We developed a review protocol according to the methodological guidance for scoping reviews [27-29], and the protocol is available via the Open Science Framework [30]. This review is reported according to PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews; Multimedia Appendix 1) guidelines [31].

An experienced research librarian helped to develop the search strategy, which included a combination of subject headings and keyword searches for the key concepts of “Social Media,” “Mental Health,” “Health Knowledge,” “Attitudes,” “Stigma,” and “Help Seeking Behavior.” A total of 5 databases were searched, including MEDLINE, Embase, PsycINFO, Web of Science, and CINAHL. Multimedia Appendix 2 shows the full search strategy for MEDLINE, which was adapted for the other databases. The gray literature, such as public mental health campaign websites identified from the review and leading mental health charity websites in the United Kingdom and other high-income countries, were searched for relevant reports not published in peer-reviewed journals. Reference lists of eligible studies and review studies were also hand-searched. We included studies published from 2004—as this was the advent of widespread social media use—to May 2024 [32].

The inclusion and exclusion criteria listed in Table 1 were applied.

Rayyan, a web-based tool, was used to screen abstracts according to the inclusion and exclusion criteria outlined in Table 1. RP reviewed all abstracts, and APK and JMS reviewed a proportion (1026/3396, 30%) of the abstracts after duplicates were removed. RP reviewed all full-text studies, and APK and JMS reviewed a proportion of full-text studies (11/69, 16%). Disagreements were resolved in discussions between reviewers and the other authors.

We extracted data using Excel (Microsoft Corp) to collect key information on study characteristics, details of the campaigns, methodological approaches, outcome measures, and key findings. To measure the success of social media campaigns, we captured outcomes derived from a version of key performance indicators and metrics related to social media use in health promotion adapted from Neiger et al [26], which includes, exposure, reach, and low-, medium-, and high-engagement (Table 2). Outcome measures capturing changes in knowledge about mental health conditions, attitudes toward mental health, stigma, and intention to seek help for mental health problems were not accounted for using the original framework, but were categorized as high engagement with a campaign, as greater knowledge and intentions are precursors to behavior change and taking actions offline [25,26,33]. We assessed changes in outcomes both before and after the campaign, as well as between individuals who were aware of the campaign and those who were not. We only reported differences that were statistically tested and for which P values, CIs, or effect sizes were available.

We conducted a narrative synthesis of study findings summarizing the characteristics of social media campaigns (aims, target audience, target location, and campaign developer, eg, charity and government) [34]. The study characteristics, such as settings, design, participants, and main findings, were summarized. We organized findings by outcome by summarizing people’s exposure to the campaign (eg, campaign awareness and views and impressions) and low-medium level engagement with the campaign on social media (eg, likes and shares on social media). To summarize the campaign’s reach, especially among underserved communities, we gathered data on the demographics of the individuals reached by the campaign, where available. We summarized how effective social media campaigns were at improving high engagement, that is, people’s knowledge, attitudes, stigma (which was largely measured by the desire for social distance), and behavior change (which included intentions to seek help for mental health, service use, and activities to enhance positive mental health).

A total of 5827 publications were identified from the database search, and a further 28 publications were identified from gray literature. After screening, 26 sources were included for data extraction and analysis (see PRISMA-ScR flow diagram in Figure 1).

Table 3 describes the characteristics of the campaigns. The campaigns included in this review were mostly undertaken in the United States (10/26, 39%), followed by the United Kingdom (8/26, 31%) and Canada (4/26, 15%). The most common social media platforms used by campaigns were Facebook (13/26, 50%), followed by X (formerly Twitter; 7/26, 27%), Instagram (4/26, 15%), and YouTube (4/26, 15%), respectively. All the campaigns that reported the social media platforms they used reported delivering content on several platforms (15/26, 58%), with most campaigns using three platforms (10/26, 39%). Most of the campaigns were developed by government agencies like local or national health or public health departments (12/26, 46%) and by charities (12/26, 46%), followed by health services (10/26, 39%), universities (4/26, 15%), and sports teams (4/26, 15%). Just over half of the campaigns were developed by multiple agencies (14/26, 54%).

