We conducted formative qualitative work with a diverse group of adolescents and young adults, HIV advocates, and clinic providers to inform campaign content. Focus group and interview guides were informed by the Health Stigma and Discrimination Framework to explore key drivers and manifestations of HIV-related stigma and desired change goals (ie, what they hope would change as a result of the campaign), key messages, and content delivery strategies [6,33,34]. Participants in these focus group and interview sessions identified 2 overarching change goals: (1) reframing perceptions of HIV and living with HIV and (2) changing attitudes and actions toward people living with HIV. Key messages to achieve these goals included conveying that U=U, HIV can affect anyone, HIV is a chronic and treatable condition, and people living with HIV desire empathy, not pity.

The DiME study team consisted of individuals from Peru and the United States: an infectious disease epidemiologist (MFF), a medical doctor and HIV program manager (RAE), a nurse (MW), a behavioral scientist and social worker (JTG), a stigma expert (KAK), a paid community manager and activist living with HIV (JV and DS), a social media content creator (JA), and research staff (AN, MC, YS, and EC). Most team members were bilingual in Spanish and English. We contracted a Peruvian content creation agency to create social media content that addressed HIV-related stigma, specifically for the Peruvian audience. While the agency had experience developing content related to social issues, including stigmatized infectious diseases (eg, tuberculosis), it did not have experience creating content related to HIV or stigma. To sensitize agency team members to these topics, we provided a virtual education session on clinical information and created an explanatory report including an overview of stigma, how HIV-related stigma manifests in Lima, and a summary of formative research findings [33,35]. Guided by this information, the agency created a campaign proposal using the slogan “Seamos una voz” (Be the voice), which focused on dispelling stereotypes and normalizing conversations around HIV while promoting empathy, not pity. Videos debunked myths (eg, HIV and AIDS are synonymous, HIV only affects the lesbian, gay, bisexual, transgender, queer or questioning, intersex, asexual, and others (LGBTQIA+) community, HIV is transmitted through touch, and people living with HIV cannot lead “normal” lives) and educated about important concepts (eg, U=U), with each video ending with the tagline “Be the voice against prejudice” or “Be the voice against discrimination.”

The content creation agency produced a draft script of each post for review by the study team and a community advisory board made up of 15 people living with HIV, 4 of whom were HIV activists. Community advisory board members were diverse in terms of age (range: approximately 19-37 years) and sexual orientation. For each piece of content, each member of the study team and advisory board completed a brief survey to assess the appropriateness of the material and suggest improvements. The study coordinator compiled this feedback and shared it with the agency. Once produced, each piece of content underwent a second review by the study team and advisory board for final approval, or, occasionally, a decision against producing a particular piece of content. The agency filmed content around Lima featuring diverse Peruvian actors and incorporated trending music, memes, sound clips, and stickers (customizable graphic elements added to videos). Of the 35 total pieces of video content produced, 7 did not pass the approval process, and they were archived: 4 because of a misalignment between content message and campaign goals, 2 because the recorded scenes were deemed unrealistic, and 1 because actors inadvertently used stigmatizing language that could not be edited. A randomized social marketing evaluation of a subset of DiME campaign content found that brief exposure to the content, as would be experienced in a social media feed, led to detectable short-term changes in HIV-related stigma among young people recruited from the general population in Lima [36].

To achieve our accounts’ target posting frequency of twice weekly, the study team produced 21 additional videos to supplement material created by the agency. These videos provided information about sexually transmitted infection testing and treatment in Lima, addressed stigmatizing comments left on the DiME account, and shared testimonials from young adults living with HIV. The study team community manager and content creator appeared in these videos.

