10. Engagement and adherence outcomes.

Study	Social Media Intervention Component	Outcome	Result	Good adherence (assessed as >70% engagement with or adherence to the social media intervention)
Ahmad2018	Facebook and WhatsApp	Participation in WhatsApp (accessing information)	96.9% of parents participated in WhatsApp.	Good adherence
		Participation in Facebook groups (accessing information)	81.3% of parents participated in Facebook.
Ahmad 2020	Web‐based	Watching videos	65% (24/37) watched 7 to 12 videos from “start to finish".	Cannot assess adherence by watching videos or responding not applicable to the appropriateness, supportiveness, and informativeness of exchanges in discussion forums.
		Exchanges on discussion forums	18.9% (7/27) of participant chose “not applicable” for questions on the appropriateness, supportiveness or informativeness of the discussion forum, indicating their nonuse.
Ashton 2017	Step count app	App use and step count uploads	Step counts were uploaded for an average of 48 (SD 19) out of the available 65 days (range of 10–65 days per participant).	48/65 days =74%, 75% read posts Good adherence
	Facebook group	Facebook posts	All participants (100%, n = 24) joined the Facebook group, with a total of 23 posts, of which 22 were posted by the moderator. Average of 20 views and 1.8 ‘likes’ per post. In total, 75% (n = 18) reported reading the weekly Facebook posts.
Bai 2015	Professional‐guided online community	Not reported	Not reported	Not reported
Baker 2011	Online discussion board	Not reported	Not reported	Not reported
Bantum 2014	Web‐based program	Number of sessions attended Website posts	Number of sessions ever attended (logged on at least once) was mean 5.3 (SD 1.28, range 0‐6) 67.0% (203/303) of participants attended all six sessions, and 86.8% (263/303) attended 4 or more. Average of 46 posts per participant over the six‐week intervention period (total of 8016 posts)	87% attended 4 or more sessions Good adherence
Bender 2017	Facebook	Wearing Ftibit	97% (21/22) wore Fitbit at least 5 days per week.	Good adherence
		Logging weight	79% (17/22) logged weight a least once per week.
		Logging food/calorie intake	89% (20/22) logged food/calorie intake at least once per week.
Booth 2018	Twitter	Not reported	Not reported	Not reported
Boyd 2019	Facebook	Session attendance	100% (n=12) of mothers in the social media group attended at least one online session. Average attendance was 83%.	Good adherence
		Comments	Average participant commenting was 73%.
Bull 2012	Facebook	Loyal visitors on Just/Us	93 individuals (10%) were “loyal” visitors who posted regularly to the page Total of 277 posts by visitors to the page during the study period.	Low adherence
Castillo 2013	Website	Not reported	Not reported	Not reported
Cavalcanti 2019	Facebook	Not reported	Not reported	Not reported
Cavallo 2012	Facebook	Posts	Intervention participants who posted more than once (n=37) had on average 8.0 Facebook interactions during the intervention. 81 (32%) interactions were to the discussion board and 130 (50%) were posts or responses to the group wall.	Cannot assess adherence by number of posts
Chai 2018	WeChat messenger	Not reported	Not reported	Not reported
Chen 2019	WeChat	Not reported	Not reported	Not reported
Cheung 2015	Facebook	Posts	Mean posts was 21.0 (SD 34.4) in the Facebook social groups. 58% (23/40) of the Facebook participants did not post any (median 0, IQR 2.3). Posts were sharing of smoking or quitting experiences (81/123 posts, 65.9%) and simple reply to the moderator’s inquiry (82/123 posts, 66.7%).	Low adherence
	WhatsApp	Posts	Mean number of posts was 55.0 (SD 50.7). The WhatsApp social groups had more moderators’ posts (median 60, IQR 25 vs median 31.5, IQR 7; P=.05) and participants’ posts than Facebook (median 35, IQR 50 vs median 6, IQR 9; P=.07). 23 of 42 (54.8%) WhatsApp participants posted 1 to 9 times in the social group (median 3, IQR 7). The majority of posts were sharing of smoking or quitting experiences (151/384 posts, 39.3%) and simple reply to the moderator’s inquiry (WhatsApp: 131/384 posts, 34.1%).
Cobb 2014	E‐mail‐, web‐, and mobile‐based intervention	Not reported	Not reported	Not reported
Coffeng 2014	Social media platform	Not reported	Not reported	Not reported
Dadkhah 2014	Facebook	Not reported	Not reported	Not reported
Dahl 2019	Website and app	Group based interactions	The frequency of participant posts (daily on 5 of 7 days in a week) did not generate much discussion between members, except for one study team.	Cannot assess adherence by frequency of posts.
