Higher rates of HIV in US black gay men may be due to smaller choice of partners and more age mixing

This article is more than 11 years old. Click here for more recent articles on this topic

Limited partner choice, wider age gaps between partners, and mistaken beliefs about HIV status in regular partners are all driving the substantially higher rates of HIV in US black gay men relative to other ethnicities, a recent study suggests.

The study also found that higher rates of HIV in black gay men were not due to higher HIV risk behaviours – indeed it found, as some other studies have done, considerably lower overall rates of unprotected sex in black men than in others.  

It also found, however, that black gay men were significantly more likely to have partners who were considerably younger or older than they were.


multivariate analysis

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

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’. 

mathematical models

A range of complex mathematical techniques which aim to simulate a sequence of likely future events, in order to estimate the impact of a health intervention or the spread of an infection.


How well something works (in a research study). See also ‘effectiveness’.

not significant

Usually means ‘not statistically significant’, meaning that the observed difference between two or more figures could have arisen by chance. 

Other studies have found this too. What this study adds is that it found that interactions of risk factors, rather than single risk factors, were the most crucial determinants of higher HIV risk in black men. For instance, while black men were less likely to have unprotected sex than other ethnicities, they were more likely to drop condom use once a relationship became long term.

Similarly, while older non-black gay men were significantly more likely to use condoms if their partners were older than they were, in black men the pattern was the opposite: younger black gay men were less likely to use condoms with significantly older partners.  

The study

The study was a relatively small one in which HIV-negative gay men under 40 who were not monogamous were asked to keep diaries of their sexual encounters over a three-month period, updating the diaries every week.

Via Facebook and the small-ads website craigslist, researchers recruited 143 gay men who were selected in order to make the group balanced in terms of race, age and location within the US. To give some specifics, 42% of the group were aged 16 to 25 (mean age 27.5); 23% were black and 37% white; 75% described themselves as gay and 20% bisexual.

Participants were asked to score their partners’ age difference from them, their gender, familiarity and their perceived HIV status.


The study found that participants had an average of about one sexual encounter a week and that 54% of these were with partners they already knew. Over a quarter (27%) of these involved unprotected anal or vaginal sex.

Participants’ age made no difference to whether they had unprotected sex, but race did. Black men were only a third as likely to report unprotected sex in any one encounter as men of other ethnicities (odds ratio 0.32, p <0.01).  In multivariate analysis, however, taking age, the sexual familiarity of the partner and their HIV status into account, this became statistically non-significant (due to factors discussed below) and the only factor that was independently associated with non-use of condoms was having a female partner: the minority of men who were bisexual were four times less likely to use a condom with a woman than a man.

Although participants were in general more likely to have sex with men of their own ethnicity than others, black men were by far the most ‘sexually homophilous’ group, to use the researchers’ term. They were eleven times more likely to have black partners than other racial groups. In contrast, Latino men were three times more likely to have Latino partners than other groups and white men two times more likely to have white partners.

Black men were 34% more likely to have unprotected sex if their partner was older than them than if they were younger, and over twice as likely if their partner was more than ten years older. This effect was not moderated by the participant’s own age.  In contrast, younger black men (under 25) were actually more likely to use condoms than non-black men if their partner was even younger than they were. 

This was exactly the opposite of the pattern in older, non-black gay men, who were nearly twice as likely to use condoms with partners older than themselves than younger ones. In HIV-negative men, this would tend to reduce the risk of acquiring HIV, whereas the pattern seen in black men would tend to increase it.

In general, the number of previous sexual encounters did not affect the likelihood of unprotected sex and, as we said previously, black men were actually less likely to have unprotected sex, averaged over all encounters. However, as their familiarity with sexual partners increased, their likelihood of unprotected sex gradually increased.

Using mathematical models to project from the actual data, the researchers found that black men were almost four times less likely to have unprotected sex with a partner they had never met before. The likelihood of unprotected sex became roughly equal between ethnicities once the participant had had sex with their partner about 20 to 30 times, and in cases where they had had sex 50 times or more with them, suggesting a quite committed relationship, black men were more than twice as likely to report unprotected sex than non-black men.

Conclusions – and dilemmas

What this suggests is that while black men do attempt to moderate their risk of HIV, and in some ways do so more consistently than other men, the much higher prevalence of HIV in the black community overwhelms these attempts.

Black men may even be trying to use strategies that are directly counter-productive. If, for instance, partner familiarity is used as a guide to whether to use condoms or not, then it could mean that having sex with one or a few regular partners would end up being more risky than having sex solely with casual partners.

The age difference data may also be crucial. Young black men were more likely than any other group to have unprotected sex with older partners – exactly the pattern that would be most likely to spread HIV rapidly between generations.

The researchers comment that, while these data suggest some fine-tuning of HIV prevention messages (such as not giving up on condom use with regular partners), the behaviour patterns revealed may also be caused by much harder-to-shift structural and social factors.

To what extent, for instance, do economic factors make a difference? Young black men in the US are highly likely to be poor, unemployed and recently imprisoned, and may have economic reasons for preferring older partners.

And what about racism? To what extent do black men choose to have sex with other black men, and to what extent does rejection by other men force them to? In addition, if it is widely known in the gay community that black men are more likely to have HIV, this will tend to strengthen the effects of racism, ‘ghettoising’ HIV.

The researchers conclude: “Future research must strive to understand the mechanisms underlying these predictors of sexual risk in order to develop efficacious intervention strategies that address the unique needs of black MSM and reduce the profound effect of HIV on this community.”


Newcomb ME, Mustanski B Racial differences in same-race partnering and the effects of sexual partnership characteristics on HIV risk in MSM: a prospective sexual diary study. JAIDS 62(3):329-333, 2013. See abstract here.