Using cutting-edge data analytical methods, Africa Health Research Institute (AHRI) scientists have developed an improved method to accurately predict where the highest rate of new HIV infections (HIV incidence) will likely occur in a community. This innovative set of measurements will help to target treatment and prevention resources and interventions where they are most needed.
In the paper, published in the scientific journal Science Translational Medicine, the researchers show that Population Viral Load (PVL) measures, which have previously been used to try to predict the intensity of new HIV infections in a particular area, do not accurately predict HIV incidence in a rural South African setting. They proposed and rigorously tested a set of modified indices, which considered variations in viral load patterns as well as the proportion of the population already infected with HIV, and found this gave a strong prediction of high incidence.
The researchers also warn that there is an urgent need to get men tested and onto treatment in South Africa.
Using population viral load (PVL) as a measure of HIV infection rates
As countries gear up to meet the UNAIDS’ 90-90-90 HIV treatment targets, accurately measuring progress and providing evidence of the impact of widespread treatment on life expectancy as well as the rate of new HIV infections is critical. One of the accepted ways of doing this is using PVL metrics. HIV viral load level – or the amount of virus present in a person’s semen or blood – is the single most important biological factor when it comes to HIV transmission. In short, the higher a person’s viral load, the more infectious they are. Based on this, calculating an average viral load for every HIV infected person living in a community over a given period could predict the potential for transmission (i.e. PVL) and thus the future rate of new HIV infections. The effectiveness of PVL measures has, however, previously been questioned – because it often draws only on data from healthcare facility records, ignoring members of a community who may be HIV infected but unaware of their status, or not on treatment. There are also concerns around the interpretation of some PVL metrics. Up until now, PVL predictions had not been compared against directly-measured HIV incidence in a rural South African community.
In this study AHRI researchers, together with colleagues from the University of KwaZulu-Natal (UKZN) and University College London (UCL), put PVL indices to the test. The strength of their study is the population-based data it used from a real-world setting in AHRI’s population intervention programme in rural KwaZulu-Natal. Researchers had data from all community members, including those who were HIV negative. They constructed viral load indices based on the viral load patterns in 2011, and then followed up with 8 732 HIV-uninfected research participants living in the programme area between 2011 and 2015.
Because data from clinics are typically used to construct community viral load measures, the researchers also compared their results from the population with clinic records to assess whether routinely collected facility-based data could be harnessed to accurately predict transmission potential.
Findings
Their results show that many of the PVL metrics currently used do not accurately predict HIV incidence in a particular community. Once they adjusted the metrics to take into account spatial variations in viral load patterns as well as proportion of HIV infections within a community the measures became highly predictive. The authors have also cautioned against using routine health facility based viral load data only to predict the transmission potential of similar rural settings, as they found this data doesn’t give an accurate prediction of HIV incidence.
They also found large differences in HIV viral load by sex at the population level. High viral loads (defined as over 50,000 copies/ml) were 40% more prevalent in men compared to women. This finding reflects observations throughout Africa showing that men are less likely than women to be tested for HIV and successfully link to HIV care, as well as being less likely to successfully adhere to treatment.
“In this era of decreasing HIV funding, it’s critical that we make the best use of the resources that are available. One way we can do this is to target areas of high transmission. Using these indices we can now identify vulnerable communities where future rates of new HIV infections will be highest – which will be useful in guiding prevention intervention strategies,” said AHRI Faculty Scientist Professor Frank Tanser, who led the study.
“It’s also critical that we find better ways of getting men onto treatment in order to reverse the HIV epidemic. One of the reasons that the rate of new infections is so high among women is because their male partners have high viral loads as a consequence of them not accessing HIV care and treatment. At AHRI we are looking at innovative solutions including financial incentives for men to test and treat, as well as gender sensitive mobile and technology options,” he added.
(Top image: An example of a dried blood spot sample (DBS), used for viral load measurements in the study. Photo by Ben Gilbert, Wellcome Trust)