Clustering of host-seeking activity of Anopheles gambiae mosquitoes at the top surface of a human-baited bed net.

BACKGROUND Knowledge of the interactions between mosquitoes and humans, and how vector control interventions affect them, is sparse. A study exploring host-seeking behaviour at a human-occupied bed net, a key event in such interactions, is reported here. METHODS Host-seeking female Anopheles gambiae activity was studied using a human-baited 'sticky-net' (a bed net without insecticide, coated with non-setting adhesive) to trap mosquitoes. The numbers and distribution of mosquitoes captured on each surface of the bed net were recorded and analysed using non-parametric statistical methods and random effects regression analysis. To confirm sticky-net reliability, the experiment was repeated using a pitched sticky-net (tilted sides converging at apex, i.e., neither horizontal nor vertical). The capture efficiency of horizontal and vertical sticky surfaces were compared, and the potential repellency of the adhesive was investigated. RESULTS In a semi-field experiment, more mosquitoes were caught on the top (74-87%) than on the sides of the net (p < 0.001). In laboratory experiments, more mosquitoes were caught on the top than on the sides in human-baited tests (p < 0.001), significantly different to unbaited controls (p < 0.001) where most mosquitoes were on the sides (p = 0.047). In both experiments, approximately 70% of mosquitoes captured on the top surface were clustered within a 90 x 90 cm (or lesser) area directly above the head and chest (p < 0.001). In pitched net tests, similar clustering occurred over the sleeper's head and chest in baited tests only (p < 0.001). Capture rates at horizontal and vertical surfaces were not significantly different and the sticky-net was not repellent. CONCLUSION This study demonstrated that An. gambiae activity occurs predominantly within a limited area of the top surface of bed nets. The results provide support for the two-in-one bed net design for managing pyrethroid-resistant vector populations. Further exploration of vector behaviour at the bed net interface could contribute to additional improvements in insecticide-treated bed net design or the development of novel vector control tools