Prevention of sexual transmission of mpox: a systematic review and qualitative evidence synthesis of approaches

Abstract Background The ongoing multi-country mpox outbreak in previously unaffected countries is primarily affecting sexual networks of men who have sex with men. Evidence is needed on the effectiveness of recommended preventive interventions. To inform WHO guidelines, a systematic review and qualitative evidence synthesis were conducted on mpox preventive behavioural interventions to reduce: (i) sexual acquisition; (ii) onward sexual transmission from confirmed/probable cases; and (iii) utility of asymptomatic testing. Methods Medline, EMBASE, PubMed, Cochrane and WHO trial databases, grey literature and conferences were searched for English-language primary research published since 1 January 2022. A reviewer team performed screening, data extraction and bias assessment. A qualitative thematic synthesis explored views and experiences of engagement in prevention in individuals at increased risk. Results There were 16 studies: 1 on contact-tracing, 2 on sexual behaviour, and 13 on asymptomatic testing. Although MPXV was detected in varying proportions of samples (0.17%–6.5%), the testing studies provide insufficient evidence to fully evaluate this strategy. For the qualitative evidence synthesis, four studies evaluated the experiences of most affected communities. Preferences about preventive interventions were shaped by: mpox information; the diversity of sexual practices; accessibility and quality of mpox testing and care; and perceived cost to wellbeing. Conclusions Evidence on the effectiveness of interventions to prevent the sexual transmission of mpox remains scarce. Limited qualitative evidence on values and preferences provides insight into factors influencing intervention acceptability. Given global and local inequities in access to vaccines and treatment, further research is needed to establish the effectiveness of additional interventions.


Introduction
In May 2022, near-simultaneous rapidly evolving outbreaks of mpox (formerly monkeypox) due to clade IIb monkeypox virus (MPXV) were reported from several countries not historically affected by sustained mpox transmissions.The World Health Organization (WHO) declared a public health emergency of international concern in July 2022 [1].As of 31 January 2024, a total of 93,921 laboratory confirmed cases and 662 probable cases, including 179 deaths, have been reported to WHO from 117 countries in all six WHO regions [2].In contrast to previous outbreaks, this multi-country mpox outbreak primarily emerged in sexual networks of sexually active men who have sex with men (MSM), and approximately 50% of persons affected are also living with HIV [3,4].Large case series and studies (predominantly in MSM) described differences in clinical features and high propensity for transmission through mucosal (especially rectal) contact during this recent outbreak [5,6].In addition, sexual transmission of the more virulent clade I MPXV in central Africa has also been documented for the first time [7].
In accordance with the International Health Regulations (IHR) [8], WHO and other public health agencies issued a series of temporary recommendations [9].People with or at risk of mpox were advised to: isolate; avoid sexual and skin-to-skin contact while symptomatic; reduce number of sex partners, particularly in certain contexts (e.g.sex-on-premises venues); and use personal protective measures [10].
As well as the above recommendations, preventive interventions could include changes in types of sexual practices; evaluation with partners (e.g.asking about symptoms or lesions); seeking testing and care as needed, and, for confirmed cases, covering any lesions, disclosure to partners, and contact tracing.Asymptomatic screening in people with other sexually transmitted infections is another potential measure to assess the status of the outbreak and prevent onward transmission.
To address evidence gaps about preventive interventions, the WHO Health Emergencies Programme commissioned a systematic review addressing three populationintervention-comparison-outcomes (PICO) questions about sexual transmission of mpox: 1. Do preventive interventions reduce the risk of mpox infection?2. Do prevention interventions reduce the risk of onward transmission from a person with mpox?
3. Does testing asymptomatic individuals at increased risk of mpox (without known exposure)reduce onward transmission?
The review concentrates specifically on prevention strategies related to sexual transmission but does not include vaccines and vaccination campaigns.This is due to the limited evidence available on the effectiveness of non-vaccine preventive (risk-limiting) interventions in sexual transmission or the natural history of mpox.
Additionally, we conducted a separate qualitative evidence synthesis to contextualise the findings from the systematic review and understand the views on and experiences of preventive interventions among members of most affected communities.Methods and findings from the two reviews are presented individually, while complementing each other in the Discussion.

Systematic review
This systematic review is based on the protocol registered PROSPERO (ref: CRD420234259060) and reported in accordance with the PRISMA 2020 checklist.

