Less HAART use, more unsafe sex, could mean 50% leap in new HIV cases in Aussie gay men

This article is more than 17 years old. Click here for more recent articles on this topic

Widespread use of HAART by HIV-positive gay men in Australia led to a reduction in new HIV infections between 1996 and 1998, conclude Australian investigators using a mathematical model. However, in an article published in the April 1st edition of the Journal of Acquired Immune Deficiency Syndromes, they argue that a year-on-year increase in the amount of unprotected sex amongst gay men, coupled with reduced use of anti-HIV therapy, could mean that annual HIV incidence increases by 50% by 2006, raising the prospect of a resurgent HIV epidemic amongst Australian gay men in the next few years. The investigators also conclude that their data show that the use of HAART alone is insufficient to eradicate HIV transmission.

Earlier studies have suggested that since the introduction of HAART there has been an increase in rates of risky sex between gay men. HIV treatments were rapidly taken up by HIV-positive gay men in Australia, and in 1996 - 97, 70% of HIV-positive gay men were taking HAART. However, concern about side-effects and the long-term tolerability of HAART has meant that an increasing number of men are either taking structured breaks from HAART, or delaying the initiation of HIV therapy, meaning that the proportion of HIV-positive gay men receiving treatment has declined from the 1996-97 high.

Investigators developed a mathematical model of HIV transmission amongst Australian gay men between 1995 and 2006. The model included an estimate of the percentage of HIV-positive gay men who were aware of their infection, the amount of unprotected anal sex gay men were engaging in, and the proportion of HIV-positive men taking anti-HIV treatment.



The total elimination of a pathogen, such as a virus, from the body. Eradication can also refer to the complete elimination of a disease from the world.

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.


In everyday language, a general movement upwards or downwards (e.g. every year there are more HIV infections). When discussing statistics, a trend often describes an apparent difference between results that is not statistically significant. 


Term used to indicate how well a particular drug is tolerated when taken by people at the usual dosage. Good tolerability means that drug side-effects do not cause people to stop using the drug.

Using their model, the investigators wished to answer two questions. First, what were the competing effects of widespread use of HAART but increasing rates of unprotected sex, and second, what trends could be predicted if HAART use fell but unsafe sex increased.

The investigators calculated that there was a slow decline in new HIV infections in 1995 and 1996, and then a steeper fall between 1996 and 1998 as the use of HAART became widespread. However, new HIV infections started to increase slowly between 1998 and 2001 as the amount of unprotected anal sex amongst gay men increased.

Four scenarios were modelled between 2001 and 2006. The first scenario envisaged stable use of HAART and stable rates of unprotected sex. In this scenario, annual HIV incidence amongst gay men would decline by 17% from 2001 levels.

Scenario two envisaged a decline in HAART usage to 50% of gay men but stable rates of unprotected anal sex. This would result in HIV incidence at best remaining stable at the 2001 level, or at worst increasing by 5%.

However, scenario three modelled for a 10% per year increase in the amount of unprotected anal sex amongst gay men, and even if HAART usage by HIV-positive gay men remained stable at 70%, this would produce a 22% increase in new HIV infections.

Furthermore, if only 50% of gay men with HIV were taking HAART and rates of risky sex increased by 10% a year, new HIV cases would increase by 53% per year on 2001 levels.

“Our model suggests that the widespread use of effective [antiretroviral therapy] led to a decline in HIV incidence among homosexual men in Australia between 1996 and 1998,” comment the investigators. However, even a modest reduction in the proportion of gay men taking HAART, accompanied by an increase in the rate of unprotected anal sex, was enough to reverse this trend. They caution that their model is very much a worst-case scenario and does not take account of harm minimisation behaviour such as negotiating unprotected sex between men of the same HIV status.

The investigators also suggest that their model shows that the use of antiretroviral drugs by large numbers of people with HIV will not be sufficient to eradicate HIV transmission. They note that HIV testing levels are very high amongst Australian gay men, and that between 80% - 90% of gay men with HIV know of their infection. Further, there is a high level of HIV knowledge amongst Australian gay men and they have access to free anti-HIV treatment and care.

“We believe that this combination of circumstances is such that if [antiretroviral therapy] could eradicate HIV transmission, then it is more than likely to be observed in the male homosexual population in Australia than any other population,” comment the investigators. However, “eradication of HIV is not occurring among homosexual men in Australia, and… HIV incidence rates may actually be increasing.”

In conclusion, the investigators caution that although HAART led to a short-term reduction in HIV incidence between 1996 and 1998, increased rates of unprotected sex and decreased use of HAART, in line with treatment guidelines, “suggest that Australia may face a resurgent HIV epidemic among homosexual men in the next few years.”

Further information on this website

New HIV infections steady in UK gay men despite introduction of HAART - news story


Clements MS et al. Modeling trends in HIV incidence among homosexual men in Australia 1995 – 2006. JAIDS 35: 401 – 406, 2004.