Get Social: Randomized Trial of a Social Network Delivered Lifestyle Intervention

Worcester Polytechnic Institute, National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)

The present study is a non-inferiority trial comparing the efficacy of a lifestyle intervention delivered entirely via an online social network to a traditional lifestyle intervention delivered via group meetings.

Lifestyle interventions have had established efficacy for over a decade but are still not widely disseminated, largely due to high cost and patient and provider burden. Online social networks are an alternative way to deliver lifestyle counseling and delivery via this modality may virtually eliminate patient visits, the main source of cost and burden in traditional modalities. Interactions in online social networks are frequent, brief, and asynchronous because users login to their online communities during downtime during work and leisure time, or when they simply feel a need for social connection. As such, social media becomes embedded into people's daily lives. This provides an opportunity to embed health behavior change programming into people's daily lives. Thus far in the literature, existing online social networks have been used as component of web- or mobile app-based lifestyle interventions but not as the primary modality for intervention delivery. The purpose of this work is to conduct a non-inferiority trial to compare a lifestyle intervention delivered entirely via private groups on the online social network Twitter to a traditional in-person group-based lifestyle intervention. Using a randomized trial (N=328), investigators will test whether a lifestyle intervention delivered via an online social network (Get Social condition) will result in a mean percent weight loss at 12 months that is not appreciably worse than the gold-standard in-person group-based lifestyle intervention (Traditional condition), i.e., the social network arm will not lose on average 2% less than the in-person arm. Secondary non-inferiority outcomes include weight loss at 12 months, and dietary intake and physical activity at 12-months. Investigators hypothesize that the Get Social condition will be less expensive than the Traditional condition. To understand for whom an online social network modality is most suited, investigators will test predictors of weight loss in the Get Social condition including engagement, age, sociability, neuroticism, openness, and smartphone and social network use. Investigators hypothesize that people who are younger, more sociable, engage more on the social network, higher in neuroticism/openness, and heavier overall smartphone and social network users will lose more weight in the Get Social condition. Findings from this study may support an intervention delivery modality that is more conducive to settings like worksites, health plans, and clinics that serve large populations but have limited space, staffing, and resources for traditional in person interventions.

Get Social participants will receive a weight loss intervention in a "protected" Twitter group. The intervention content will be structured to deliver in an online context. The online coaches will post daily content, respond to questions, address concerns, and encourage engagement. The weight loss intervention is based on the Diabetes Prevention Program (DPP), an evidence-based weight loss program focused on lifestyle changes. The goals for the intervention are 175 minutes of moderate physical activity per week and an overall weight loss of 7%. Each participant will get an individualized calorie goal that would facilitate a 1-2 lbs. weight loss weekly. Participants will also be encouraged to use an app such as MyFitnessPal to track daily diet.

Participants will attend weight loss groups weekly for 8 weeks, then biweekly for 16 weeks, then monthly between months 6 and 12. The weight loss intervention is based on the Diabetes Prevention Program (DPP), an evidence-based weight loss program focused on lifestyle changes. The goals for the intervention are 175 minutes of moderate physical activity per week and an overall weight loss of 7%. Each participant will get an individualized calorie goal that would facilitate a 1-2 lbs. weight loss weekly. Participants will also be encouraged to use an app such as MyFitnessPal to track daily diet.

A smartphone is needed to use MyFitnessPal for all participants and to access Twitter for the Get Social participants

Total time to deliver the intervention and administrative time was tracked during the study for each condition and cost was calculated using national median salary for dietitians & nutritionists and administrative assistants & secretaries, and national fringe data for all civilian workers across all industries from Bureau of Labor Statistics. Hourly rate based on salary plus fringe divided by 2080 hours (40 hours/week x 52 weeks). Administrative tasks charged at administrative assistant salary rate, all other treatment delivery tasks charged at dietitian rate. Dollars per participant was calculated by dividing total cost for each condition by the number of participants in that condition.

Administrative and intervention-delivery was measured. Amount of administrative and intervention delivery time per participant was calculated.

