Research has shown that men who have sex with men (MSM) in Vietnam are at high risk of HIV as a result of multiple sexual partners, low condom use and low perceptions of risk.1^–3 Although MSM were ignored by the government in the past, the revised law on HIV/AIDS passed by the Vietnam National Assembly in 2006 specifically lists homosexuals among the high-risk groups prioritised for HIV prevention programming.4

There are few data on the prevalence of HIV among MSM in Vietnam. Sentinel surveillance does not include MSM and homosexual sex is not usually disclosed as a risk factor when HIV cases are reported. The Vietnam Ministry of Health reports that the majority of HIV infections are transmitted through intravenous drug use and between female sex workers (FSW) and their clients.5 However, more than one-third of infections in the country have risk factors that are either unknown or unreported, whereas MSM and homosexual sex are not mentioned in national statistics.

HIV prevalence among MSM in Vietnam has only been measured in the urban areas of Ho Chi Minh city (HCMC) and Hanoi. The HIV prevalence rate was 8% in 2004 among a convenience sample of 600 MSM tested in HCMC in 2004.6 Further surveys in 2006 showed MSM HIV prevalence rates of 5.3% in HCMC and 9.4% in Hanoi.7 Although 73% of the population of Vietnam lives in rural areas,8 no data exist on HIV prevalence among MSM outside of the country’s two largest cities.

Nha Trang city is situated on the south-central coast of Vietnam, 500 km north of HCMC. It is the capital of Khanh Hoa province and has a population of approximately 300 000 people. Khanh Hoa province, with a population of more than 1 million, is one of the 10 provinces in Vietnam with the highest prevalence rates of HIV.5 By the end of 2005, a total of 1706 cases of HIV infection had been reported to the provincial health department (unpublished data, Khanh Hoa Health Department). Previous qualitative research conducted in Nha Trang city showed that MSM engaged in very risky sexual behaviour and had poor knowledge about HIV transmission and prevention.9

Since 2003, the Khanh Hoa provincial health service has implemented the men’s sexual health programme with the goal of decreasing HIV and sexually transmitted infection (STI) rates among high-risk men, including MSM in Nha Trang.10 In an effort to expand the programme to provide outreach services in the rural areas, an assessment was carried out to determine HIV prevalence and risk behaviour among MSM throughout the province.

METHODS

The geographical area of the assessment was all five districts in Khanh Hoa province: one urban area (Nha Trang city) and four rural districts (Dien Khanh, Cam Ranh, Ninh Hoa, and Van Ninh). Recruitment was stratified such that 70–80% of the responses would represent the rural districts. All data were collected from July to September 2005.

Inclusion criteria for the assessment were men who had sex with another man at least once in the previous year, Vietnamese citizenship, age 15 years or above and giving voluntary consent to participate. Participants gave oral consent after reading an information sheet describing the purpose and procedures of the assessment. Staff members were present to answer questions or provide additional information. In order to keep all information anonymous, neither names nor any other identifying information were recorded. The study was approved by the scientific and ethical review committee at the Khanh Hoa Department of Health.

Quantitative data were collected using a standardised questionnaire (see questionnaire form published online only). The survey instrument contained 107 questions divided into 10 sections: demographics, sexual health, sexual behaviour, regular sexual partners, casual female partners, oral sex, anal sex with casual male partners, sex with sex workers, sex work and knowledge about HIV transmission and prevention. Three millilitres of blood was taken for an HIV test (Genscreen; Biorad, Hercules, California, USA), performed at the Preventive Medicine Centre in Nha Trang. Positive results were confirmed with two additional HIV tests following standard procedures dictated by the Vietnam Ministry of Health. HIV testing was linked to the survey questionnaire through a serial number. Participants were encouraged to go to the local voluntary HIV testing centre, where they could receive HIV counselling and their test result by providing the study serial number. All participants were provided with condoms, water-based lubricant, HIV prevention brochures and referrals to the services of the provincial MSM outreach programme.

