Abstract
ABSTRACT
Background
The growing attention to non‐malarial febrile illnesses (NMFIs), driven by the decline in malaria cases across the African continent, has revealed pathogens previously overshadowed by the burden of malaria. Recent outbreaks, the resurgence of diseases and the rising antimicrobial resistance burden, compounded with data gaps in previous studies, have underscored the need to closely examine this area in recent times.
Methods
We searched for articles through PubMed, Scopus, ScienceDirect and Web of Science databases to identify the common causative agents of NMFIs and estimate their proportional morbidity rates (PMRs) in Africa, stratifying agents across setting (rural vs. urban), population status (inpatient vs. outpatient) and outbreak status. We included studies that enrolled patients presenting with acute undifferentiated fever defined as fever without an identified alternative cause.
Results
We analysed 101 studies comprising 68,905 febrile patients across 20 African countries, most with high risk of bias. Viruses were most frequently reported, with dengue and chikungunya viruses showing high PMR estimates of 6.8% (95% CI: 2.7%–16.2%) and 24.1% (95% CI: 7.0%–57.1%) respectively. Bacteria were less common (pooled PMR:
K. pneumoniae
,
S. aureus
,
E. coli
and
S. pneumoniae
most frequently reported and often resistant to penicillins, fluoroquinolones and 3rd‐generation cephalosporins. Among outpatients, viral infections dominated (82.9%), particularly dengue and chikungunya in urban and outbreak contexts. In rural areas,
S. pneumoniae
,
C. burnetii
and
Borrelia
spp. were common bacterial causes, while dengue, Crimean‐Congo haemorrhagic fever and Rift Valley fever viruses predominated. Despite the predominance of molecular methods, most studies included the use of both molecular and serological methods.
Conclusion
The causative agents of NMFIs in Africa are enormous and context‐specific. These findings highlight the need to move beyond malaria‐centric fever management toward context‐specific diagnostic and surveillance strategies. Methodological harmonisation is necessary to yield reliable summary estimates in future studies.