Systematic review and meta-analysis for the occurrence of Crimean-Congo Hemorrhagic Fever, Tularemia and Rift Valley Fever in pastoralist systems in Africa

Dickson Nyaguthii^1,2*Joshua Orungo Onono²Peter Kimeli²Eric Fevre^1,3Daniel Nthiwa^1,4Mathew Muturi⁵Michael E. von Fricken^6,7David Redding⁸Ellen P Carlin⁹Victor Ofula¹⁰Dino Martins¹¹Francesco Pietro Fava¹²James Hassel^1,7

• ¹International Livestock Research Institute (ILRI), Nairobi, Kenya
• ²University of Nairobi, Nairobi, Kenya
• ³University of Liverpool, Liverpool, United Kingdom
• ⁴University of Embu, Embu, Kenya
• ⁵Zoonotic Disease Unit, Nairobi, Kenya
• ⁶University of Florida, Florida, United States
• ⁷Smithsonian Institution, Washington, United States
• ⁸Natural History Museum, London, United Kingdom
• ⁹Parapet Science and Policy Consulting, Washington, United States
• ¹⁰Kenya Medical Research Institute, Nairobi, Kenya
• ¹¹Turkana basin Institute, New York, United States
• ¹²Università degli Studi di Milano, Milan, Italy

Pastoralism is a livestock production system practiced in areas with harsh environmental conditions. It is characterized by low investment in animal health, with increased risk of outbreaks of animal diseases. Zoonoses such as Crimean Congo haemorrhagic fever (CCHF), Rift Valley fever (RVF) and tularaemia, may be spread through livestock product value chains. Published papers on the three diseases were searched from PubMed, Web of Science, Scopus and Google scholar databases using predetermined search terms. After their eligibility for inclusion was examined, full texts of the included articles were downloaded and data extracted. Data included variables on publication details, laboratory methods and measures of disease occurrence with the studies also being grouped according to livestock value chain nodes. Random effects models were fitted to pool the study prevalences. Egger's test and symmetry of funnel plots were used to investigate publication bias. A random forests algorithm was used to determine relevant moderators of prevalence and a mixed effects model to examine relevance of selected moderators. RVF was the most studied pathogen (22/34; 64.7%) followed by CCHF (10/34; 29.4%) with 2/34(5.9%) papers describing the occurrence of tularaemia in pastoral settings in Africa. Pooled prevalence for RVF was highest in humans 29% (95% CI: 7% -69%), then camels 19% (95% CI: 7% -43%) and lowest in goats at 6% (95% CI: 4% -10%). CCHF prevalence was highest in camels at 48% (95% CI: 8% -91%) and lowest in humans at 6% (95% CI: 2% -19%). Prevalence was highest in livestock farms for RVF (13% (95% CI: 10% -16%) and CCHF 15% (95% CI: 4% -44%). Females were more likely to be positive 5.20 (95% CI: 3.09 -8.76 p<0.01) while mixed herds were more likely to be positive for RVF (33.34 (95% CI: 0.72 -1548.64, p= 0.0734). Two papers investigated tularaemia with pooled positivity being 2% (95% CI: 0% -8%). This review provides evidence that CCHF, RVF, and tularaemia pathogens are endemic in pastoral areas. Enhanced surveillance for these pathogens along livestock value chain nodes to mitigate risk of their spread is warranted.