Design of surveillance plans for Rift Valley fever in two different ecological zones in southern Africa

Abstract

Rift Valley fever (RVF) is an economically important, vector-borne transboundary livestock disease caused by a zoonotic arbovirus and is characterized by irregular epidemics occurring under favourable climatic and environmental conditions. A prominent feature of RVF is the occurrence of large abortion storms in pregnant livestock and mass mortality of susceptible newborn animals in affected regions, along with human illness in high risk groups in close contact with infectious livestock or competent arthropod vectors during RVF epidemics. Rift Valley fever virus (RVFV) has in recent decades spread beyond historical endemic foci in sub-Saharan Africa to trigger epidemics in livestock and human populations further afield and is now considered a major veterinary public health, food security and potential bioterrorism threat. Disease surveillance and timely reporting of epizootic or epidemic disease events, e.g. outbreaks of RVF remain suboptimal in many countries in Africa. In the context of southern Africa, two key problems constrain the development of effective surveillance systems for RVF, namely: 1. A lack of knowledge of the epidemiology of RVF; and 2. A lack of well-designed surveillance plans, tailored to regions. The epidemiology of RVF is highly complex and shaped not only by macro-scale climatic events but also location-specific idiosyncrasies in the vector-borne disease complex such as local climatic, environmental, vector and vertebrate host variables and agent characteristics. Regional differences in RVF ecology highlight the challenges associated with the accurate, timely identification and description of key determinants leading to outbreaks of RFV in different ecological zones, which complicate and often delay the implementation of risk mitigation measures. Following a comprehensive literature review of the epidemiology of RVF in sub-Saharan Africa – with particular reference to Mozambique and South Africa – this dissertation aims to address the above two key problems in relation to RVF surveillance by means of: 1. A systematic description of factors known or suspected to be relevant to RVF ecology in two different ecological zones in southern Africa; and 2. The design of two active, hazard-specific surveillance plans focused on the risk-based sampling of sentinel livestock and arthropod vectors to describe the baseline level, distribution and impact of RVF at selected sentinel sites in each ecological zone. The plan for one ecological zone incorporates a surveillance component for the detection of clinical RVF infection in human patients, whereas both plans suggest a framework for the monitoring of key environmental variables as well as the active sampling of RVFV reservoir candidate vertebrate wildlife. The ultimate aim, although beyond the scope of this dissertation, is the creation of two integrated surveillance systems where all of the above-mentioned data can be incorporated and analyzed in a geographic information system (GIS) platform. If implemented correctly, the respective plans should produce surveillance system outputs that can be used to support the timely, cost-effective implementation of risk mitigation measures when surveillance estimates cross defined thresholds, and that can contribute to the development of predictive risk models for each ecological zone.