The Epidemiology of Zika, Chikungunya and Dengue Viruses in Africa: A Systematic Review and Meta-analysis of Prevalence, Geographic Distribution and Risk Factors
DOI:
https://doi.org/10.33003/sajols-2026-0403-52Keywords:
Dengue; Chikungunya; Zika; Africa; Meta-analysis; AedesAbstract
Dengue, Chikungunya and Zika viruses circulate widely in Africa, yet their burden remains poorly quantified because febrile illness is commonly attributed to malaria and laboratory confirmation is rare. This review estimated the pooled prevalence of the three infections and examined how prevalence varies by place, period, diagnostic method and population group. Six databases were searched from 1 January 2000 to 15 June 2026. Observational studies reporting laboratory-confirmed infection among people resident in Africa were eligible. Two reviewers screened and extracted data in duplicate and appraised studies with the Joanna Briggs Institute checklist. Prevalence was pooled using random-effects models after Freeman–Tukey double arcsine transformation. Heterogeneity, subgroup effects, meta-regression, publication bias and leave-one-out sensitivity analyses were examined. One hundred and twenty-nine studies (198,742 participants; 32 countries) were included. Pooled prevalence was 21.8% (95% CI 17.4–26.5) for Dengue, 19.4% (15.1–24.1) for Chikungunya and 9.6% (6.2–13.6) for Zika, with I² above 98% throughout. Estimates were highest in West and East Africa and lowest in North and Southern Africa, and rose across successive publication periods. IgG assays returned two to three times the prevalence obtained by RT-PCR or plaque reduction neutralisation. Febrile patients and outbreak settings gave the highest values. Dengue–chikungunya co-infection was 3.1%. Urban residence, household water storage, absent window screens and recent travel were consistently associated with infection. These three arboviruses are common across much of Africa, but the evidence base is uneven and heavily dependent on serology. Confirmatory testing, standardised protocols and wider geographic coverage are needed.