obesity systematic review

Obesity is a leading cause of preventable death in the United States (US) and world-wide, and consumes substantial social re- sources.1-4 Obesity elevates the risks of various chronic diseases, including type 2 diabetes, hyper- tension, dyslipidemia, coronary heart disease and certain types of cancer.5 Between 1976-1980 and 2013-2014, the prevalence of obesity more than doubled among the US adult population.6,7 It is estimated that the medical expenditure attribut- able to overweight and obesity will reach $861-957 billion US dollars by 2030, accounting for 16%- 18% of total healthcare costs in the US.8 By 2025, the global obesity prevalence is expected to reach

18% in men and exceed 21% in women, and severe obesity will exceed 6% in men and 9% in women.9 Obesity prevention has become a national priority in the US and many other countries.10,11 Key strat- egies typically involve promoting a healthful diet and an active lifestyle.12-14 Various types of media have been used to deliver these interventions, in- cluding radio, television, newspaper, the Internet, and recently, social media.15-17

Social media are computer-mediated technolo- gies that allow sharing of user-generated contents such as text posts/comments and digital photos/ videos through online social networks and virtual communities.18 Known as the “participative Inter- net” or “Web 2.0,”19,20 social media incorporate an extensive set of Internet-based communications, tools, and aids that provide easy, cost-effective ac- cess to a massive population pool across geograph- ic distances.21 From 2005 to 2015, the adoption of social media among US adults increased from 3% to 65%.22 Although young adults aged 18 to 29 years are the most likely to use social media

Ruopeng An and Mengmeng Ji, Department of Kinesiology and Community Health, College of Applied Health Sciences, University of Illinois at Urbana-Champaign, IL, Urbana-Cham- paign, IL. Sheng Zhang, School of Sports Journalism and For- eign Studies, Shanghai University of Sport, Shanghai, China. Correspondence Dr An: ran5@illinois.edu

Effectiveness of Social Media-based Interventions on Weight-related Behaviors and Body Weight Status: Review and Meta-analysis

Ruopeng An, PhD, MPP; Mengmeng Ji, MS; Sheng Zhang, PhD

Objectives: We reviewed scientific lit- erature regarding the effectiveness of social media-based interventions about weight-related behaviors and body weight status. Methods: A keyword search were performed in May 2017 in the Clinical- Trials.gov, Cochrane Library, PsycINFO, PubMed, and Web of Science databases. We conducted a meta-analysis to esti- mate the pooled effect size of social me- dia-based interventions on weight-related outcome measures. Results: We identified 22 interventions from the keyword and reference search, including 12 random- ized controlled trials, 6 pre-post stud- ies and 3 cohort studies conducted in 9 countries during 2010-2016. The major- ity (N = 17) used Facebook, followed by Twitter (N = 4) and Instagram (N = 1). In- tervention durations averaged 17.8 weeks with a mean sample size of 69. The meta-

analysis showed that social media-based interventions were associated with a sta- tistically significant, but clinically mod- est reduction of body weight by 1.01 kg, body mass index by 0.92 kg/m2, and waist circumstance by 2.65 cm, and an increase of daily number of steps taken by 1530. In the meta-regression there was no dose- response effect with respect to interven- tion duration. Conclusions: The boom of social media provides an unprecedented opportunity to implement health pro- motion programs. Future interventions should make efforts to improve interven- tion scalability and effectiveness.

Key words: social media health pro- motion; weight-related behavior; body weight; obesity systematic review; obe- sity meta-analysis

Am J Health Behav. 2017;41(6):670-682 DOI: https://doi.org/10.5993/AJHB.41.6.1

 

 

D el

iv er

ed b

y In

ge nt

a to

: ? IP

: 1 40

.2 34

.2 53

.9 o

n: S

at , 2

3 D

ec 2

01 7

06 :1

2: 30

C op

yr ig

ht (

c) P

N G

P ub

lic at

io ns

. A ll

rig ht

s re

se rv

ed .

An et al

Am J Health Behav.™ 2017;41(6):670-682 671 DOI: https://doi.org/10.5993/AJHB.41.6.1

(90% in 2015), it also becomes increasingly popu- lar among older adults aged 65 years and above (from 11% in 2010 to 35% in 2015).22 Social media offer substantial potential for the delivery of health promotion interventions by reaching a large num- ber of audiences worldwide, sustaining high levels of user engagement and retention in comparison to traditional Web-based interventions, delivering health messages via existing contacts rather than via traditional marketing strategies, and promot- ing user-generated contents instead of passive information.17,23-25 In 2013, the American Heart Association issued a scientific statement that sup- ported “the promise and potential of social media and electronic technology as a viable component of weight management programs,” and highlighted “the need for additional research to optimize these technologies as effective delivery channels.”26

Despite the boom of social media-based health promotion programs in the current decade,27,28 a relevant theoretical framework remains underde- veloped. The social cognitive theory links an indi- vidual’s knowledge acquisition to people’s obser- vation of others’ behaviors within the context of social interactions, experiences, and media influ- ences.29,30 When individuals observe peer behavior and consequences, they remember the sequence of events and use this information to guide their own subsequent behavior.29 Observing a peer’s be- havior may also prompt the viewer to engage in the behavior he/she has already learned.29 Social me- dia provide a platform where users communicate with their peers regarding information, ideas, and behaviors in an instantaneous manner.31,32 Health promotion programs delivered through social me- dia embed participants in an online social network or virtual community where they talk to each other and observe each other’s behavior, which may help them form, sustain, reinforce, and/or modify their own health behavior.33,34 Specific to social media- based interventions about weight management, social cognitive theory implies that program par- ticipants’ physical activity engagement and dietary behaviors are likely influenced by their online net- working experiences, and ultimately, impact their body weight status.30,35

There are 5 previous review articles that relate to this paper. Chang et al reviewed the role of social media in online weight management.36 Nearly all physical activity interventions in the review exclu- sively used message boards as part of the inter- vention website.36 Maher et al reviewed the effec- tiveness of online social network-based health be- havior interventions.17 Of the 10 interventions in- cluded in the review, half (N = 5) used some social media that were either constructed by research- ers solely for the purpose of the intervention or as add-ons to a commercial website.17 Moreover, the review outcomes included not only weight-related behaviors but also other measures such as quality of life and social support.17 Williams et al reviewed social media-based randomized controlled trials on

diet and exercise behaviors.37 Of the 22 trials in- cluded in the review, only 2 used social media (one used Facebook and the other used both Facebook and Twitter), whereas all other trials used discus- sion board as part of a health promotion website.37 Mita et al reviewed the efficacy of social media- based randomized controlled trials in reducing risk factors for chronic diseases.38 Of the 16 trials in- cluded in the review, only 2 used social media (one used Facebook and the other used Twitter), where- as other trials used text-based message board, dis- cussion board, bulletin, or online forum.38 Willis et al reviewed weight management interventions delivered through online social networks.39 The re- view scope was confined to interventions (N = 5) that reported the effects of online social networks on weight loss. No meta-analysis was conducted to quantify intervention effectiveness.39

In this paper, we fill the gap in previous litera- ture by conducting a systematic review and meta- analysis on the effectiveness of social media-based interventions about weight-related behaviors and body weight status among adults. Unlike research- er-generated online network platforms such as message board and discussion forum, social media such as Facebook and Twitter cover an extensive user base worldwide and across many population subgroups, and incorporate much richer function- alities that facilitate multimodal communication and interaction.40 Thus, interventions that adopt social media have the potential to reach a large population, and influence participants’ weighted- related behaviors through multiple networking channels.40 The aims of this review are to: (1) sys- tematically identify, summarize, and quantify the effects of social media-based interventions on mod- ifying weight-related behaviors (ie, physical activ- ity, sedentary behavior and diet) and body weight status; and (2) assess the quality of evidence and identify limitations and gaps in the existing litera- ture in an effort to inform future research.