MSM were recruited for the quantitative data collection using respondent-driven sampling.11 Two seeds were enrolled from each of the five districts. All seeds were volunteer or paid peer educators working with the MSM outreach programme. The sample size of 300 was calculated to give an estimate of HIV prevalence of plus or minus 2.5%, assuming a sample prevalence of 5%. Recruitment was stopped after five waves of recruits when the final sample size was reached. Participants received an incentive of 30 000 Vietnamese dong (VND), approximately US$1.90, to complete the questionnaire. Each participant could recruit up to three additional subjects and received 15 000 VND (US$0.95) for each successful recruitment.

The questionnaires were administered in each of the five districts in order to minimise the travel time required for participation and eliminate any potential bias for non-participation from the more distant rural areas. A total of 330 men was screened and 295 (89%) were included in the data analysis; 33 were excluded based on enrollment criteria and two did not fully complete the interview questionnaire.

All data presented are crude unadjusted results. Statistical analysis was conducted using SPSS version 12.0. Statistically significant differences were determined using a two-sided t test for independent samples for continuous variables and χ² testing for categorical variables.

RESULTS

The geographical distribution of the participants was 27% from Nha Trang city and 73% from the four rural districts (Cam Ranh 26%, Dien Khanh 19%, Ninh Hoa 12%, Van Ninh 17%). The median age was 24 years and the median number of years of education was eight in both the urban and rural areas. Additional demographic information is listed in table 1. A higher proportion of rural MSM identified as heterosexual, but this difference was not statistically significant.

HIV prevalence

All 295 participants were tested for HIV. None of the MSM tested positive, giving an HIV prevalence rate of 0% (95% CI 0 to 1.24%).

Risk behaviour

Sexual risk behaviour is summarised in table 2. Urban MSM had their first sexual experience at a younger mean age (17.7 years) than rural MSM (19 years) (p = 0.001). Almost half (48%) of all subjects had anal sex with casual male partners in the previous 6 months, with no difference between the rural and urban men. Only five (1.7%) of the total sample had ever used intravenous drugs. Urban MSM were more likely to have ever injected drugs (4/79, 5.1%) than rural men (1/216, 0.5%) (p = 0.007).

Commercial sex

Urban MSM were significantly more likely to have ever had sex with a sex worker and to have had sex with a male sex worker (MSW) or FSW in the last 6 months (table 2). Although 33% of urban MSM and only 23% of rural MSM reported ever having sex to earn money, the difference was not significantly different. MSW served a mean of five and a median of three clients in the last 6 months (range 1–60), with no differences in the urban and rural populations. Of the 56 MSW who reported numbers of clients, only eight (14%) reported an average of more than one client per month.

The amount of money MSW reported earning was lower for the rural men: the majority (78%) of rural MSW reported less than 50 000 VND (US$3.15) per client, whereas 73% of urban MSW reported earning more than 50 000 VND and 23% reported more than 200 000 VND (US$12.60) from each client.

Condom and lubricant use

Data on condom and lubricant use are listed in table 3. Only 26% of rural MSM used water-based lubricant with condoms, significantly less than the urban MSM. Condom use with oral sex was uniformly infrequent. Of those MSM who had sex with FSW, condom use during the last episode of vaginal sex was 60%. Condom use at the last anal sex with male partners was higher at 75–89%. Condom use for the last insertive anal sex with male clients was reported as only 36% among rural MSW, significantly lower than the rate of condom use reported by urban MSW.

Knowledge and perceptions of risk

Both rural and urban MSM were aware of HIV transmission through unprotected sex, but rural men were less likely to know that HIV could be transmitted from mother to child or by sharing intravenous needles (table 4). Rural MSM were also more likely to believe mistakenly that HIV could be transmitted through casual contact or mosquito bites. Rural MSM were significantly less likely to know that anal sex had a higher risk of transmission than oral sex.