We contracted a Peruvian influencer management agency to identify and recruit influencers to participate in the campaign. We presented the agency with the stigma report, the campaign’s objectives, and the scope of work for influencer participation. We sought a diverse group of influencers in terms of gender, sexual orientation, and the types of content they produce. The agency proposed a diverse list of widely recognized top Peruvian influencers. The final list of paid participating influencers included 5 from this list and 2 suggested by the study team: 4 LGBTQIA+ lifestyle influencers, 1 physician, 1 reality television star, and 1 newscaster, with follower counts ranging from 117,000 to 23.8 million. Paid influencers were provided with a digital pamphlet containing information on HIV-related stigma and guidance on appropriate language to use when discussing HIV in their videos. Influencers created video scripts, which were subsequently reviewed and edited by the study team and community advisory board before filming and posting content to their platforms. Payments ranged from US $1000 to US $3200 per post. Seventy-eight microinfluencers, defined as influencers with more modest and niche followings, were identified by the management company based on perceived willingness or past campaign participation. Invited microinfluencers had 1000 to 689,000 followers and were invited to join the campaign voluntarily by reposting campaign content or by creating content based on the pamphlet shared by the study team.

We adhere to Guidelines and Checklist for the Reporting on Digital Health Implementations (iCHECK-DH; Multimedia Appendix 1) [37]. Social media indicators (ie, followers, views, likes, comments, shares, and saves) were monitored and recorded weekly by the study team during the campaign. The Instagram professional dashboard and TikTok business suite were used to extract demographic information and performance insights (ie, views and video watch time). Key performance indicators included (1) reach (total number of views), (2) interaction (total number of likes, comments, shares, and saves), and (3) engagement rate (total number of likes, comments, shares, and saves, divided by the total number of views).

Top-performing videos were considered to be those with the greatest interaction at campaign end. All direct message exchanges to the DiME account on both platforms were categorically analyzed. Following analysis, the study team purposively selected representative examples from each category for presentation in the manuscript and translated messages from Spanish to English. A sentiment analysis was performed on all user comments left on campaign content. Three study team members (AN, MC, and MFF) manually categorized comments as positive, negative, or neutral, generally and with respect to the content, using coding definitions developed by the study team based on an initial review of all comments posted (Multimedia Appendix 2). Positive comments praised the DiME initiative and expressed resonating with content, while negative comments expressed stigmatizing opinions or questioned the normalization of HIV as a chronic and treatable condition. Neutral comments included questions and answers regarding HIV, superfluous information, greetings, and ambiguous emojis. Comments were divided among the 3 reviewers (AN, MC, and MFF) for coding, with uncertainties discussed as a group to reach consensus. One study team member (AN) reviewed all final codes for consistency. To summarize sentiment polarity (ie, the balance between positive and negative sentiment), we calculated the ratio of positive to negative comments.

After completion of the campaign, all participating paid and microinfluencers were invited to complete a questionnaire asking them about their experience participating in the campaign (Multimedia Appendix 3). Responses were summarized.

This work was approved by the Vía Libre Bioethics Committee in Peru (CIB-8326) and by the Longwood Campus Institutional Review Board (IRB) at Harvard Medical School (IRB22-0741). Both ethical reviews covered all components of the analysis. As the IRB determined this component of the research to be exempt, influencers who completed the questionnaire were not asked to provide written consent but were informed that participation was voluntary. No compensation was provided for study participation. Prior to analysis, all data were deidentified.

The DiME campaign ran from July 8, 2024, to December 12, 2024, on the Instagram and TikTok social media platforms under the account name @dime.peru. These platforms were chosen for campaign implementation as they are widely used by adolescents and young adults in Peru and were considered appropriate channels for reaching the campaign’s intended audience [15,16,33]. Forty-eight videos were posted on Instagram and 49 on TikTok, with an approximate posting frequency of twice weekly. For 35% (17/48) of Instagram posts and 31% (15/49) of TikTok posts, paid boosts were used to increase reach, with a total of approximately US $564 spent across both platforms. Separately, the 7 collaborating paid influencers contributed a total of 13 videos to the campaign. Influencers posted content to their own social media accounts and directed their followers to the DiME account. Six of these videos were posted in collaboration with the DiME account, appearing simultaneously on the influencer’s page and the DiME Instagram account (collaborations were not possible on TikTok during the campaign). The DiME community manager monitored comments and direct messages, and provided timely responses to questions, misinformation, and requests for support.

