Rift valley fever risk mapping and modelling in Tanzania

Thumbnail Image



Journal Title

Journal ISSN

Volume Title


Sokoine University of Agriculture


Rift Valley fever (RVF) was first reported in Tanzania in 1930 and the last outbreak occurred in the country in 2006/07. Besides the long history of RVF in the country, little is known about its spatial and temporal epidemiology and habitat suitability for its occurrence. This study was conducted to determine potential risk factors and develop the country RVF risk map. Enzyme-linked immunosorbent assay was used to examine the presence of antibodies specific to RVF virus (RVFV) in serum samples from domestic ruminants, humans and wild animals. Logistic regression modelling was used to analyze RVF outbreak data and RVFV seropositivity. Space-time permutation and MaxEnt modelling were used to identify clusters and habitat suitability for RVF occurrence, respectively. Between 1930 and 2007, there were a total of 10 RVF outbreaks with overlapping of clusters that continuously covered more parts of the country. Overall, the seroprevalence of IgG specific to RVFV in domestic ruminants (n = 1435) was 25.8% (95% CI: 23.52, 28.05) and in humans (n = 541) was 10.7% (95% CI: 8.11, 13.34). The IgG specific to RVFV was detected in nine (n = 22) and one (n = 3) serum samples from African buffalo and African elephant, respectively. The potential risk factors for RVF occurrence included eastern Rift Valley ecosystem (OR = 6.14, CI: 1.96, 19.28), rainfall during the previous two months >405.4mm (OR = 12.36, CI: 3.06, 49.88), clay (OR =8.76, CI: 2.5, 30.5) and loam (OR = 8.8, CI: 2.0, 37.8) soil texture, introduction of domestic ruminants into the herd (OR = 5.08, CI: 2.74, 9.44; p< 0.001), human contact with aborted foetus materials (OR = 2.89, CI: 1.48, 5.60), human participation in the slaughtering of animals (OR = 2.65, CI: 1.39, 5.04), human having consumed meat from dead animals (OR = 2.06, CI: 1.05, 4.00). The findings of this study have shown that the north-eastern, central and lake zones of the country have larger amount of suitable habitat for RVF occurrence than the north-western and southern zones. These research findingsiii can be applied to guide risk-based cost-effective RVF surveillance and interventions strategies in the country.


PhD Thesis


Rift valley fever, risk mapping, Tanzania, risk modelling, temporal epidemiology, spatial epidemiology