Emergence of behavioural avoidance strategies of malaria vectors in areas of high LLIN coverage in Tanzania

Abstract

Despite significant reductions in malaria transmission across Africa since 2000, progress is stalling. This has been attributed to the development of insecticide resistance and behavioural adaptations in malaria vectors. Whilst insecticide resistance has been widely investigated, there is poorer understanding of the emergence, dynamics and impact of mosquito behavioural adaptations. We conducted a longitudinal investigation of malaria vector host choice over 3 years and resting behaviour over 4 years following a mass long-lasting insecticidal nets (LLINs) distribution in Tanzania. By pairing observations of mosquito ecology with environmental monitoring, we quantified longitudinal shifts in host-choice and resting behaviour that are consistent with adaptation to evade LLINs. The density of An. funestus s.l., declined significantly through time. In tandem, An. arabiensis and An. funestus s.l. exhibited an increased rate of outdoor relative to indoor resting; with An. arabiensis reducing the proportion of blood meals taken from humans in favour of cattle. By accounting for environmental variation, this study detected clear evidence of intra-specific shifts in mosquito behaviour that could be obscured in shorter-term or temporally-coarse surveys. This highlights the importance of mosquito behavioural adaptations to vector control, and the value of longer-term behavioural studies.

Figure 1

Predicted mean mosquito abundance per trap per night with 95% confidence interval (a) host-seeking An. funestus s.s. from January 2012 to May 2015 indoors (b) resting An. arabiensis and An. funestus s.s. indoors and An. arabiensis in animal sheds from January 2012 to May 2015 and (c) Human blood index of An. arabiensis overall and in animal sheds from January 2012 to June 2014. Non-significant effects were not retained in the best model; therefore, predictions for these were not available. Raw data is shown in Supplementary Figs. S1 and S2.

Figure 2

Study site in the Kilombero Valley in Kilombero and Ulanga districts, Tanzania, showing Ifakara and the four study villages as well as the weather stations. Entomological and environmental data was collected for all four villages. (Generated by ArcGIS 10.2, https://www.esri.com/software/arcgis/arcgis-for-desktop).