During the 5-month campaign, the DiME account gained 2206 followers on Instagram and 11,800 followers on TikTok. According to platform analytics obtained in December 2024 for the last month of the campaign, approximately 46% of followers had Peru-based accounts (with the next largest group of followers coming from Mexico, followed by those from neighboring countries), 64% were male, and 56% were aged 18‐34 years. Median reach per post after the first 4 weeks of the campaign was 470 on Instagram and 994 on TikTok, and increased to 4011 (IQR 1739‐13,286) and 3203 (IQR 1594‐15,100), respectively, at campaign end. Median interactions per post increased from 30 on Instagram and 54 on TikTok within the first 4 weeks, to 86 (IQR 57-206) and 110 (IQR 76-363), respectively, at campaign end. Our overall engagement rate at campaign end was 1.65% on Instagram and 6.89% on TikTok. Influencer content generated over 2.8 million views and 160,000 interactions. Metrics reported in Table 1 reflect numbers through the campaign end in December 2024.

Of the 1462 total comments received on DiME posts produced by the creation agency and study team, 684 (47%) were positive, 546 (37%) were neutral, and 232 (16%) were negative. A small percentage of comments (81/1462, 6%) conveyed antistigma messaging using hostility or negativity toward stigmatizing opinions. We classified these as positive if they endorsed the content messaging. The resulting sentiment polarity ratio (positive-to-negative) was 2.95:1, indicating nearly 3 times as many positive as negative comments. DiME account followers frequently responded to negative comments with their own messages of education:

Influencers received a total of 1462 comments on their content, 64.8% (947/1462) of which were positive. Many of their followers congratulated them for informing their audiences about HIV: “[There should be] more influencers like you who don’t just entertain but also inform.” or “I learned more in two minutes with [influencer name] than in five years of school!”

We received 512 direct messages to our accounts corresponding to the following categories: requests for HIV and sexually transmitted infection (STI) testing locations (306/512, 59.8%); questions regarding HIV testing protocols, transmission, and treatment (65/512, 12.6%); sharing of personal experiences (8/512, 1.6%); and other (133/512, 26.0%). The other classification was used for messages praising the DiME campaign, messages asking where DiME was physically located, and greetings without follow-up, among others. Messages originated from 37 localities in Lima, 16 Peruvian provinces, and 7 other Latin American countries. Using the messaging platform, the DiME community manager directed users to the public health testing facility in their locality, answered user questions, and/or provided support and encouragement. Illustrative messages are included in Table 2.

Metrics for top-performing content are shown in Tables 3 and 4. Of the content created by the agency and study team, the top post depicted actors representing a young male same-sex couple in a sero-discordant relationship where one partner is living with HIV while the other is not. They were filmed in Lima, dispelling myths about sexual relationships while living with HIV and encouraging their peers to examine preconceived notions. This video was posted on November 8 and “went viral,” rapidly gaining substantially higher reach and engagement than typical campaign content. It received 388,276 views, 17 times the average number of views for DiME content, and 48,646 interactions, 80 times the average interactions. Other top-performing posts, as defined by the number of total interactions, dealt with rights in the workplace for young adults living with HIV, the difference between HIV and AIDS, and the stereotype that adolescents and young adults living with HIV cannot live like their peers. These videos received more than 65,000 views and 2000 interactions in total.

The 5 top-performing influencer pieces were all posted on TikTok and included videos created by lifestyle influencers in the LGBTQIA+ community, a male social media newscaster, and a female reality television celebrity. These videos all received at least 180,000 views and 8000 interactions.

Figure 1 provides a comparison between reach and interaction metrics for the top-performing videos created by the agency and study team and by influencers. Influencer numbers were greater for all categories with the exception of comments received.

Thirty of 78 unpaid microinfluencers (38% of total invited) shared a total of 27 and 9 posts on Instagram, created by the agency and influencers, respectively. Three influencers independently created and posted their own content about HIV-related stigma and directed their followers to the DiME account. Seven microinfluencers that were not part of the project also reposted campaign content.

