Use of social media for sexual health promotion: a scoping review

Background In order to prevent sexually transmitted infections (STIs), the World Health Organization recommends educating people on sexual health. With more than 2 billion active users worldwide, online social media potentially represent powerful channels for health promotion, including sexual health. Objective To review the scientific literature on the use of online social media for sexual health promotion. Design A search was conducted of scientific and medical databases, and grey literature was also included. The selected publications were classified according to their study designs, sexual health promotion main subject, target audience age, and social media use. Results Fifty-one publications were included; 4 publications presenting randomized intervention studies, 39 non-randomized intervention studies, and 8 observational studies. In 29 publications (56.9%), the main subject of the sexual health promotion was ‘general’ or to increase STI testing. Thirty publications (58.8%) specifically focused on youth or young people (aged 11–29 years). Fourteen publications that used social media either as unique channels for sexual health promotion interventions or as a tool supporting the sexual health promotion reported an effect on behavior (27%), and two of those studies found a reduction in the number of positive chlamydia and gonorrhea cases linked to social media intervention. Forty-four publications (86.3%) involved Facebook in some way. Conclusions Although billions of people worldwide actively use social media, we identified only 51 publications on the use of social media for promoting sexual health. About a quarter of the publications have identified promising results, and the evidence for positive effects of social media interventions for promoting sexual health is increasing. There is a need for more studies that explicitly discuss their theoretical framework, and that have strong research designs, in order to further increase the evidence base of the field.


Introduction
It is estimated that about a million people globally acquire a sexually transmitted infection (STI) every day, including many who are infected with the human immunodeficiency virus (HIV) (1,2). Individuals who have STIs may experience acute or chronic symptoms. Women with STIs in particular are at risk for pelvic inflammatory disease, cervical cancer, infertility, ectopic pregnancy, and transmitting STIs to their children during birth (1,2). In order to prevent STIs, minimize the effects among infected people, and to reduce the enormous burden that STIs represent for developed and developing countries, the World Health Organization (WHO) has published the Global Strategy for the Prevention and Control of Sexually Transmitted Infections, where the need for educating people on sexual health is emphasized (1,2). In order to achieve this educational goal, the WHO points to the importance of choosing the communication channels that most effectively reach the target population (1,2).
Online social media, such as Facebook or Twitter, have become extremely popular worldwide and might therefore be powerful channels for reaching many people. Since their launch, the adoption of these technologies has been steeply increasing, surpassing 2 billion active users worldwide in 2015. Facebook is the most used social media channel, with more than 1.5 billion users (3).
Because social media are popular and frequently used by many people of various ages worldwide, there is potential for the media to be used for health promotion (4Á7), including for potentially sensitive and stigmatizing subjects such as those related to sexual health (8Á10). Previous reviews have addressed the use of computer-based technologies in general for sexual health promotion but not specifically the social media (11Á14). And some reviews have analyzed the use of social media for health promotion in general, including sexual health (8,10). We have identified prior reviews that addressed the use of online social media specifically for sexual health education or sexual health promotion. One was an early (2011) review of the literature that found that 71% of the earliest promotion activities used Facebook, 30% targeted young people, and 25% specifically addressed HIV (15). A more recent review focused specifically on the impact of social media interventions targeting adolescents and young adults. The study reported that online social media can increase knowledge regarding STI prevention in this population, but the evidence was weaker regarding the effects on behavior change (16). Another review studied the viability of social media as tools that health care professionals can provide to adolescents (17). The authors highlighted the need for further studies on how to use these technologies to educate adolescents about STIs (17).
The number of publications and programs describing social media as a means for promoting sexual health is rapidly increasing. There is a need for an updated review of the literature that is not restricted to youth or to peerreviewed publications, specific uses of social media, or to specific types of outcomes or study designs, that examines the literature regarding the use of social media for STI prevention and health promotion. The objective of this scoping review is to describe the scientific literature on the use of online social media for sexual health education and sexual health promotion.

