Consistent decline in partner numbers in US gay men in last decade, but no change in condom use

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

Data from two national sex surveys in the United States show that gay and bisexual men (men who have sex with men, MSM) reported significantly fewer sexual partners in the previous year in a survey conducted between 2006 and 2010 than they did in one conducted in 2002. This decline was consistent across most ethnicities and age groups, but was particularly marked, and statistically significant, in younger men aged under 24.

In contrast, the proportion who reported having condomless anal sex at least once in the previous year did not change between surveys. In the minority of men who also had sex with women, condom use fell markedly, but on the other hand the proportion of MSM who also had sex with women fell too.   

The proportion of men who tested for HIV or for sexually transmitted infections (STIs) in the last year did not change, although the proportion who had never tested for HIV fell.

The survey

The data come from the last two National Surveys of Family Growth (NSFGs). The NSFG is a survey of 15 to 44-year-olds; participants are contacted at random by phone but due to lower contact/response rates, people under 24, black people and Hispanic people are ‘oversampled’, i.e. a higher proportion are initially contacted than are in the general population.


statistical significance

Statistical tests are used to judge whether the results of a study could be due to chance and would not be confirmed if the study was repeated. If result is probably not due to chance, the results are ‘statistically significant’. 


Having sex without condoms, which used to be called ‘unprotected’ or ‘unsafe’ sex. However, it is now recognised that PrEP and U=U are effective HIV prevention tools, without condoms being required. Nonethless, PrEP and U=U do not protect against other STIs. 

risky behaviour

In HIV, refers to any behaviour or action that increases an individual’s probability of acquiring or transmitting HIV, such as having unprotected sex, having multiple partners or sharing drug injection equipment.

NSFGs used to be conducted every three to seven years, but in 2006 a decision was taken to conduct interviews (by voice-assisted automated computer interview) continuously. This study therefore compared figures from interviews conducted in 2002 with ones conducted in 2006 to 2010. 

NSFG interviewed 4928 and 10403 men in 2002 and 2006 to 2010, respectively. Of these, 197 and 272 reported having a male sexual partner in the last year – 2.7 and 2.1% respectively (this difference was not statistically significant, p = 0.1).

The results

The mean number of male sexual partners MSM reported in the previous year fell significantly from 2.9 to 2.3 between the two surveys (p = 0.035) and was more marked in men under 24 years old (mean 2.9 to 2.1 partners, p = 0.027). The number of partners also fell in men aged 35 to 44 from 3.0 to 2.2, though this was not quite statistically significant (p = 0.07).

The fall in the number of partners was statistically significant in men with incomes under 150% of the US federal poverty level (3.0 to 2.1) and in men living in suburban metropolitan areas (3.2 to 2.1) but not in city-centre areas (2.6 in both surveys). There were declines in partner numbers in white (3.0 to 2.5) and black (2.4 to 1.9) men, though these did not reach statistical significance. In general though, there was a consistent picture of fewer partners among most groups.

There were no changes in condom use for anal sex. In 2002, 57% of men had not used a condom the last time they had sex and in 2006 to 2010 the proportion was 58%. In the minority of men who also had sex with women, the proportion who had not used a condom the last time they had vaginal sex was 46% in 2002 but had become 67% by 2006 tp 2010, and this difference was statistically significant (p = 0.04). However, the proportion of MSM who had had female partners also decreased from 38 to 25% (p = 0.03).

One other notable difference was that fewer men reported transactional sex (sex for money or drugs) in the last year (down from 15 to 3%) and fewer men said they had injected drugs or had had sex with someone who had injected drugs (from 12 to 5%).

HIV and STI testing in the last year did not increase. In 2002 and 2006 to 2010, 41% of men said they had had an HIV test in the last year and in the case of STI check-ups 38% reported having one in 2002 and 39% in 2006 to 2010. The proportion of men who had never had an HIV test, however, fell from 25 to 15%.

Conclusions and comments

The researchers comment on the fact that HIV prevalence and the incidence of STIs increased in gay men during a period when numbers of partners and some other sexual risk behaviours were falling. They note that there have been previous studies in Seattle and Peru where STI incidence and/or HIV diagnoses have remained high even though sexual risk indicators in gay men have fallen. Studies of young black gay men in the US, including one recently presented at the 20th Conference on Retroviruses and Opportunistic Infections (CROI), have consistently shown that they tend to have fewer partners despite considerably higher HIV incidence.

The researchers speculate that this may be due to ‘network factors’: factors about partners that are not captured by the individual risk behaviour focus of most studies. For instance, some studies have found that black gay men tend to restrict sex to partners of their own ethnicity and are also more likely to have sex with men a number of years older or younger than themselves. Both of these would tend to concentrate HIV infection within the black gay community.   

Whether these are the main drivers of US black men’s greater vulnerability to HIV infection, another interesting aspect of this study is that gay men appear to have taken steps that could reduce their HIV risk by using a method that has received little emphasis in HIV prevention programmes for gay men – reducing their number of partners – while not increasing condom use, which has received the most emphasis.


Leichliter JS et al. Temporal trends in sexual behaviour among men who have sex with men in the United States, 2002 to 2006-10. J Acquir Immun Defic Syndr, early online publication, DOI: 10.1097/QAI.0b013e31828e0cfc, 2013.