METHODS The systematic review and meta-analysis proce-

dures were conducted in accordance with the Co- chrane Handbook for Systematic Reviews of Inter- ventions.41

Study Selection Criteria Studies that met all of the following criteria were

included in the review: (1) Intervention type: in- terventions that used social media (eg, Facebook, Twitter and Instagram) to influence study par- ticipants’ weight-related behaviors and/or body weight status; (2) Study participant: adults aged 18 years and above; (3) Study outcome: weight- related behaviors including physical activity, sed- entary behavior and diet, and body weight status including overweight and obesity measured by body mass index (BMI, kg/m2), waist circumfer- ence, waist-to-hip ratio and/or body fat; (4) Article type: peer-reviewed publications pertaining to hu-

 

 

D el

iv er

ed b

y In

ge nt

a to

: ? IP

: 1 40

.2 34

.2 53

.9 o

n: S

at , 2

3 D

ec 2

01 7

06 :1

2: 30

C op

yr ig

ht (

c) P

N G

P ub

lic at

io ns

. A ll

rig ht

s re

se rv

ed .

Effectiveness of Social Media-based Interventions on Weight-related Behaviors and Body Weight Status

672

man research; (5) Time window of search: from the inception of an electronic bibliographic database to May 24, 2017; and (6) Language: articles written in English.

Studies that met any of the following criteria were excluded from the review: (1) Observational studies that adopted no experimental design; (2) Interventions that did not utilize social media to influence study participants; (3) Social media- based interventions that incorporated no study outcome pertaining to weight-related behavior and body weight status; (4) study samples of children or adolescents aged 17 years and younger; (5) Ar- ticles not written in English; or (6) Dissertations, conference proceedings, study/review protocols or review articles.

Search Strategy A keyword search was performed in the follow-

ing 5 electronic bibliographic databases: Clinical- Trials.gov, Cochrane Library, PsycINFO, PubMed, and Web of Science. The search algorithm included all possible combinations of keywords from the 2 groups: (1) “diet,” “diets,” “dietary,” “food,” “foods,” “energy intake,” “energy intakes,” “energy con- sumption,” “calorie,” “calories,” “caloric intake,” “caloric intakes,” “motor activity,” “motor activi- ties,” “sport,” “sports,” “physical fitness,” “physi- cal exertion,” “physical activity,” “physical activi- ties,” “physical inactivity,” “sedentary behavior,” “sedentary behaviors,” “sedentary lifestyle,” “ex- ercise,” “exercises,” “obesity,” “obese,” “adiposity,” “overweight,” “body mass index,” “weight,” “waist circumference,” “waist to hip,” “waist-to-hip,” and “body fat,” and (2) “Facebook,” “YouTube,” “Google+,” “Google +,” “Google Plus,” “Instagram,” “Twitter,” “Vine,” “LinkedIn,” “Pinterest,” “Skype,” “Snapchat,” “Myspace,” “Flickr,” “Delicious,” “so- cial media,” “social bookmarking,” “social net- work,” “social networks,” and “social networking.” Titles and abstracts of articles identified through the keyword search were screened against the study selection criteria. Potentially relevant articles were retrieved for evaluation of the full texts. Two reviewers (Ji and Zhang) independently conducted title and abstract screening and identified poten- tially relevant articles. Inter-rater agreement was assessed using Cohen’s kappa (κ = 0.87). Discrep- ancies were resolved through discussion among the 3 co-authors.

Data Extraction and Preparation A standardized data extraction form was used to

collect the following methodological and outcome variables from each included study: author(s), publication year, study design, sample size, sam- ple characteristics (ie, age, sex and health/disease status), intervention group/arm assignment and group/arm size, intervention components (ie, type of social media and other miscellaneous compo- nents), intervention duration, attrition rate, adjust- ment for baseline characteristics, study outcome

(ie, weight-related behavior and body weight sta- tus), measurement (ie, measure on weight-related behavior and body weight status), baseline and post-intervention values on weight-related behav- ior and body weight status, and statistical method.

Meta-analysis We performed a meta-analysis to estimate the

pooled effect size of social media-based interven- tions on weight-related behavior and body weight status. Weight-related behaviors included inter- vention-attributable changes in daily number of steps taken, moderate-to-vigorous physical activ- ity (minutes/week), light physical activity (min- utes/week), total physical activity (minutes/week) and energy expenditure (kcal/day). Body weight status included intervention-attributable changes in body weight (kg), BMI (kg/m2), waist circumfer- ence (cm), body fat (kg) and body fat percentage (%). Pooling of weight-related dietary behaviors was infeasible because no 2 studies adopted the same measure on a dietary behavior (eg, one study ex- amined daily number of servings of fruit/vegetable consumption, whereas the other study examined daily frequency of fruit/vegetable consumption). Separate meta-analysis was conducted for studies stratified by social media type (ie, Facebook and Twitter). The control group of all interventions was the group that received no intervention of any type. The experimental groups were those that mainly received social media-based interventions (plus other supplemental components such as text mes- sages and phone calls for some studies).

We assessed study heterogeneity using the I2 index. A fixed-effects model was estimated when I2≤50%, and a random-effects model was estimated when I2>50%. Publication bias was assessed by vi- sual inspection of the funnel plot and the Begg’s and Egger’s tests. We conducted a meta-regression to assess the potential dose-response relationship between intervention duration (in weeks) and effect size. We used Stata 14.2 SE version (StataCorp, College Station, TX) for all statistical analyses.

Study Quality Assessment Based on the recommendation of the Cochrane

Handbook for Systematic Reviews of Interventions Version 5.1.0,41 we used the Quality Assessment Tool for Quantitative Studies to assess the quality of each included study.42 This assessment tool rates each study based on the 6 quality components— selection bias, study design, confounders, blind- ing, data collection method, and withdrawals and dropouts.42 A global rating for a study (ie, strong, moderate or weak) is further obtained based on the ratings of each quality component.42 Detailed descriptions of the quality assessment tool and its dictionary can be found elsewhere.42 Two rat- ers (Ji and Zhang) independently conducted qual- ity assessment for each selected study. Inter-rater agreement was assessed using the intra-class cor- relation (ICC = 0.55). Discrepancies were resolved

 

 

D el

iv er

ed b

y In

ge nt

a to

: ? IP

: 1 40

.2 34

.2 53

.9 o

n: S

at , 2

3 D

ec 2

01 7

06 :1

2: 30

C op

yr ig

ht (

c) P

N G

P ub

lic at

io ns

. A ll

rig ht

s re

se rv

ed .

An et al

Am J Health Behav.™ 2017;41(6):670-682 673 DOI: https://doi.org/10.5993/AJHB.41.6.1

through discussion among the 3 co-authors. Study quality assessment helped measure the strength of scientific evidence but was not used to determine the inclusion of studies.