Methods
To analyze the use of online social media for sexual health education or sexual health promotion, we used a systematic approach, drawing on Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (18). A full electronic search strategy covered all the studies published until the end of October 2015 involving the terms 'sexual health promotion' or 'sexual health education' in combination with the following words: 'social media'; 'social networking'; 'Facebook'; 'Twitter'; 'YouTube'; 'Instagram'; and 'Snapchat'. The search was performed in the following multidisciplinary databases: Embase, Pubmed (MeSH terms and text word), PsychINFO, Applied Social Sciences Index and Abstracts (ASSIA), ProQuest Health and Medical Complete, British Nursing Index, Computer and Information System Abstracts, and MEDLINE (Ovid). Publications were included in the review if they: 1) were empirical studies reporting results, 2) described studies that used social media as a tool for sexual health promotion or education, and 3) were written in English. Papers that did not meet all three criteria were excluded.
Following the search, duplicates were removed. Thereafter, all the titles and abstracts were examined by one reviewer (EG) to determine if the papers met the inclusion criteria; doubts regarding their inclusion/exclusion were discussed and agreed with a second reviewer (RW). Subsequently, full-text articles of the selected studies were retrieved and rigorously examined to sort out any remaining papers that did not meet the criteria. The articles selected for full review were classified according to their study designs (i.e. randomized or, non-randomized intervention studies or observational studies); main sexual health promotion subject (incurable STI, HIV; curable STIs, i.e. chlamydia/syphilis/gonorrhea/human papillomavirus (HPV); or STI prevention or sexual health in general); target audience age (specifically youths or young people; adults; or unspecified/general); sexual preference (straight/unspecified/all; men who have sex with men [MSM]; lesbian, gay, bisexual, and transgender [LGBT]); and social media use (as a unique channel for the promotion or as a tool supporting sexual health promotion). Data were extracted by one reviewer (EG) and verified by a second reviewer (RW).

Results
Sample A total of 9,462 publications were identified; the search strategy and its results are summarized in the Supplementary file. Fifty-one of these publications met the inclusion criteria (19Á69) (see Fig. 1). Forty-one of these 51 publications corresponded to unique studies; in addition, three publications belonged to Project HOPE (26,40,43); three were part of the Get Yourself Tested campaign (19,37,50); two papers referred to the FaceSpace Project (23,33); and two studies reported on an HIV self-test campaign on Grindr. All the included studies were carried out between 2008 and 2015.

Designs of included publications
Among the 51 included publications, 4 reported on randomized intervention studies, 39 reported on nonrandomized interventions, and 8 reported on observational studies. A summary of the publications included in this review is presented in Table 1.

Randomized studies
Only four of the included papers had a design that involved a randomization procedure (22,26,38,40). Two of these publications, belonging to the Project HOPE study, described an intervention in which participants were randomized to receive either peer-delivered HIVrelated information or general health information through Facebook for 12 weeks, framed within a social network intervention and, specifically, a peer-delivered intervention (26,40). The study reported that peer leaders felt more comfortable discussing sexual health on social media (93.3% and 100% vs. 68.8% and 93.8%, pB0.05) and also found a positive association between participation in the group and the likelihood of HIV testing (26,40). Another study randomized the participants to an intervention through a Facebook page on youth health information or to a News page on Facebook for 2 months (22). The study found a higher tendency to use condoms in the intervention group at 2 months (intervention 68% versus control 56%, pB0.04) and more frequent protection in sex acts (intervention 63% versus control 57%, Studies included in qualitative synthesis (n = 51)  LGBT youth in same-sex relationship (16Á20 years old)   LGBT, lesbian, gay, bisexual, and transgender; MSM, men who have sex with men; STI, sexually transmitted infection. Social media for sexual health promotion pB0.03) but no lasting effect at 6 months (p00.86) (22). Another study randomized health centers to offer a reminder service to increase HPV vaccination completion (including messages sent through Facebook) or to schedule routine follow-up (38). The intervention group did not increase vaccine completion rates (38). All the randomized studies were carried out in the US. It is difficult to compare the effects of these randomized studies because they measure outcomes as different as feeling comfortable discussing sexual health on social media and rates of HIV testing (26,40), condom use (22), and HPV vaccination completion rates (38). Two of the three studies did report some type of positive statistically significant outcome from social media interventions.