Three paid influencers and 2 microinfluencers responded to our postcampaign survey (5/78, 6% of those invited). Three reported prior experience participating in campaigns to raise awareness about social issues, but none had experience with HIV-related campaigns. All reported accepting the invitation to participate in the campaign because it aligned with their personal values. These survey respondents reported mostly positive audience reactions to campaign content. Two indicated that participating in the campaign positively affected their image on social media, while 3 mentioned the campaign had neither a positive nor negative effect on their image. Three provided recommendations for improving communication with contracted influencers and social media followers, including considerations related to HIV-related terminology. All reported willingness to participate in future HIV-related social media campaigns.

Social media has become an effective and scalable tool for distributing public health information [38-40]; however, campaigns explicitly seeking to address HIV-related stigma are lacking. The DiME social media campaign sought to reduce HIV-related stigmatizing opinions and behaviors among adolescents and young adults generally, and self-stigma in adolescents and young adults living with HIV in Peru by addressing stereotypes and promoting empathy through educational messages, interviews, and testimonials. A diverse group of actors and spokespeople from the local community established the trust necessary to build a committed following, and content filmed in Lima enabled the audience to connect with the campaign messaging, identifying with the local context. Reach was substantially higher for videos created and shared by influencers compared to agency and study team videos, suggesting that influencers can feasibly support dissemination of messages intended to normalize discussions of HIV. Consistent with the Health Stigma and Discrimination Framework, the use of social media platforms and influencers provided a mechanism to address the drivers of stigma by promoting accurate information and encouraging empathy toward people living with HIV. Though this report focused on campaign implementation, subsequent analyses will examine whether exposure to the campaign reduced HIV-related stigma among young adults living with HIV.

Our experience suggests that collaborating with widely recognized paid influencers and unpaid microinfluencers on HIV-related social media campaigns is both feasible and advantageous. Working with a local influencer management agency allowed us to efficiently identify and recruit strategic personalities willing to participate. The iterative production process allowed influencers to incorporate their established communication styles and relationships with followers while ensuring that content appropriately addressed stigmatizing opinions. Microinfluencers’ sharing of DiME content further increased campaign visibility. Additionally, the fact that several unaffiliated microinfluencers also reposted content suggests that collaborating with influencers may organically extend message dissemination beyond formal campaign partnerships. These observations align with findings from previous public health campaigns that suggest influencers can positively communicate health information and increase engagement among recipient audiences [19,29,41,42]. In many parts of Latin America, conversation around HIV continues to be taboo; reaching large audiences who are not routinely exposed to HIV-related information is of utmost importance for stigma reduction [43]. We found that influencers were a highly effective means of reaching a wider and more diverse audience than could otherwise be achieved. Influencers were, in general, receptive to instruction for respectful presentation of HIV-related content, and their agreement to address these topics, despite potential negative comments, is a step toward broader public interest in HIV education. However, it is important that content creators receive training on HIV and HIV-related stigma prior to production in order to ensure that the information they share is communicated in a way that reduces stigma rather than reinforces it, while maintaining the creator’s personal style.

Establishing trust between a campaign and its audience should be a central objective of social media initiatives [44]. Previous research has shown that digital platforms can serve as trusted spaces where adolescents and young adults living with HIV have the confidence to discuss stigma and seek support [45-47]. At the end of the 5-month campaign, we had built a community on both platforms where followers shared personal experiences and asked for advice through direct messages or public comments. We believe the role of the community manager was important for obtaining this trust. This position was filled by a person living with HIV with experience in peer counseling and HIV communication, bringing lived experience and promoting authentic connection with the audience. This seemed to strengthen the campaign’s credibility and enhance the relatability of interactions with users. Preserving a balance between the desire to quickly gain followers and extend reach and cultivating human connection through slower community building is crucial to maintaining the relational dimension that enables stigma reduction. The process of trust-building took time. For the first half of our campaign period, reach and engagement were much lower than at campaign end. As such, determining the appropriate duration for public health social media campaigns should account for the time needed to cultivate an active and engaged following. Organizations should also have a transition plan or taper phase for campaign end, as once the account becomes a trusted source of information, followers expect constant engagement. We allocated resources to continue maintaining the DiME account for one year after the official campaign end, although at a reduced frequency compared with the active campaign period.