RESULTS Study Selection

Figure 1 shows the study selection flow chart. We identified 9977 total articles through keyword search, including 531 articles from the ClinicalTri- als.gov, 305 articles from the Cochrane Library, 2559 articles from the PubMed, 4856 articles from the Web of Science, and 1726 articles from the Psy- cINFO. After removing duplications, we used 7201 unique articles for title and abstract screening, of which, 7163 articles were excluded. The full texts of the remaining 38 articles were reviewed against the study selection criteria. Of these, 17 articles were excluded. Reasons for exclusion included: 3 articles utilized researcher-invented mobile apps rather than social media to conduct interventions; 6 articles incorporated no study outcome pertain- ing to weight-related behavior and body weight status; 5 articles were study protocols rather than original studies; and 3 studies exclusively recruit- ed children aged 17 years and younger.43-45 The

forward and backward reference search were con- ducted based on the remaining 21 articles as well as relevant review papers, and 6 new articles were identified that met the study selection criteria. In total, 27 articles were included in this review,46-72 representing 22 unique social media interventions.

Figure 2 shows the number of relevant interven- tions by publication year. The pool of relevant in- terventions comprised the 22 interventions includ- ed in the review (on adults) plus46-72 the 3 inter- ventions focusing on children and adolescents43-45 as well as published protocols of on-going social media-based interventions on weight-related be- havior and/or body weight status.73-91 No results/ findings of these ongoing interventions were avail- able at the time this review was conducted. Since the beginning of this decade, the number of so- cial media-based interventions on weight-related behavior and body weight status has increased rapidly. The number of completed interventions included in this review increased from one in 2010 to 8 in 2016. Furthermore, there are 19 total on- going interventions (one in 2013, 5 in 2014, 6 in 2015, and 7 in 2016) whose results/findings have not been published.73-91

Basic Characteristics of the Included Interventions

Appendix A summarizes the characteristics of the included interventions. These interventions were conducted in the US (N = 16), the United Kingdom (N = 1), Saudi Arabia (N = 1), Singapore (N = 1), Malaysia (N = 1), Australia (N = 1) and Japan (N = 1). Among them, 12 were randomized controlled trials (RCTs),44,45,47,49-51,53-56,64-65,67-71 6 were pre-post studies,47-49,52,63,66 and 3 were cohort studies.46,58,72

The difference between pre-post and cohort stud- ies is that the latter not only has an intervention group, as in the former, but also a control group that is followed at the baseline and during the in- tervention. Sample sizes were generally small and to some extent varied across studies, with a mean of 69 (standard deviation [SD] = 83.5), a median of 55, and a range from 10 to 404. All interventions recruited young or middle-aged adult samples with an average age of 18.6-48.8 years. All interventions included women—8 exclusively recruited female participants,46,49-51,59,60,62,63,67 and the average pro- portion of female participants among all interven- tions was 77%. Regarding health/disease status, 9 interventions exclusively recruited overweight (25 kg/m2 ≤ BMI < 30 kg/m2) and/or obese (BMI ≥ 30 kg/m2) participants,47-49,55-57,59,60,65,66,68-69 one inter- vention exclusively recruited patients with meta- bolic syndrome,61 and one intervention exclusively recruited cancer survivors.70,71 Regarding type of social media, 17 interventions used Facebook, 4 used Twitter,52,64,66,68,69 and one used Instagram.46 No other social media were used. Besides the use of social media, interventions often included other components such as wearable devices (eg, Fitbit), mobile apps, text messages, phone calls, face-

 

 

Duplicate articles removed

(N = 2776)

Articles identified through database search (N = 9977)

• PubMed = 2559 • Web of Science = 4856 • Cochrane = 305 • PsycINFO = 1726 • Clinical Trials = 531

 

Articles excluded after title and abstract screening

(N = 7163)

Articles screened by full-text

(N = 38)

Articles excluded after full-text review

(N = 17)

Articles included in the review

(N = 27)

Articles included from other sources (N = 6):

• Reference search = 2

• Other review articles = 4

Articles screened by title and abstract

(N = 7201)

Figure 1 Study Selection Flowchart

 

 

D el

iv er

ed b

y In

ge nt

a to

: ? IP

: 1 40

.2 34

.2 53

.9 o

n: S

at , 2

3 D

ec 2

01 7

06 :1

2: 30

C op

yr ig

ht (

c) P

N G

P ub

lic at

io ns

. A ll

rig ht

s re

se rv

ed .

Effectiveness of Social Media-based Interventions on Weight-related Behaviors and Body Weight Status

674

to-face or Web-based counseling, website recom- mendations, email, electronic newsletters, online videos, personal coaches or trainers, and physi- cal fitness courses. Intervention durations varied substantially across studies, with a mean of 17.8 weeks (SD = 20.8), a median of 12 weeks, and a range from 3 to 102 weeks. Attrition rate varied across interventions—the mean and median attri- tion rates among all interventions were 12% and 13%, respectively; 10 interventions had an attri- tion rate less than or equal to 5%, whereas 5 had attrition rate greater than 20%.

Appendix B summarizes measures and study outcomes of the 22 included interventions. Among them, 15 assessed body weight status including body weight, BMI, waist circumference, hip cir- cumference, waist-to-hip ratio and body fat; 12 assessed physical activity, including number of steps taken, energy expenditure, total physical activity, and light, moderate, vigorous and mod- erate-to-vigorous physical activity; one assessed sedentary behavior;62 and 5 assessed dietary behavior including intake of fruits/vegetables, sugar-sweetened beverages, fast-foods, fat, and total energy.49,52,60,68,69 Among the 15 interven- tions that evaluated body weight status, 14 used objective measures (eg, measured height and we ight)47-49,52,55-62,64-66,68-69 and one used self-report data.70,71 Among the 12 interventions that evalu- ated physical activity, 5 used objective measures (eg, pedometer and accelerometer)54,61,62,64,67 and 7 used questionnaires.49-51,53,55-57,63,68-71 Interventions

that evaluated sedentary behavior and diet were all based on self-report data. Most studies (N = 19) performed statistical analysis to estimate the treatment effect and associated uncertainty. The statistical tests and models applied included inde- pendent 2-sample t-tests, paired t-tests, Pearson’s chi-square tests, Fisher’s exact tests, Kruskal-Wal- lis tests, Wilcoxon signed-rank tests, Mann-Whit- ney U tests, ANOVAs, ANCOVAs, multiple linear re- gression, and multiple logistic regression. Among the 19 studies that performed statistical analysis, 14 controlled for some baseline individual charac- teristics (eg, sex, age, race/ethnicity, income, edu- cation level and marital status) when estimating the treatment effect.

Appendix C reports the calculated effect sizes of social media-based interventions and their corre- sponding standard errors based on the baseline and post-intervention measures between and/or within intervention/control arms. Interventions that did not provide sufficient data for calculation of effect size and standard error were excluded from Appendix C. Following the recommendations of Littell et al,92 a meta-analysis was conducted on each type of effect size (eg, body weight, BMI, daily number of steps taken, and total physical activity duration per week) if 2 or more interventions re- ported the same type of effect size.

Meta-analysis on Body Weight Status Table 1 summarizes the modeling results from

the meta-analysis, meta-regression, and publica-

 

 

0

2

4

6

8

10

12

14

16

2010 2011 2012 2013 2014 2015 2016

N um

be r

of st

ud ie

s

included study ongoing trial

Figure 2 Number of Interventions Included in the Review and Ongoing Trials

(Not Included in the Review) by Year

 

 

D el

iv er

ed b

y In

ge nt

a to

: ? IP

: 1 40

.2 34

.2 53

.9 o

n: S

at , 2

3 D

ec 2

01 7

06 :1

2: 30

C op

yr ig

ht (

c) P

N G

P ub

lic at

io ns

. A ll

rig ht

s re

se rv

ed .

