Men who have sex with men (MSM) bear a disproportionate burden of HIV worldwide, representing an estimated 7.7% of the global epidemic.¹ In Vietnam, the government has prioritized HIV prevention among MSM as part of its efforts to end the HIV/AIDS epidemic by 2030.² Recent data show an alarming rise in HIV prevalence and incidence among MSM in various provinces.^3,4,5 HIV prevalence among MSM rose from 1.7% in 2005 to 2.4% in 2013, then to 13.4% in 2020.⁶ Hanoi follows this tendency. According to a 2016 study, 6.7%⁷ of MSM participants had HIV infection, rising to 10.9% by 2020.⁸ The rapid increase in HIV transmission necessitates a comprehensive strategy to deliver effective HIV prevention and care services to MSM in Vietnam.

HIV testing is the gateway to treatment for individuals who test positive and effective prevention such as PrEP for those who test negative. MSM and other populations at risk of HIV acquisition should test often. For example, the US Centers for Disease Control and Prevention (CDC) recommends that MSM should be tested for HIV at least annually.⁹ Similarly, the Vietnamese government promotes high HIV testing coverage and high frequency by focusing on key populations including MSM, their sexual partners and social networks, and offering self-testing.^10,11 Frequent testing is crucial for early diagnosis of HIV infection to maximize the benefit of antiretroviral therapy (ART) in reducing morbidity, mortality, and onward transmission. HIV testing also provides an opportunity to identify MSM who are currently uninfected who can benefit from effective prevention interventions, such as PrEP, to maintain their HIV-negative status.

Despite some observed improvements, there is a persistent gap in the level of HIV status awareness among MSM in Vietnam. A recent report from the Joint United Nations Program on HIV/AIDS (UNAIDS) indicated that 83.4% of MSM in Vietnam in 2022 reported having tested negative for HIV in the past 12 months or knew that they had been living with HIV,¹² which is an increase from 65% in 2019 in a previous UNAIDS report.¹³ The National Sentinel Surveillance System of Vietnam has also observed a downward trend in the proportion of MSM who have never tested for HIV, from 39.9% in 2015 to 22.1% in 2020.¹⁴ Persistently high rates of never testing among MSM creates a significant barrier to reaching the first of the three 95:95:95 targets set in Vietnam’s national strategy to end the AIDS epidemic by 2030. That is, to obtain better than 95% HIV status awareness, we must identify which MSM groups are most likely to have never been tested. Thereby, programs can prioritize interventions and maximize early detection of HIV among MSM to reduce morbidity, mortality and onward transmission while also seizing the opportunity to help MSM who are not living with HIV but at risk to initiate PrEP and other enhanced preventive services.

To provide data to help guide HIV-testing campaigns and raise care and prevention awareness, we analyzed baseline data from a large community-recruited cohort study of MSM in Hanoi, Vietnam. Our analysis has three goals. First, we wanted to know how many MSM had never been tested for HIV prior to enrollment and were therefore unaware of their status. Second, we sought to characterize groups of MSM most likely to have never been tested for HIV to better understand the causes of low testing prevalence and tailor programs to their needs. Finally, we identified interactions with healthcare that could have led to HIV testing to quantify missed opportunities for HIV testing among MSM.

A total of 1,893 MSM were recruited, including 796 (42.0%) through TLS, 295 (15.6%) through RDS, and 802 (42.4%) through IBS (Table 1). Participants were young, with 15.4% aged 16–19 years and 47.5% aged 20–24 years. More than half (58.5%) resided in Hanoi for five or more years. Nearly one-fourth (22.5%) had high school education or lower. Most (92.0%) had never been married to a woman, and few (7.3%) had children. Nearly four-fifths (79.5%) earned less than the gross regional domestic product (GRDP) per capita of Hanoi.

The proportion of MSM who reported having never been tested for HIV in their lifetime was 39.4%. Among MSM who reported ever testing for HIV (n=1,148), 77.7% had been tested in the past 12 months, with 74.1% of tests conducted in healthcare settings, 19.0% performed with assistance from a peer educator, and 6.0% performed using a self-test kit at home. The self-reported results of the last HIV test indicated that 6.5% were positive, 88.5% were negative, 3.8% were unknown, and 1.2% were indeterminate. Few MSM (1.4%) reported ever injecting non-prescribed drugs; 5.4% reported ever having a partner who was living with HIV. Nearly one-fifth (19.8%) reported having sex with someone they had met online.

Table 2 presents demographic characteristics and risk behaviors of the 1,893 MSM participants by ever (n=1,148) versus never (n=745) testing for HIV in their lifetime. In bivariate analysis, MSM recruited by TLS had a higher proportion of individuals who had never tested (44.1%) compared to those recruited through RDS (36.3%) or IBS (35.8%) (p=0.002). Regarding demographic characteristics, the percentage of MSM who had never tested for HIV was higher among MSM who were younger (60.6% for those aged 16–19 years, p<0.001), had lived in Hanoi for less than 5 years (45.8%, p<0.001), had a high school education or lower (47.4%, p<0.001), had children (47.5%, p=0.042), and had lower income (42.5% for those earning under GRDP per capita, p<0.001). MSM who had ever had a male sex partner known to be living with HIV were less likely to have never tested (13.6%) compared to those who did not (39.4%) or did not know or provide an answer (42.7%, p<0.001). The internalized stigma score was lower among MSM who had never been tested for HIV compared to those who had been tested (19.3 vs. 19.9, p=0.012).