Non-randomized intervention studies
Thirty-nine of the selected publications had a nonrandomized design. Eighteen of them only showed data regarding project reach and engagement (number of users, time spent, etc.) (23, 28, 30, 43, 48, 54, 56, 59Á69), while another 10 publications additionally showed data on STI incidence and testing (19, 24, 36, 37, 47, 49, 51Á53, 55). The designs in 12 of the publications combined quantitative data on technology use and also questionnaires or opinions of project users regarding their satisfaction, knowledge, engagement, or behavior change (intention to test) (29,30,33,34,39,41,42,46,50,58,63,67). The remaining study had an observational approach and analyzed the strategies for successful user engagement in some Facebook and Twitter profiles undertaking sexual health promotion (45).
Regarding the effect of these non-randomized interventions, two studies reported that chlamydia and gonorrhea infections were reduced by 54 and 51%, respectively, after the intervention (24,49), while the number of syphilis cases was increased by 78% in another study as a result of an increased number of STIs tests (47). Three studies reported increases in the number of patients presenting for STI testing by 71% (19), 122% (47), and up to 128% (37); and one study found a significant increase in intention to test from 22.3 to 43.9% (30). As a result of the intervention, 34 and 249 program users, respectively, attended health services (52, 63) and 667 asked for a self-test (53). Some non-randomized intervention studies also reported an increase in condom use (74% vs. 56%) (55) and (73% vs. 43%) (67); a 23% self-reported increase in condom utilization (24); or a significantly increased intention to use condoms (34.2% vs. 26.2%) (30).
Only five of these non-randomized intervention studies or programs referred to the theoretical models they used as a framework for behavior change intervention (24,30,41,58,65). The chosen approaches were Kelly's popular opinion leader model; game-based learning with participatory approach; Pender's health promotion model; peer educa-tion models; and a community engagement model. Two projects referring to the Kelly's popular opinion leader model and Pender's health promotion model, respectively, reported positive results regarding an increase in intention to test (43.9% vs. 22.3%) and in intention to use condoms (34.2% vs. 26.2%) (30); 23% self-reported an increase in condom utilization, and 54% reported a reduction in positive chlamydia cases among 15Á17 years olds (24).

Observational studies
Eight studies were observational*where the subjects participated in focus groups or answered questionnaires or surveys (20,21,25,27,31,35,44,57). In all the studies with young people, social media were reported to be pervasive, and the study participants reacted positively to using new technologies for sexual health promotion or education (20,21,25,27,31). In the observational studies with adults, the importance of considering privacy, stigma, and social norms was emphasized (44,57), and in this sense, links to social media profiles were not considered to be appealing (35). No theoretical framework was reported in any of the observational studies included in this review.

Target audience age
Among the 51 included publications, 30 (58.8%) specifically focused on sexual health education for youth or young people (11Á29 years), while 11 publications (21.6%) explicitly targeted adults (including young adults and middle-aged and older adults). The 10 remaining publications (19.6%) were not directed to any specific age group ( Table 2).

Use of social media
In 23 of the included studies (45%), social media was the core or the unique channel used for sexual health promotion. These studies involved Facebook, either used separately (22,24,30,36,39,40,43,56) or in addition to other social media channels (such as Twitter, Flickr, YouTube, etc.) (20,23,26,33,34,44,45,55,64,66). Three studies used the geosocial networking app Grindr (51Á53); one was carried out on YouTube (62); and one publication referred to two purpose-developed online social networks (57). Eight publications that used social media as unique channels for sexual health promotion interventions reported an effect on behavior (three randomized trials and five non-randomized interventions). The reported effects were: increased condom use (74% vs. 56% before the intervention; and 68% in the intervention group versus 56% in the control group, pB0.05), 23% self-reported an increase in condom use or intention to use condoms (34.2% vs. 26.2%); 34 social media users utilized health services; 667 users requested an HIV self-test and indicated an increased intention to test (43.9% vs. 22.3%); users were more comfortable using social media to discuss sexual health topics (pB0.05) (22,24,26,30,40,52,53,55). One of these publications also found a 54% reduction in the number of positive chlamydia cases linked to social media intervention in a specific age group (24).
In 28 of the publications (55%), social media was considered as a resource supporting another sexual health promotion channel. Twenty of these 28 publications considered more than one social media channel as a supporting resource for the promotion (i.e. Twitter, Instagram, YouTube, Whatsapp, Vkontakte) or did not specify a unique resource (19,21,25,28,29,31,35,37,42,47,49,54, 59Á61, 63, 65, 67Á69). Only one of these studies did not refer specifically to Facebook. In this latter study, the sexual health promotion was carried out on YouTube, and the comments on the study were tracked though Twitter (19). When a unique social media channel to support the promotion was chosen, Facebook was the medium of choice in seven publications (32,38,41,46,48,50,58) and Twitter in one study (27). Six publications using online social media as a tool sup-porting sexual health promotion reported an effect on behavior, all of them non-randomized interventions. The effects were 73% self-reporting condom use (versus 43% before the intervention) (67); 249 new users utilizing health services (63); and an increase in STI testing by 71 to 128% (19,37,47). Two publications reported a reduction in the number of cases of gonorrhea by 51% and also a 78% increase in syphilis cases as a result of the increase in syphilis tests (47,49).

