Targeting HIV "Hotspots" May Not Be Effective for Ending the Epidemic

HIV prevention investigators sometimes identify particular regions with a high density of HIV prevalence as “hotspots.” These hotspots are often believed to be disproportionately involved in new HIV transmission compared with other areas and are subjected to targeted interventions.

Investigators of a new study, conducted in Uganda and published in The Lancet HIV, cast doubt on many assumptions around HIV hotspots. Their research shows that some hotspots transfer very few infections to nearby communities and instead receive more infections from outside.

The investigators studied fishing communities on the shores of Lake Victoria, Uganda, where around 40% of the population are living with HIV. This area has some of the highest HIV prevalence in the world. Yet the investigators were surprised to find that more HIV infections were traceable to inland communities with lower HIV prevalence than the hotspot fishing communities.

"Our finding shows that HIV disease dynamics are not as obvious as they may seem, and advises caution against equating and stigmatizing hotspots as population groups that drive HIV spread in Africa," Oliver Ratmann, PhD, a study author and lecturer at the Imperial College London Department of Mathematics, said in a press release.

Data were collected from a population-based survey of all individuals aged 15-49 years in Rakai, Uganda, which took place from August 2010 to January 2015.

The investigators deep-sequenced viral RNA from participants with HIV who did not have prior use of antiretroviral therapy to map the chain of infection events.

A total of 25,882 individuals participated in the population survey, an estimated 75.7% of the lakeside population and 16.2% of the inland population in the Rakai region. In total, 5142 participants were found to be HIV positive, 2703 occurred in inland settings and 2439 occurred in fishing communities.

Of those who were found HIV positive, 3878 did not report antiretroviral therapy treatment, and 2652 of those participants had deep-sequencing results sufficient for phylogenetic analysis.

Ultimately, 446 transmission networks were reconstructed, with 293 linked pairs featuring the inferred direction of transmission.

“Adjusting for incomplete sampling, an estimated 5.7% (95% CI, 4.4—7.3) of transmissions occurred within lakeside areas, 89.2% (95% CI, 86.0–91.8) within inland areas, 1.3% (95% CI, 0.6–2.6) from lakeside to inland areas, and 3.7% (95% CI, 2.3–5.8) from inland to lakeside areas,” study authors wrote.

Only 5.4% of infections occurred between the Lake Victoria fishing community hotspots and inland communities. Among these infections, almost 3 times as many occurred as transmissions from inland communities into hotspot communities.

The investigators speculated that the results may be rooted in migration dynamics which are specific to local economies and social context.

In order to see how generalizable study results are, the investigators are repeating their research in other Lake Victoria communities. However, regardless of those results, this study reveals that particular social contexts may produce unexpected dynamics in terms of disease transmission and so may require individuated public health approaches.

“This result suggests that targeted interventions to these hotspots will not alone control the epidemic in inland populations, where most transmissions occur, the authors concluded. “Thus, geographical targeting of high prevalence areas might not be effective for broader epidemic control depending on underlying epidemic dynamics.”