Eligibility criteria
In consultation with information specialists, we conducted a single search covering all questions (Appendix A).Primary research studies, case series and letters containing reports on primary research published since January 2022 were included.Individual case reports, modelling and animal studies, and pre-prints were excluded.Papers related to vaccine interventions were excluded because a separate review is in process to support guidelines on vaccination.The search was limited to English language due to time constraints.Inclusion and exclusion criteria are described in Appendix A.

Information sources
The following databases were searched for papers from 1 January 2022 to 4 May 2023: Medline, EMBASE and PubMed (last searched 3 May 2023); Cochrane Library (for systematic reviews) and WHO trials (last searched 4 May 2023); PROSPERO reviews database.We reviewed available conference abstracts from 2022 and 2023for relevant meetings: International AIDS Conference; European AIDS Conference; Conference on Retroviruses and Opportunistic Infections; ID Week. 1 Backward and forward citation tracking was conducted for all included papers, last conducted on 6 July 2023 using Web of Science.The search was updated on 11 January 2024.

Search strategy
For the full search strategies for all databases, see Appendix B.

Selection process
Using the reference management system Rayyan [11], an initial reviewer (CM) excluded duplicate records and those that did not address the research questions.Two independent reviewers (CM and IW) reviewed titles and abstracts of all remaining results.Full articles of selected abstracts were reviewed by both reviewers against the inclusion criteria.Disagreements were discussed and reconciled by consensus and with a third reviewer (SP).The PRISMA flowchart of the study selection process is shown in Figure 1.

Data collection process
Two reviewers (CM and IW) independently extracted study data, including study design, population (including demographic data; HIV and MPVX vaccination statuses where available), intervention details, outcome measures, key outcomes and limitations.

Risk of bias assessment
We assessed the risk of bias using tools developed by the CLARITY Group at McMaster University for case-control and cohort studies [12,13].Two reviewers, CM and IW, conducted independent assessments and collaboratively resolved any conflicts.

Synthesis
A narrative synthesis of the available evidence is presented.Due to heterogeneity across studies, we could not perform a meta-analysis.

Qualitative evidence synthesis
PubMed and APAPsycInfo were searched for qualitative papers reporting on primary research from 1 January 2022 to 16 January 2024.We reviewed conference abstracts from 2022 and 2023, where available online, for the following relevant meetings: International AIDS Conference; European AIDS Conference; Conference on Retroviruses and Opportunistic Infections; British HIV Association (BHIVA); British Association for Sexual Health and HIV (BASHH).

Search strategy
The search strategy was kept intentionally broad to allow for inclusion of all potentially relevant qualitative studies.For the full search strategies for all databases, see Appendix C.

Selection process
An initial reviewer (RH) excluded duplicate records and those that did not address the research question (What are the views and experiences of interventions to prevent sexual transmission of mpox among communities at risk of infection?).RH and SP independently reviewed titles and abstracts of all remaining results against the exclusion criteria (Appendix C), then reviewed full text of selected papers and reconciled differences through discussion.

Quality assessment
Quality was assessed by both reviewers, using 12 criteria adapted from the methodology used for the evidence review [14].Due to the small numbers, all identified studies were included in the review.

Data collection and synthesis process
The degree of authors' interpretation in the qualitative studies varied, with some adopting a descriptive narrative with few second-order constructs [15].For our synthesis, themes identified by the study authors were extracted from the studies, and then further detail from the original texts was iteratively added to identify the key concepts within each study.Extracted data were synthesised thematically: study themes were grouped into 'descriptive' themes and then further synthesised to develop 'analytic' themes.

Systematic review
The initial search found 1,379 results from databases and 25 conference abstracts.Three papers were identified by forward and backward citation tracking.A further 922 papers were screened and 8 were added when the search was repeated on 11 January 2024 including reports from China.Overall, a total of 17records were included in this review, including 1 conference abstract.
Manuscripts almost exclusively described higher income countries, and were heterogenous in design, with varied populations and outcomes of interest.No studies evaluated interventions to reduce onward transmission for people with confirmed mpox (question 2).Characteristics of included studies are detailed below for questions 1 and 3.
We found no randomised controlled trials, and six prospective cohort studies.The remaining papers were retrospective observational studies.For quality assessment, see Appendix D.