Overview
The use of the Internet for health purposes has been increasing for a long time (70,71), and many health services around the world now offer Internet-based services (16,72). The development of online social media is more recent, but these media have become very popular (3), offering a strong potential for health-related use and also within the field of sexual health (15). The present review shows that although online social media have been used in the sexual health promotion research field, the number of scientific studies is still relatively modest; we were able to identify 51 publications that fit the inclusion criteria.
The use of social media for sexual health promotion is a rapidly emerging field. Although some reviews have previously been published dealing with the topic (10Á17), these are either several years old, or they only report on selected target groups (i.e. youth, high-risk men, etc.). The four RCTs (Randomized Controlled Trials) reported on in the present review are relatively new and therefore were not included in several of the prior reviews. Our updated review includes recent publications covering all target groups. It also covers relevant sexual health promotion projects that have been carried out by NGOs, many of which have not been included in previous review papers.
There is a need to include randomization procedures and to discuss theoretical frameworks A main finding in this review was that 14 of the 51 publications (27%) reported a behavior change effect regarding sexual health that was linked to social media. However, only three of these publications were randomized studies (22,26,40), that is, studies with a design that allowed for controlled measurements of the effects of the interventions. Two of these quality studies (three publications) reported important results linked to the use of social media with regard to users feeling comfortable with sexual health topics on these channels (26,40), a short-term increase in condom use, and increased HIV testing rates (22). Another randomized trial did not find any benefit in sending reminders to complete a HPV vaccination schedule with the use of several channels, including social media (38). Although the four studies had a strong RCT design, they all reported on various limitations that might have impacted the findings, including a reliance on self-reporting (17), small sample sizes (21,35), and low completion rates (33). The different outcome measures used in the randomized studies make it difficult to compare the outcomes of the different social media interventions in these studies. However, we find it promising that two of the three studies did report positive statistically significant differences between the intervention and control groups. Only 4 of the 14 studies reporting behavior effects have been contextualized within a theoretical framework or model or have referred explicitly to a model. More studies that systematically evaluate interventions with a randomized control condition and that refer to a theoretical framework clearly are needed to increase the evidence and move the field forward.
A further 39 studies were non-randomized intervention studies or programs, a design that allowed for the testing out of various interventions, but where it would be somewhat more difficult to make strong claims about the effects of the interventions themselves (because there was no randomization). The non-randomized intervention studies seemed to have had positive results in terms of increases in rates of STI testing, higher condom use, and intention to test, better sexual health knowledge, and a potential to change behavior linked to the intervention (19,24,30,37,47,52,53,55,63,67), and even a change in curable STIs rates (24,47,49). Many of the studies reported that large numbers of people had visited educational sites or utilized the online interventions (23, 28, 29, 32Á34, 39, 43, 45, 48, 50, 54, 56, 59Á69).
Eight studies were observational and, although these designs may provide important insights*for instance, relating to willingness to use or obstacles to use online social media for sexual health (20,21,25,27,31,35,44,57), they do not allow for strong claims about the effects of social media on STI prevention and sexual health promotion. This means that, although there have been some studies examining the effects of the use of social media in this field, there still is a lack of studies with a more rigorous design allowing for stronger claims about the effects of such interventions.
It must be highlighted that only seven publications referred to some type of theoretical framework or model upon which the interventions were based. More knowledge is needed regarding the mechanisms that promote sexual health, and explicitly describing the theoretical underpinning of studies is important in order to further the knowledge base of any research field (73). The theoretical frameworks or models that were mentioned in these sexual health promotion studies were: a social network intervention framework (peer-delivered intervention) (26,40), game-based learning with a participatory design approach (41), Kelly's popular opinion leader model (30), peer education models (58), Pender's health promotion model (24), and a community engagement model (65).
Most of the publications focused on general sexual health promotion, targeted youth, and used Facebook The review also demonstrated that the majority of studies involving online social media for sexual health dealt with the topic in more of a general way or focused on increased STI testing. Only 15 of the publications dealt with the incurable STI, HIV, and seven dealt with curable STIs (chlamydia, syphilis, gonorrhea, or HPV). About 60% of the included publications specifically targeted young people (up to 29 years old), and these publications were mainly on STI health promotion in general. The studies that explicitly targeted adults were directed toward MSM/LGBT and mostly focused on HIV. We have found only a few intervention projects using social media for additional sexual health promotion topics, such as abortion and teenage pregnancy. These projects were carried out by NGOs. We believe that, in addition to health promotion on STIs, there especially is a need for further studies examining the effects of social media interventions for health promotion in other sexual health domains, such as maternal health, contraception, or female genital mutilation.
Although it is important to reach people at a young age to prevent early infection with STIs, this does not mean that other age groups should not be targeted. The problem of STIs primarily impacts young people and young adults, but it is also important to avoid STIs among adults older than 30. Although young people have been early adopters of online social media (74), today online social media are popular in all age groups, which means that online social media could be a good channel for sexual health promotion among older adults as well.
Studies from other health care fields utilizing social media have demonstrated positive effects on health outcomes dealing with wellness, obesity, and the management of chronic diseases (75). Thus, there is mounting evidence suggesting that online social media have a great potential in the field of health information and also with respect to some other types of health interventions. In the sexual health field, Facebook seems to be the most frequently used online social media channel, possibly because of the very high number of users (3). In this review, 44 papers (86%) involved Facebook as a channel for sexual health promotion*either as the only online social medium or in combination with other online social media. This review found that the use of Facebook is slightly higher than the 71% reported by Gold et al. (15) in a review of social media use for sexual health promotion published in 2011, when Facebook had 500 million active users worldwide (15).
Facebook and other social media are increasingly being used to reach specific populations for health promotion purposes. As evidence of the efficacy of this type of intervention for health promotion is increasing, organizations belonging to the public sector, NGOs, and other stakeholders might be encouraged to use social media more often for sexual health promotion and sexual health education.
Although the WHO reports that the majority of people suffering from STIs are located in Southeast Asia, sub-Saharan Africa, Latin America, and the Caribbean (1), the higher proportion (40) of the publications included in this review targeted populations from the US, Canada, Europe, Australia, and New Zealand. Only 10 of the publications reported on studies that had been carried out in other parts of the world, such as in Africa (Nigeria and Ghana) (60, 66, 67, 69), Southeast Asia (30,41,57), the Middle East (including Pakistan) (59,61), and in South America (Peru) (43). One publication involved 79 countries (29). Most of the sexual health education interventions carried out in developing countries are organized by NGOs. There is a discrepancy as to where the largest number of people affected by STIs live, the worldwide distribution of social media users, and where the most studies are carried out. This may in part represent a question of resources that are available for sexual health promotion because about 55% of all worldwide social media users are located in Asia and the Middle East; 13% in Central and South America; and 6% in Africa; while only about a quarter of all social media users are distributed in Europe, North America, and Oceania (76). However, other issues, including social and cultural factors, are also relevant. Social and cultural factors are of importance regarding how sensitive topics are discussed in general (77,78), and these factors might therefore influence the acceptability of using online social media for sexual health topics in different parts of the world.