Daly 2017	Facebook	Not reported	Not reported	Not reported
Dehlendorf	Online community	Logins	Participants had an average of 7.90 logins during the study period (SD 6.89), with number of logins ranging from 0–63. 59 participants (12.1%) never logged in. Number of logins and percentage of participants who never logged in were not differential by arm.	Cannot assess adherence by number of logins
Duncan 2014	Interactive web site	Self‐monitoring entries	Median number of self‐monitoring entries per week at 3 months and 9 months was 1.00 (IQR 20.0) and 1.00 (IQR 21.5). The average number of self‐monitoring entries at 3 months and 9 months was 16.69 (SE 2.38) and 22.51 (SE 3.79), respectively. Following the initial reduction in usage between week 1 and week 3, usage continued to decline throughout the intervention period. No measure of usage of the IT‐based platform was associated with any of the behaviours examined.	Cannot assess adherence based on median entries
		“Mate” feature	21 participants (10%) used the Mate feature. No participants using this feature had more than 1 mate.
Edney 2020	App	Logging steps	Participants logged steps for an average of 72 (SD=35) out of 100 days in the gamified app.	72/100 days of logged steps = 72%. 89/100 days of gamified usage = 89%. Good adherence
		Usage	The gamified group used the additional gamified app features an average of 89 (SD=118) times during the 100‐day intervention.
Fiks 2017	Facebook	Facebook Posts	100% of participants joined and posted in their assigned Facebook group. Group members posted a mean of 30 times per group per week, which is more than twice the rate of posting that we had defined as ‘‘active engagement’’ (an average of 2 posts/group/day, or 14/week). In regression analyses, individual participation was inversely associated with weight‐for‐length z‐score (P = 0.007), but not other parent or child characteristics. Participants were most active in the groups around the perinatal period. During the prenatal curriculum (7 weeks), there were 1953 participant posts across the 4 groups; then 1802 from 0 to 3 months postnatal, 1074 from 3 to 6 months, and from 6 to 9 months, when curriculum content was posted less frequently, there were 553.	100% joined and posted in the group which met study authors' criteria for active engagement. Good adherence
		Posts	Conversations about feeding (63 total conversations across all groups; 31 were mother initiated and not moderator initiated), infant sleep (29 total, 9 mother initiated), infant activity (40 total, 12 mother initiated), and maternal well‐being (51 total, 19 mother initiated).
Glanz 2017	Website and interactive web site	Usage	Over the study period, the interactive web site offered 59 blog entries and 31 chat sessions. Participants contributed 90 comments and questions but a majority of the interaction was between participants and the research team rather than between participants.	Cannot assess adherence by number of comments
Gnagnarella 2016	Website	Participation – ‘social actions’	Participants actively participated to the discussion forum, blog and content with 293 total social actions including 97 messages or comments.	Cannot assess adherence by 'social actions'
Godino 2016	Facebook	Engagement	Median (IQR) engagement with the intervention declined over time: 98 (9–265) interactions at 6 months, 76 (0–222) at 12 months, 41 (0–198) at 18 months, and 12 (0–161) at 24 months. Participants with high levels of engagement did not achieve greater weight loss than participants with low levels of engagement (P > 0·05 at all time points).	Cannot assess adherence by interactions
Graham 2018	Online social network	Community use	The social networking feature increased community use by 15.3% (P < .001). In the arms receiving nicotine replacement therapy, the social network feature increased community use by a total of 14% points (P < .001).	Cannot assess adherence by community use
Graham 2011	Interactive web site	Not reported	Not reported	Not reported
George 2013	Facebook	Not reported	Not reported	Not reported
Greene 2013	Online social network	Not reported	Not reported	Not reported
Hammersley 2019	Facebook	Module completion	At least 5 of the 6 modules were completed by 29 participants (69%). 26 participants (61.90%) completed all 6 modules.	Cannot assess adherence by module completion.
Herring 2014	Facebook	Not reported	Not reported	Not reported
Herring 2017	Facebook	Facebook posts	11% of participants commented or “liked” posts, the average number of weekly coach posts was 1.7 ± 0.9, which waned over time.	Low adherence
Horvath 2013	Interactive web site	Intervention use	Participants updated their medication graph 3,935 times.	Cannot assess adherence by medication graph updates
Hutchesson 2018	Web‐based (website, app, email, text messages and social media)	Website logins	96.6% of participants (n = 28) logged‐on to the website.	Good adherence for logins to website
		Webpage visits	Webpage visits was 52 ± 29, ranging from 0–135. Most participants (89.6%) accessed the website in the first week of the program, with the proportion logging‐in declining thereafter.
