Back to contents

Love is a danger

Gus Cairns, Michael Carter
Published: 01 June 2009

American researchers have reported that gay men in the US are more likely to contract HIV from their main sexual partner than from casual sex.  What are the implications for HIV prevention messages there? ask Gus Cairns and Michael Carter.

In a country where sex between men and women is the main transmission route for HIV, the biggest increase in a woman’s risk of getting HIV happens when she gets married. That’s one reason why sexual abstinence campaigns don’t work: most women with HIV were abstinent - till their wedding night.

But in countries where most transmission is between gay men there has been an underlying assumption that it’s all fuelled by casual sex. This idea was recently challenged by a US study1, which calculated that two-thirds of gay men who acquire HIV do so from their regular partner.

The study didn’t look at actual HIV transmission episodes in a group of men. Instead, it analysed data from the National HIV Behavioral Surveillance System (NHBS), a study of 3652 HIV-negative and positive gay men in five US cities between 2003 and 2005.2 This looked at the number of sexual partners men had in a year, both main and casual; the perceived HIV status of their partners; the proportion having anal sex who didn’t use a condom last time they had sex; and whether they were insertive (‘top’) or receptive (‘bottom’). The NHBS study also took anonymous saliva samples for HIV screening from half of the men they studied. It could therefore tell whether gay men’s knowledge of their HIV status was accurate.

Because the NHBS only collected data about the last sexual episode, one missing thing needed to know was how many times men had actually had sex during the year with main and casual partners. They therefore added in data from a previous study, the Vaccine Preparedness Study, which recruited 3617 gay men in six US cities between 1995 and 1997.

Encouraging couples to test, or re-test, for HIV the moment they get involved with someone emotionally has several benefits.

With all these data, the researchers were able to work out what proportion of HIV infections must have come from regular partners, and the answer was around two-thirds – 68%. Even when they tested their model to destruction by putting in the minimum possible estimated rate of unsafe sex with main partners and the maximum possible for casual partners, still more than half of HIV infections came from main partners.

There were several reasons for this. Firstly, men had more actual occasions of sex with their main partner. In raw numbers there were about 195,000 occasions of sex between main partners and about 173,000 episodes of casual sex. So gay men don’t exactly behave like Indian brides, but they still have more sex (about 80 times a year on average) with their regular partner than with casual partners.

Secondly, they were more likely to have anal sex with main, rather than casual partners and were more likely to be the receptive partner too. Forty per cent had been anally receptive last time they had sex with their boyfriend (though they might have been active too) but only a quarter of them the last time they had casual sex.

Thirdly, they were considerably less likely to use condoms with main than casual partners. Men used condoms for anal sex (in either role) 72% of the time with casual partners but only 43% of the time with main partners. Add all these together and you get a lot more instances of unsafe sex with main partners per year than with casual partners.

HIV prevalence was around 25% in both casual and main partners. Men were pretty bad at either knowing or guessing their own or their partners’ HIV status. When the researchers looked at actual HIV transmissions, they worked out that only 16% of men who caught HIV had known their partner had HIV.

In contrast 46% of men who acquired HIV caught it from someone they had believed to be HIV-negative, and 55% of transmissions from main partners came from men who believed themselves to be negative. In casual sex, 63% of transmissions happened in situations where the partner’s status had been unknown.

Men with diagnosed HIV infection were therefore the group least likely to be responsible for HIV transmission. This was because, when one partner was known to have HIV, couples were no less likely to use condoms with their main partner than during casual sex.

There were a couple of other standout findings from the study. Firstly, although being the receptive partner (the ‘bottom’) accounted for nearly 70% of infections, 28% of infections were acquired by the insertive partner (the ‘top’). This is more than predicted from previous studies, though some HIV campaigners did comment that some men might have lied about being a ‘bottom’ as it is still stigmatised. Note also that the study found that 2 to 3% of infections came from oral sex – a figure that backs up earlier studies.

Secondly, the annual HIV incidence – the proportion of this group that acquires HIV every year – was calculated to be 2.2%. Given that this is ten times the observed mortality rate among gay men in the Vaccine Preparedness Study,3 the number of gay men in the USA with HIV will continue to grow, at least until significant numbers start dying of old age, unless HIV incidence is cut drastically.

This study obviously has huge implications for HIV prevention messages. It’s well known that people use condoms much less with their main partner and for many couples this is because condoms are seen as a sign of distrust. Trying to get regular partners who don’t use condoms already to use them can be a challenge.

Encouraging couples to test, or re-test, for HIV the moment they get involved with someone emotionally has several benefits. If one partner does test positive for HIV it not only means they are in a much better position to look after their own health, but successful treatment will also reduce their infectiousness. And regardless of test results, testing for HIV is a powerful prevention tool, bringing people into contact with sexual health services, but most importantly, opening up a discussion between partners about sexual health and safer sex.

Exactly which prevention message is most appropriate, however, depends on one question not asked by the survey. In the words of Roger Tatoud, Senior Programme Manager for the International HIV Clinical Trials Research Office at Imperial College in London, “If most people get HIV from their main partner, where did he get it from?” This is also a dilemma faced by researchers into heterosexual transmission. There are two theories:

The ‘serial monogamy’ theory is based on the idea that committed relationships among gay men and especially young gay men might not last very long, and this is borne out by the finding that the men averaged 1.4 ‘main partners’ a year. If the average interval between HIV tests or, for those who’ve never tested, between first sex and the most recent, is longer than or near the average length of committed relationships, then there are going to be a lot of men entering new relationships who don't know they have HIV. Message: test as soon as you get involved.

The ‘concurrency’ theory, on the other hand, assumes there is far more sex going on outside a lot of main relationships than the researchers captured, both because respondents didn’t tell them about it (we know people lie about sex to even the most confidential survey) or because the researchers made assumptions about what ‘main relationships’ mean to gay men, for instance that emotional commitment means monogamy, and so didn’t ask clearly enough about it. Message: have an agreement to maintain safer sex outside your main relationship.

Whichever theory is right, this study challenges assumptions that offering gay men marriage and the chance to settle down with the dog and Radio 4 might mean less HIV transmission, and will cause some rethinking in the world of HIV prevention.


1. Sullivan PS et al. Estimating the proportion of HIV transmissions from main sex partners among men who have sex with men in five US cities. AIDS 23(9):1153-1162, 2009.

2. Centers for Disease Control and Prevention. HIV prevalence, undiagnosed infection, and HIV testing among men who have sex with men: five US cities, June 2004-April 2005. MMWR 52:597-601, 2005.

3. Seage GR et al. Are US populations appropriate for trials of human Immunodeficiency Virus Vaccine? The HIVNET Vaccine Preparedness Study. American Journal of Epidemiology 153(7):619-627, 2001.

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.
Community Consensus Statement on Access to HIV Treatment and its Use for Prevention

Together, we can make it happen

We can end HIV soon if people have equal access to HIV drugs as treatment and as PrEP, and have free choice over whether to take them.

Launched today, the Community Consensus Statement is a basic set of principles aimed at making sure that happens.

The Community Consensus Statement is a joint initiative of AVAC, EATG, MSMGF, GNP+, HIV i-Base, the International HIV/AIDS Alliance, ITPC and NAM/aidsmap

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.

NAM’s information is intended to support, rather than replace, consultation with a healthcare professional. Talk to your doctor or another member of your healthcare team for advice tailored to your situation.