Leukemia comprises a heterogeneous group of blood cancers arising in bone marrow and other blood-forming tissues, characterized by uncontrolled proliferation of immature blood cells. Classification by cell lineage (lymphoid or myeloid) and by clinical course (acute or chronic) guides diagnosis, treatment and prognosis. Global trends show changing incidence and burden over recent decades; accurate regional estimates for Africa are limited by uneven cancer registration and data collection. The source study aimed to produce a pooled estimate of leukemia prevalence in Africa and to identify factors associated with leukemia by conducting a systematic review and meta-analysis following PRISMA guidelines.
The review followed PRISMA reporting standards. A comprehensive literature search strategy targeted multiple bibliographic databases, including PubMed/MEDLINE, Scopus and ScienceDirect, supplemented by Google Scholar and manual searches to identify eligible observational studies reporting leukemia prevalence and associated factors in African populations.
A total of 17,546 records were initially retrieved across the searched databases and supplementary searches. After screening and application of inclusion criteria, fifteen studies were retained for quantitative synthesis. These included studies collectively recruited 42,884 participants for analysis. The selection process and flow were reported in the source.
Extracted data included study characteristics, sample sizes, number of leukemia cases, population type (for example, pediatric versus general population), publication year and reported associations between potential predictors and leukemia. The source indicates that study-level quality assessment occurred, though details of the specific quality tool or scoring thresholds were provided in the manuscript rather than reproduced here.
Statistical pooling used STATA version 11 and a random-effects model to calculate pooled prevalence estimates with 95% confidence intervals. Between-study heterogeneity was quantified using Higgins’ I2 statistic. Meta-regression was employed to explore potential sources of heterogeneity. Publication bias was evaluated both visually with funnel plots and statistically with Egger’s weighted regression test; in the source a p-value < 0.05 was considered evidence of significant publication bias.
Fifteen studies met the inclusion criteria and were pooled for meta-analysis, representing 42,884 individuals across multiple African countries. The included studies varied by country, sample size, population type (children versus general population), and publication year, contributing to the observed heterogeneity in prevalence estimates.
The pooled prevalence of leukemia across the included African studies was estimated at 5.09% (95% CI: 4.02% to 6.17%). The pooled estimate was associated with high heterogeneity (I2 = 97.1%), indicating substantial variation in reported prevalence across the studies.
To address heterogeneity, subgroup analyses were performed. By country, the pooled prevalence ranged notably: Nigeria had the highest reported pooled prevalence at 10.58%, while Zambia had the lowest at 0.58%.
When stratified by publication year, studies conducted before 2015 had a pooled prevalence of 4.16%, compared with 6.67% for studies conducted after 2016.
Sample-size stratification indicated that smaller studies (fewer than 384 participants) yielded a higher pooled prevalence of 7.99%, whereas studies with larger samples (≥385 participants) showed a pooled prevalence of 3.61%.
By population type, pooled estimates were 3.38% in pediatric populations and 7.15% in general-population samples.
Among reported associations, two factors were identified as statistically significant predictors in the pooled analysis. Rural residency was strongly associated with leukemia (pooled OR 56.98, 95% CI: 9.97–326.70). Older age was also associated with higher odds of leukemia (pooled OR 3.90, 95% CI: 1.39–10.98). The source reports these pooled association estimates as part of its results.
Heterogeneity across included studies was substantial (I2 = 97.1%). Meta-regression was used to explore possible sources of heterogeneity, including study-level characteristics such as country, sample size and year of publication; the source documents that these analytic methods were applied, though specific meta-regression coefficients are reported in the original manuscript. Publication bias was assessed visually with funnel plots and tested using Egger’s test; a p-value threshold of <0.05 was used to indicate significant bias.
The pooled prevalence estimate of 5.09% indicates a measurable burden of leukemia across the included African studies, with pronounced variation by country, study period, sample size and population sampled. The identification of rural residency and older age as significant predictors highlights potential demographic and contextual factors relevant to leukemia epidemiology in Africa. The high between-study heterogeneity and differences by sample size and region underscore limitations in available data and the challenge of producing a single, uniform prevalence figure for the continent.
This systematic review and meta-analysis reported a pooled leukemia prevalence of 5.09% across 15 studies (42,884 participants) in Africa. Rural residency and older age were identified as significant predictors. The authors emphasize that leukemia represents a public health concern in the region, while the high heterogeneity across studies and differences by country, study size and population suggest that local surveillance and improved cancer registry data are needed to refine burden estimates and inform policy. The source provides all data and analytical results within the published manuscript.