		App usage	58.6% participants used the Easy Diet Diary application to self‐monitor their food (164 ± 312 entries), exercise (6.7 ± 11.1 entries) and/or weight (1.1 ± 2.2 entries). Engagement with the app declined until around Week 10, where it appears a more consistent group of users was established with 10.3% of participants using the app for nine of the remaining weeks.
		Opening email newsletters	On average, participants opened 10.0 ± 4.4 of the 19 email newsletters.
		Reading text‐messages	52.4% of participant reported reading text‐messages regularly.
		Facebook posts	The mean number of posts by participants was 1.8 ± 2.5.
		Facebook comments	The mean number of comments by participants was 12.4 ± 19.8.
		Instagram likes	There were 54 ‘likes’ with a mean number of likes of 1.9 ± 4.7 per participant, with no comments made
Hwang 2013	Online discussion forum	Forum messages	60% of participants posted at least one message; 27 posted only one message.	Low adherence
Jane 2017	Facebook	N/A	N/A	Not reported
Joseph 2015	Facebook	Not reported	Not reported	Not reported
Kernot 2019	Facebook and app	App visits	Participants visited the app on average 26 times (95% confidence interval [CI] = 21.5 to 30.5).	48/50 days of logged steps = 96%. Good adherence. Cannot assess adherence for other outcomes.
		Logging steps	Participants logged steps for 48/50 days (95% CI = 45.9 to 50.0) on average.
		Wall posts	Mean number of posts on the group message wall was 9 (95% CI = 5.9 to 12.1).
		Vitrual gifts	The mean number of virtual gifts sent to teammates was 7 (95% CI = 4.2 to 9.8).
Kim 2019	Naver Band	Not reported	Not reported	Not reported
Kolt 2017	Interactive web site	Average days with step comment per week	At 12‐18 months, interactive web site compared to basic web site: mean 0.03 (SD 0.15) vs. mean 0.06 (SD 0.39)	Cannot assess adherence based on days with comments
Koufopoulos 2016	Online community	Comments or posts	Total of 861 comments or posts left by users on intervention web site. 33 of 82 participants (40%) posted something on the site that was not purely a preventer post (i.e. posts about symptoms, questions about asthma, answer or reply comments) and there were a mean 3.24 (SD 0.94) non‐preventer posts over the 9 weeks of the intervention. Eight of 82 participants (10%) explicitly asked questions of the community and tended to post more frequently (mean number of non‐preventer posts in this subgroup was 5.38, SD 3.50).	Low adherence
		Site adherence	41.7% adherence in the control condition compared to 11.3% site adherence in the intervention condition at week 1.
Lau 2012	Interactive web site	Not reported	Not reported	Not reported
Li 2017	Tencent QQ	Posted three good things	23.2% (n=23) posted their three good things messages <1 day/week on QQ, 31.3% (n=31) posted 1‐4 day/week, and 16.2% (n=16) posted ≥5 days/week.	Low adherence.
		Reviewed posts by others	31.3% (n=31) reviewed messages posted by other members <1 day/week, 32.3% (n=32) 1‐4 day/week, and 26.3% (n=29) ≥5 days/week.
		Provided positive feedback	67.7% (n=67) provided positive feedbacks to other group members during the intervention period (55.6% provided 1‐10/week and 12.1% provided >10/week).
Li 2020	WeChat	Not reported	Not reported	Not reported
Liao 2020	WhatsApp	Participation in the discussion group	Overall, 58.1% (93/160) of the participants who joined the WhatsApp discussion groups participated in the online discussion, on average 3.08 posts (SD 5.90) per participant. A total of 434 posts from participants were retrieved over eight weeks, on average 13.6 posts per group per week.	Good adherence for reading the discussion posts. Low adherence for participation in discussion group.