Retention will be calculated as the percentage of participants who complete the final weigh-in in each condition.

Acceptability and their satisfaction with intervention was measured by asking participants "If given the opportunity to continue participating in the program, how willing would you be to continue?" Response options were from 1 ("Definitely would") to 5 ("Definitely would not"). The outcome is the number of positive responses which include responses of 1 or 2 ("Probably would") on this measure.

Burden was measured through the question "Participating in the program was time consuming." Response options were 1 ("Strongly agree") to 5 ("Strongly disagree"). We are reporting number of participants who agreed with the statement by responding with a 1 ("Strongly agree") or 2 ("Agree")

National Cancer Institute's (NCI) automated self-administered 24-hour dietary recall (ASA24™) was used to measure caloric intake.

National Cancer Institute's (NCI) automated self-administered 24-hour dietary recall (ASA24™) measured caloric intake. Difference in caloric intake during baseline and 6 month evaluations was calculated.

National Cancer Institute's (NCI) automated self-administered 24-hour dietary recall (ASA24™) will record dietary intake. Change in caloric intake between baseline and 12 months was calculated.

Arizona Physical Activity Questionnaire will be used to provide estimates of total energy expenditure to calculate number of participants reaching the guideline of 150 minutes per week of moderate to vigorous physical activity.

Arizona Physical Activity Questionnaire will be used to provide estimates of total energy expenditure to calculate number of participants reaching the guideline of 150 minutes per week of moderate to vigorous physical activity.

Arizona Physical Activity Questionnaire will be used to provide estimates of total energy expenditure to calculate number of participants reaching the guideline of 150 minutes per week of moderate to vigorous physical activity.

Engagement will be measured by tracking group attendance and quantifying communication in the online group throughout the intervention in terms of percent of intervention modules participated in.

Engagement will be measured by tracking group attendance and quantifying communication in the online group throughout the intervention in terms of percent of intervention modules participated in.

Weight Management Support Inventory to assess social support for weight management. The directions ask respondents to rate on a 5-point scale how frequently they experienced a certain interaction with other people in the past 4 weeks, ranging from 1 (never) to 5 (daily). The items also ask how helpful that event was when it occurred, rated on a 5-point scale ranging from 1 (not helpful) to 5 (extremely helpful). Higher scores indicate more frequent or more helpful social support events. We asked this about four potential social groups: family, friends, friends on Twitter, and friends on Facebook.

Weight Management Support Inventory to assess social support for weight management. The directions ask respondents to rate on a 5-point scale how frequently they experienced a certain interaction with other people in the past 4 weeks, ranging from 1 (never) to 5 (daily). The items also ask how helpful that event was when it occurred, rated on a 5-point scale ranging from 1 (not helpful) to 5 (extremely helpful). Higher scores indicate more frequent or more helpful social support events. We asked this about four potential social groups: family, friends, friends on Twitter, and friends on Facebook.

Weight Management Support Inventory to assess social support for weight management. The directions ask respondents to rate on a 5-point scale how frequently they experienced a certain interaction with other people in the past 4 weeks, ranging from 1 (never) to 5 (daily). The items also ask how helpful that event was when it occurred, rated on a 5-point scale ranging from 1 (not helpful) to 5 (extremely helpful). Higher scores indicate more frequent or more helpful social support events. We asked this about four potential social groups: family, friends, friends on Twitter, and friends on Facebook.

Openness to Experience and Neuroticism/Emotional Stability were measured using the Ten-Item Personality Inventory. The Ten-Item Personality Inventory consists of 10 items that measure the five dimensions of the Five Factor model of personality. Participants are asked to respond to a series of words with personality descriptions with the prompt "I see myself as" with response options of 1 ("Disagree Strongly") to 7 ("Agree Strongly"). Opposite items are reverse coded. Items for Emotional Stability/Neuroticism are "anxious, easily upset" and "calm, emotionally stable". Items for Openness to experience are "open to new experiences, complex" or "conventional, uncreative." The average of the scores represents each dimension with higher scores indicating higher levels of Openness or Emotional Stability (and lower levels of Neuroticism).