Table 3 presents the results of the multivariable logistic regression models characterizing the associations with never testing for HIV in the lifetime. The likelihood of never testing was higher (adjusted odds ratio [aOR] >1.0) among MSM who did not remember or know if they had ever had a male partner living with HIV (aOR 1.41, 95% CI 1.07–1.84). Factors associated with a lower likelihood of never testing (i.e., aOR <1.0 or equivalently greater likelihood of ever testing) were older age groups (e.g., aOR for 30 years and older versus 16 to 19 years 0.58, 95% CI 0.36–0.95), university education (versus high school education or lower; aOR 0.69, 95% CI 0.51–0.94), higher monthly income (aOR for each one million VND increase 0.97, 95% CI 0.95–0.99), having a male partner living with HIV (aOR 0.33, 95% CI 0.18–0.62), higher internalized stigma score (aOR for each one-point increase 0.98, 95% CI 0.95 – 1.00), and being recruited by the IBS method (aOR 0.71, 95% CI 0.55 – 0.91).

Missed opportunities for HIV testing among the 745 MSM who reported never being tested for HIV in their lifetime are shown in Figure 1. Regarding healthcare providers, the most common potential missed opportunities were when MSM who had never tested received a free condom from a health educator or peer educator (35.3%), tested for an STI (23.3%), or disclosed MSM status to healthcare workers (16.4%). Additionally, 13.0% of MSM who had never been tested for HIV reported ever having been diagnosed with an STI, 9.9% received a free condom from a health clinic or hospital staff, and 3.5% received a condom from an HIV testing center worker. With respect to experiences of MSM that constitute potential missed opportunities for seeking HIV testing, the most common factors were fear of seeking health care because of their homosexuality (82.7%) and having been refused health care because of their homosexuality (76.2%). Nearly one in five MSM who had never tested reported not seeking care for their STI symptoms at a public health facility (18.8%), having sex with someone they met online (18.3%), and not seeking STI care at a private clinic (7.8%). Few MSM who had never been tested reported ever having a partner living with HIV (2.1%) or ever sharing injecting equipment (0.4%).

Our study, conducted among a diverse sample of MSM in Hanoi, found that nearly two in five (39.4%) had never tested for HIV. Notably, our estimate of never testing is higher than found among MSM in the National Sentinel Surveillance System of Vietnam, at 22.1% in 2020.¹⁴ There are several possible reasons for this difference, including the sampling methods, study locations, and year. Because our cohort study asks for more details on HIV testing, including barriers and experiences of stigma related to healthcare, our less optimistic estimate may unfortunately be more accurate. That is, the short rapid surveys of the National Sentinel Surveillance System may be more subject to social desirability response bias favoring reporting prior testing. Globally, our estimate of never testing for HIV among MSM falls at an intermediate level. For example, compared to countries in Latin America, which has also been witnessing a resurgence in HIV infection among young MSM,²⁰ our estimated proportion of never testing is lower than that in Brazil and Peru, where nearly half of MSM have never tested for HIV.²¹ However, our level of never testing among MSM is higher than that in developed countries such as the US (13%)²² and Japan (16.5%).²³

In Vietnam, the government can continue its efforts to address the proportion of MSM who have never been tested for HIV, hopefully continuing the downward trend in noted in the National Sentinel Surveillance System.¹⁴ This is crucial to ensure the success of the national strategy to end the HIV/AIDS epidemic by 2030 through its “Three Zero’s” campaign: zero new infections, zero deaths from HIV, and zero stigma on HIV.²⁴ A central part of the strategy to reach zero new infections is to achieve UNAIDS’ 95–95-95 goals. That is, to halt viral transmission, 95% of people living with HIV are diagnosed, of whom 95% are on antiretroviral treatment, and of whom 95% are virally suppressed. The first 95 target is threatened by the high level of never testing among MSM as seen in our data. Further, the second and third 95 targets will not have their full impact if the first is not reached, which in turn means that epidemic control is not achieved. Our estimate for never testing MSM in Hanoi is particularly dismaying in the face of policies, strategies, and action plans that have been implemented in Vietnam over the last several years after recognizing MSM as the population expected to account for the largest proportion of new HIV infections in the coming years.²⁵

