Mortality rates of children with HIV may be underreported due to lack of patient follow-up and the usage of routine program data.
It is vital to have accurate counts of the number of children living with HIV (CHIV) and the number of deaths among them. Having a clear estimate of these numbers ensures proper resource allocation and helps track progress toward the United Nations Program on HIV/AIDS (UNAIDS) 90-90-90 goal.
When these numbers cannot be directly estimated, statistical methods are employed. A recent investigation published in the Journal of the International AIDS Society saw the need to conduct a current estimation of CHIV global mortalities due to recent HIV developments such as improved mother-to-child transmission prevention, early infant diagnosis and PCR testing at birth, earlier ART (antiretroviral therapy) initiation, and universal treatment.
Additionally, the investigators anticipated that a considerable number of deaths among CHIV go unrecorded, due to ART programs in Africa failing to link patient records to vital registration or just general lack of patient follow-up.
The study sought to estimate the temporal trends of all-cause mortality rates among CHIV younger than 15 and on ART in various regions in and outside of Africa. The investigators included 45711 CHIV on ART from the years 2004-2017 at 72 programs in Africa, Asia-Pacifica, and Latin America.
They conducted separate analyses for the data subsets, distinguishing the four sub-Saharan African regions from Asia-Pacific and Latin America. Within each group of regions, they separated children younger than five years of age from children 5-14 years old.
For all geographic regions, the investigators used routine-care International epidemiology Databases to Evaluate AIDS (IeDEA) observational data from participating treatment programs, and in addition, for southern Africa they used novel IeDEA data from a recent tracing study to thoroughly cover CHIV deaths that may have been missed by programs.
Describing their methodology, the investigators said, “We performed two distinct analyses to produce: (i) ‘unadjusted’ estimates of mortality, based on the routine data only; and (ii) for the African regions, ‘adjusted’ estimates that incorporated the tracing study data. Secondary sensitivity analyses were also performed to investigate the impact of selected analysis decisions.”
The investigators utilized Poisson regression to estimate CHIV mortality by age, sex, CD4 at ART start, duration on ART, geographic region, and year.
After adjusting for mortality in those classified lost to follow-up (LTFU), the results showed a substantial increase in mortality estimates for African regions: “The aggregate mortality rate was 9.4 deaths per 100 person-years (95% CI: 6.3 to 13.4), much higher than in the routine data and varying substantially by program.”
Higher mortality was correlated with shorter duration on ART (three times higher for ART duration less than a month versus a year or more on ART) and younger age (mortality doubled in CHIV younger than that one year versus five years or older). Additionally, female CHIV had half the mortality rate of males.
The investigators found that mortality rates of CHIV have substantially decreased over time. Compared to 2005, mortality rates in 2017 were 60% lower in Africa and 73% lower outside of Africa. However, the use of the novel IeDEA tracing study data confirmed the researchers’ suspicion that routine program data has severely underestimated CHIV mortality rates.
Overall, the lowered global mortality rate of CHIV on ART is reassuring, and a testament to the importance of ensuring early diagnosis and treatment, as well as follow-up.