Results for question 1: do preventive interventions decrease the risk of MPVX infection?
We identified three studies meeting the inclusion criteria for question 1 (see Table 1).One retrospective observational study conducted in the US [16] examined contact tracing, while the other two cross-sectional studies [17,18] investigated specific sexual behaviours.All three studies included exclusively male participants, predominantly MSM, who self-reported their sexual behaviour.Sample sizes ranged from 472 to 7538 participants.
The study by Cope et al. [16] investigated the effectiveness of contact tracing in curbing mpox transmission among MSM, particularly after the expansion of access to mpox vaccines.They found a significant reduction in proportion of MSM with mpox who provided location details for their sexual contacts after the vaccination campaign began.This resulted in lower contact identification rates later on in the outbreak and once the vaccination campaign had taken place.The study authors suggested that increased workload on healthcare providers combined with an initial surge in reported contacts (possibly linked to desire to access vaccines) created an unsustainable baseline for later tracing efforts.This suggests that solely relying on contact tracing may be insufficient to curb mpox transmission.
The study by Ghosn et al. [17] was part of an ongoing randomised open-label trial for a sexually transmitted infection treatment in MSM, coinciding with the onset of the mpox outbreak in France when the first case was identified on 19 May 2022.Subsequently, on 11 July 2022, the mpox vaccine was recommended for all MSM and at-risk populations.The study aimed to assess mpox incidence among at-risk MSM in the trial before and after vaccination, identify mpox-associated characteristics, evaluate vaccine effectiveness, and analyse selfreported sexual behaviour modification on mpox incidence including number of sexual partners in the last 3 months.Additionally, the study compared these characteristics between mpox cases and mpox-free controls in the study.Among the 472 participants included in the study, 16.3% (77) experienced mpox, with only one case occurring after the vaccination campaign began, and the vaccination status for this case was unknown.Following the introduction of the vaccine, vaccination coverage increased from 0% to 87%, and a significant decrease in the number of sexual partners in the mpoxfree control group was observed both before and after the vaccination campaign.However, adjusted analysis revealed that only the introduction of the vaccine significantly reduced mpox infections.The study authors concluded that sexual behaviour change did not appear to reduce mpox incidence in that population.
Du et al. [18] investigated the association between sexual behaviour and mpox prevalence amongst men.
They developed an index of safe sex behaviour based on three behaviours: condomless anal sex, commercial sex, and group sex.Participants (n ¼ 7538) were categorised into three groups: high safe sexual behaviour (no engagement in any of the behaviours), moderate safe behaviour (engagement in one or two behaviours), and low safe behaviour (engagement in all three behaviours).The overall prevalence of mpox was found to be 0.73%.Upon further analysis, the crude mpox prevalence was In departure from the previous study, the authors concluded that reducing engagement in low safe behaviour is important to controlling transmission.
Five prospective observational studies assessed presence of MPXV in individuals at higher risk of acquiring mpox [21,23,26,27,31].One study both examined stored samples and tested individuals attending a sexual health clinic [22].
The outcomes reported either the number of asymptomatic individuals with MPXV infection or the presence of MPXV in the swab samples.Only three studies [23,28,29] evaluated both male and female participants for presence of MPXV in the swab samples.Included participants ranged from 7 days to 81 years of age (one study included a neonate [29]).
Four studies [21,24,26,30] did not identify any MPXVpositive cases, and seven studies [19,20,22,23,25,27,31] reported prevalence in their respective study populations (range 0.17%-6.5%).Several studies identified presymptomatic and asymptomatic infection: Mizushima et al. [31] report that of five positive results from 1346 participants, three remained asymptomatic, one developed symptoms three days after testing, and one later reported that he had had mild symptoms prior to sample collection; Thomassen et al. [27] report one positive result of 224 participants, who developed symptoms three days later; Agusti et al. [23] report 7 positive results among 113 subjects, of whom 2 remained asymptomatic throughout follow up.
Two studies [28,29] reported MPXV prevalence in both individuals and swab samples, with prevalence of 0.7% and 8% respectively in individual samples; and MPXV prevalence of 0.9% and 7% respectively in swab samples.Oropharyngeal and anorectal swab samples showed the highest yield.
Two studies [23,27] found that asymptomatic screening was feasible and acceptable to study participants, including one [23] that found that self-sampling was feasible and acceptable to a MSM with higher exposure to mpox and transgender women in Spain.

