Age and gender profiles of HIV infection burden and viraemia: novel metrics for HIV epidemic control in African populations with high antiretroviral therapy coverage

Andrea Brizzi, Joseph Kagaayi, Robert Ssekubugu, Lucie Abeler‐Dörner, Alexandra Blenkinsop, David Bonsall, Larry W. Chang, Christophe Fraser, Ronald M. Galiwango, Godfrey Kigozi, Imogen Kyle, Mélodie Monod, Gertrude Nakigozi, Fred Nalugoda, Joseph G. Rosen, Oliver Laeyendecker, Thomas C. Quinn, M. Kate Grabowski, Steven J. Reynolds, Oliver Ratmann

TL;DR

While global HIV targets (the "95-95-95" goals) focus on the percentage of people treated within specific groups, this study argues that these metrics can be misleading. By analyzing longitudinal data from Uganda, researchers show that the groups furthest from clinical targets are often not the ones driving the total "burden of viraemia" in a population; instead, demographic "heavy hitters"—like women aged 25-29 and men aged 30-34—contribute most to the pool of transmissible virus, even when they are close to meeting global treatment goals.

What Problem Does This Paper Address?

The international community measures progress against HIV using the UNAIDS "95-95-95" targets: 95% of people living with HIV (PLHIV) should know their status, 95% of those should be on antiretroviral therapy (ART), and 95% of those on ART should achieve viral suppression.

However, these are relative metrics. They highlight inequalities in healthcare access but ignore the absolute number of people with detectable virus (viraemia) in a community. If a small demographic group has poor viral suppression, they might appear as a high priority on a 95-95-95 dashboard, even if they represent only a tiny fraction of the total viraemic population. Conversely, a large demographic group might meet the 95-95-95 targets but still harbor the majority of the community's viraemia simply due to their sheer size. This paper proposes "burden profiles" as a more effective metric for directing public health interventions to where they can most effectively reduce transmission.

Methods

The researchers utilized longitudinal data from the Rakai Community Cohort Study (RCCS) in Uganda, spanning 2013 to 2019. This period coincided with the massive scale-up of "Universal Test and Treat" (UTT) programs.

Key methodological features include:

  • Granular Analysis: Unlike standard reports that use 5-year age blocks, this study used 1-year age bands to identify precise peaks in infection and viraemia.
  • Bayesian Modeling: They employed non-parametric Bayesian models to estimate HIV seroprevalence and population-level viral suppression, accounting for non-participation by using data from first-time participants as a proxy.
  • Geographic Comparison: The study compared "inland" agrarian communities (typical population pyramids) with "fishing" communities (atypical, male-dominated, hyper-endemic areas).
  • Standardization: All estimates were standardized to census data to ensure the results reflected the actual underlying population structure.

Datasets

The study relied on the Rakai Community Cohort Study (RCCS):

  • Sample Size: 37,743 unique individuals participated at least once.
  • Scope: 40 communities (36 inland, 4 fishing) in south-central Uganda.
  • Data Points: 13,844 viral load measurements from 6,677 participants living with HIV.
  • Demographics: Individuals aged 15–49 years.

Key Insights and Ideas

  • The 95-95-95 Paradox: In inland communities, young people (aged 15-24) were the furthest from achieving viral suppression targets. However, because their absolute numbers of PLHIV are relatively low, they only accounted for about 20% of the total viraemic burden.
  • Demographic Dominance: The "burden of viraemia" is concentrated in slightly older groups. Women aged 25-29 and men aged 30-34 contributed the most to the total number of viraemic individuals in 2019, despite these groups being very close to (or having already achieved) the 95-95-95 targets.
  • Masking Effects: Standard 5-year age banding can mask the actual age groups where viraemia peaks. For example, in inland communities, viraemia peaked specifically in men aged 23-27.
  • Community Differences: In fishing communities, viraemia was 2.5 times higher than in inland areas. While suppression rates improved faster in fishing communities, the high baseline prevalence meant they still carried a disproportionate burden of the regional epidemic.

Key Results

  • Viraemia Decline: Between 2013 and 2019, the proportion of the total population with viraemia dropped from 4.9% to 1.9% in inland communities and from 19.1% to 4.7% in fishing communities.
  • Gender Disparity: In 2019, HIV seroprevalence was significantly higher in women than men (1.78-fold higher in inland areas; 1.53-fold in fishing areas).
  • The "Who" of Viraemia: In inland communities, women aged 20-29 and men aged 25-34 each contributed roughly 10% to the total viraemic population.
  • Target Achievement: By 2019, fishing communities had surpassed the 95-95-95 suppression thresholds, yet they still maintained a higher absolute burden of viraemia than inland communities that had not yet met the targets.

Limitations and Future Work

  • Age Constraints: The RCCS only tracks individuals aged 15–49. The study could not account for the growing burden among those over 50 or children under 15.
  • Viraemia vs. Transmission: While viraemia is a strong predictor of transmission, it is not a 1:1 map. Behavioral factors (number of partners, etc.) also dictate transmission risk, which this specific metric doesn't capture.
  • Generalizability: While the patterns are typical for Eastern and Southern Africa, the data is specific to Uganda.
  • Future Direction: The authors suggest that future surveillance must prioritize population censuses alongside clinical data to calculate these "burden profiles" and better direct "Test and Treat" resources.

Summary updated

Abstract

The 95-95-95 targets provide a useful benchmark for monitoring progress towards HIV epidemic control, but do not contextualize underlying population structures and so may direct interventions towards groups that represent a marginal fraction of the population with viraemia.
BibTeX
@article{brizzi2025age,
  author = {Andrea Brizzi and Joseph Kagaayi and Robert Ssekubugu and Lucie Abeler‐Dörner and Alexandra Blenkinsop and David Bonsall and Larry W. Chang and Christophe Fraser and Ronald M. Galiwango and Godfrey Kigozi and Imogen Kyle and Mélodie Monod and Gertrude Nakigozi and Fred Nalugoda and Joseph G. Rosen and Oliver Laeyendecker and Thomas C. Quinn and M. Kate Grabowski and Steven J. Reynolds and Oliver Ratmann and Andrea Brizzi and Joseph Kagaayi and Robert Ssekubugu and Lucie Abeler‐Dörner and Alexandra Blenkinsop and David Bonsall and Larry W. Chang and Christophe Fraser and Ronald M. Galiwango and Godfrey Kigozi and Imogen Kyle and Mélodie Monod and Gertrude Nakigozi and Fred Nalugoda and Joseph G. Rosen and Oliver Laeyendecker and Thomas C. Quinn and M. Kate Grabowski and Steven J. Reynolds and Oliver Ratmann and Oliver Ratmann},
  title = {{Age and gender profiles of HIV infection burden and viraemia: novel metrics for HIV epidemic control in African populations with high antiretroviral therapy coverage}},
  journal = {International Journal of Epidemiology},
  year = {2025},
  doi = {10.1093/ije/dyaf126},
}

Loading related content...

Figure