal progress such as weight change or minutes of PA.

If an index participant is randomized to Friend/Family Share OFF, the text messages from the bot will prompt the participant and friend/family member to communicate about the LM program; those messages will be standardized, except for names, and will not be tailored to index participant progress. Index participants will be instructed to refrain from sharing any device data with friends or family members.

Weight will be measured using a wireless scale. 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) remove errant weights (e.g., >1 kg change in 1 day).

Minutes per week of moderate-to-vigorous PA will be measured by using FitBit. Steps per day from the Fitbit also will be examined.

Average calorie intake over 7 days will be derived from the Fitbit digital food log as a secondary outcome.

(mediator, moderator) will be measured with the Supportive Accountability Measure and the Supportive Accountability Inventory

(Mediator, moderator) will be measured with items, adapted from previous research, that ask index participants to rate how helpful each program component is reflecting on how their progress compares to their goals, planning short-term behaviors and strategy use, and evaluating if current goals should be adjusted.

(Mediator, moderator) will be objectively measured as the number of days in which digital devices were used to measure weight, record eating, or PA.

(Mediator, moderator) will be measured with the Iowa-Netherlands Comparison Orientation Measure, with items modified to be focused on the lifestyle modification group.

(Mediator, moderator) assessed with the Social Support for Healthy Behaviors Measure. Index participants will report the frequency with which friends and family members engaged in behaviors that support or undermine healthy eating and physical activity in the past month.

(Moderator, moderator) will be measured with items adapted from the client version of the Working Alliance Inventory-Bond Subscale. Ratings will be completed separately for the relationship with the coach, the lifestyle modification group, and the designated friend/family support person, yielding scores for each party

(moderator) will be measured with items, adapted from previous research, that assess for weight self-stigma.

(mediator, moderator) Items to characterize the friend/family member relationship, such as their age, gender, body size, perceived healthfulness of diet, interest in weight loss, frequency of communication, comfort discussion lifestyle modification with them, reactions to weight-related communication

(mediator, moderator) Items to characterize the small group dynamic, including frequency of communication and reactions to communication

(moderator) Items to characterize the participant, including age, sex, race, ethnicity, education level, household composition

(moderator) Coach level of behavioral weight loss experience (# of sessions previously conducted) and level of education will be calculated

(moderator) Whether text messages were delivered to app (Wave 1 and 2) or via SMS (Wave 3 and 4) will be examined to determine if this impacts intervention efficacy

Inclusion Criteria: Individuals must be of overweight or obese BMI (27-50 kg/m2) Individuals must be adults (aged 18-70) Able to engage in physical activity (defined as walking two city blocks without stopping) Access and willingness to use a smartphone and internet Has one adult friend or family member who indicates willingness to serve in a support role Satisfactory completion of all enrollment procedures English Speaking Exclusion Criteria: Medical or psychiatric 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, or limit ability to comply with the program or participate appropriately in group-based treatment Currently pregnant or breastfeeding, or planning to become pregnant in the next 24 months Use of insulin or a medication that can cause significant change in weight History of bariatric surgery Weight loss of greater than or equal to 5% in the previous 3 months Weight loss of greater than or equal to 10% in the past 3 years that is currently being maintained