Gay men’s HIV risk varies hugely between individuals and over time, cohort study finds

Implications for PrEP programming
This article is more than 10 years old. Click here for more recent articles on this topic

A study that looked at the way risk of HIV transmission changed over time in a group of gay men during a six- to eight-year period has found that there was vast variation in the degree of risk men subjected themselves to, the length of time they were at risk and, as a result, HIV incidence.

The researchers analysed the number of times cohort members took sexual risks over the study period (by allocating a "risk score" to each six-month period) and found that men's risk scores tended to be consistent, and to fall into three different groups.  It found that one-in-seven men belonged to a very high risk group, a third of whom became infected with HIV over the study period. Just under a quarter belonged to a moderate risk group, of whom 10% became HIV positive.

The other two-thirds were at low risk of HIV, except for short periods; 3% of them acquired HIV. Being in the one-third of the cohort that did take more risks was associated with being white, having a high income, and being younger; in addition, being in the most at-risk one-seventh of the group was associated with depression and taking recreational drugs.


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.


A mental health problem causing long-lasting low mood that interferes with everyday life.


Cost-effectiveness analyses compare the financial cost of providing health interventions with their health benefit in order to assess whether interventions provide value for money. As well as the cost of providing medical care now, analyses may take into account savings on future health spending (because a person’s health has improved) and the economic contribution a healthy person could make to society.

longitudinal study

A study in which information is collected on people over several weeks, months or years. People may be followed forward in time (a prospective study), or information may be collected on past events (a retrospective study).


Receptive anal intercourse refers to the act of being penetrated during anal intercourse. The receptive partner is the ‘bottom’.

The authors specifically did this cohort analysis because they wanted better information that could help in the targeting of pre-exposure prophylaxis (PrEP) at the right groups: one of the reasons this prevention method has taken off slowly in the US and not yet received approval elsewhere is concern about its cost. Cost-effectiveness studies suggest that PrEP will only be economical if taken by people with the highest risk of HIV infection (see this report for one example).

It is, however, of broader interest, as the first-ever study to demonstrate a relationship between specific characteristics and what the authors call “risk trajectories” – longitudinal patterns of risk over time.

The study

The group selected for study was a subgroup of the US Multicentre AIDS Cohort Study (MACS), which is described in this recent report on life expectancy.

For the purposes of this study, the researchers took a look at members of the study who were HIV negative and recruited between 2001 and 2003, and included every man who had been seen between January 2003 and September 2004 and had made at least one other study visit up to September 2011. The length of “trajectory time” over which men were studied ranged from six years, three months to eight years, nine months.

At each visit, men answered a questionnaire about their sexual risk behaviour during the previous six months and their behaviour was then given a "sexual risk score" according to the most HIV-risky sexual activity the person had taken part in during that time. These ranged from zero points for no anal sex at all to six points for unprotected receptive anal intercourse with at least one HIV-positive or unknown-status partner. The degree of risk was broadly in accordance with a previous study on how risky specific activities were in gay men (see this report). Partners' viral load, if known, was not a category captured in this study.

There were 419 men in the study group, which was racially diverse (38% white, 42% black) and with a median age of 38 (20% under 30). On average, 11% of the group reported having had the highest-risk category of sex since their previous study visit.

One feature of this study group is that a much higher proportion were reporting not having anal sex at all than in many other studies: 43% reported they had not had anal sex in the last six months on their initial visit, and this rose to 56% at the last visit. This may be due to the ageing of the cohort but may also be due to higher-risk men dropping out of the study (people who were young, on low income or black were more likely to drop out).

As noted above, the participants fell into three risk groups according to their average sexual risk score, and how consistent it was over time: 63% low risk, 23% moderate risk and 14% high risk. People generally remained in the same risk group. The likelihood of someone in the low-risk group having an episode of high-risk behaviour over the following six months was only 0.9%. In comparison, 29% of the moderate-risk group were likely to have a high-risk episode in the following six months (this likelihood declined to 17% over the course of the study) and 71% of the high-risk group. This means that in the high-risk group, if they had high-risk sex in one six-month period, only 29% did not have high-risk sex in the following six-month period. There was no-one in the lowest-risk group who had high-risk behaviour for two consecutive six-month periods: conversely, 48% of men in the moderate-risk and 93% of men in the high-risk group had high-risk behaviour for at least two consecutive six-month intervals over the study.

The average length of time over which people stayed in the high-risk category was one year in the moderate-risk group and two years in the high-risk group; in other words, even members of the high-risk group did not stay high risk all the time; they went in and out of being at risk of HIV as their relationship status changed or as they adopted different sexual practices.

This is best illustrated by a selection of 15 different members of the cohort, five from each of three risk categories, where periods of high-risk sex are red, periods of low risk or no sex are blue, and missing visits are blank:

The factor most strongly associated with being in the high-risk group was having an annual income over US$20,000 – people in the higher-income category were nearly five times more likely to be in the highest risk group than others – higher income was also associated with being in the moderate-risk group, though not as strongly. The factor most strongly associated with being in either the moderate- or the high-risk group (i.e. not in the low-risk group) was being white as opposed to being black or Hispanic. White men were 3.9 times more likely to be in the moderate-risk and 3.7 times more likely to be in the high-risk group.

Youth was also strongly associated with being in the moderate- or high-risk groups, with a 6% decline in the risk of being in the moderate group for every year older, and an 8% decline in the likelihood of being in the high-risk group.

Depression and substance use were associated with being in the high-risk group (2.4 and 2.0 times more likely respectively), but not with being in the moderate-risk group.

The biggest limitation of this study is that members of the MACS study are likely to be highly motivated gay men who may take more account of sexual risk than others: as already noted, an untypically large proportion did not have anal sex. This may mean that in more typical gay populations, high-risk behaviour is more common.

Nonetheless, this is the first study ever to document gay men’s “risk careers” over a long period of time and in such detail. Its findings, as the researchers say, “could enable clinicians to efficiently screen and identify [gay men] who exhibit ‘seasons of risk’ for potential PrEP use”.


Pines HA et al. Sexual risk trajectories among MSM in the United States: implications for pre-exposure prophylaxis delivery. JAIDS, published ahead of print, December 2013.

This news report is also available in Russian.