There were 5 campaigns that were evaluated by multiple studies: Time to Change, WhyWeRise, Act Belong Commit related campaigns, In One Voice and Action Minded, and WhatMakesUs related campaigns (see Table 4 for a summary of the social media activities of these campaigns).

Most studies evaluating social media-based mental health campaigns have used serial cross-sectional surveys to assess campaign effectiveness by measuring exposure, reach, and engagement. Campaign awareness varied across campaigns and by demographic factors, including ethnicity, gender, age, and socioeconomic group. Younger people and women were most consistently reached by public mental health campaigns. Most studies measured changes in knowledge, stigma, and attitudes, and over half of all studies reported on behavior change outcomes such as help-seeking. Stigma and attitudes toward mental health were found to improve most post campaign compared to precampaign, and behavior change outcomes showed the least improvement. However, post campaign, those who were aware of campaigns were much more likely than those who were unaware to change their behaviors, such as seeking help for mental health concerns, engaging in activities for positive mental health, and intending to seek help.

This review showed that mental health knowledge, stigma, and attitudes generally improved before and after campaigns. These findings could reflect general societal shifts in these outcomes over the same period rather than be directly attributable to the campaigns. For example, many high-income countries have seen a general trend of reduced mental health stigma over the past 30 years, especially concerning common mental health problems like depression [62,63]. However, some studies found that campaign awareness was associated with negative attitudes about mental health. These included believing that those who have had a mental illness will never contribute much to society, that a person with a mental illness is a danger to society, and a desire to delay seeking mental health treatment out of fear of others finding out [42,43,45]. These findings came from the US WhyWeRise campaign, and it was suggested that the reasons for these attitudes could be due to the campaign inadvertently focusing more attention on some of the challenges that can be associated with mental health problems rather than empowerment [45]. Previous research also indicates that improving knowledge and awareness might not have long-lasting changes in attitudes, stigma, and behavior change [48,64]. Furthermore, analysis of social and traditional media over the past 20 years has shown that the media generally has failed to move away from negative stereotypes, such as associating some mental health disorders with danger and fear [65,66]. Although these campaigns likely aimed to challenge these perceptions, they seem to have not always succeeded in combating the stereotypes perpetuated by other media sources. Research suggests that future public mental health campaigns should move away from promoting a biomedical explanation of mental health problems that can perpetuate the othering of individuals affected by mental distress. Instead, campaigns should consider the public understanding of mental health and frame mental health problems as a response to social, economic, political, and biological factors [64,67]. Research on identifying effective and ineffective content and messaging strategies is crucial for framing campaigns appropriately and avoiding the perpetuation of negative stereotypes [68].

All evaluations that measured behavior change showed that those who were campaign aware were more likely to change their health behaviors after the campaigns than those who were unaware of the campaigns. These behaviors included self-reported intentions to seek help and engage in activities to enhance mental health, but also objective measures, such as an increase in the use of mental health outpatient services, helplines, and websites. However, we cannot be certain the behavior changes were due to the campaigns, as they did not account for all confounding factors. For example, the campaign-aware group might have been more likely to seek help because they were experiencing more distress than those who were unaware of the campaigns. Additionally, when behavior change was measured before and after the campaigns, few evaluations found any change in behaviors. In some evaluations of UK campaigns, help-seeking intentions decreased, possibly due to increased difficulty accessing mental health care in many parts of the United Kingdom [37,61,69]. This suggests that being aware of campaigns may be key to achieving behavior change. However, these behavior changes tend not to be sustained and need to be supported by improved access to services. Individual behavior change alone is insufficient; changes in services are also necessary for a sustained impact on mental health. Fragmented systems and long waiting lists pose a barrier to behavior change, as they can inhibit opportunities, as suggested by the Capability, Opportunity, Motivation, and Behavior Change model [25]. Public health campaigns alone cannot address these issues.

