The global burden of sickle cell disease in children under five years of age: a systematic review and meta-analysis

Abstract

Background: Sickle cell disease (SCD) is a common haematological disorder, affecting millions of people worldwide. It is most prevalent in malarial endemic areas in the tropics where outcomes are often poor due to resource constraints, resulting in most children dying before reaching adulthood. As increasing progress is made towards reducing under 5 mortality from infectious causes, non-communicable diseases (NCDs) including SCD have risen to the forefront of the global health agenda. Despite this, the global mortality burden of SCD remains poorly understood. This study aimed to estimate the incidence and mortality of SCD in children under 5 years of age in order to inform policy and develop sustainable strategies to improve outcomes.

Methodology: We performed a systematic literature search of Medline, EMBASE, Journals@Ovid, and Web of Science for studies on the incidence and mortality of SCD in children under 5, with search dates set from January 1980 and July 2017. We conducted random effects meta-analysis to obtain pooled meta-estimates of birth prevalence and mortality rates globally, and for each World Health Organization (WHO) region.

Results: 67 papers were found with relevant data. 52 contained data on incidence and prevalence and 15 contained data on mortality. The overall pooled estimate of mortality from the limited data available was 0.64 per 100 years of child observation (95% CI = 0.28-1.00) with the highest rate seen in Africa 7.3 (95% CI = 4.03-10.57). The global meta-estimate for the birth prevalence of homozygous sickle cell disease was 112 per 100 000 live births (95% CI = 101-123) with a birth prevalence in Africa of 1125 per 100 000 (95% CI = 680.43-1570.54) compared with 43.12 per 100 000 (95% CI = 30.31-55.92) in Europe.

Conclusion: There were a number of limitations in the depth and breadth of available data however it is clear that both the highest prevalence and highest mortality of SCD is in Africa. In order to address this burden, there is a need for national comprehensive newborn screening to identify patients, and the development of holistic SCD care programmes to provide therapeutics and education for families and children with SCD. This targeted funding should form part of a broader increased global focus on NCDs in childhood.

Figure 1

PRISMA flowchart detailing study selection.

Figure 2

Pooled birth prevalence (per 100 000 live births) of sickle cell disease homozygotes globally and by world regions.

Figure 3

Distribution of birth prevalence (per 100 000 live births) of sickle cell disease homozygotes by world regions. Note: The boxes represent the interquartile range of birth prevalence where the middle 50% (25-75%) of data are distributed; the bars represent birth prevalence outside the middle 50% (<25% or >75%); the dots represent specific birth prevalence which were a lot higher than normally observed (outliers) and the lower, middle and upper horizontal lines represent the minimum, median and maximum birth prevalence (excluding outliers), respectively.

Figure 4

Pooled birth prevalence (per 100 000 live births) of sickle cell disease heterozygotes by world regions.

Figure 5

Distribution of birth prevalence (per 100 000 live births) of sickle cell disease heterozygotes by world regions. Note: The boxes represent the interquartile range of birth prevalence where the middle 50% (25-75%) of data are distributed; the bars represent birth prevalence outside the middle 50% (<25% or >75%); the dots represent specific birth prevalence which were a lot higher than normally observed (outliers) and the lower, middle and upper horizontal lines represent the minimum, median and maximum birth prevalence (excluding outliers), respectively.

Figure 6

Pooled mortality rate (per 100 child-years of observation) due to sickle cell disease across world regions.

Figure 7

Distribution of sickle-cell disease mortality (per 100 child-years of observation) over world regions. Note: The boxes represent the interquartile range of mortality rate where the middle 50% (25-75%) of data are distributed; the bars represent mortality rate outside the middle 50% (< 25% or > 75%); the dots represent specific mortality rate which were a lot higher than normally observed (outliers) and the lower, middle and upper horizontal lines represent the minimum, median and maximum mortality rate (excluding outliers), respectively.