How do teens born with HIV fare in different parts of the world?
Adolescents born with HIV in sub-Saharan Africa, Asia and Latin America started treatment years later and were several times more likely to die in early adolescence than those in Europe.
Source
The epidemiology of adolescents living with perinatally acquired HIV: A cross-region global cohort analysis
Study at a glance
- Design
- Cohort — Retrospective analysis of pooled individual data from 12 paediatric HIV cohort networks (CIPHER); competing-risks cumulative incidence and Cox models of mortality from age 10 to 15, with Europe as reference.
- N
- N=38187 · 38,187 adolescents with perinatally acquired HIV (in care before and after age 10) drawn from 183,119 children in the pooled dataset.
- Population
- Adolescents aged 10-19 with perinatally acquired HIV in 51 countries across Europe, North America, South America and the Caribbean, South and Southeast Asia, and sub-Saharan Africa.
- Outcome
- Characteristics at first visit, ART start, age 10 and last visit (age, CD4, height-for-age), and mortality, transfer out and loss to follow-up between ages 10 and 15.
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
The researchers combined records from 12 paediatric HIV research and care networks into one dataset and selected children who were in HIV care before age 10 and again after, as a proxy for infection at birth. They compared age at first care and treatment start, immune status (CD4) and growth across regions, country income groups and birth years. They then estimated the chance of dying, transferring to another clinic, or being lost to follow-up between ages 10 and 15, treating these as competing outcomes.
What they found
Of the 38,187 adolescents, 79% lived in sub-Saharan Africa. Median age at starting antiretroviral therapy was 0.9 years in North America but 7.9 years in sub-Saharan Africa, and many adolescents outside Europe and North America remained stunted. By age 15, cumulative mortality was 2.6% overall, from 0.8% in Europe to 4.4% in South America and the Caribbean; compared with Europe, the unadjusted mortality hazard ratio was 4.35 in sub-Saharan Africa and 6.07 in South America and the Caribbean. Children born in 2000-2005 had lower mortality than earlier birth cohorts, but loss to follow-up more than doubled.
The limits
What it doesn't show
Deaths among adolescents lost to follow-up were probably missed, especially in sub-Saharan Africa where loss to follow-up was highest, so regional mortality differences are uncertain; sensitivity analyses showed estimates could shift greatly depending on assumptions. Perinatal infection was inferred from entering care before age 10 rather than recorded, Nigeria was not included, and North America was represented only by the US. The data come from clinics that may offer better care than average, and about 44% of cohorts only enrolled children already on ART, so treatment coverage and survival are likely overestimated.
Key terms
- Perinatally acquired HIV
- HIV transmitted from mother to child during pregnancy, birth or breastfeeding.
- Antiretroviral therapy (ART)
- A combination of drugs that suppresses HIV replication, allowing the immune system to recover and preventing progression to AIDS.
- Loss to follow-up (LTFU)
- When a patient stops attending care and their outcome is unknown; here, no visit for more than a year.
- Competing risks
- Outcomes that prevent another outcome from being observed, such as transfer or loss to follow-up preventing observation of death at that clinic.
- Height-for-age Z-score (HAZ)
- How a child's height compares with WHO reference children of the same age and sex; below -2 indicates stunting.
- Hazard ratio
- The ratio of the instantaneous rate of an event in one group relative to a reference group over follow-up.
Flashcards
0 of 11 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
Which region contributed most of the adolescents in this cohort?
Common questions
Why does loss to follow-up matter so much for mortality estimates?
If adolescents who stop coming to clinic have actually died, those deaths are never recorded. In regions with high loss to follow-up, recorded mortality may be far too low; the authors showed that plausible assumptions could raise estimated mortality in sub-Saharan Africa dramatically.
How did the researchers decide who had perinatally acquired HIV?
Transmission route is rarely recorded, so they used a proxy: children in HIV care before age 10 and still in care after age 10, excluding known non-vertical infections. This may miss children diagnosed after age 10.
Is outcome getting better over time?
Partly. Children born in 2000-2005 started treatment younger with better CD4 counts and had lower mortality, but they were still often stunted and were more often lost to follow-up.
More on Infectious disease