We found that campaign awareness across these social media campaigns was more common for younger age groups and women. This is likely partly because most social media users are younger and, therefore, more exposed to the campaigns. However, evidence suggests that globally, men are more likely than women to use platforms like Facebook, which were used by many of the campaigns in this review [70]. It may be that more women are campaign aware because data show that this group has experienced the sharpest increases in mental health problems over the last 20 years in many high-income countries, so these campaigns are of more relevance to them, and they are more aware [71-73]. However, there is a real need to reach young men, as they have significantly higher suicide rates compared to women [74]. Some studies also showed success at targeting populations who are unequally affected by mental health or have poor access to services, such as ethnic minorities and those in lower socioeconomic groups. However, the reach of campaigns was not reported in most studies, so it is unclear if they are reaching those who might benefit most and the extent to which they can address inequalities in mental health access and support. Campaigns that seemed to successfully target these populations were those that used different languages, popular sports teams, and mental health activists to share and create information [44,45,50]. In total, 1 study found that there was less improvement in changes in attitudes and behaviors post campaign in African American respondents compared to White respondents in the United States, suggesting that some campaigns may have less positive impact on underserved groups. Future campaigns should leverage the benefits of social media to reach different underserved groups using content from influencers and activists popular with the target groups.

This was a scoping review, so we have not assessed included studies for quality. However, this design was necessary to scope out what is known and what has been done in a short time period across a wide range of literature, to understand the effectiveness of social media–based public mental health campaigns. As we have only included studies written in English and focused on high-income settings, the findings of this review may not be generalizable to non-English-speaking countries. We chose only to include studies from high-income countries to more easily compare campaigns across countries with similar public health resources, so these findings may not be relevant to low-middle-income countries. The field of social media research is expanding rapidly, and this scoping review only captures a snapshot of evidence related to social media campaigns across platforms that were most popular during the period of 2008-2022. More recent campaigns will likely adapt their methods and messaging to newer platforms such as TikTok, and this may prove to have a different impact on outcomes. All studies used cross-sectional surveys and had multiple campaign components, so it is not possible to know whether changes that occurred in the outcomes were caused by the social media campaigns or some other factor that changed over time. The studies that compared those who were campaign aware and not aware had issues with confounding variables, making it difficult to attribute changes solely to campaign awareness. Studies also assessed outcomes over varying lengths of time, and some used nonstandardized measures of knowledge, attitudes, and stigma, which makes it difficult to compare outcomes across campaigns. Many studies used self-report measures to evaluate behavior change, such as their intention to visit a mental health practitioner, so it may not reflect actual changes in behavior. The review was limited because it excluded qualitative studies that would have provided deeper insights into how the campaign messaging is understood, which aspects are most impactful, and why some outcomes may be negative.

This scoping review underscores the potential of social media campaigns in improving knowledge, attitudes toward mental health, and stigma, and promoting help-seeking behavior. It highlights the potential importance of campaign awareness in contributing to behavior change and the need for more targeted campaigns to reach underserved communities, who may benefit from better mental health information, access to resources, and a focus on empowerment. However, we still have questions regarding whether these social media campaigns achieve more than traditional media campaigns and if we are fully capitalizing on their potential for extended reach. Future research should more consistently measure reach and behavior change outcomes to understand which groups benefit from these campaigns and how these campaigns affect help-seeking behaviors. Future campaigns need to consider how the public understands mental health, how to present information to not perpetuate mental health stigma, and how campaigns could be designed to achieve sustained benefits. Furthermore, as the field of social media research continues to evolve rapidly, ongoing evaluation of new platforms and campaign strategies will be crucial to maximize the impact of these interventions on public mental health.