Among the lessons learned was the importance of developing content weeks in advance of the target posting date, as the nature of HIV and stigma information requires adequate time for thoughtful input regarding how content is perceived. Video first drafts were more effective than written scripts at conveying appropriate tone and visual appeal. Furthermore, language used in campaign content is critical [48]. Following the recommendation of a participating influencer, we sought to strike a balance between destigmatizing language and language understood by a lay population. For example, in Spanish, “to infect” (contagiar) is more common colloquially than the preferred term “to transmit” (transmitir). Though adolescents and young adults living with HIV may perceive the term “infect” to be derogatory, “transmit” is less commonly used and may hinder comprehension or relatability. These nuances underscore the importance of convening advisory committees of young people living with HIV and young people from the intended recipient population (ie, those not known to be living with HIV) to help strike the balance of comprehension, relatability, and respect. If possible, young adults living with HIV should accompany video recording sessions to ensure that the inadvertent use of stigmatizing language can be corrected in real time and that recorded situations mirror realistic interactions to the greatest extent possible. We were unable to incorporate this practice during the entire campaign period.

Several limitations should be considered when interpreting our implementation, reach, and engagement metrics. Though a strength of this work was our constant communication and collaboration with an advisory committee of adolescents and young adults living with HIV, this lengthened the content creation process, which led to a relatively low posting frequency of 2 videos per week. Social media algorithms reward high posting frequencies [49,50]. We chose not to create all content in advance of the campaign in order to leverage social media trends. However, the length of the content validation process meant that trends sometimes became obsolete by the time content was approved. Additionally, exact reach and engagement are difficult to assess. Adolescents and young adults may have multiple social media accounts on the same platform, inflating unique reach numbers. Stigma may hinder young people from following an HIV-related account or interacting with its content, although they have viewed it. This raises the possibility that meaningful engagement is greater than suggested by the metrics provided by campaign platforms. The occasional use of paid boosts also makes it difficult to distinguish between organic versus paid reach and engagement. Moreover, baseline differences in social media use and engagement across demographic groups may influence exposure to and interaction with campaign content; however, these factors were not considered in this study. Regarding influencer collaboration, we did not collect data on reasons for nonparticipation among microinfluencers and only 5 influencers responded to our postcampaign survey. Therefore, our understanding of influencer experiences and perspectives is limited to a small subset of participants. Future studies involving influencers could consider including an individual interview in the influencer contract. We also recognize a lack of metrics to assess the qualitative impact of users’ interactions with the DiME community manager (eg, changes in perceptions or opinions). This paves the way for future studies that measure the narrative transformation of social media users. Finally, we acknowledge that the DiME campaign was designed specifically to address stigma among social media users in Peru; however, we believe that the underlying processes used to design and implement the campaign and engage influencers are broadly generalizable.

Social media has the power to educate and cultivate empathy [51], and the potential to positively influence behavior among adolescents and young adults who are just beginning to navigate the health care system independently [52,53]. The DiME campaign used educational videos, testimonials from young people living with HIV, and influencer content to bring HIV education to this population in an accessible and engaging way. We found TikTok to be an especially promising platform for disseminating antistigma information to adolescents and young adults; the campaign gained nearly 12,000 TikTok followers in 5 months and its engagement rate of 6.89% exceeded the average 2.3% reported for the education sector on this platform [54]. However, both DiME accounts have continued to grow since the campaign ended and have garnered a place in the HIV space in Peru, with tags from other organizations’ HIV-related content. Direct messages indicate there is a critical need for messaging to reduce HIV-related discrimination. Future campaigns should consider creating content for older populations or specific groups, such as health care providers, from whom stigma and discrimination may be especially harmful [55,56].

The DiME social media campaign was designed to reduce HIV-related stigma by challenging stereotypes about HIV and promoting empathy toward people living with HIV through short, engaging videos tailored to adolescents and young adults in Lima, Peru. Collaboration with popular local influencers to create and disseminate content was a novel approach in this context and served to increase campaign reach and engagement. Influencers received mostly positive feedback from their followers. To our knowledge, the DiME campaign is among the first to explicitly address HIV-related stigma via a community-based social media campaign in Latin America, adding to the growing evidence that digital platforms are promising tools for stigma reduction across languages, cultures, and health conditions.