Knowledge about MSM in general and the risk of HIV was associated with safer sexual behaviour. Those rural men who agreed with the statement “MSM are at higher risk than other people in Vietnam” were more likely to have used condoms the last time they had receptive anal sex with casual partners when compared with men who disagreed with the statement or were not sure (40/45 (89%) vs 39/54 (72%), p = 0.04). Rural MSW who agreed with the statement were more likely to have used condoms for the last insertive anal sex with male clients (10/13 (77%) vs 6/31 (19%), p<0.001), although there was no difference in condom use for the last receptive anal sex with male clients (data not shown).

Access to services and information

The most commonly cited sources for information on HIV/AIDS were books and newspapers, television and healthcare workers. Only 3% of MSM noted the internet as a source of information, although 30% reported that they used the internet to meet other MSM and 30% of MSW used the internet to meet customers.

The majority of MSM said they would be comfortable discussing their sexual orientation and behaviour with healthcare workers: 22% under any circumstances and another 76% if the situation felt comfortable to them, with no significant differences between the urban and rural men.

Of all the MSM who responded, 85% of urban MSM and 74% of rural MSM had been in contact with the MSM peer educators from Nha Trang. Only 10 respondents (3.4%) had previously been tested for HIV before participating in the research: all had been in contact with the peer educators.

DISCUSSION

None of the MSM who participated in this survey tested positive for HIV antibodies. Although a prevalence of zero in one survey cannot rule out any HIV among MSM in Khanh Hoa province, it does indicate that the overall prevalence of HIV in this population is very low.

This survey showed that MSM exist in rural areas of Vietnam, are relatively easy to contact and are willing to answer questions about their sexual orientation and behaviour. Most also noted that, under the right circumstances, they would be willing to talk openly and honestly about their sexuality with counsellors and health workers. In the family and community, however, most MSM keep their sexual orientation and homosexual behaviour hidden. MSM in rural areas are more likely to identify themselves as heterosexual or bisexual and less likely to see themselves as homosexual.

Fewer rural MSM have been in contact with peer educators, although the fact that 74% of rural MSM report peer educator contact shows that the peer educator programme has made significant efforts to reach the rural communities. There is evidence that contact with peer educators may be associated with safer sexual practices.

MSM in rural areas are less knowledgeable about HIV transmission and prevention than MSM in the city. Public health campaigns in Vietnam have been very successful in informing the population about the routes of HIV transmission. However, this knowledge is often superficial and is not accompanied by detailed knowledge about prevention methods or the risks of specific sexual behaviours. In this survey, rural MSM were less likely to know that HIV cannot be transmitted through casual contact or that anal sex carries a higher risk of transmission than other forms of sex.

Perception about risk can be an important factor in determining risk behaviour. Previous research has shown that perceptions about the risk of HIV among MSM in general are positively associated with safer sexual behaviour among MSM in HCMC.1 Among MSM in the rural areas of Khanh Hoa, the belief that MSM in general have a higher risk of HIV was also associated with increased condom use. However, rural MSM are less knowledgeable about the higher risk of HIV among MSM than are MSM in Nha Trang.

The rural areas of Khanh Hoa province have a population of MSM that can be easily reached through peer educators. These men engage in high-risk sexual behaviour with multiple sexual partners. There are significant gaps in their knowledge about HIV transmission and prevention. Rural MSM need more information on the risks of homosexual sex, particularly anal sex and methods to decrease the risk of HIV and STI transmission. Knowledge about and access to water-based lubricant also needs to be improved.

New methods for delivering information should be considered. The internet is widely available in most towns in Vietnam and is often used by MSM to meet friends and potential sex partners. Internet sites could be one way to provide information to some MSM who otherwise might be difficult to reach.

Although this assessment was limited to Khanh Hoa province, the results may be applicable to other provinces in Vietnam. As in Khanh Hoa, more than 70% of the population of Vietnam live in rural communities. MSM may be less visible in rural areas, but they exist in sizeable numbers and if sexually active are at risk of HIV and STI. This assessment represents the first effort to gather information about MSM in rural Vietnam: more research is needed in order to understand better the specific risks of rural MSM and to develop HIV prevention interventions targeting these populations. All existing MSM programmes in Vietnam are located in urban areas: these programmes should make more efforts to reach out to the rural areas in their provinces.