Article Text

Short report
Men who have sex with men in southern Vietnam report high levels of substance use and sexual risk behaviours but underutilise HIV testing services: a cross-sectional study
  1. Quang Duy Pham1,2,
  2. Thuong Vu Nguyen1,
  3. Phuc Duy Nguyen1,
  4. San Hoang Le1,
  5. Anh Tho Tran3,
  6. Long Thanh Nguyen4,
  7. David Peter Wilson2,
  8. Lei Zhang2
  1. 1Department for Disease Control and Prevention, Pasteur Institute, Ho Chi Minh City, Vietnam
  2. 2Surveillance and Evaluation Program for Public Health, Kirby Institute, UNSW Australia, Sydney, New South Wales, Australia
  3. 3Department of Health Services, Long Xuyen, An Giang, Vietnam
  4. 4Ministry of Health, Hanoi, Vietnam
  1. Correspondence to Dr Lei Zhang, Kirby Institute, UNSW Australia, Sydney NSW 2052, Australia; lzhang{at} and Dr Thuong Vu Nguyen, Pasteur Institute, 167 Pasteur Street, 8 Ward, 3 District, Ho Chi Minh City, Vietnam; nguyenthuong{at}


Objectives This study aims to investigate the levels and correlates of unprotected anal intercourse (UAI), drug and alcohol use, and HIV testing among Vietnamese men who have sex with men (MSM).

Methods A total of 381 MSM were recruited in a community-based cross-sectional survey in two towns (Long Xuyen and Chau Doc) in An Giang province in 2009 by using a two-stage cluster sampling. Face-to-face interviews were conducted to elicit respondents’ characteristics. Regression analysis was used to determine the correlates of key behavioural outcomes.

Results In the month before being interviewed, 19.9% respondents had used drugs (13.6% injected), 25.2% had consumed alcohol daily and 33.9% had a UAI with ≥2 male partners. Only 19.2% were tested for HIV in the 12 months before being interviewed. Injecting drug use was significantly associated with having sexual partners who also inject, whereas daily alcohol consumption was associated with an ever-married/cohabiting with women, being transgender and having had at least three male partners in the previous 3 months. Transactional sex, weekly alcohol use, early sexual debut and perception of being at higher risk of HIV infection were correlates of UAI in multiple partnerships. MSM who self-identified as not being gay and those who perceived themselves to be at low risk of HIV infection were less likely to test for HIV.

Conclusions Due to the scarcity of effective MSM-targeted prevention programmes, it is likely that substance use, risky sexual behaviours and low testing uptake may substantially contribute to the spread of HIV among Vietnamese MSM sampled. Harm reduction programmes targeting MSM, and in particular injecting MSM, should be rolled-out in this province.

  • HIV

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Vietnam is facing a rapid surge of HIV epidemic among men who have sex with men (MSM) since the mid-2000s.1 A third of MSM report unprotected anal intercourse (UAI) in the last sexual act with another man.2 The proportion of self-reported illicit drug use has increased from 21.0% and 22.8% in 2006 to 25.8% and 31.8% in 2009 in Ho Chi Minh City and in Hanoi, respectively.2 ,3 Vietnamese MSM have difficulty gaining access to HIV prevention programmes due to considerable stigma and discrimination. In An Giang, which is a typical semirural province in the Mekong Delta in southern Vietnam, although an increased awareness of HIV risk behaviours in MSM has been reported over the past 5 years, harm reduction programmes for MSM have not been a public health focus (personal communication, Anh Tran Tho).

This study aims to examine the prevalence of, and factors associated with, the use of substances, risky sexual behaviours and uptake of HIV testing services among MSM population in An Giang province. The study findings are expected to provide an insight for targeted public health responses against HIV transmission among MSM in this location.


Detailed sampling method and measurement procedures of the study have been described elsewhere.4 Briefly, using the method of Kelsey and colleagues,5 we estimated an adequate sample size to be 341. This calculation took into consideration the estimated ratio of the number of non-injecting to injecting MSM (13), the prevalence level of HIV in non-injecting MSM (7%) and an OR (4.0)3 to ensure sufficient statistical power. By including an estimated 10% refusal rate for blood collection and damaged specimens, this resulted in an effective sampling size of 380. Eligible study respondents were male persons, at least 15 years of age, having had anal and/or oral sex with another man within 12 months prior to the survey, and who had been living in study locations for at least 1 month. A total of 381 MSM respondents in two (Long Xuyen and Chau Doc) of the 11 districts in An Giang province were recruited in this community-based cross-sectional study between August and December 2009 by using a two-stage cluster sampling. Face-to-face interviews were held with the eligible respondents to elicit study information. After the interviews, blood and urine samples were collected for testing for HIV, syphilis, Chlamydia trachomatis and Neisseria gonorrhoea. The study was approved by Institutional Review Boards of the Pasteur Institute, Ho Chi Minh City, Vietnam (reference number: 316/QD-PAS), and the University of New South Wales Australia, Sydney, Australia (reference number: HC14161).