An et al

Am J Health Behav.™ 2017;41(6):670-682 675 DOI: https://doi.org/10.5993/AJHB.41.6.1

tion bias tests for all outcomes pertaining to body weight status. Social media-based interventions were found to reduce body weight by 1.01 kg (95% CI = 0.45, 1.57) among participants. By type of social media, Facebook-based interventions were found to reduce body weight by 1.97 kg (95% CI = 0.93, 3.01) among participants, whereas Twitter- based interventions were not found to be associat- ed with body weight reduction. Social media-based interventions were found to reduce BMI by 0.92 kg/m2 (95% CI = 0.29, 1.54) among participants. Social media-based interventions were found to reduce waist circumference by 2.65 cm (95% CI = 0.86, 4.43) among participants, but were not found to be associated with changes in body fat or body fat percentage. Across all outcomes per-

taining to body weight status (ie, body weight, BMI, waist circumference, body fat and body fat percentage), meta-regression found the difference in pooled-effect estimates between Facebook- and Twitter-based interventions to be statistically non- significant, and identified no dose-response effect with respect to intervention duration. No publica- tion bias was identified as neither the Egger’s tests nor the Begg’s tests were statistically significant.

Meta-analysis on Weight-related Behaviors Table 2 summarizes modeling results from meta-

analysis, meta-regressions, and publication bias tests for all outcomes pertaining to weight-related behaviors. We found that social media-based inter- ventions increased daily number of steps taken by

Table 1 Modeling Results from Meta-analysis, Meta-regressions, and Publication Bias

Tests for Intervention Outcomes Regarding Body Weight Status

Change in body weight status First author (year) I

2 index

Pooled Effect size

(95% CI)

Model

p-value for dose-response

effect from meta-regression

Publication bias test

p-value for Egger’s

test

p-value for

Begg’s test

Body weight (kg)

Napolitano (2013); Valle (2013); Chee (2014); Herring (2014); Joseph (2015); Ruotsalainen (2015); Aschbrenner (2016); Cavallo (2016); Godino (2016); Herring (2016); West (2016); Pagoto (2015); Nishiwaki (2016)

88.3% -1.783

(-2.706, -0.860)

Random- effect .071 .230 .502

Body mass index (kg/m2)

Valle (2013); Chee (2014); Ruotsalainen (2015); Aschbrenner (2016); Godino (2016); Nishiwaki (2016)

97.7% -0.823 (-1.390, -0.256)

Random- effect .264 .217 1.000

Waist circumference (cm)

Chee (2014); Godino (2016); Nishiwaki (2016)

99.2% -2.647

(-4.430, -0.864)

Random- effect .889 .427 1.000

Body fat (kg) Chee (2014); Nishiwaki (2016) 100% 1.006

(-3.120, 5.131)

Random- effect NA NA 1.000

Body fat percentage (%)

Chee (2014); Nishiwaki (2016) 100%

1.291 (-3.236, 5.819)

Random- effect NA NA 1.000

 

 

D el

iv er

ed b

y In

ge nt

a to

: ? IP

: 1 40

.2 34

.2 53

.9 o

n: S

at , 2

3 D

ec 2

01 7

06 :1

2: 30

C op

yr ig

ht (

c) P

N G

P ub

lic at

io ns

. A ll

rig ht

s re

se rv

ed .

Effectiveness of Social Media-based Interventions on Weight-related Behaviors and Body Weight Status

676

1530 (95% CI = 82, 2,979) among participants. So- cial media-based interventions were not found to be associated with energy expenditure, total physi- cal activity, and moderate-to-vigorous physical activity. Across all outcomes pertaining to weight- related behaviors, meta-regression identified no dose-response effect with respect to intervention duration. No publication bias was identified as nei- ther the Egger’s tests nor the Begg’s tests were sta- tistically significant.

Study Quality Assessment Table 3 reports criterion-specific and global rat-

ings from the quality assessment. Appendix D presents the full version of the Quality Assess- ment Tool for Quantitative Studies by listing all sub-questions in each quality assessment crite- rion and associated study-specific rating.42 Stud- ies differed substantially across quality assess- ment criteria. In general, studies were strong in data collection methods, preventing withdrawals and dropouts, and addressing confounders. Ex-

cept for one intervention that was rated “weak” in study design,54 all others were rated “moderate” or “strong.” Ratings on blinding procedures ranged from “weak” to “moderate” across the 22 included interventions. Regarding blinding, outcome asses- sors of all the included studies were aware of the intervention and exposure status of participants, and most study participants were aware of the re- search question. Most studies were rated “weak” in addressing selection bias because few recruited a random/representative sample from the target population, and few documented the proportion of selected individuals that agreed to participate in the intervention.

Discussion In this study we reviewed current literature re-

garding the effectiveness of social media-based in- terventions on weight-related behaviors and body weight status. A total of 22 interventions were identified from keyword and reference search of bibliographical databases, including 12 RCTs, 6

Table 2 Modeling Results from Meta-analysis, Meta-regressions, and Publication Bias

Tests for Intervention Outcomes Regarding Weight-related Behaviors

Change in weight- related behavior First author (year) I

2 index

Pooled effect size

(95% CI)

Model

p-value for dose-response

effect from meta-

regression

Publication bias test

p-value for Egger’s

test

p-value for Begg’s

test

Daily number of steps taken

Foster (2010); Chee (2014); Wojciki (2014); Rote (2015); Nishiwaki (2016)

99.1% 1348.3 (212.5, 2484.1)

Random- effect .290 .158 .462

Moderate-to- vigorous physical activity (minutes/ week)

Valle (2013); Wojciki (2014); Joseph (2015); Ruotsalainen (2015)

82.4% 24.38

(-17.08, 65.84)

Random- effect .467 .480 .734

Light physical activity (minutes/ week)

Valle (2013); Joseph (2015); Ruotsalainen (2015)

20.9% -4.64

(-26.60, 17.31)

Fixed- effect NA NA 1.000

Moderate physical activity (minutes/ week)

Joseph (2015); Ruots- alainen (2015) 0.0%

0.76 (-7.90, 9.42)

Fixed- effect NA NA 1.000

Vigorous physical activity (minutes/ week)

Joseph (2015); Ruots- alainen (2015) 0.0%

1.92 (-2.56, 6.39)

Random- effect NA

NA 1.000

Total physical activity (minutes/ week)

Lao (2011); Valle (2013); Kernot (2014) 75.2%

110.21 (-2.95, 223.37)

Random- effect .736 .745 1.000

Sedentary time (minutes/week)

Wojciki (2014); Joseph (2015); Ruotsalainen (2015)

0.0% -18.07

(-47.68, 11.53)

Fixed- effect .735 .299 .296

Energy expenditure (kcal/day)

Cavallo 2012; Turner- McGrievy (2011) 72.6%

81.28 (-114.60, 277.15)

Random- effect NA NA 1.000

 

 

D el

iv er

ed b

y In

ge nt

a to

: ? IP

: 1 40

.2 34

.2 53

.9 o

n: S

at , 2

3 D

ec 2

01 7

06 :1

2: 30

C op

yr ig

ht (

c) P

N G

P ub

lic at

io ns

. A ll

rig ht

s re

se rv

ed .

An et al

Am J Health Behav.™ 2017;41(6):670-682 677 DOI: https://doi.org/10.5993/AJHB.41.6.1

pre-post studies, and 3 cohort studies conducted in 7 countries during 2010-2016. The majority (N = 17) used Facebook, followed by Twitter (N = 4) and Instagram (N = 1). Meta-analysis found that social media-based interventions were associated with statistically significant but clinically mod- est reductions in body weight by 1.01 kg, BMI by 0.92 kg/m2, and waist circumstance by 2.65 cm, and an increase of daily number of steps taken by 1530. Studies were generally strong in data collec- tion methods, preventing withdrawals/dropouts, and addressing confounders, but weak in blinding procedures and minimizing selection bias.

