Estimating mosquito bionomics parameters with a hierarchical Bayesian model

Abstract

BackgroundThe malaria transmission potential and the vulnerability of Anopheles mosquitoes to different vector control methods depend, among other factors, on the endophily, endophagy, anthropophagy and survival of each species. Local information on these bionomic parameters is generally unavailable. MethodsTo address this, we estimated species-specific values of these parameters using an augmented version of the global database of bionomics data by Massey et al. (2016). We applied inclusion and exclusion criteria to select eligible studies with relevant experimental designs that minimise bias from collection methods for parous, sac, endophagy, and endophily rates as well as for the resting duration. For the human blood index (HBI), we separated data from indoor and outdoor collections. We fitted hierarchical Bayesian models with levels based on Anopheles taxonomy to estimate these quantities. Based on the estimated bionomics, we quantified the expected vectorial capacity reduction after the introduction of a pyrethroid-pyrrole insecticide-treated net (ITN) for 57 Anopheles species. ResultsWe identified 26 eligible studies for endophagy and 61 for the parous rate, leading to a Bayesian posterior average for the Anopheles genus of 42% (95% credible interval: 18-70) and 55% (32-77) respectively. HBI values widely varied depending on the location of collection, except for some species showing strong anthropophilic behaviours. Resting duration was estimated to be 2.1 days (1.2 - 4.8) at the genus level. Few studies were available to estimate the sac and endophily rates, which prevented us from deriving precise estimates for the whole Anopheles genus. Our estimates of the vectorial capacity reduction following the introduction of a pyrrole-pyrethroid ITN ranged between 48% and 76% across species, highlighting the important differences among mosquito species in vulnerability to vector control interventions. ConclusionThis work demonstrates how data from both Anopheles species complexes and individual species can be leveraged to generate species-specific estimates of bionomic parameters, capturing the local characteristics and behaviour of malaria vectors. The dataset is readily updatable as new data become available. However, more frequent and standardised field surveys are still needed to accurately characterise local vector behaviour.