This study is the first published report describing how SNSs are being used for health promotion, in this case, sexual health promotion. Although there are many examples of SNSs being used for sexual health promotion, most activities are unreported in the scientific literature and the number and activity of end-users varies greatly. Knowing the scale and the scope of the current level of health promotion using SNSs is a key first step in designing more effective health promotion interventions in this new medium.
For the moment, it appears the use of SNSs for sexual health promotion is not widespread: most activities are from the United States, largely target young people and primarily focus on having an organisational or programme presence on SNSs. These outcomes are perhaps not surprising given the emergence of SNSs and the high internet penetration in the United States, the initial young user-base of SNSs  and the reality that many organisations may have viewed SNSs (at least initially) as simply an additional online location in which to have a presence, alongside their organisational website. However, as SNSs become more widely used, it is likely that they will also be increasingly used in more diverse ways for health promotion, including for the delivery of campaigns and interventions (now that there is an established user base) and for targeting sub-populations other than young people.
The dominance of three SNSs (Facebook, MySpace and Twitter) within the health promotion activities identified is partly a reflection on our search strategy (which specifically sought out activities on Facebook and MySpace) and also a reflection of the current market share of these SNSs. The advantage of using these established SNSs is that the target audience is already present and interacting with their social networks, unlike creating a custom SNS that must first attract end-users before it can reach individuals for health promotion. However utilising an established SNS can restrict how the health promotion activity is presented, the content that can be provided under each SNS's 'acceptable use' policy, and ownership of online content, which may affect the delivery and fidelity of health promotion activities.
Defining features of Web 2.0 include generation of content by end-users and online social engagement . There was great diversity in popularity and the extent of online interaction among the health promotion activities identified. The most popular health promotion activities had thousands of end-users, with regular posts by owners and end-users each week. Nonetheless, many health promotion activities were inactive, particularly those using MySpace. There seems little purpose in having a relatively 'static' presence on a SNS, with few posts and end-user interactions, in addition to an organisational or campaign 'Web 1.0' website.
From reviewing the health promotion activities identified, it appears that some organisations have simply broadened their online presence into SNSs with relatively minimal effort, using similar content to their existing websites and making little attempt to encourage social activity and engagement. However other organisations appear to have 'purpose built' their presence on SNSs, providing regular updates and delivering content specifically designed for each SNS used. Often, but not always, the most popular sites are also those with the most active online communities. Online social activity does not always happen naturally;  future investigations should focus on the most popular and active health promotion activities on SNSs in order to better understand the content, features and approaches that successfully encourage social engagement. These elements could then be used to develop more engaging interventions, which may be more effective as interaction is known to promote deeper learning and understanding .
SNSs are constantly evolving. This creates challenges for health promoters, for example when the functionality of SNSs change, or when end-users migrate from one SNS to another. In this review, the high proportion of dormant health promotion activities using MySpace may be a reflection of the more general migration of users from MySpace to Facebook [62–64]. Organisations need to be flexible in responding to this evolution in order to maximise the value of health promotion activities using SNSs. For example, from 2009 Facebook allowed external websites to use Facebook logins and access content from Facebook which has been very popular [65, 66]. Thus it is now possible to deliver health promotion activities using functions (and audience reach) of SNSs, without the site actually being hosted on an external commercial platform.
A comprehensive overview of existing sexual health promotion activities using SNSs required us to search electronic sources and SNSs themselves, as well as the published scientific literature. That so little was available in the published scientific literature was most likely a reflection of the rapid emergence and uptake of SNSs, coupled with the time involved in obtaining funding, implementing and evaluating activities using SNSs, and publishing the results. An additional impediment to the scientific publication of health promotion activities using SNSs may be the lack of consensus regarding appropriate evaluation frameworks for these activities . However searches of electronic sources and SNSs bring their own challenges, such as the restricted search capabilities, the inability to replicate searches (see limitations), the incompleteness of information within health promotion activities identified and the unmanageably large number of records retrieved. Given that the need to use electronic sources to produce comprehensive scientific reviews is unlikely to abate, it would be useful to establish 'best practice' guidelines to inform future searches of these contemporary information sources. Such guidelines could include processes for archiving search results for future reference (for example, printing results to PDF).
This review has several limitations. Primarily, the methods for searching electronic sources and SNSs are not well established, and it is likely that some sexual health promotion activities using SNSs were not identified due to the number of search terms and searches possible. As "sexual health" and "health promotion" involve a broad range of topics and activities, we were forced to make choices about which search terms could be used in each data source. However, we attempted to maximise coverage by searching within key SNSs as well as using multiple electronic data sources and multiple search terms. In addition, the searches conducted are not replicable because online content and search algorithms are constantly changing. The search strategy developed also limited the likelihood that campaigns using SNSs primarily for 'viral marketing' would be identified (although one such campaign was identified and included). Only English language sources were searched, which biased results towards health promotion activities from English speaking countries. Due to the large number of records retrieved and time limitations, each record was assessed by only one screener. However the two screeners regularly consulted each other when it was unclear whether the record met the inclusion criteria, and in case of disagreement consulted with a third individual (author AP). For practical reasons, measure of reach was limited to number of end-users of each health promotion activity, while the measure of online social engagement was limited to user posts within a short time period. Although these metrics have clear limitations, it has been argued that online usage statistics are important because they currently offer the one standardised and comparable metric for internet-based interventions and have been associated with positive outcomes across a range of health conditions .
This study focused on providing an overview of the current use of SNSs for sexual health promotion; it did not aim to assess the impact of individual health promotion activities. Process and impact evaluations of individual health promotion activities using SNSs should consider inclusion of measures such as:
Characteristics of end-users - demographics, health knowledge, attitudes and behaviours;
Quantity of interactions - number of interactions with end-users;
Quality of interactions - content analysis of interactions to assess relevance and utility;
Message spread - number of 'shares' and 'retweets' of site content (and characteristics of secondary recipients of site content, if possible);
Impact of activity on health knowledge, attitudes and behaviour; and
Cost-effectiveness of activities, particularly in comparison to the cost and effectiveness of delivering health promotion interventions via more traditional channels.