The study outcome variables were injecting drug use, daily alcohol use, having multiple UAI partners (defined as having UAI with at least two male sexual partners) in the previous month and having been tested for HIV in the previous 12 months. Prevalence ratios were used to measure the magnitude of the association between each of the four outcome variables with demographic, sexual and drug-using characteristics identified in multivariate regression models with backward elimination. All statistical analyses were performed using Stata 12 (Stata Corporation, College Station, Texas, USA).


All sample demographic characteristics have been reported previously.4 In brief, 73.2% of respondents were below 25 years of age, 36.6% reported completion of primary education or were illiterate, 7.9% had ever-married a woman, and 40.4% did not self-identify as gay.

Overall, 76 of the 381 MSM respondents (19.9%) reported drug use in the previous month, including 24 (6.3%) non-injecting and 52 (13.6%) injecting subjects. Among the 52 injectors, more than a fifth (21.2%) had either receptive or distributive sharing in the previous month. Multivariate regression analysis showed that having male sexual partners who also injected and having non-paying female partners in the previous 12 months were associated with an increased likelihood of injecting drug use in the previous month, but weekly alcohol consumption was inversely associated with this indicator. The likelihood of injecting drug use also varied with age, peaking in the mid-20s (data not shown). Furthermore, 183 (48.0%) respondents consumed alcohol one to six times per week and 96 (25.2%) of them consumed it every day in the month prior to the survey. Factors associated with daily alcohol consumption included being transgender-identified, having three or more male sexual partners in the previous 3 months, and being married to, separated from or cohabiting with a woman (table 1).

Table 1

Results of bivariate and multivariate regression analysis: proportions of and factors associated with injecting drug use, daily alcohol use, multiple UAI partners and HIV testing among MSM

We found that over a third (33.9%) of respondents had multiple UAI partners in the previous month. Alcohol consumption at least once per week, having sexual debut before age 17, participating in transactional same-sex activities in the previous 12 months and a self-perception of being at high risk of HIV infection were significantly associated with a greater likelihood of having multiple UAI partners among MSM. Despite high levels of the HIV risk behaviours, approximately a quarter (24.9%) of respondents had ever received an HIV test. In all, 73 individuals (19.2%) were tested for HIV within 12 months prior to the survey. Not identifying as gay and a self-perception of being at low risk of HIV infection significantly reduced the likelihood of HIV testing in the previous 12 months, whereas illicit drug use in the month prior to the interview increased the chance of HIV testing (table 1).


We found that although substance use and HIV risk-taking behaviours were commonly reported among the MSM participants, only a small proportion accessed HIV testing services. This is the first study among MSM in a semiurban setting in South Vietnam. The findings have been derived from a well-designed community-based study and provide important insights about addictive and sexual behaviours among MSM in the region. However, they should be interpreted with caution. First, our purposive two-stage sampling approach may not have reached MSM who did not attend the mapped venues, limiting the generalisability of the sample. Second, as MSM respondents were recruited from MSM hotspots in two towns, these respondents may over-represent a subgroup at higher levels of UAI and alcohol consumption. Third, the number of multiple UAI partners and daily alcohol consumption may not reflect the actual levels of high-risk sex and addiction to alcohol among MSM at the study location.

