Fig 1.
Systematic review process to select the studies relevant for meta-analysis of NMFIs in Africa.
Table 1.
Variables extracted from the included studies and their detailed description.
Fig 2.
Number of studies per country included in meta-analysis of NMFIs in Africa.
Studies on populations from multiple countries were treated as separate studies. *Tallies from Tanzania and Zanzibar were combined in the image (OpenStreetMap contributors, http://geoportal.icpac.net/layers/geonode%3Aafr_g2014_2013_0).
Fig 3.
Aetiologic agents directly and/or indirectly detected in fever patients from different African regions.
Proportions of positive cases were calculated by dividing the number of cases diagnosed by any kind of laboratory method by the total number of samples tested in each African region.
Fig 4.
Chord diagrams presenting the samples tested for the diagnosis of (A) bacterial infections and (B) viral, parasitic, and fungal infections.
Arc lengths represent the total numbers in each category. Blood samples comprised serum and plasma. Nasal swabs also included oro- and nasopharyngeal swabs and aspirates.
Table 2.
Most common aetiologic agents found in the reviewed literature to cause non-malarial febrile illnesses (NMFIs) based on number of publications.
Fig 5.
Heat map showing the occurrence of clinical signs and symptoms in patients infected with specific agents.
(shading represents the proportion of positive cases (%)). Clinical data were only added to the heat map when linked to a particular pathogen.
Table 3.
Summary statistics of the studies investigating the agents included in meta-regression analyses.
Table 4.
Results of the final multivariable meta-regression models of aetiologic agents.
Statistical significance was set at p value <0.05.