"The number of gay and bisexual men contracting HIV rose…because of an increase in numbers having unprotected sex," The Guardian reports.
The story is based on a study that used UK data on HIV and sexual risk behaviour among men who have sex with men (MSM). The data was used to build computer models to estimate the impact of various factors on HIV rates since the 1980s.
Although this model cannot predict all the factors that play a role in HIV incidence among MSM, it is a useful tool for policy makers to assess which preventative strategies work and which could potentially have the greatest impact.
This research highlights the important role condom use has played in reducing HIV rates. It is hoped it will encourage MSM to have regular HIV tests and continue to use condoms to protect themselves and others from HIV.
Where did the story come from?
The study was carried out by researchers from University College London, the Health Protection Agency (HPA) and other research centres in the UK and Denmark, and was funded by the UK National Institute for Health Research (NIHR).
It was published in the peer-reviewed open access journal, PLoS ONE.
The Independent, BBC and The Guardian covered this story well.
What kind of research was this?
This research was a study modelling HIV incidence in the UK in men who have sex with men (MSM). Modelling studies are useful for looking at how different factors could influence patterns of disease, for example. They can also help policy makers make decisions about how to use resources to their greatest effect. Models work on the basis of various assumptions, and their accuracy depends on how precise these assumptions are.
The researchers wanted to understand which specific factors affect HIV incidence so prevention efforts can be improved. They say that although levels of antiretroviral therapy (ART) have increased among MSM with HIV, the number of new HIV infections has not decreased. For example, in 2010 more than 3,000 MSM were diagnosed with HIV, which is reported to be the highest number since the start of the HIV epidemic in the late 1970s and early 1980s.
What did the research involve?
The researchers looked at ART use, unprotected sex and HIV testing, and how these impacted HIV trends in MSM over the past 30 years in the UK.
They used comprehensive HIV data routinely collected from the UK ("surveillance data"), data on self-reported condom use among MSM, and other information so they could build complex computer models to simulate the following:
- sexual risk behaviour
- HIV transmission
- HIV progression (the extent to which the infection has damaged the immune system)
- the effect of ART in MSM on HIV incidence in the UK from 1980-2010
The researchers made various assumptions, including that all transmissions took place through unsafe (condomless) sex, and that after HIV diagnosis a proportion of men substantially reduce unsafe sex with short-term partners.
For each factor they put into the model, they ran the model with a range of possible values. They then looked at which combination of values resulted in a model that best fit what was actually seen in the UK population between 1980 and 2010.
The researchers also investigated hypothetical scenarios, such as what would have happened to HIV incidence if ART had never been introduced.
What were the basic results?
The researchers found that they could generate a model that was generally consistent with the trends seen in HIV in the UK between 1980 and 2010.
The main findings of the model were:
- The model suggested that after high HIV incidence in the early 1980s, there was a decline in sexual risk behaviour and a resulting reduction in HIV incidence.
- The model only matched the data if sexual risk behaviour increased after the introduction of effective ART, from an estimated 35% of men having unsafe anal sex with a partner of unknown or negative HIV status in the past year, to 44% in 2010. This represented an absolute increase of 9%, or a 26% relative increase. This was associated with an increase in the incidence of HIV, from an average of three new cases per 1,000 people per year in 1990-97 to about 4.5 new cases per 1,000 people per year in 1998-2010. Men with undiagnosed HIV were the main sources of new infections, with a smaller proportion from men who had been diagnosed but were not receiving ART. The smallest proportion was from men who had been diagnosed with HIV and received ART.
- The researchers found that if ART had never been introduced, HIV incidence would be higher (a 68% increase on 2006-10 figures).
- If all condom use ceased, HIV incidence would have been 424% higher.
- If ART had been provided to all people diagnosed with HIV from 2001 onwards, there would have been a 32% lower incidence of HIV between 2006 and 2010.
- Rates would also have been reduced if there had been more testing (targeting men who have had unprotected sex in the past three months), as more men could be diagnosed and treated with ART.
- If 68% of men had been tested each year by 2010, compared with the 25% observed, the incidence of HIV would have been 25% lower.
- If there were higher testing rates and ART was started at diagnosis, incidence would have been reduced by 62%.
How did the researchers interpret the results?
The researchers conclude that ART has almost certainly reduced the incidence of HIV in the UK between 1980 and 2010.
They say that their findings suggest that a modest increase in condomless sex among MSM after the introduction of ART is responsible for the net increase in incidence of HIV in the UK, so increased condom use should be encouraged.
The model also suggests that much higher rates of HIV testing, along with starting ART at the time of diagnosis, would be likely to lead to considerable reductions in HIV incidence.
The study has identified two main factors that have impacted HIV rates among MSM in the UK – condom use and antiretroviral therapy (ART).
It also found that if there was more HIV testing and ART was started immediately after diagnosis, the rates of the disease could be reduced further.
Other studies, such as surveys of MSM, have also found an increase in condomless sex after the introduction of effective ART.
In the UK, ART is usually only started once a person's CD4 cell count (a measure of immune function) falls to below 350 cells/mm3. The authors note that randomised controlled trials (RCTs) have not yet reliably assessed the balance of the benefits and risks of starting ART soon after diagnosis, but trials are ongoing.
The researchers state that their modelling of the impact of immediate treatment with ART after HIV diagnosis (without waiting for white blood cell count to drop to a specified level) presumes that this would not lead to an increase in condomless sex. They suggest that a negative impact on condom use seems unlikely.
This model benefits from having a large amount of UK data available on HIV-related trends. But, as with all modelling studies, it is impossible to take all potential factors into account. For example, the current model did not factor in the risks associated with unprotected oral sex. The assumptions the model is based on may also not happen in the real world, which affects how seriously we can take the predictions the model makes.
But these kinds of models are helpful for policy makers to assess what the effects of preventative strategies are likely to have been, and which strategies may potentially have the greatest impact.
For the individual, this research highlights that our most effective weapon against HIV is a cheap piece of latex – the (not-so) humble condom. As well as providing effective protection against HIV, when used correctly the condom can also protect against other STIs, such as gonorrhoea and chlamydia.
It is hoped that this study will encourage people at risk of HIV – especially men who have sex with men – to continue to use condoms to protect themselves and others from HIV, as well as having regular HIV testing.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
The Guardian, 16 February 2013
BBC News, 16 February 2013
Daily Mail, 15 February 2013
The Independent, 16 February 2013
Links to the science
PLoS One. Published online February 15 2013