One in five children on ART still lost to follow-up, while disparities between low- and middle-income countries in sub-Saharan Africa persist

Image by Elizabeth Glaser Pediatric AIDS Foundation/DFID - UK Department for International Development. Creative Commons licence.

The proportion of children under three months of age starting antiretroviral therapy (ART) doubled from 5% to 10% between 2006 and 2017, while death rates were halved, according to findings from over 32,000 children living with HIV from 16 countries in sub-Saharan Africa. However, one in five children were still lost to follow-up and the study highlighted persisting disparities between low- and middle-income countries in sub-Saharan Africa.

Dr Victoria Iyun and colleagues from the International epidemiology Databases to Evaluate AIDS (leDEA) collaboration report these findings in The Lancet HIV, on differences in outcomes across time periods in this multi-regional cohort of children who started ART before the age of five.

Drs Joseph Rujumba and Grace Ndeezi from Makerere University, Uganda, note in an accompanying comment, that the decline in mortality “reaffirms the value of ART scale-up on child survival…and should be a motivator for countries in sub-Saharan Africa to increase universal ART programmes for children further, to realise an AIDS-free generation.”


The World Health Organization recommended universal ART for children under two in 2010 and children under five in 2013. Considerable progress has been made since then. However, scale-up of routine paediatric care programmes continues to fall far behind adult HIV care in the region. Barriers include delayed diagnosis, poor linkage to care, late start of ART, little access to free HIV services and poor retention in care.


The authors used data from IeDEA to describe chronological trends in characteristics of children under five when starting ART, and to compare outcomes over time and across country income groups.

They undertook a retrospective cohort analysis of data from 45 paediatric sites in 16 low-, lower-middle- and upper-middle-income countries in sub-Saharan Africa. Countries were Benin, Burundi, Côte d’Ivoire, Democratic Republic of the Congo, Ghana, Kenya, Lesotho, Malawi, Mali, Mozambique, Rwanda, South Africa, Togo, Uganda, Zambia and Zimbabwe.

Outcomes were trends in characteristics at ART start (age, weight, height and CD4%), death within 24 months of starting ART and loss to follow-up.


Of the 32,221 children living with HIV and starting ART before the age of five between 2006 and 2017, approximately half were female and over 75% came from the IeDEA southern region. The majority (75%) were enrolled from urban or semi-urban sites.

Median age at the start of ART remained stable at 20.4 months in 2006-10 and 19.2 months for 2014-17. The median age was lowest in upper-middle-income countries at 13.2 months, compared to 24.2 months in low-income countries and 22.6 months in lower-middle-income countries.

Nonetheless, the proportion starting ART under three months of age increased from 5.1% to 10.0% over the same time period. Starting ART under three months of age is considered a proxy for early infant diagnosis.

Most infants starting ART under three months of age were in South Africa, the only upper-middle-income country included, where over 80% uptake of early infant diagnosis and universal treatment has been reached, the authors say.

Even in the most recent time period, approximately half of all children were immunosuppressed, underweight or had stunted growth when starting ART. Proportions were similar across countries in all income groups.

"Attention should be focused on children who have fallen through gaps in care."

Severe immune suppression in children decreased over time, but still remained very high: 75% in 2006-10 and 55% in 2014-17. The proportion of children severely underweight decreased from 33% to 24%, but the proportion with stunting only declined from 44% to 38%. 

Death rates at 24 months on ART decreased from 6.5% in 2006-10 to 2.9% in 2014-17. Overall, 23% were lost to follow-up at 24 months, with no change seen over time. If more than half of children lost to follow-up died, the improvement in mortality would be lost.

The risk of death varied according to country income group. Assuming all loss to follow-up was due to death, children in upper-middle-income countries had the lowest risk of death relative to low-income countries (aHR: 0.64, 95% CI: 0.59-0.68). Children in lower-middle-income countries had a greater risk of death than those in low-income countries (aHR: 1.31, 95% CI: 1.24-1.38).

Strengthening scale-up 

Rujumba and Ndeezi note these findings highlight serious gaps in scale-up for children living with HIV that need urgent attention: the high proportion who had severe immune suppression were underweight or had stunted growth before starting ART; and the persistence of loss to follow-up. These are known risk factors for high mortality in children supporting the need for strengthening early diagnosis and treatment before advanced disease begins.


middle income countries

The World Bank classifies countries according to their income: low, lower-middle, upper-middle and high. There are around 50 lower-middle income countries (mostly in Africa and Asia) and around 60 upper-middle income countries (in Africa, Eastern Europe, Asia, Latin America and the Caribbean).

retention in care

A patient’s regular and ongoing engagement with medical care at a health care facility. 

low income countries

The World Bank classifies countries according to their income: low, lower-middle, upper-middle and high. While the majority of the approximately 30 countries that are ranked as low income are in sub-Saharan Africa, many African countries including Kenya, Nigeria, South Africa and Zambia are in the middle-income brackets. 

linkage to care

Refers to an individual’s entry into specialist HIV care after being diagnosed with HIV. 

loss to follow up

In a research study, participants who drop out before the end of the study. In routine clinical care, patients who do not attend medical appointments and who cannot be contacted.

The researchers say that attention should be focused on children who have fallen through gaps in care because of late diagnosis, delayed turnaround of results and poor engagement in care. They suggest testing infants in immunisation or outpatient clinics whose mothers may have missed prevention of mother-to-child transmission.

Rujumba and Ndeezi support a holistic approach with particular attention to improving linkage to care and tracing of caregivers to improve their engagement in care, notably in the poorest countries. “Child and caregiver circumstances have implications for retention in care for children living with HIV,” they note. However, they were not assessed by the authors and are often neglected by donors and programme implementers.

Community health workers “should actively be engaged as building blocks to identify, link and support children living with HIV.”

Improving outcomes for children living with HIV, Rujumba and Ndeezi conclude, requires implementation of programmes that also include nutrition monitoring, innovative and locally led interventions, and incorporating operational research to continuously improve interventions.


Iyun V et al. Variations in the characteristics and outcomes of children living with HIV following universal ART in sub-Saharan Africa (2006-17): a retrospective cohort study. Lancet HIV, online ahead of print, April 2021 (open access).

Rujumba J and Ndeezi G. Considerations for strengthening ART scale-up for children. Lancet HIV, online ahead of print, April 2021 (open access).

Full image credit: 'PMTCT basics'. Image by Elizabeth Glaser Pediatric AIDS Foundation/DFID - UK Department for International Development. Available at under Creative Commons licence CC BY-NC-ND 2.0.