Participants were asked which social networks they had an account on and number with a Twitter account was reported.

Participants were asked which social networks they had an account on and number with a Twitter account was reported.

The average of two blood pressure readings was used as the measure of blood pressure. If participants only had one blood pressure reading that was used as the measure of blood pressure.

The average of two blood pressure readings was used as the measure of blood pressure. If participants only had one blood pressure reading that was used as the measure of blood pressure.

The average of two blood pressure readings was used as the measure of blood pressure. If participants only had one blood pressure reading that was used as the measure of blood pressure.

Participants completed the four item STOP sleep apnea questionnaire. Participants who answered yes to 0-1 items were classified as low risk for sleep apnea. Participants who answered yes to 2-4 items were classified as high risk for sleep apnea.

Participants completed the four item STOP sleep apnea questionnaire. Participants who answered yes to 0-1 items were classified as low risk for sleep apnea. Participants who answered yes to 2-4 items were classified as high risk for sleep apnea.

Participants completed the four item STOP sleep apnea questionnaire. Participants who answered yes to 0-1 items were classified as low risk for sleep apnea. Participants who answered yes to 2-4 items were classified as high risk for sleep apnea.

To measure sleep quality, participants completed the Insomnia Severity Index. Higher scores (maximum 28) represent increased levels of insomnia, from 0-7 ("Absence of insomnia"), through 8-14 ("Sub-threshold insomnia"), 15-21 ("Moderate insomnia"), and 22-28 ("Severe insomnia").

To measure sleep quality, participants completed the Insomnia Severity Index. Higher scores (maximum 28) represent increased levels of insomnia, from 0-7 ("Absence of insomnia"), through 8-14 ("Sub-threshold insomnia"), 15-21 ("Moderate insomnia"), and 22-28 ("Severe insomnia").

To measure sleep quality, participants completed the Insomnia Severity Index. Higher scores (maximum 28) represent increased levels of insomnia, from 0-7 ("Absence of insomnia"), through 8-14 ("Sub-threshold insomnia"), 15-21 ("Moderate insomnia"), and 22-28 ("Severe insomnia").

To measure sleep quality we used select questions from the Pittsburgh Sleep Quality Index which allowed the calculation of sleep duration.

To measure sleep quality we used select questions from the Pittsburgh Sleep Quality Index which allowed the calculation of sleep duration.

To measure sleep quality we used select questions from the Pittsburgh Sleep Quality Index which allowed the calculation of sleep duration.

Participants were asked whether they had ever been diagnosed with sleep apnea with response options Yes or No.

Participants were asked 1. Are you being treated for sleep apnea presently? (Yes/No) and 2. Have you ever received treatment for sleep apnea?

Participants were asked: Have you ever been diagnosed with sleep apnea? If they answered yes, they were asked if they were currently being treated for sleep apnea.

Participants were asked: Have you ever been diagnosed with sleep apnea? If they answered yes, they were asked if they were currently being treated for sleep apnea.

To measure depressive symptoms, participants completed the Beck Depression Inventory. Total scores can be interpreted as 1-10 ("normal ups and downs"), 11-16 ("Mild mood disturbance"), 17-20 ("Borderline clinical depression"), 21-30 ("Moderate depression"), 31-40 ("Severe depression"), 41+ ("Extreme depression").

To measure depressive symptoms, participants completed the Beck Depression Inventory. Total scores can be interpreted as 1-10 ("normal ups and downs"), 11-16 ("Mild mood disturbance"), 17-20 ("Borderline clinical depression"), 21-30 ("Moderate depression"), 31-40 ("Severe depression"), 41+ ("Extreme depression").

To measure depressive symptoms, participants completed the Beck Depression Inventory. Total scores can be interpreted as 1-10 ("normal ups and downs"), 11-16 ("Mild mood disturbance"), 17-20 ("Borderline clinical depression"), 21-30 ("Moderate depression"), 31-40 ("Severe depression"), 41+ ("Extreme depression").