Qualitative evidence synthesis findings
After screening, four qualitative interview-based studies were found to have addressed the question: 'What are the views and experiences of interventions to prevent sexual transmission of mpox among communities at risk of infection?' Study characteristics for three cross-sectional [32][33][34] and one longitudinal [35] are given in Table 3.Two European studies collected data in the initial phases of the outbreak [33,34]; one study in Australia was conducted after the initial outbreak [35]; and one in China was conducted at a time when reported cases were still increasing [32].
All studies included participants' views on mpox prevention measures, focusing particularly on: behaviour prior to the onset of symptoms [33], behaviour and practices from the time of symptom onset to recovery [32], participants' experiences of treatment and care [35], and engagement in primary and/or secondary prevention measures [34].
Evidence synthesis resulted in four the mes: provision of accurate information on mpox; the diversity of sexual practices and settings; the importance of accessible and high-quality mpox testing and patient care for secondary prevention; and how acceptability of an intervention relates to the perceived cost to wellbeing.

Effective mpox prevention depends on the provision of accurate information
Participants understood sex to be the primary transmission route for mpox, and that increased sexual activity and in particular settings could increase risk of mpox acquisition [32][33][34].However, one study [33] highlighted that network-level factors that made MSM especially susceptible to mpox were unclear to participants.This meant that those who limited their sexual activity perceived themselves as having low risk of infection (and, in the case of those who acquired mpox, did not immediately recognise their symptoms).
All four studies highlighted that a lack of information about the range of mpox symptoms made self-diagnosis challenging for participants who acquired mpox: mpox symptoms, especially when mild, were confused with other STIs or skin conditions, leading to delays in seeking treatment.Perceptions of mpox severity and ability to treat mpox also influenced people's experiences of mpox and their decisions to engage in prevention behaviours.

Not all sexual practices and settings afford opportunities for standard prevention approaches
Participants discussed how having sex with anonymous or unknown partners meant acquiring details for contact tracing which could be difficult or impossible, and the expectations of 'contact tracers' were unrealistic [33,34].Similarly, checking sexual partners for possible mpox symptoms was also seen as impractical.Participants considered abstinence-based measures unfeasible, preferring safer sex and risk mitigation advice [34].
MSM are motivated to engage in prevention measures, but accessibility and quality of testing and patient care may reduce engagement All study authors indicated that MSM taking part in their studies were highly engaged with sexual health services and motivated to seek testing and care for possible STIs.However, there were challenges in accessing testing and accurate diagnosis, necessitating multiple visits and insistence on specialist referrals at times.
Once diagnosed, many participants felt that the predominant focus of clinicians was infection control, rather than care [32,35].Participants in China found mandatory hospital quarantine excessive in the context of lack of mpox treatment, and that self-isolation at home would have been preferable [32].Self-isolation was seen as particularly challenging for those in precarious employment, and for those in shared accommodation [32,34].

Acceptability is relative to perceived cost for individuals' wellbeing
Three studies [32,34,35] described acceptability of different mpox prevention measures, indicating that acceptability of a measure was relative to the participant's perception of its cost to their wellbeing.
Participants weighed up the risks of acquiring mpox against cost to their wellbeing from engaging in prevention measures, which was influenced by their experience of other health issues, the perceived severity of mpox(including concerns about scarring), and their degree of  knowledge about mpox [32,34,35].Those considering sex-on-premises spaces as central to their social and sexual lives deemed it unfeasible to avoid sex altogether [34].Concerns about mpox stigma were a barrier to engagement in prevention measures requiring disclosure (e.g.testing, contact-tracing) [32,34].Anticipated mpox stigma was related to individuals' feelings about their sexuality-some were worried about engaging in secondary prevention measures which could lead to direct or indirect disclosure of their sexuality or to them being labelled as 'promiscuous'.In May et al. some participants described their prevention measures to avoid disclosurefor example, attending work while symptomatic but once any visible symptoms (i.e. on face, arms, etc.) had passed [34].
Lastly, participants with experience of mpox noted that emotional and practical support from family and friends, and positive experiences of care from clinicians, were vital to recovery and to engaging in prevention measures such as self-isolation.

