High HIV incidence for Thai men who have sex with men; many acquiring HIV in their early twenties

This article is more than 12 years old. Click here for more recent articles on this topic

Among young Thai men who have sex with men, 6 in 100 acquire HIV each year. With the average age at infection being 26, this explosive epidemic is affecting a far younger group of men than the gay epidemics in Western countries. These are the headline findings of the first three years of a study to monitor HIV incidence, reported by Frits van Griensven to the Eighteenth International AIDS Conference in Vienna on Tuesday.

More encouragingly, a second study from Thailand, reported the same day, suggested that HIV prevalence could be declining in men who have sex with men, after having peaked at about 30% in 2007.

Thailand is often seen as a lesson in early and effective HIV control. However, until recently, efforts have largely concentrated on commercial sex workers, their clients and injecting drug users. The ongoing spread of HIV among men who have sex with men (MSM) was largely ignored until a much overdue study of Bangkok MSM in 2005 found an HIV prevalence of 17%.


odds ratio (OR)

Comparing one group with another, expresses differences in the odds of something happening. An odds ratio above 1 means something is more likely to happen in the group of interest; an odds ratio below 1 means it is less likely to happen. Similar to ‘relative risk’. 

multivariate analysis

An extension of multivariable analysis that is used to model two or more outcomes at the same time.

oral sex

Kissing, licking or sucking another person's genitals, i.e. fellatio, cunnilingus, a blow job, giving head.


Refers to the mouth, for example a medicine taken by mouth.

person years

In a study “100 person years of follow-up” could mean that information was collected on 100 people for one year, or on 50 people for two years each, or on ten people over ten years. In practice, each person’s duration of follow-up is likely to be different.

In the incidence study, a cohort of 1292 Bangkok MSM was recruited, which is being followed at four-monthly intervals. Recruitment began in April 2006 and was completed by January 2008. Participants are all Thai nationals, male at birth, resident in the Bangkok area, aged 18 or over, and have had anal or oral sex with a man in the six months before recruitment. Participants were recruited from the sexual health clinic where follow-up visits were conducted, through a website, from a range of bars, saunas and parks used by MSM, and through community organisations.

Men recruited were relatively young (73% aged 29 and under), and were frequently born outside of Bangkok (64%). A majority were employed, well educated and identified as homosexual or gay.

In this sample, 22% of men were HIV-positive at baseline but were generally unaware of their infection. After three years of study, another 135 men had acquired HIV. In a 12-month period, the annual rate of HIV acquisition (incidence) was 5.9 per 100 person years. This did not vary between years 1, 2 or 3 of the study.

The mean age at seroconversion was 26.4, and the median was 26. “This means that 50% of the men who became infected during follow-up were younger than 26 years, and we have seen quite a large number of cases where men got infected when they were 18 or 19 years of age,” van Griensven commented. He noted that in cohort studies in other parts of the world, men are typically in their thirties when they seroconvert.

Among those men who were initially HIV-negative, it was the youngest men who were actually most likely to have acquired HIV by the end of the study. Using the Kaplan-Meier method to estimate the chances of remaining HIV-negative after four years, 89% of those aged 30 or over would be negative, in contrast to 71% of those aged 18 to 21 at recruitment.

This suggests that men are most vulnerable to infection when they are younger – men who did not acquire HIV at a young age are less likely to acquire it when they are older. van Griensven said that behavioural surveys showed that on average younger men had more drug use, unprotected sex and sexual partners.

The age difference is also illustrated by the odds ratios for having either prevalent or incident HIV infection. Prevalence relates to an infection acquired at any time in the past, so older men were more likely to have prevalent HIV. A man aged 30 or over was three times more likely to have prevalent HIV than a man aged 18 to 21 (odds ratio 3.2)..

On the other hand, incidence refers only to newly acquired infections. A man aged 18 to 21 was also almost three times more likely to have incident HIV than a man aged over 30 (odds ratio 2.7).

As well as age, other factors associated with incident (new) infections in the multivariate analysis were using drugs for sexual pleasure (primarily crystal meth, odds ratio 2.9); using taking the receptive position (odds ratio 1.9); inconsistent condom use (odds ratio 5.3 but with wide 95% confidence intervals, 1.3 to 21.8); past herpes infection (odds ratio 1.6) and past syphilis infection (odds ratio 2.5).

A second study from Thailand...suggested that HIV prevalence could be declining in men who have sex with men, after having peaked at about 30% in 2007

In response to a question, van Griensven stated that his study could not indicate whether incidence is rising or falling in Bangkok. However, a separate study of HIV prevalence among Bangkok men who have sex with men may suggest that prevalence is in fact falling.

Nonetheless, the survey method (a series of cross-sectional surveys, conducted at two-year intervals) is not the most reliable way of measuring these changes. Slightly different recruitment methods used for each survey could result in a different profile of men being recruited to different surveys, with the result that apparent changes in HIV prevalence may not be genuine.

The surveys have been conducted over an eight-year period, in 2003, 2005, 2007 and 2009. Men aged 15 and over, who report anal or oral sex with another man in the past six months, are recruited at entertainment venues, saunas and parks.

Overall HIV prevalence rates were 17% in 2003; 28% in 2005; 31% in 2007; and 25% in 2009. The increase from 2003 to 2005 was statistically significant, whereas the apparent decrease from 2007 to 2009 was not (p=0.22).

For those under the age of 22, HIV prevalence was 13% in 2003; 22% in 2005; 22% in 2007; and 13% in 2009.

Regarding risk behaviours, almost all 2003 survey respondents reported that they had engaged in anal sex at some point in their life, but only about two-thirds of respondents did so in 2009 (98 vs 62%; p<0.001).

On the other hand, almost no men in 2003 reported using drugs in the three months preceding the survey, but in 2009, one in five had done so (4 vs 20%; p< 0.001). Drug use during last sex also increased, from 0.7 to 4.2% (p< 0.001).

Consistent condom use in the past three months remained steady at around 64% – 68% across survey rounds.

Brazil: prevalence and misconceptions

Also at the conference, Ligia Kerr reported on a preliminary study of HIV prevalence in men who have sex with men in Brazil. The national average prevalence of 13% in urban areas and 11% nationally, with prevalence as high as 20% in one city.

Against this background of high HIV prevalence, knowledge of HIV risks and risk perception was low, she reported. Only 47% of men interviewed had correct knowledge of HIV transmission. Three quarters of men thought that they were at low risk of HIV infection, although almost half of men who reported a recent casual sex partner had not used a condom in anal intercourse. The proportion was similar in men with a regular sexual partner.

Further information

Presentations by the speakers and their related abstracts are available on the official conference website.


Van Griensven F et al. Three years of follow-up in the Bangkok MSM cohort : evidence of an explosive epidemic of HIV infection. Eighteenth International AIDS Conference, Vienna, abstract TUAC0301, 2010.

Kladsawas K et al. Trends in HIV prevalence and risk behaviour among men who have sex with men (MSM) in Bangkok, Thailand, 2003 to 2009. Eighteenth International AIDS Conference, Vienna, abstract TUAC0203, 2010.

Kerr L et al. MSM in Brazil: baseline national data for prevalence of HIV. Eighteenth International AIDS Conference, Vienna, abstract TUAC0206, 2010.