Participants were asked "Have you ever posted about or had discussions about your weight loss or desire to lose weight?" Number who answered yes are reported.

Participants were asked "Have you ever posted about or had discussions about your weight loss or desire to lose weight?" Number who answered yes are reported.

Participants were asked "Have you ever posted about or had discussions about your weight loss or desire to lose weight?" Number who answered yes are reported.

The intervention was divided into different content units delivered either during a meeting (Traditional Condition) or a series of tweets (Get Social Condition). For the Get Social Condition, tweets with intervention content were produced to meet all objectives for each content unit and programmed to go out at set times for each group so 100% of the objectives for each content unit were met. For the Traditional Condition, checklists were created with objectives for each content unit and staff listened to 2 to 3 randomly selected group recordings for each study wave to determine if these objectives were met by the interventionist. Treatment Fidelity was defined as the percentage of objectives delivered by the interventionist for each wave.

To measure sleep quality we used select questions from the Pittsburgh Sleep Quality Index to calculate habitual sleep efficiency (the percentage of time in bed spent sleeping).

To measure sleep quality we used select questions from the Pittsburgh Sleep Quality Index which allowed the calculation of habitual sleep efficiency (the percentage of time in bed spent sleeping).

To measure sleep quality we used select questions from the Pittsburgh Sleep Quality Index which allowed the calculation of habitual sleep efficiency (the percentage of time in bed spent sleeping).

Inclusion: Smartphone users Ages 18-65 Body Mass Index (BMI) 27-45 Must currently log into a social media platform on a daily basis basis and engage (like, reply, post) at least 4 days per week. Participants will be excluded if they: Do not have a smartphone; Are unable to get medical clearance from their Primary Care Physician (PCP); Have plans to move during study; Are not interested in losing weight; Have Type 1 or uncontrolled 2 Diabetes (as determined by PCP); Have medical conditions that would prevent increasing physical activity or making dietary changes; Are pregnant/lactating or plans to become pregnant during study; Are currently taking medication affecting weight; Currently participating in a weight loss program or using an online social network to assist them with weight loss (e.g. Weight Watchers, Sparkpeople, etc); Unable to walk at least ¼ mile unaided without stopping; Experienced a weight loss of 5% or more in past 3 month; A history of/or plans on having bariatric surgery; Did not complete the baseline measures; Not willing to use Twitter, or if a current Twitter user, not willing to create a new account specifically for study purposes; Participated in another weight loss study under the direction of the PI of this study; Current smoker (smokes 3 or more cigarettes per day); Unavailable to attend weekly group meetings; Prefers one condition over another; Score of 30 or higher on the Beck Depression Inventory (BDI) or endorsement of 2 or 3 on BDI #9 indicating suicidal thoughts; Presence of binge eating disorder Did not complete the orientation webinar; Are unable to provide consent due to mental illness or a cognitive impairment; Does not speak English; or Are a prisoner.

Jake-Schoffman DE, Brown SD, Baiocchi M, Bibeau JL, Daubenmier J, Ferrara A, Galarce MN, Hartogensis W, Hecht FM, Hedderson MM, Moran PJ, Pagoto SL, Tsai AL, Waring ME, Kiernan M. Methods-Motivational Interviewing Approach for Enhanced Retention and Attendance. Am J Prev Med. 2021 Oct;61(4):606-617. doi: 10.1016/j.amepre.2021.04.005.

Wang ML, Waring ME, Jake-Schoffman DE, Oleski JL, Michaels Z, Goetz JM, Lemon SC, Ma Y, Pagoto SL. Clinic Versus Online Social Network-Delivered Lifestyle Interventions: Protocol for the Get Social Noninferiority Randomized Controlled Trial. JMIR Res Protoc. 2017 Dec 11;6(12):e243. doi: 10.2196/resprot.8068.