Impact case study database

This case study relates to research conducted at Lincoln since July 2017 working with an international, interdisciplinary team of researchers from the UK, Zambia and Canada, and partner organisations (Zambian Ministry of Health [MoH]; Zambezi Water Resources Management Authority [WARMA], University of Barotseland [UBL] and the Zambezi EcoHealth Partnership [ZEP]). The research to which this case study refers is published in Cross et al (2020) [3.1] and Smith et al (2020) [3.2], with an accompanying popular article in Smith and Thomas (2020) [3.3].

Research context

Controlling the mosquitoes that transmit malaria (vectors) is one of the main ways to prevent and reduce the disease. Concerted national malaria control efforts in Zambia (such as bednets and indoor residual spraying of houses with insecticides), have led to a nationwide decline in parasite prevalence among children under 5 from 22% to 9% between 2006 and 2018. However, in some areas, such as Western Province (population 1.1m people) these interventions have not led to reductions and prevalence amongst under-5s in 2019 was 10%, twice as high as 2010 levels. Ecologically, the region is dominated by the Barotse floodplain, comprising a vast network of wetlands flooded seasonally by the Zambezi River and its tributaries: we investigated the relationship between this environment and transmission in order to understand the increase in malaria in the Province.

Malaria vectors are mosquito species of the genus Anopheles, with a few considered ‘primary vectors’ (human specialists, highly efficient vectors of human malaria parasites, main role in transmission) whereas others are ‘secondary vectors’ (take blood from a range of vertebrates, inefficient vectors, minor role in transmission). This is an important distinction in public health as primary vector species in sub-Saharan Africa tend to bite humans at night and indoors, so are susceptible to bednets and indoor residual spraying of insecticides, whereas secondary vectors predominantly bite humans outdoors during the day and at dusk, and so may evade these interventions. Traditionally, malaria control in sub-Saharan Africa has not been concerned with riverine environments as they rarely provide suitable breeding habitat for primary vectors, which breed in small ephemeral water bodies (e.g. puddles) associated with seasonal rain. Water in riverine habitats tend to be too deep, cold and fast flowing for primary vector species.

UoL research

To address these issues, Lincoln research combined field data collection in Sep-Nov 2017 to Apr-June 2019 in wet and dry seasons (Zambezi bathymetry, mosquito surveys stratified spatially by hydrogeomorphology, remote sensing ground control); time series radar and optical remote sensing of habitats, LISFLOOD modelling of the Zambezi seasonal inundation; large sample, high-throughput genetic sequencing of mosquitoes and climate-transmission modelling. This facilitated detailed comparison of taxon-specific abundance patterns in vectors between ecological zones differentiated by hydrological controls.

We discovered [3.1] both an unexpectedly high species diversity of vectors in Western Province and that almost all were secondary vector species, previously thought incapable of maintaining high levels of transmission. We also showed these species were breeding in habitats unsuitable for primary species, being widespread and abundant in floodplains along river margins and associated hydromorphological features. We concluded that these extensive habitats therefore support residual transmission, over much longer seasons, and that in many areas significant levels of transmission is by secondary vectors. This provided an explanation for the continued high prevalence of malaria in the Province and has important consequences for future vector control by the MoH.

It was therefore important to map these hydroclimatic zones across Zambia. To address this problem, we used downscaled daily climate predictions from seven GCMs to run a continental-scale hydrological model for a process-based representation of mosquito breeding habitat availability. A complex pattern of malaria suitability emerges as water is routed through drainage networks and river corridors serve as year-round transmission foci [3.2, 3.3]. Using this model, we produced (report to MoH) national maps of malaria transmission by season length and estimates for each province of the areas and populations at risk in these zones, including under projected climate change [3.2, 3.4].