We also identified demographic characteristics associated with an increased likelihood of never testing for HIV among MSM, including younger age, lower education level, lower income, recruitment via the IBS method, and a higher level of internalized stigma. These findings are consistent with those of previous studies from other countries in Europe, North America, and Asia.^{21,22,26–30,31s,32s} The high level of never testing among younger MSM (60.6% of those aged 16 to 19 years) merits particular attention. The finding may be partly explained by older men having a longer time to test within their lifetimes and by the shorter period of risk for young MSM. However, case surveillance data confirmed a pattern of rising new infections among youth in Vietnam. Among the 9,025 new HIV infections reported in the first ten months of 2022, the majority (81.6%) were transmitted through sex, and nearly half (48.6%) occurred among individuals aged 16 and 29 years.^33s These findings highlight the importance of providing comprehensive sexual and reproductive health information to teenagers that recognizes diverse sexual orientations and gender identities. Communication and education campaigns, whether conducted in schools or through social media, can enhance teenagers’ understanding of gender, risk behaviors, and self-health care. Additionally, these campaigns could play an important role in combating homosexuality-related stigma among MSM. Consequently, MSM may reduce high-risk sexual behaviors and actively seek HIV prevention and care services. In Vietnam, communication campaigns tailored to meet the needs of MSM and other LGBT communities have been implemented in collaboration with non-governmental organizations and have achieved significant success.^34s However, expanding these campaigns to reach multisite or national levels would have a greater impact.

Our study identified multiple missed opportunities for HIV testing that could originate from healthcare and prevention service providers, as well as experiences of MSM that identified barriers to testing and events that could prompt test seeking. Several types of missed opportunities were particularly poignant in the context of sexual health, such as when distributing condoms, screening for STIs, and treatment of STIs. Further, a substantial proportion of MSM who had never been tested for HIV disclosed to their provider that they were MSM. These missed opportunities might be most easily addressed through policies to routinely conduct provider-initiated HIV testing in diverse settings. Routine or universal HIV testing may also reduce stigma as a barrier to testing for MSM, as it does not require disclosure of MSM status. It was particularly disappointing that the vast majority of MSM who had never tested reported experiencing stigma and even refusal of health services because of their homosexuality. In recognition of the impact of stigma on HIV prevention, the government of Vietnam has taken proactive measures to address enacted stigma towards MSM by healthcare providers. Sensitization training programs have been implemented to familiarize healthcare staff with the concepts of sexual orientation, gender identity, and gender expression (SOGIE), and how to incorporate these principles into care services for key populations (KP), including MSM. In addition, national guidelines for the implementation of KP-friendly services have been issued to ensure comprehensive and inclusive healthcare for all individuals.

The stigma-related barrier to testing was further underscored by MSM who reported having symptoms of STI but did not seek treatment. Thus, health education to reduce stigma needs to reach healthcare providers in addition to encouraging MSM experiencing STI symptoms to seek care. HIV self-testing may provide another avenue to reach MSM who have never been tested when stigma from providers is a substantial barrier. Our finding that only 6% of previously tested MSM had used an at-home self-test indicates an untapped potential to distribute self-tests at venues, through peers, or through online-initiated delivery. Opportunities to increase HIV testing that we found have been common to other studies^35s–38s and have led to the proposal of closer integration of STI and HIV testing, encouraging self-testing, improving education on HIV testing, and expanding community-based HIV testing. The last strategy was introduced in Vietnam in 2018 to increase the reach of HIV testing for key populations.³⁹

We recognize our study has limitations. First, our study sample may not be representative of MSM in Vietnam or Hanoi. Nonetheless, the strength of our sample is that it was not dependent on the recruitment of MSM seen at facilities, such as STI clinics. Moreover, the diverse sampling methods (i.e., through peer referral and at physical and online venues) meant that we were not dependent on the biases of one type of recruitment. For example, MSM who do not attend physical venues to find partners can be reached through online recruitment, whereas peer referral may reach MSM who are not connected to either online or physical venues. Second, the required commitment of enrolling in a longitudinal cohort may result in participation bias. Third, there may be social desirability response biases, such as previously testing or denying engaging in stigmatizing risk behaviors. To mitigate bias, we used ACASI, which can improve the accuracy of self-responses to sensitive questions.^40s Lastly, we recognize the age of our data, collected from 2017 to 2019. Conditions related to HIV testing may have changed over time, including those due to external factors, such as COVID-19. More recent data on HIV testing rates would indeed be valuable to better reflect the current situation. Nonetheless, our data collected greater detailed information about the barriers to HIV testing than is possible in the National Sentinel Surveillance System and are the most recent available of its kind. We also point to the need for a baseline trajectory, including one preceding COVID-19, against which to assess future progress.

Despite these limitations, our data draw attention to the large gap in HIV testing for Vietnam’s priority MSM population. Simultaneously, we identified practical opportunities to close this testing gap. These include maximizing provider-initiated testing within all interactions with MSM and providing education to reduce stigma. HIV testing may also be expanded at venues where MSM can be reached, particularly if peer-delivered and performed with community consultation.^{39s,41s,42s,43s} Finally, it is important to maximize HIV testing opportunities for young MSM for whom HIV testing coverage is low and the potential for future HIV transmission and acquisition is high in Vietnam.