Aedes aegypti population sampling using BG-Sentinel traps in north Queensland Australia: statistical considerations for trap deployment and sampling strategy

Abstract

BG-Sentinel mosquito traps were trialed as a tool for the rapid assessment (24-h collections) and routine monitoring (72-h collections) of adult Aedes aegypti L. populations in north Queensland. Analysis of Ae. aegypti collections using BG-Sentinels set in suburban Cairns for 24 h permitted the calculation of sample size for a range of precision levels. Clusters of houses with BG-Sentinels operating continuously for 15 d, with collections every 72 h, also permitted required sample size calculation. Evidence of Ae. aegypti spatial clustering at the house scale was revealed, with statistically significant effects detected for all collection days. Less variation was detected at each trap location, with only nine of 32 trap locations revealing significant clustering over time. Trap-out effects through continuous BG-Sentinel operation at a fixed location were absent. The findings support fixed position sampling at 72-h intervals for routine monitoring of Ae. aegypti populations in Cairns. Despite the relationship between collections of adult vectors and the incidence of disease remaining unknown, BG-Sentinel collections provide an alternative and less labor-intensive abundance measure for assessing risk of dengue virus transmission and success of dengue vector control programs.