Uncontrolled HIV causes the majority of deaths in children with HIV in western Kenya

USAID/Wendy Stone. Creative Commons licence. Image is for illustrative purposes only.

The vast majority of deaths amongst children with HIV in western Kenya are a result of uncontrolled HIV infection, according to a mortality study published in the January issue of AIDS.  The study also found that one in six deaths in children under five in Western Kenya are caused by HIV, with around 1% of all children born in the region dying as a result of HIV infection.

"These findings underscore the need to strengthen identification of children living with HIV, prompt ART initiation, aggressive treatment monitoring as well as screening and early management of malnutrition among children living with HIV” state the authors.

While deaths in children under five years old have reduced dramatically in Kenya over the last two decades, regional disparities persist. In western Kenya, deaths among children under five were twice the national average in 2014. The relatively high mortality rates in this region have been attributed to the disproportionate rates of infectious diseases, including HIV and malaria, among children.



Any lung infection that causes inflammation. The infecting organism may be bacteria (such as Streptococcus pneumoniae), a virus (such as influenza), a fungus (such as Pneumocystis pneumonia or PCP) or something else. The disease is sometimes characterised by where the infection was acquired: in the community, in hospital or in a nursing home.


A serious disease caused by a parasite that commonly infects a certain type of mosquito which feeds on humans. People who get malaria are typically very sick with high fevers, shaking chills, and flu-like illness. 

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).


Studies aim to give information that will be applicable to a large group of people (e.g. adults with diagnosed HIV in the UK). Because it is impractical to conduct a study with such a large group, only a sub-group (a sample) takes part in a study. This isn’t a problem as long as the characteristics of the sample are similar to those of the wider group (e.g. in terms of age, gender, CD4 count and years since diagnosis).


Muscle and fat loss.


However, determining cause of death can be difficult in Kenya (and many other low- and middle-income countries) due to limited diagnostic tools, widespread multi-morbidity, and the low number of autopsies conducted. Cause of death information is often gathered through verbal autopsies – detailed interviews conducted with family members after a death – but the reliability of this data is questionable.

Minimally-invasive tissue sampling – where a small sample is taken from a body and then tested for various pathogens – offers increased accuracy in determining cause of death in children compared to verbal autopsies alone. It has also been found to be more acceptable to families than full autopsies.

The Child Health and Mortality Prevention Surveillance (CHAMPS) study uses this approach, along with a range of other data, to identify and monitor causes of death in children under five in Kenya. Dr Dickens Onyango and colleagues therefore analysed CHAMPS data to estimate HIV cause-specific mortality rates in children under five years of age and to understand how the causes of death differed by HIV status.

The CHAMPS study

CHAMPS collects mortality data from two sites in western Kenya with a combined population of 22,270 children under the age of five – Karemo, a rural area in Siaya County, and Manyatta, and urban informal settlement in Kisumu County. When CHAMPS is notified of a death, they aim to collect a range of data including a verbal autopsy, medical records (including maternal medical records if the child was under 12 months), and a tissue sample for testing. A panel of experts evaluates this data and assigns cause(s) of death based on internationally agreed definitions. If multiple illnesses contributed to death, these are arranged in a causal sequence as ‘underlying’, ‘morbid’ and ‘immediate’ causes of death.

Underlying causes of death are those that triggered the chain of events that led to death. Immediate causes of death are the illnesses that directly caused death. If there is an illness or condition by which the underlying cause of death brings about the immediate cause of death, this is listed as a morbid cause of death.

HIV is considered an underlying cause of death when opportunistic infections or AIDS-defining illnesses were present at the time of death. When HIV is listed as the cause of death, this is broken down by sub-categories – for example, where there are signs of malnutrition the underlying cause of death would be listed as ‘HIV disease resulting in wasting syndrome’.

Mortality rates

Overall and HIV-specific mortality rates were estimated based on total and HIV-specific deaths among children aged under five in the CHAMPS study area between January 2018 and December 2019.

The overall under-five mortality rate for the region was 63.4 deaths per 1000 live births. The HIV-specific mortality rate for the region was 9.4 deaths per 1000 live births. This means that around one in six deaths in children under five in Western Kenya are caused by HIV, and around 1% of all children born in the region die as a result of HIV infection.

Causes of death

This analysis included 176 children aged 28 days to under 5 years at time of death who were enrolled in CHAMPS between May 2017 and September 2020 and had sufficient data for the analysis.

Of these, 92 (52%) were aged below 12 months, 100 (57%) had died in a health facility, and 25 (14%) had HIV. HIV status was determined through an HIV PCR test of sampled tissue, as the researchers were not always able to access medical records which would identify if a child had been diagnosed with and treated for HIV.

Children with HIV were more likely to have had a mother with only primary school-level education or less (p = 0.02).

Amongst the children with HIV, only one had a viral load below 1000. The median viral load was 112,205 (interquartile range 9349-2,670,143).

Causes of death were determined for 171 (97%) of the children included in the study.

HIV was recorded as the underlying cause of death for 24 (96%) of the children with HIV. Fifteen (60%) of these children had their underlying cause of death recorded as HIV resulting in wasting syndrome.

Amongst the 146 children without HIV, the leading underlying causes of death were malnutrition (23%), malaria (23%), pneumonia (10%), and gastroenteritis (7%).

Most of the children in the study had immediate causes of death that were different from the underlying causes of death. Immediate causes of death in children with and without HIV were similar, with malaria, sepsis and pneumonia being the most common in both groups. However, children with HIV were twice as likely to die of pneumonia compared to children without HIV (33% vs 16% respectively, p = 0.05).

Viruses (especially cytomegalovirus) were the predominant cause of pneumonia in children with HIV, whereas bacterial pathogens were most common in children without HIV. The authors note that in Kenya, antibiotics are recommended for treatment of pneumonia in children aged under 5 years yet 30% of HIV-positive children with pneumonia fail treatment in 48 hours. This is probably because antibiotics target bacterial instead of viral infections.  

Children with HIV were more likely to have at least one morbid condition than children without HIV (44% vs 24% respectively, p = 0.05). Malnutrition was the most common morbid condition reported for both children with and without HIV.  

Overall, malnutrition was more likely to play a role in the death of a child with HIV compared to a child without HIV. Malnutrition was recorded as an underlying, morbid, or immediate cause of death for 56% of children with HIV compared to 32% of children without HIV (p = 0.04).

Nearly all deaths were preventable

Devastatingly, the authors conclude that “nearly all deaths were due to causes for which effective interventions are widely available in Kenya”.

It is likely that most of the children who died as a result of uncontrolled HIV were either not diagnosed or not receiving ART, highlighting the urgent need to improve access to testing and treatment for children with HIV in this region. Early identification and care of malnourished children with HIV could also help to dramatically reduce deaths.

Finally, the authors recommend that pneumonia treatment guidelines in Kenya be reviewed to include greater emphasis on treating viral pneumonia, especially in children with HIV.


Onyango, DO et al. Causes of death in HIV-infected and HIV-uninfected children in the Child Health and Mortality Prevention Surveillance study—Kenya. AIDS 36: 59-68, 2022.

doi: 10.1097/QAD.0000000000003086

Full image credit: Little girl with health card, Kenya. USAID/Wendy Stone. Image is for illustrative purposes only. Available at www.flickr.com/photos/usaidkenya/6126989408/ under a Creative Commons licence CC BY-NC 2.0.