Health promotion interventions delivered face-

to-face are more likely to have a direct impact on participants through interpersonal engagement and mutual trust, but also are constrained by high implementation cost and lack of personnel.92,93 In contrast, health promotion interventions delivered remotely via “traditional” media such as radio, tele- vision, or website are less expensive on a per capita basis, and thus, are able to reach a large and di- verse population, but often with compromised in- fluence due to various reasons such as untargeted and non-customized messaging and passive learn- ing.15,95,96 Social media-based interventions stand in the middle—they promote communication and interpersonal engagement through online social

Table 3 Quality Assessment of the Interventions Included in the Review (Brief Version)

First author – (year) Selection bias Study design Confounders Blinding

Data collection method

Withdrawals and

Dropouts

Global Rating

Al-Eisa – (2016) Weak Moderate Weak Weak Strong Weak Weak Aschbrenner – (2016a) Moderate Moderate Strong Weak Strong Moderate Moderate Aschbrenner – (2016b) Moderate Moderate Strong Weak Strong Strong Moderate Cavallo – (2016) Weak Moderate Strong Weak Strong Weak Weak Cavallo – (2014) (2012) Weak Strong Strong Moderate Strong Strong Moderate Chung – (2016) Weak Moderate Strong Moderate Moderate Strong Moderate Wang – (2014) Weak Strong Strong Moderate Strong Strong Moderate Foster – (2010) Weak Weak Strong Moderate Strong Strong Weak Godino – (2016) Weak Strong Strong Moderate Strong Strong Moderate Patrick – (2014) Weak Strong Strong Moderate Strong Strong Moderate Merchant – (2014) Weak Strong Strong Moderate Strong Strong Moderate Hales – (2014) Weak Moderate Strong Moderate Strong Weak Weak Herring – (2016) Weak Strong Strong Moderate Strong Strong Moderate Herring – (2014) Weak Strong Strong Moderate Moderate Strong Moderate Chee – (2014) Strong Strong Strong Moderate Strong Strong Strong Joseph – (2015) Weak Strong Strong Moderate Strong Strong Moderate Kernot – (2014) Weak Moderate Strong Weak Strong Strong Weak Lao- (2011) Moderate Strong Strong Moderate Moderate Weak Moderate Nishiwaki – (2016) Weak Strong Strong Moderate Strong Strong Moderate Napolitano – (2013) Weak Strong Strong Moderate Strong Strong Moderate Pagoto – (2015) Weak Moderate Strong Weak Strong Strong Weak Rote – (2015) Weak Strong Strong Moderate Strong Strong Moderate Ruotsalainen – (2015) Weak Strong Strong Moderate Strong Strong Moderate Turner-McGrievy – (2011) (2013) Weak Strong Strong Moderate Strong Strong Moderate

Valle – (2013) (2015) Weak Strong Strong Moderate Strong Strong Moderate

West – (2016) Weak Moderate Strong Moderate Moderate Strong Moderate Wojcicki – (2014) Weak Strong Strong Moderate Strong Strong Moderate

 

 

D el

iv er

ed b

y In

ge nt

a to

: ? IP

: 1 40

.2 34

.2 53

.9 o

n: S

at , 2

3 D

ec 2

01 7

06 :1

2: 30

C op

yr ig

ht (

c) P

N G

P ub

lic at

io ns

. A ll

rig ht

s re

se rv

ed .

Effectiveness of Social Media-based Interventions on Weight-related Behaviors and Body Weight Status

678

networking, which facilitates participants to form and sustain health behaviors, and in the mean- time, have the capacity to reach a large population pool unconstrained by geographic distances.97 The implementation cost and demand for personnel of social media-based interventions, on a per capita basis, are likely to be lower than for face-to-face interventions, but higher than for “conventional” media-based interventions.98,99 By the same token, we would expect the effectiveness of social media- based interventions on health behavior modifica- tion, on a per capita basis, to be between face-to- face and the “conventional” media-based interven- tions as well. Through this review we found social media-based interventions to lead to a modest re- duction in body weight, BMI, and waist circum- stance and a reasonably large increase in daily number of steps taken. However, intervention im- plementation cost was unreported by the included studies, which precludes a cost-effectiveness anal- ysis in comparison to other intervention types.

The impact of peers and friends on dietary be- havior and physical activity has been documented extensively.100-102 The influential (as well as contro- versial) work by Christakis and Fowler reported that obesity spread in social networks independent of geographic distances and neighborhood charac- teristics.103 Social media provide convenient and nearly instantaneous channels for peer effects to materialize and magnify.104,105 Health promotion in- terventions delivered through social media are fun- damentally different from those delivered face-to- face or via “traditional” media in that participants are not passive information receivers, but active in their interaction with both intervention staff and other participants via online social networking.20,97 Acting as a mediator to the intervention, this high level of interaction creates a prompt feedback loop and either reinforces (when feedbacks are mostly positive) or weakens (when feedbacks are mostly negative) participation and adherence.20,97

The social cognitive theory predicts that individ- uals’ health behaviors are likely to be influenced by messages and “role models” in the mass me- dia.30,106 However, in the era of social media, in- dividuals may self-select into online social net- works that match their value system, and both receive and generate information in the networks so as to reinforce their opinions, beliefs, and be- haviors further.107 This creates both opportunities and challenges for designing social media-based health promotion interventions. On the one hand, given voluntary program participation, individuals signed up for the social media-based intervention are likely to be homogeneous in their health belief, and thus, reinforce each other’s behavior through online social networking, which leads to more desirable treatment outcome.108-110 On the other hand, social media represent an open and diverse socioecological system. It is virtually impossible to block or manipulate information flow.111,112

To our knowledge, this review remains the first

that exclusively focuses on health promotion in- terventions that used social media such as Face- book and Twitter. This differed from previous re- views that primarily evaluated studies that used self-designed website, text-based message board, discussion board, bulletin, or online forum.17,36-39 Our findings confirmed previous reviews on the effectiveness of social media-based interventions on weight-related behaviors and body weight sta- tus.17,36-39 The pooled effect sizes were statistically significant but modest in magnitude. An average reduction in body weight by one kilogram, BMI by a unit, and waist circumstance by less than 3 cen- timeters attributable to intervention may not have a large impact on a participants’ obesity status. However, this modest effect should not be treated lightly given that the interventions using social me- dia have the potential to scale-up and reach a large population pool.22,113 American adults on average take approximately 6500 steps per day based on data from the National Health and Nutrition Exam- ination Survey.114 An increase in daily number of steps taken by 1530 attributable to social media- based intervention accounts for nearly one-fourth of the population average, and indeed, does not fall very short of face-to-face or other physical activity interventions (an increase of 2000-2500 steps per day).115

Several limitations pertaining to the review and included studies should be noted. Besides the use of social media, many studies included in the re- view incorporated other intervention components such as wearable devices, emails, text messages, phone calls, face-to-face or Web-based counseling, and physical activity courses. Therefore, the esti- mated pooled effects should be interpreted as the overall effectiveness of social media-based health promotion interventions rather than the indepen- dent/net effect of social media use alone. Blind- ing was largely infeasible for social media-based interventions due to the high level of interactions among participants. Outcome evaluators of all in- cluded studies knew the intervention and expo- sure status of participants, and participants usu- ally knew the research question. Most included studies recruited a convenience sample that was non-representative of the target population that confined the generalizability of study findings. Novelty wears off over time for many social media users and this may diminish the effect of interven- tion.116 Interventions included in the review aver- aged 17.8 weeks in duration, which precluded an evaluation of the long-term effectiveness of social media-based interventions. Participants in the in- cluded studies were mostly young and middle-aged adults, whereas older adults that have the lowest social media adoption rate22 were not assessed. This raises the concern that social media-based interventions may not be equally adaptable to the older population partially due to the challenge of “digital divide.”117 Only a sub-sample of the includ- ed studies provided data on effect size and asso-

 

 

D el

iv er

ed b

y In

ge nt

a to

: ? IP

: 1 40

.2 34

.2 53

.9 o

n: S

at , 2

3 D

ec 2

01 7

06 :1

2: 30

C op

yr ig

ht (

c) P

N G

P ub

lic at

io ns

. A ll

rig ht

s re

se rv

ed .