Limitations
Only 51 publications matched the inclusion criteria in this review. Although we believe that the search terms used allowed us to capture relevant publications in this field, we might have missed publications that were not identified with the search terms or that were not published in the journals or databases that were searched. Because we searched the websites of the NGOs manually, some projects might have been missed as might publications from NGOs whose websites were not specifically searched. Some sexual health promotion interventions and programs using social media were excluded from our review because they did not report any study outcomes.
The included publications had heterogeneous designs and purposes and reported different types of outcomes. The quality of the studies varied, and there were only four publications that reported on studies with a strong randomized design. However, these studies did report findings with important implications, including increased condom use and HIV testing. Due to the small number of studies included in this review that reported effects, and the differences among the included studies, it was not possible to compare effect sizes or to conduct a meta-analysis.

Conclusions
Some studies have used online social media for sexual health promotion or STI prevention*either as a sole intervention or in combination with other interventions. Only a handful of studies have used a controlled randomization procedure. About a quarter of the publications identified promising results, and the evidence for the positive effect of social media interventions for promoting sexual health is increasing. However, there is a need for more studies with strong designs to increase the weight of the evidence and to move this emerging field forward.

Authors' contributions
EG designed the study, performed the searches, analyzed the data, drafted and revised the manuscript, and approved the final manuscript. RW designed the study, analyzed the data, drafted and revised the manuscript, and approved the final manuscript.