‘We have come a long way’: Empirical data support UNAIDS HIV incidence estimates
Key takeaways:
- Empirical incidence data support UNAIDS estimates that HIV incidence has significantly declined over the year.
- Data also showed HIV incidence has declined 3% faster among young adults.
SAN FRANCISCO — HIV incidence in sub-Saharan Africa has declined substantially, according to both UNAIDS estimates and empirical data pulled from decades of studies, researchers found.
“We estimate HIV incidence using prevalence and ART coverage, and we have a lot of that data over time from a lot of sites,” Oliver Stevens, technical analyst at the Imperial College London, told Healio. “Prevalence is a measure of how much HIV an area has at a given time. If you have two of those measures over time, you should be able to back calculate what incidence was between those two time points. That's sort of like the core idea.”
Actual data on incidence, he said, however, is “really hard to come by.”
“The experiments are really expensive to run, and they're incredibly manpower intensive,” he said.
According to the study, UNAIDS estimates — which have been derived by fitting mathematical models to national surveillance data — show that HIV incidence in sub-Saharan African has significantly declined since 2000.
Stevens and colleagues aimed to assess whether the level and age distribution of HIV incidence from mathematical-model derived estimates are consistent with empirical HIV incidence observations. To do so, the team matched empirical HIV incidence observations from cohort studies, intervention trials and household surveys in sub-Saharan Africa all conducted between 1990 and 2022 to UNAIDS HIV incidence estimates.
Data showed that HIV incidence among adults aged 15 to 49 declined by 75% to 90% between 2010 and 2023.
Stevens added, though, that there are still certain populations that experience a disproportionately high incidence of HIV — including people participating in intervention trials, pregnant and postpartum women and people participating in other studies with inclusion criteria.
Study data showed that studies among pregnant women (incidence rate ratio [IRR] = 2.83) and control arms of intervention trials (IRR = 2.54), for example, had significantly higher incidence than modeled incidence estimates (IRR = 2.83).
“That's important because it means the study is no longer representative of everyone who lives in that area,” he said.
Additionally, Stevens said that aging patterns observed in cohort studies showed that even though there have been “really steep declines” in incidence among all adults, those declines are not being experienced equally across age groups. According to the study, incidence declined 3% faster per year among young adults aged 15 to 24 years vs. those aged 25 years and older (95% CI, 2%-4%).
“Incidence in the younger age group [of] adults under 25 is falling even faster,” he said. “What that means is that the average age of HIV incidence is moving older, so you're becoming increasingly older when you get infected.”
Based on all of the study findings though, Stevens said that the empirical incidence data do support the reported declines in HIV incidence from UNAIDS.
“We have come incredibly far,” he said. “Obviously, we're talking a lot about funding at the moment and it's a very charged issue. However, with all the investment that has been obtained over the last 10, 15, 20 years, these incidence declines are enormous.”
Collapse
Stevens O, et al. Abstract 126. Presented at: Conference on Retroviruses and Opportunistic Infections; March 9-12, 2025; San Francisco.
Read more about