		Reading discussion posts	Of the 117 participants of the intervention groups who completed the outcome assessment, 115 (98.3%) reported reading the discussion posts at least several times a week during the intervention period and 105 (89.7%) had read more than one half of all discussion posts
Linden 2017	Web‐based support	Number of entries to the self‐care diary	Median (min.; max.): 1 (0; 5850)	Low adherence
Looyestyn 2018	Facebook	Facebook interaction	The mean total number of interactions with the Facebook group in response to posts was 34.7 (SD 40.7; median 15 [interquartile range 62.3]; range 0‐158). 11/41 (27%) participants had more than 50 interactions	Cannot assess adherence by number of interactions
Lytle 2017	Social network web site.	Logins	50%–60% of participants logged in at the beginning of the study and self‐monitored their weight at least once a month; by 24 months only about 30% were regularly logging their weight.	Low adherence
Maher 2015	Facebook	Logging steps	28 (55%) logged steps for all 50 days of the program, 35 (69%) logged steps for 36 days or more. These steps were logged across a mean of 18 unique logins (SD 13.3, range 0‐46).	Low adherence
		Discussion posts	Participants made a mean of 2.7 wall posts to their team discussion wall (SD 3.4, range 0‐13).
Mailey 2016	Website discussion board and podcasts	Website engagement	Discussion board use was higher in the enhanced group throughout the 8‐week intervention.	Low adherence
		Podcast engagement	85% (n = 37) reported listening to the podcasts during weeks 1 and 2, compared to 55% during weeks 7 and 8.
Mascarenhas 2018	Google Hangout	Session attendance	Women in the intervention arm (n=30) attended 2.8 (SD 1.17) group video sessions per week on average for over 8 weeks. Participants attended 3.3 sessions per week in the first half of the study and 2.4 sessions per week in the second half.	Cannot assess adherence by number of sessions attended per week.
Morris 2015	Online social network	Overall engagement	Engagement was higher for the online social network group (mean 137.10, SD 20.93) than the control (expressive writing) (mean 122.29, SD 20.81).	Cannot assess adherence by number of posts.
		Posts	Intervention group mean 2.72 (SD 2.76) compared to control mean 6.62 (SD 8.76)
Moy 2015	Online social network	Online community forum	83.8% of the participants (129/154) used the online community forum at some point during the 12‐month study; 66.2% (102/154) of participants directly viewed an online community forum thread or entry, and 17.5% (27/154) of participants posted a new topic or a reply at least once.	Low adherence (17.5% posted)
Moy 2010	Interactive web site	Not reported	Not reported	Not reported
Nam 2015	Twitter	Tweets	Mean total tweets 79.66, relational tweets 47.48, self‐contained tweet 19.17, Twitter usability 4.06 (SD: 1.45)	Cannot assess adherence by number of tweets
Nam 2020	App	Not reported	Not reported	Not reported
Namkoong 2017	Facebook	Facebook use	37/101 (36.6%) posted at least one campaign idea, 50/101 (49.5%) made comments, and 74 /101(73.3%) “liked” others’ posts during the 3 weeks of the intervention period. There was a total of 160 posts (M = 1.29, SD = 3.04), 1,015 comments (M = 10.05, SD = 30.85), and 9,046 “likes” (M = 89.56, SD = 217.87).	Good adherence
Napolitano 2013	Facebook Plus	Likes	22.2% (4/18) of the participants ‘‘liked’’ the study‐related posts on the Facebook group. These participants ‘‘liked’’ the post once throughout their programs	>70% posted or commented at least once. Good adherence
		Posts	77.8% (14/18) posted or commented on the study‐related content at least once. The average number of posts per person among those who posted or commented was 1.3
		RSVP	72.2% (13/18) responded to the event invitations at least once during the program. The average number of RSVPs within those who responded was 8.56.
O’Neil 2016	Interactive web site	Not reported	Not reported	Not reported
Owen 2015	Interactive web site	Social networking component	Discussion board all users (n=296): mean 142.9 (SD 243) Personal pages all users (n=296): mean 132.4 (SD 243) Webmail all users (n=296): mean 1.8 (SD 6.2)	Low adherence
Petrella 2017	Online social network	Not reported	Not reported	Not reported
Ramo 2015	Facebook	Comments	77% (n = 192) commented at least once to their Facebook group. Median commenting was 13 [interquartile range (IQR) = 1–66], and among those who commented at least once was 31 (IQR = 7–84); 101 participants (40.6%) commented at least once during a live counselling session. Groups with an incentive had higher comment volume than no‐incentive, and monthly and weekly incentives had the highest volumes (none: median = 5; IQR = 0–25; monthly: median = 30; IQR: 2–87; weekly: median = 31; IQR: 2–94; daily: median = 11; IQR = 1–50).	Good adherence
Ramo 2018b	Facebook	Comments	Among TSP participants, 77% (n = 192) commented at least once to their Facebook group. Median commenting among the full TSP sample was 13 [interquartile range (IQR) = 1–66], and among those who commented at least once was 31 (IQR = 7–84). Comments were greater among those in pre‐contemplation (median = 22; IQR = 3–82) and preparation (median = 23; IQR = 2–70) than contemplation (median = 7; IQR = 0–57). Groups with an incentive had higher comment volume than no‐incentive, and monthly and weekly incentives had the highest volumes (none: median = 5; IQR = 0–25; monthly: median = 30; IQR: 2–87; weekly: median = 31; IQR: 2–94; daily: median = 11; IQR = 1–50).	Good adherence
		CBT session attendance	Ten (15.9% of participants in preparation) participated in one or more CBT treatment sessions during the 90‐day intervention period.