Discussion
Behaviour change adaptations and other public health interventions were recommended to reduce transmission of mpox [10], including prior to and following mpox vaccine introduction in some settings.The search strategy for this review was thus formulated to assess the benefits of a range of preventive interventions other than vaccines and perceptions of them among persons at risk.Research beyond this review does suggest that those at risk of acquiring mpox have been modifying their behaviour in response to the outbreak [32][33][34][35][36][37][38].For some, these changes have endured: a WHO survey of people at increased risk of mpox acquisition found 35% of participants who had changed their sexual behaviour because of mpox concerns said they were still doing so a year after the outbreak was identified [39].
However, our review has so far found limited quantitative evidence evaluating the effectiveness of preventive interventions on sexual transmission of mpox, for either acquisition or onward transmission.This may reflect the challenges in conducting evaluation of behaviour change and other public health interventions during a rapidly developing outbreak of an emergingand in some contexts, poorly understood -infectious disease.
We found only limited data on the benefits of contact tracing [16], with the sharing of information about contacts by patients declining once vaccines become available.Furthermore, qualitative evidence offers important context regarding the limits of contact tracing due to concern about stigma and in some instances due to the anonymous nature of some sexual contacts [33].
Evidence on the utility of asymptomatic testing(question 3) was also limited.Studies reported on the identification of MPXVDNA in persons who were asymptomatic or presymptomatic and the potential for early transmission, underscoring that there may be benefits in raising awareness and intensifying case finding during an outbreak and supporting partner notification where possible.However, there were no studies evaluating the impact of asymptomatic testing interventions in relation to overall transmission.Asymptomatic testing studies were significantly confounded by the timing in relationship to the respective outbreak, lack of controls, small sample sizes, retrospective study design, short study or recruitment periods, and recruitment from a single site.Where asymptomatic testing was seen to be feasible and/or acceptable [23,27], studies did not evaluate cost effectiveness for their respective target populations.In addition, no study identified levels of disease prevalence in which asymptomatic testing may be a useful tool.
Nonetheless, against the background of limited evaluative, quantitative studies, our qualitative evidence synthesis indicates that most affected communities were engaged with preventive and harm reduction strategies during the outbreak.Interventions that increase knowledge of mpox and reduce stigma are most helpful to prevention efforts, as are positive experiences in healthcare.Focussing interventions towards MSM through respectful and non-discriminatory partnering with MSM community organisations and promoting peer-to-peer information sharing was considered reasonable by participants.

Limitations and conclusions
This review is subject to several limitations.The evidence was largely descriptive, rather than evaluative.The heterogeneity of study design and findings made pooling data unfeasible.Studies answering Question 3 relied mostly on self-reporting of symptoms, and there were possible under-estimates of asymptomatic transmission where studies were conducted after the peak of the outbreak in their country.This led to later prospective studies included and finding of low prevalence of asymptomatic infection.Finally, while there were only four qualitative studies addressing our question, these were from significantly different geographic contexts, enhancing global understanding of themes identified.
The primary focus of this systematic review was finding evidence about strategies to reduce sexual transmission of mpox.As a result, the published literature predominantly emerged from non-endemic regions experiencing new outbreaks related to sexual transmission and most of the reviewed literature came from higher income settings.This may limit the transferability of our findings to lower-resource settings and countries where mpox is endemic.As the epidemiology of mpox globally changes however, we believe our findings can have relevance in any context where mpox is transmitted through sexual contact, particularly amongst MSM.
In conclusion, although the mpox public health emergency of international concern was declared over in May 2023, cases continue to be reported worldwide, with the highest number of confirmed cases in the WHO Region of the Americas and the European Region [40].Many lower and middle-income countries including endemic countries in Africa still do not have sufficient diagnostic capacity or any access to vaccines or antivirals outside clinical trials.There is a continuing need for global cooperation to ensure equity of access to diagnostics, vaccines and antivirals to all who need them.
Neither natural immunity from previous MPXV infection nor post-vaccination immunity are entirely effective at protecting individuals from MPXV re-infection, although they may limit illness duration and severity [41].WHO has highlighted the need for further research to better understand the underlying mechanisms in recurrent mpox disease [42].
Collectively, these issues underscore the importance of identifying other effective preventive measures for preventing transmission in different contexts which are also acceptable to communities at risk of mpox.Working with communities to improve mpox education is important to support decision-making around risk and finding equitable ways to reduce the perceived and actual social and economic costs of prevention interventions remains paramount.Such complementary measures should also be explored, such as offering harm reduction advice, providing financial or other forms of support to patients, and proposing public health measures appropriate to different local contexts.

Note 1 .
ID Week is the joint annual meeting of the Infectious Diseases Society of America (IDSA), Society for Healthcare Epidemiology of America (SHEA), the HIV Medicine Association (HIVMA), the Pediatric Infectious Diseases Society (PIDS), and the Society of Infectious Diseases Pharmacists (SIDP).

Table 1 .
Characteristics of included studies for question 1: do preventive interventions decrease the risk of MPVX infection?

Table 2 .
Characteristics and findings of included studies for question 3: does testing asymptomatic individuals at risk of mpox without known exposure to MPXV reduce onward transmission?.

Table 3 .
Characteristics of studies included in qualitative evidence synthesis.