In May 2022, the United States (US) experienced a Clade II mpox (previously known as monkeypox) outbreak, and by August 2022, the US Department of Health and Human Services officially declared a public health emergency¹. The peak of mpox cases occurred on August 1, 2022, with 645 cases in the US reported to Centers for Disease Control and Prevention (CDC)². By November 2022, the seven-day average of mpox cases ranged from 16–39 cases but social media campaigns and vaccination efforts were ongoing to prevent another surge. There are two types of mpox virus: Clade I, which causes deaths in up to 10% of infected people, and Clade II, which leads to less severe mpox infections and is associated with the 2022 outbreak³. Gay, bisexual, and other men who have sex with men (GBMSM) and racial and ethnic minorities were disproportionately affected by the mpox outbreak in the US⁴. The ease with which Clade II mpox virus spreads through direct contact, such as during sex or other close skin-to-skin contact, poses a significant challenge for containment efforts, particularly in communities where risk factors for transmission are prevalent⁵. While the incidence of cases continues to be lower than the height of the 2022 epidemic, new Clade II mpox cases continue to be identified in the US^6–8. Moreover, a recent increase in mpox cases in the Democratic Republic of Congo (DRC) has included the first documented sexual transmission of Clade I mpox virus and highlights the risk for another global spread of mpox, including to the US⁹.

Public health efforts have focused on health education messaging and using vaccination as an important tool in preventing the spread of mpox. The CDC and the Advisory Committee on Immunization Practices recommend the JYNNEOS vaccine for people at increased risk for mpox infection^10,11. The JYNNEOS vaccine is a two-dose series, with each dose administered four weeks apart subcutaneously or intradermally; persons vaccinated with JYNNEOS have been found to be 9.6 times less likely to contract mpox than those who were unvaccinated¹². During the first few months of the 2022 US mpox outbreak, vaccine stock was limited, the demand was overwhelming, and Black and Hispanic persons were found to be less likely to be vaccinated despite being disproportionately affected^13,14. Additionally, an analysis published in June 2023 found that even after vaccine rollout initiatives were implemented to expand mpox vaccine access among minority populations, three-quarters of vaccine-eligible Black persons and American Indian or Alaska Native persons remained unvaccinated¹⁵.

Collecting qualitative information from individuals who may be more affected by the mpox epidemic was critical to understanding their perspectives, experiences, and potential reasons for not getting the mpox vaccine. During the 2022 outbreak, this information needed to be collected rapidly to inform public health response. To better understand mpox vaccine uptake and hesitancy among vaccine-eligible persons, the National Coalition of STD Directors (NCSD) and CDC used Remesh, a web-based artificial intelligence (AI) platform, to conduct a virtual focus group with unvaccinated GBMSM in November 2022¹⁶.

A 60-minute virtual focus group was held through Remesh on November 16, 2022. Purposive sampling was used to ensure racial and ethnic diversity among participants. Recruitment was conducted through market research panels that partner with Remesh. Participants were eligible if they identified as a cisgender or transgender man, were aged 18–55, reported having sex with another man in the past 6 months, and had not received any dose of the JYNNEOS vaccine. Participants who completed the full focus group were offered an incentive of $25 value in an award of their choosing (e.g., cash, gift card, etc.). This activity was determined to be non-research and exempt from undergoing a review from CDC’s Institutional Review Board and was conducted consistent with applicable federal law and center policy (e.g., 45 CFR part 46.102.l.2; 21 CFR part 56; 42 USC §241.d; 5 USC §552a; 44 USC §3501 et seq).

Participants provided written informed consent before taking part in the study. Participation was voluntary and respondents could skip questions or exit the focus group at any time. The focus group included 62 open and close-ended questions and presented images and graphics for participants to respond to through a chat-based conversation. The development of the questions was informed by the Health Belief Model (HBM), a theoretical framework that focuses on how individuals’ perceived beliefs influence their health-related behaviors¹⁷. Following HBM’s constructs, questions were included on participants’ perceived risk to mpox, willingness to get the mpox vaccine, and perceived barriers preventing them from getting the vaccine. The first 12 questions served as an initial screener and collected information on demographics, recent sexual history, and experience getting tested and/or vaccinated for mpox. Participants who reported having previously received any dose of the mpox vaccine were discontinued from completing the rest of the questions.

The AI system used machine learning and natural language processing in real-time to analyze participant responses and provide results immediately to the moderator. While participants did not engage directly with each other, they could see everyone’s responses and engage with the input from others based on questions asked from the moderator (i.e., based on responses, the moderator could then ask participants to indicate their level of agreement with others’ input). Quantitative data could be viewed in bar or pie chart format with frequencies provided and the platform allowed for cross-tabulation of respondent’s answers to different questions. Qualitative data were thematically coded and categorized in real time by the AI platform. To confirm the results from the Remesh system, the first author (C.G.) downloaded all data from the AI system and performed independent replication in Excel, including double coding the open-ended responses and validating results, categories, and emerging themes. The codes generated by Remesh were used as an initial deductive codebook and additional inductive codes were applied during analysis. Due to the inherent nature and limitations of the AI system’s ability to capture all emergent themes from the qualitative data, additional codes and themes were identified through human review. To ensure consistency of the coding process, two coauthors (G.C.H., F.L.) met with C.G. to review the codes and themes, resolve any discrepancies internally and with the AI system coding, and reach consensus.