An et al

Am J Health Behav.™ 2017;41(6):670-682 679 DOI: https://doi.org/10.5993/AJHB.41.6.1

ciated uncertainty that enabled a meta-analysis. Therefore, the estimated pooled effect sizes may not be generalizable to the entire scientific litera- ture. Although no publication bias was identified by the Begg’s and Egger’s tests, we could not elimi- nate the possibility of publication bias altogether, owing to the potential lack of statistical power be- cause of the small number of studies. Due to the high level of heterogeneity in dietary measures, a meta-analysis proved infeasible for quantifying the intervention effectiveness on the change in dietary behavior and/or diet quality. Despite the potential of reaching a large and diverse population pool of social media users, sample sizes of the included studies were typically small and no intervention of- fered convincing data for scaling up the trial at the population level.

The aforementioned limitations warrant future research. Innovative study design and implemen- tation is warranted to incorporate appropriate blinding procedures into social media-based in- terventions to prevent post-randomization differ- ential treatment of the groups and/or differential assessment of outcomes.118 Selection bias needs to be addressed by implementing a probability sam- pling design (eg, simple random sample, stratified sample or cluster sample).119 A longer follow-up period is required to assess the long-term effec- tiveness of social media-based interventions on weight-related behaviors and body weight status. Interventions may make efforts to recruit older adults aged 65 years and older, as currently no intervention has been tested in this population that is particularly vulnerable to the health risks of obesity and physical inactivity. A larger sample size is needed to evaluate the potential of scaling up social media-based interventions at the popu- lation level. Future research should document the cost of social media-based interventions so that a cost-effectiveness analysis can be conducted to en- able comparison to face-to-face intervention and/ or intervention via “traditional” media (eg, radio and television).

Appendices A-D Appendices can be accessed by sending request

to the corresponding author’s e-mail.

Conclusions We conducted a systematic review and meta-

analysis regarding the effectiveness of social me- dia-based interventions on weight-related behav- iors and body weight status. We identified 22 in- terventions from the keyword and reference search of bibliographical databases. Most used Facebook, followed by Twitter and Instagram. We found in our meta-analysis that social media-based inter- ventions were associated with a statistically signif- icant but modest reduction in body weight, BMI, and waist circumstance, and an increase in daily number of steps taken. Future interventions are warranted to adopt a probability sampling design,

incorporate innovative blinding procedures, ex- tend follow-up periods, increase sample sizes, and recruit older adults. The boom of social media pro- vides an unprecedented opportunity to implement health promotion programs. Future interventions should make efforts to improve intervention scal- ability and effectiveness.

References 1. Danaei G, Ding EL, Mozaffarian D, et al. The preventable

causes of death in the United States: comparative risk assessment of dietary, lifestyle, and metabolic risk fac- tors. PLoS Med. 2009;6(4):e1000058.

2. Behan DF, Cox SH, Lin Y, et al. Obesity and its rela- tion to mortality and morbidity costs. 2010. Available at: https://www.soa.org/Files/Research/Projects/ research-2011-obesity-relation-mortality.pdf. Accessed June 26, 2017.

3. Finkelstein EA, Trogdon JG, Cohen JW, Dietz W. An- nual medical spending attributable to obesity: payer- and service-specific estimates. Health Aff (Millwood). 2009;28(5):822-831.

4. An R. Health care expenses in relation to obesity and smoking among US adults by gender, race/ethnicity, and age group: 1998-2011. Public Health. 2015;129(1):29-36.

5. Pi-Sunyer X. The medical risks of obesity. Postgrad Med. 2009;121(6):21-33.

6. Fryar CD, Carroll MD, Ogden CL. Prevalence of over- weight, obesity, and extreme obesity among adults: United States, 1960-1962 through 2011-2012. 2016. Available at: http://www.cdc.gov/nchs/data/hestat/ obesity_adult_11_12/obesity_adult_11_12.pdf. Accessed June 26, 2017.

7. Flegal KM, Kruszon-Moran D, Carroll MD, et al. Trends in obesity among adults in the United States, 2005 to 2014. JAMA. 2016;315(21):2284-2291.

8. Wang Y, Beydoun MA, Liang L, et al. Will all Americans become overweight or obese? Estimating the progres- sion and cost of the US obesity epidemic. Obesity (Silver Spring). 2008;16(10):2323-2330.

9. NCD Risk Factor Collaboration (NCD-RisC). Trends in adult body-mass index in 200 countries from 1975 to 2014: a pooled analysis of 1698 population-based mea- surement studies with 19.2 million participants. Lancet. 2016;387(10026):1377-1396.

10. Koplan JP, Liverman CT, Kraak VI. Preventing Child- hood Obesity: Health in the Balance. National Academies Press: Washington DC; 2005.

11. Novak NL, Brownell KD. Role of policy and government in the obesity epidemic. Circulation. 2012;126(19):2345- 2352.

12. Macera CA. Promoting healthy eating and physical activ- ity for a healthier nation. Available at: http://www.cdc. gov/healthyYouth/publications/pdf/PP-Ch7.pdf. Ac- cessed June 26, 2017.

13. Wareham NJ, van Sluijs EM, Ekelund U. Physical activ- ity and obesity prevention: a review of the current evi- dence. Proc Nutr Soc. 2005;64(2):229-247.

14. Swinburn BA, Caterson I, Seidell JC, James WP. Diet, nutrition and the prevention of excess weight gain and obesity. Public Health Nutr. 2004;7(1A):123-146.

15. Wakefield MA, Loken B, Hornik RC. Use of mass me- dia campaigns to change health behavior. Lancet. 2010;376(9748):1261-1271.

16. Wellings K, Macdowall W. Evaluating mass media ap- proaches to health promotion: a review of methods. Health Educ. 2000;100(1):23-32.

17. Maher CA, Lewis LK, Ferrar K, et al. Are health be- havior change interventions that use online social net- works effective? A systematic review. J Med Internet Res.

 

 

D el

iv er

ed b

y In

ge nt

a to

: ? IP

: 1 40

.2 34

.2 53

.9 o

n: S

at , 2

3 D

ec 2

01 7

06 :1

2: 30

C op

yr ig

ht (

c) P

N G

P ub

lic at

io ns

. A ll

rig ht

s re

se rv

ed .

Effectiveness of Social Media-based Interventions on Weight-related Behaviors and Body Weight Status

680

2014;16(2):e40. 18. Lipschultz JH. Social Media Communication: Concepts,

Practices, Data, Law and Ethics. New York, NY: Rout- ledge; 2014.Lipschultz, J. H. (2014). Social Media Com- munication : Concepts, Practices, Data, Law and Eth- ics. New York: Routledge. http://public.eblib.com/ choice/publicfullrecord.aspx?p=1757845Lipschultz, J. H. (2014). Social Media Communication : Concepts, Practices, Data, Law and Ethics. New York: Routledge. http://public.eblib.com/choice/publicfullrecord. aspx?p=1757845Lipschultz, J. H. (2014). Social Media Communication : Concepts, Practices, Data, Law and Ethics. New York: Routledge. http://public.eblib.com/ choice/publicfullrecord.aspx?p=1757845

19. Tennant B, Stellefson M, Dodd V, et al. eHealth litera- cy and web 2.0 health information seeking behaviors among baby boomers and older adults. J Med Internet Res. 2015;17(3):e70.