Less than a quarter of men in our sample had ever received a HIV test. Compared with their transgender/gay-identified counterparts, ‘straight’ MSM who do not identify as gay are less likely to test for HIV despite higher infection risks.4 This suggests that most HIV-infected ‘straight’ MSM may not be aware of their infection, representing a substantial risk of HIV transmission to their female partners. Unexpectedly, respondents with a moderate self-perception of HIV risk are less likely to test for HIV than those perceiving themselves as ‘no risk’. This may be explained in part by a prior ethnographic study which indicated that along with prejudice and discrimination, a fear of knowing their HIV status may be a key barrier to accessing HIV testing services among MSM.6

This study reports a high level of substance use among MSM in An Giang province. The proportion of injecting drug use among MSM surveyed is substantially higher than that in other Vietnamese settings,2 indicating the urgency for timely harm reduction programmes among MSM in this location. Our study further highlights the frequent sharing of injecting equipment among injecting MSM, suggesting an elevated risk of HIV transmission due to the overlapping risky sexual and injecting behaviour in this MSM subgroup. This is consistent with our previous finding of a higher HIV prevalence among injecting MSM (20.6%) than non-injecting MSM (3.5%).4

Alcohol use is highly prevalent among MSM in Vietnam. Subgroups, such as transgender-identified and married MSM, are subjected to more frequent alcohol consumption. Stigma and discrimination against transgender individuals may have led to the elevated daily alcohol consumption as a result of anxiety and depression.7 Similarly, influenced by traditional family values, Vietnamese MSM enter their first marriage in their early 30s, and in fear of their homosexuality being exposed, also experience similar levels of depression, stress and other psychological disorders.8 Consistent with previous studies,9 we demonstrate that alcohol consumption contributes to a higher likelihood of having multiple UAI partners among Vietnamese MSM, suggesting drinking may be central in facilitating social networking and same-sex partnerships.

Transactional sexual activities are significantly associated with a higher rate of multiple UAI partners among MSM. This is consistent with findings in other Asian contexts in which MSM who sell sex for money are less likely to use condoms during anal intercourse.10 Interestingly, our study demonstrates a significant association between having multiple UAI partners and self-perception of high risk HIV infection, indicating the underlying unwillingness or inability of MSM to use condoms despite an awareness of potential HIV infection.

The present study has important implications for HIV prevention, linkage to healthcare and future behavioural research among MSM in Vietnam. In addition to conventional strategies such as condom distribution and health education, interventions which significantly reduce substance use and increase utilisation of HIV testing services should be prioritised. Further scale-up of drug treatment through a community-based approach is highly recommended for drug-using MSM. Condoms and lubricants may be distributed through convenient channels such as peer educators, fixed boxes in gay venues and free exchange vouchers at local pharmacies. Social marketing campaigns promoting HIV testing and services targeting MSM should also be further expanded. Future research should focus on identifying the underlying factors leading to substance use, UAI and low HIV testing coverage and potential overlapping risk behavioural patterns among MSM in Vietnam.

Key messages

  • Levels of substance use and sexual risk behaviours are high in Vietnamese MSM.

  • Transactional sex and alcohol consumption contribute to unprotected sex.

  • Access to HIV testing remains insufficient.

  • MSM who do not identify as gay have a greater HIV prevalence but are less likely to test for HIV.


The authors are grateful to the respondents of the study. We thank the peer educators, data collectors and leaders at the Preventive Medicine Centre in Long Xuyen and Chau Doc; Anh Hoang Mai and Hung V. Do in the provincial project management unit of the Vietnam HIV/AIDS Preventive Project in An Giang; Huong Thu Thi Phan, Vietnam Authority of HIV/AIDS Control; Huu Ngoc Tran, Nghia Van Khuu, Cuong Quoc Hoang, Phuong Kim Thi Tran and other staff members at the Pasteur Institute in HCMC for their contribution. We appreciate Ms Louisa Wright at UNSW Australia for editing and proofreading our manuscript. The views expressed in this paper are those of the authors and do not necessarily represent the position of the authors’ institutions of the Australian or Vietnamese governments.


Supplementary materials

  • Abstract in Vietnamese

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • Handling editor Jackie A Cassell

  • Contributors QDP led the design of the study, performed all statistical analysis, interpreted the data and drafted the manuscript. ATT, SHL and LTN assisted in the field implementation of the study and the interpretation of findings. PDN assisted in data entry and data cleaning. TVN, DPW and LZ reviewed the manuscript and contributed towards critical revision of the manuscript. All authors have approved the final manuscript.

  • Funding This work is supported by the Vietnam HIV/AIDS Preventive Project in An Giang, Vietnam under the 2009 grant 04/HDTN/BQL-AG.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval The Pasteur Institute, Ho Chi Minh City, Vietnam, and the University of New South Wales, Sydney, New South Wales, Australia.

  • Provenance and peer review Not commissioned; externally peer reviewed.