The findings from this study demonstrate that many unvaccinated GBMSM who may be at increased risk for mpox either wanted the vaccine or, with appropriate health messaging, may be open to getting the vaccine. Participants were particularly interested in the statistic around the risk reduction the mpox vaccine offers compared to those who are unvaccinated. Notably, at the beginning of the focus group, 22 participants were undecided about getting the mpox vaccine, and after receiving health messaging that addressed their concerns, 12 of those participants said the messaging made them reconsider getting the vaccine. When asked to share what other information about mpox is needed, most responses related to wanting more research, data, and transparency about the vaccine.

Examining the mpox epidemic through a historical lens is critical to better understanding of and ability to mitigate the stigma that impacts GBMSM seeking and receiving healthcare²⁰. When HIV was identified in the early 1980s and predominantly impacted GBMSM, intense homophobia and stigma greatly hindered the public health response and affected access to care^20,21. To this day, stigma remains one of the biggest challenges to fighting HIV and remains a prevalent challenge for GBMSM seeking healthcare -- this may well have contributed to participants’ concerns over being judged by their healthcare provider if they were to ask for the mpox vaccine or visible scarring that may occur after receiving it^20,22. Creating a stigma-free space between healthcare providers and GBMSM is crucial to bridging gaps in essential care, especially during times of national epidemics. Additionally, participants who had attempted but were unable to get the mpox vaccine noted barriers such as low vaccine availability and scheduling issues, which are consistent with other studies^18,19. At the time of this study, vaccine stock was low, but as of April 2024, the JYNNEOS vaccine is commercially available in the U.S. and can be accessed at local pharmacies and clinics²³. Offering the mpox vaccine in alternative or non-clinical locations, such as community-based locations or Pride events, in addition to traditional sites, like pharmacies and doctor’s offices, may help increase overall access to the vaccine, improve cultural competency among providers, and improve vaccine uptake among GBMSM.

Results from this study were used to directly inform outreach and health communications messaging during the mpox response, including social media content and a toolkit released by NCSD^24,25. The concerns over scarring caused by the vaccine and positive responses regarding the options for the vaccine injection placement led to health messaging emphasizing that patients could ask providers for the mpox vaccine in their shoulder or upper back. This finding is consistent with other studies that found that stigma associated with the mpox vaccine, particularly the fear of being publicly outed as “gay” from forearm scarring following the injection, contributed to people feeling hesitant to get the vaccine¹⁹. Additionally, health messaging emphasizing that the mpox vaccine is a two-dose series and that both shots had the best efficacy for protection was incorporated into outreach materials^24,25. Lessons learned from this study may be of particular concern given the recent increase in Clade I mpox virus in the DRC, including its severity and sequelae, and its unprecedented spread through sexual transmission, including among GBMSM⁹.

Even with targeted recruitment, the sample size of this study is too small to be generalizable and the representation of minority participants was very low, which limited the external validity of these results. However, conducting this virtual focus group during the height of the 2022 mpox epidemic allowed us to collect real-time information from individuals who may have been most impacted and quickly inform response efforts in a way that would have been limited by traditional methods. Future studies should prioritize recruitment from Black, Asian, American Indian or Native Alaskan, and Hispanic individuals to ensure their perspectives are shared. Individuals within the GBMSM community have been historically reluctant to participate in qualitative studies due to a myriad of factors, including stigma, societal discrimination, and privacy concerns²⁶. The anonymity and convenience of an online platform mitigated many of these challenges and allowed a safe space for GBMSM to share their experiences and perspectives without fear of judgement. Because participation was based on an individual’s ability to access and navigate and engage with an AI system, selection bias may have occurred. However, logistical challenges with in-person recruitment and data collection were mitigated by allowing GBMSM to contribute their experiences in locations convenient to them at a time where their perspectives and experienced needed to be heard. Finally, we were unable to collect information about whether participants later received the mpox vaccine and were thus unable to determine whether the health messaging was effective in increasing mpox vaccine uptake among individuals who were previously undecided or against getting it. However, as noted previously, results were used to inform a communications toolkit which has been widely disseminated and used as a model for mpox outreach communications in the response.

During an outbreak, many unvaccinated GBMSM who may be at increased risk for mpox either want the vaccine or, with appropriate health messaging, may be open to getting the vaccine. The use of Remesh allowed us to conduct a virtual focus group with GBMSM during the height of the 2022 mpox epidemic and produced quick results that were used to inform and change ongoing response efforts and health communications messaging. Messaging about vaccine efficacy, potential side effects, and how to access the vaccine may improve vaccine uptake especially as cases continue to occur. Improving healthcare provider cultural competency and expanding vaccine availability at both traditional (e.g., pharmacies, doctor’s offices) and non-traditional (e.g., community-based organizations, Pride events) spaces may also improve uptake.