Ren 2019	WeChat	Not reported	Not reported	Not reported
Richardson 2010	Online social network	Posts	65% (165/254) of participants were using the online community, either as posters or “lurkers” (readers who did not post).	Low adherence
		Step data	The group with access to the online community uploaded valid pedometer data on more days than the control group (87% compared to 75% of days).
Rote 2015	Facebook	Facebook posts	Total of 196 posts (range 6‐32 per participant). Average of 7.1 (SD 4.5) posts per participant) Week 1 – 46 posts, week 2 – 30 posts, week 8 – 13 posts	Cannot assess adherence based on posts
Rouf 2020	Facebook	Seeing and reading Facebook posts	On average, Facebook posts were seen by 19.9 (SD=3.6) for the Facebook only group and 26.9 (SD = 5.0) for the Facebook plus text group. 21.7% (n=10) and 45.5% (n=20) of participants reported reading the Facebook posts all the time for the Facebook group and Facebook plus text group, respectively. 37.0% (n=17) and 25% (n=11) reported reading the messages more than once a week for the Facebook group and Facebook plus text group, respectively.	Low adherence for reading Facebook posts all of the time. Cannot assess adherence for other outcomes.
		Facebook likes	On average, Facebook likes were 1.1 (SD 1.4) for the Facebook only group and 3.6 (2.4) for the Facebook plus text group.
		Facebook comments	In the Facebook group, 3 participants made comments on posts (mean 0.1, SD 0.5), whereas 4 participants in the Facebook plus text group commented on posts (mean 0.2, SD 0.9).
		Facebook sharing	No participants shared their own content in the Facebook group for the Facebook group only. However, 9% (n=4) participants in the Facebook plus text group shared their own content in the Facebook group.
		Reading text messages	79.5% (n=20) of participants reported reading the text messages for the Facebook plus text group all the time and 15.9% reported more than once per week.
		Text replies	For the Facebook plus text group, the mean number of replies from participants was 3.8 out of a maximum 21 (range 1‐18). Of 75 participants, 12 made no reply texts. The highest number of replies was to the yes/no response as to whether they had set a goal on the app (n=22).
Rovniak 2016	Online social network	Engagement	Participants in the social network group performed a mean of 3.9 (SD = 2.6) of the 12 social networking activities targeted over the 12‐week intervention. Between the 3‐ and 9‐month assessments, 8.7 % of participants posted a discussion comment on the study’s online networking site; no other online networking activities were observed.	Low adherence
Schaller 2017	Online social network	Not reported	Not reported	Not reported
Schneider 2015	Online social network	Comments	42.9% reported that they never commented on the Meetup web site.	Low adherence
		Step uploads	39.3% of participants reporting that they used activity monitoring 5 times or less and 35.7% reporting daily or almost daily use; 60.7% of participants reported that they never recorded their steps on the web site
Stoddard 2008	Online social network	Time on Website	The average number of minutes spent for either condition was 14.4 minutes (n = 1083). Time on the web site was higher for those assigned to the social network condition (min 18.0, number 526) versus the control (min 11.1, number 557), those who reported abstinence (vs still smoking), and those who made a serious quit attempt by abstaining for at least 24 hours (vs those who did not). The longest time spent on the web site (30 minutes) was for those whose quit attempt began in the 5 days prior to registering for the study.	Low adherence
		Bulletin Board Use	Only 242 opted to look at the Bulletin Board feature by clicking on the link, and of those visiting the link, 81/242 either selected an individual message to view or posted a message. This low utilization rate 81/684 (11.8%) limited our ability to analyse the impact of bulletin board use on cessation.