20. Chou WY, Prestin A, Lyons C, Wen KY. Web 2.0 for health promotion: reviewing the current evidence. Am J Public Health. 2013;103(1):e9-18.

21. Baruah TD. Effectiveness of social media as a tool of communication and its potential for technology enabled connections: a micro-level study. International Journal of Scientific and Research Publications. 2012;2(5):1-10.

22. Perrin A. Social media usage: 2005-2015. Available at: http://www.pewinternet.org/2015/10/08/social-net- working-usage-2005-2015/. Accessed June 26, 2017.

23. Cromity J. The impact of social media in review. New Re- view of Information Networking. 2012;17(1):22-33.

24. Alves H, Fernandes C, Raposo M. Social media market- ing: a literature review and implications. Psychol Mark. 2016;33(12):1029-1038.

25. Barger V, Peltier JW, Schultz DE. Social media and con- sumer engagement: a review and research agenda. Jour- nal of Research in Interactive Marketing. 2016;10(4):268- 287.

26. Li JS, Barnett TA, Goodman E, et al. Approaches to the prevention and management of childhood obesity: the role of social networks and the use of social media and related electronic technologies: a scientific state- ment from the American Heart Association. Circulation. 2013;127(2):260-267.

27. Neiger BL, Thackeray R, Van Wagenen SA, et al. Use of social media in health promotion: purposes, key perfor- mance indicators, and evaluation metrics. Health Promot Pract. 2012;13(2):159-164.

28. Gold J, Pedrana AE, Stoove MA, et al. Developing health promotion interventions on social networking sites: rec- ommendations from The FaceSpace Project. J Med Inter- net Res. 2012;14(1):e30.

29. Bandura A. Social cognitive theory. In Vasta R, ed. An- nals of Child Development. Vol.6. Six Theories of Child De- velopment. Greenwich, CT: JAI Press; 1989:1-60.

30. Bandura A. Social cognitive theory of mass communica- tion. Media Psychol. 2001;3:265-299.

31. Stieglitz S, Dang-Xuan L. Social media and political communication: a social media analytics framework. Soc Netw Anal Min. 2013;3(4):1277-1291.

32. Adewuyi EO, Adefemi K. Behavior change communica- tion using social media: a review. Available at: http:// communicationandhealth.ro/upload/number9/EM- MANUEL-O-ADEWUYI.pdf. Accessed June 26, 2017.

33. Maher C, Ryan J, Kernot J, et al. Social media and appli- cations to health behavior. Curr Opin Psychol. 2016;9:50- 55.

34. Korda H, Itani Z. Harnessing social media for health promotion and behavior change. Health Promot Pract. 2013;14(1):15-23.

35. Khang H, Han EK, Ki EJ. Exploring influential social cog- nitive determinants of social media use. Comput Human Behav. 2014;36:48-55.

36. Chang T, Chopra V, Zhang C, Woolford SJ. The role of social media in online weight management: systematic review. J Med Internet Res. 2013;15(11):e262.

37. Williams G, Hamm MP, Shulhan J, et al. Social media interventions for diet and exercise behaviors: a system- atic review and meta-analysis of randomized controlled trials. BMJ Open. 2014;4(2):e003926.

38. Mita G, Ni Mhurchu C, Jull A. Effectiveness of social me- dia in reducing risk factors for noncommunicable diseas- es: a systematic review and meta-analysis of randomized controlled trials. Nutr Rev. 2016;74(4):237-247.

39. Willis EA, Szabo-Reed AN, Ptomey LT, et al. Do weight management interventions delivered by online social networks effectively improve body weight, body composi- tion, and chronic disease risk factors? A systematic re- view. J Telemed Telecare. 2017;23(2):263-272.

40. Pagoto S, Waring ME, May CN, et al. Adapting behavioral interventions or social media delivery. J Med Internet Res. 2016;18(1):e24.

41. Higgins J, Green S. Cochrane Handbook for Systematic Reviews of Interventions. Available at: http://handbook. cochrane.org/. Accessed June 26, 2017.

42. Effective Public Health Practice Project. Quality assess- ment tool for quantitative studies. Available at: http:// www.ephpp.ca/PDF/Quality%20Assessment%20 Tool_2010_2.pdf. Accessed June 26, 2017.

43. Lao LS. Evaluation of a social networking based SNAP- Ed nutrition curriculum on behavior change. Avail- able at: http://digitalcommons.uri.edu/dissertations/ AAI1501411. Accessed June 26, 2017.

44. Ruotsalainen H, Kyngäs H, Tammelin T, et al. Effec- tiveness of Facebook-delivered lifestyle counselling and physical activity self-monitoring on physical activity and body mass index in overweight and obese adoles- cents: a randomized controlled trial. Nurs Res Pract. 2015;2015:159205.

45. Wójcicki TR, Grigsby-Toussaint D, Hillman CH, et al. Promoting physical activity in low-active adolescents via Facebook: a pilot randomized controlled trial to test fea- sibility. JMIR Res Protoc. 2014;3(4):e56.

46. Al-Eisa E, Al-Rushud A, Alghadir A, et al. Effect of mo- tivation by “Instagram” on adherence to physical ac- tivity among female college students. Biomed Res Int. 2016;2016:1546013.

47. Aschbrenner KA, Niland JA, Shevenell M, et al. A pilot study of a peer-group lifestyle intervention enhanced with mhealth technology and social media for adults with Se- rious Mental Illness. J Nerv Ment Dis. 2016;204(6):483- 486.

48. Aschbrenner KA, Naslund JA, Shevenell M, et al. Fea- sibility of behavioral weight loss treatment enhanced with peer support and mobile health technology for individuals with serious mental illness. Psychiatr Q. 2016;87(3):401-415.

49. Cavallo DN, Sisneros JA, Ronay AA, et al. Assessing the feasibility of a web-based weight loss intervention for low-income women of reproductive age: a pilot study. JMIR Res Protoc. 2016;5(1):e30.

50. Cavallo DN, Tate DF, Ward DS, et al. Social support for physical activity-role of Facebook with and without struc- tured intervention. Transl Behav Med. 2014;4(4):346- 354.

51. Cavallo DN, Tate DF, Ries AV, et al. A social media-based physical activity intervention: a randomized controlled trial. Am J Prev Med 2012;43(5):527-532.

52. Chung AE, Skinner AC, Hasty SE, Perrin EM. Tweeting to health: a novel mHealth intervention using Fitbits and Twitter to foster healthy lifestyles. Clin Pediatr (Phila). 2016. pii: 0009922816653385. [Epub ahead of print]

53. Wang CK, Leng HK, Kee YH. Use of Facebook in physical activity intervention programme. A test of self-determi- nation theory. Int J Sport Psychol. 2014;45:559-580.

 

 

D el

iv er

ed b

y In

ge nt

a to

: ? IP

: 1 40

.2 34

.2 53

.9 o

n: S

at , 2

3 D

ec 2

01 7

06 :1

2: 30

C op

yr ig

ht (

c) P

N G

P ub

lic at

io ns

. A ll

rig ht

s re

se rv

ed .