Sun 2017	Facebook	Facebook use	159 Facebook interactions were recorded among 96 users. During the 6‐week intervention period, there were 144 “likes,” 13 comments, and 2 participant‐initiated posts. The posts on the Facebook group were viewed, on average, by 77% (74/96) of participants.	Cannot assess adherence by views
Turner‐McGrievy 2011	Twitter and podcasts	Twitter posts	94% (n = 44) posted at least once to Twitter and 64% (n = 30) posted at least weekly during the first 3 months and 28% (n = 13) posted weekly or more during months 3–6. Participants made a mean of 2.1 (SD 3.1) posts to Twitter per week, with significantly more posts being made in the first 3 months (2.8, SD 3.6 posts/week) than in months 3–6 (1.3, SD 3.0 posts/week; P < .001).	Good adherence
		Podcasts	Mean number of podcasts downloaded at 0–3 months podcast only group: 14.5 (7.6) or podcasts plus Twitter 16.4 (7.2), at 3–6 months podcast only: 8.2 (8.6), podcast plus Twitter: 9.0 (9.1) P‐value.67 Number of downloads per podcast episode per person at 0–3 months podcast only: 1.51 (0.65), podcast plus Twitter 2.00 (0.52), at 3–6 months podcast only: 0.66 (0.15), podcast plus Twitter: 0.87 (0.20)
Valle 2013	Facebook	Facebook posts	Participants posted a total of 153 Facebook comments to the group wall over 12 weeks compared with 188 comments by control group participants. There was no difference in mean number of posts over the 12‐week program (intervention=3.4 (4.6) vs. control=4.6 (7.8), and almost half of participants in both groups made two or more Facebook posts over the course of the study period (intervention, 48.9 % (n=22); control, 48.8 % (n=20))	62.5% of the intervention group Low adherence
Vandelanotte 2017	Interactive web site	Not reported	Not reported	Not reported
Vogel 2019	Facebook	Comment volume	Comment volume mean (SD) for the SGM‐tailored group (POP) was 60.2 (SD 41.0) and 78.4 (SD 35.1) for the TSP‐SGM group. On average, TSP‐SGM participants responded to posts with significantly more comments throughout the intervention than did POP participants.	Cannot assess frequency by mean volume of comments
Wan 2017	Online social support	Not reported	Not reported	Not reported
Wang 2015	Facebook	Not reported	Not reported	Not reported
Washington 2017	Facebook	Not reported	Not reported	Not reported
Willcox 2017	Facebook and interactive web site	Facebook Page	Fourteen participants (31%) joined the private Facebook page. These women recorded 558 views, a mean of 40 per participant, submitted seven posts and two questions, and indicated 18 likes.	Low adherence
Willis 2017	Online social network	Likes	The average number of ‘likes’ per person each week was 1.3 on study‐related posts and posts by a peer.	Cannot assess adherence by number of likes/comments
		Comments	The average number of comments per person each week was 3.2
Yang 2019	WeChat	Number of mindfulness sessions	The mean (SD) number of mindfulness meditations per week was 3.25 (1.45) times over the 8‐week intervention.	Cannot assess adherence by frequency or minutes of mediations per week.
		Minutes spent on meditation	The mean (SD) number of minutes spent on meditation each time was 21.23 (16.16) minutes.
Young 2013	Facebook	Messages	Participants who responded to messages during period 1: intervention 54/57 (94.7%), control 38/55 (69.1%) Period 2: intervention 52/57 (91.2%), control 34/55 (61.8%) Period 3: intervention 44/57 (77.2%), control 30/55 (54.5%)	Good adherence
		Wall posts	Participants who responded to wall posts during period 1: intervention 39/57 (68.4%), control 40/55 (72.7%) Period 2: intervention 26/57 (45.6%), control 34/55 (61.8%) Period 3: intervention 27/57 (47.4%), control 28/55 (50.9%)
		Chat	Participants who responded to chat during period 1: intervention 43/57 (75.4%), control 14/55 (25.5%) Period 2: intervention 41/57 (71.9%), control 11/55 (20.0%) Period 3: intervention 36/57 (63.2%), control 11/55 (20.0%)
Young 2015	Facebook	Not reported	Not reported	Not reported
Zhang 2016	Online social network	Website Engagement	Average of 22.8 logins (SD = 47.2) across all conditions during the program. Participants in the social support condition generated 81 online messages and participants in the combined condition generated 80 messages.	Cannot assess adherence by number of logins