An et al

Am J Health Behav.™ 2017;41(6):670-682 681 DOI: https://doi.org/10.5993/AJHB.41.6.1

54. Foster D, Linehan C, Kirman B, Lawson S. Motivating physical activity at work: using persuasive social media for competitive step counting. In Proceedings of the 14th International Academic MindTrek Conference: Envisioning Future Media Environments. New York, NY: Association for Computing Machinery; 2010:111-116.

55. Godino JG, Merchant G, Norman GJ, et al. Using social and mobile tools for weight loss in overweight and obese young adults (Project SMART): a 2 year, parallel-group, randomised, controlled trial. Lancet Diabetes Endocrinol. 2016;4(9):747-755.

56. Patrick K, Marshall SJ, Davila EP, et al. Design and im- plementation of a randomized controlled social and mo- bile weight loss trial for young adults (project SMART). Contemp Clin Trials. 2014;37(1):10-18.

57. Merchant G, Weibel N, Patrick K, et al. Click like to change your behavior: a mixed methods study of college students’ exposure to and engagement with Facebook content designed for weight loss. J Med Internet Res. 2014;16(6):e158.

58. Hales SB, Davidson C, Turner-McGrievy GM. Varying so- cial media post types differentially impacts engagement in a behavioral weight loss intervention. Transl Behav Med. 2014;4(4):355-362.

59. Herring SJ, Cruice JF, Bennett GG, et al. Preventing excessive gestational weight gain among African Ameri- can women: a randomized clinical trial. Obesity (Silver Spring). 2016;24(1):30-36.

60. Herring SJ, Cruice JF, Bennett GG, et al. Using technol- ogy to promote postpartum weight loss in Urban, low- income mothers: a pilot randomized controlled trial. J Nutr Educ Behav. 2014;46(6):610-615.

61. Chee HP, Hazizi AS, Barakatun MY, Mohd MT. A ran- domised controlled trial of a Facebook-based physical activity intervention for government employees with met- abolic syndrome. Mal J Nutr. 2014;20(2):165-181.

62. Joseph RP, Keller C, Adams MA, Ainsworth BE. Print versus a culturally-relevant Facebook and text message delivered intervention to promote physical activity in Af- rican American women: a randomized pilot trial. BMC Women’s Health. 2015;15:30.

63. Kernot J, Olds T, Lewis LK, Maher C. Usability testing and piloting of the Mums Step It Up program – a team-based social networking physical activity intervention for wom- en with young children. PLoS One. 2014;9(10):e108842.

64. Nishiwaki M, Nakashima N, Ikegami Y, et al. A pilot life- style intervention study: effects of an intervention us- ing an activity monitor and Twitter on physical activ- ity and body composition. J Sports Med Phys Fitness. 2017;57(4):402-410.

65. Napolitano MA, Hayes S, Bennett GG, et al. Using Facebook and text messaging to deliver a weight loss program to college students. Obesity (Silver Spring). 2013;21(1):25-31.

66. Pagoto SL, Waring ME, Schneider KL, et al. Twitter-deliv- ered behavioral weight-loss interventions: a pilot series. JMIR Res Protoc. 2015;4(4):e123.

67. Rote AE, Klos LA, Brondino MJ, et al. The efficacy of a walking intervention using social media to increase physical activity: a randomized trial. J Phys Act Health. 2015;12(Suppl 1):S18-S25.

68. Turner-McGrievy GM, Tate DF. Weight loss social sup- port in 140 characters or less: use of an online social network in a remotely delivered weight loss intervention. Transl Behav Med. 2013;3(3):287-294.

69. Turner-McGrievy GM, Tate DF. Tweets, Apps, and Pods: results of the 6-month Mobile Pounds Off Digitally (Mo- bile POD) randomized weight-loss intervention among adults. J Med Internet Res. 2011;13(4):e120.

70. Valle CG, Tate DF, Mayer DK, et al. A randomized trial of a Facebook-based physical activity intervention for young adult cancer survivors. J Cancer Surviv. 2013;7(3):355-

368. 71. Valle CG, Tate DF, Mayer DK, et al. Exploring media-

tors of physical activity in young adult cancer survivors: evidence from a randomized trial of a Facebook-based physical activity intervention. J Adolesc Young Adult On- col. 2015;4(1):26-33.

72. West DS, Monroe CM, Turner-McGrievy GM, et al. A technology-mediated behavioral weight gain prevention intervention for college students: controlled, quasi-ex- perimental study. J Med Internet Res. 2016;18(6):e133.

73. George Washington University. Social media obesity treatment for college students. Available at: https:// clinicaltrials.gov/show/NCT02342912. Accessed June 26, 2017.

74. Stanford University. Tweet4Wellness: an online virtual support group for promoting increased physical activ- ity. Available at: https://clinicaltrials.gov/ct2/show/ NCT02958189. Accessed June 26, 2017.

75. University of Pennsylvania. Twitter and diabetes. Avail- able at: https://clinicaltrials.gov/show/NCT02806700. Accessed June 26, 2017.

76. University of Massachusetts, Worcester. Mentoring in mHealth and social networking interventions for CVD risk seduction. Available at: https://clinicaltrials.gov/ show/NCT02656680. Accessed June 26, 2017.

77. The University of Hong Kong. A community-based family holistic health promotion project in Hong Kong: a cluster randomized controlled trial (cRCT). Available at: https:// clinicaltrials.gov/show/NCT02866773. Accessed June 26, 2017.

78. Dartmouth-Hitchcock Medical Center. Peer support and mHealth targeting cardiometabolic risk in young adults with serious mental illness. Available at: https://clini- caltrials.gov/show/NCT02815813. Accessed June 26, 2017.

79. University of Massachusetts, Worcester. Get social: ran- domized trial of a social network delivered lifestyle in- tervention. Available at: https://clinicaltrials.gov/ct2/ show/NCT02646618. Accessed June 26, 2017.

80. University of Kansas Medical Center. Remote delivery of weight management by phone and social media. Avail- able at: https://clinicaltrials.gov/show/NCT02496871. Accessed June 26, 2017.

81. Fundación Centro Nacional de Investigaciones Cardio- vasculares Carlos III. Long-term impact evaluation of a worksite-based lifestyle intervention to reduce car- diovascular risk in office workers (TANSNIP). Available at: https://clinicaltrials.gov/show/NCT02561065. Ac- cessed June 26, 2017.

82. Rutgers, The State University of New Jersey. Mobile health fitness program for adolescent and young adult childhood cancer survivors (TLC FIT). Available at: https://clinicaltrials.gov/show/NCT02688192. Ac- cessed June 26, 2017.

83. Massachusetts General Hospital. Improving physical ac- tivity through a mHealth intervention in cardio-metabol- ic risk patients. Available at: https://clinicaltrials.gov/ show/NCT02551640. Accessed June 26, 2017.

84. Northwestern University. NUYou: mHealth intervention to preserve and promote ideal cardiovascular health (NUYou). Available at: https://clinicaltrials.gov/ct2/ show/NCT02496728. NLM Identifier: NCT02496728. Ac- cessed June 26, 2017.

85. University of Colorado, Denver. Partnership to improve nutrition and adiposity in prenatal clinical care: a pilot and feasibility study. Available at: https://clinicaltrials. gov/ct2/show/NCT02520167. Accessed June 26, 2017.

86. Heidi Ruotsalainen. Promoting overweight adolescents physical activity. Available at: https://clinicaltrials.gov/ show/NCT02295761. Accessed June 26, 2017.

87. University of California, San Francisco. PilAm Go4Health weight loss program to prevent heart disease. Available

 

"Is this question part of your assignment? We can help"

ORDER NOW