Comparison of the variability in mortality data generated by CDC bottle bioassay, WHO tube test, and topical application bioassay using Aedes aegypti mosquitoes

Abstract Background Insecticide resistance remains a major public health problem. Resistance surveillance is critical for effective vector control and resistance management planning. Commonly used insecticide susceptibility bioassays for mosquitoes are the CDC bottle bioassay and the WHO tube test. Less commonly used in the field but considered the gold standard for assessing insecticide susceptibility in the development of novel insecticides is the topical application bioassay. Each of these bioassays has critical differences in how they assess insecticide susceptibility that impacts their ability to differentiate between resistant and susceptible populations or determine different levels of resistance intensity. Methods We compared the CDC bottle bioassay, the WHO tube test, and the topical application bioassay in establishing the dose–response against deltamethrin (DM) using the DM-resistant Aedes aegypti strain MC1. Mosquitoes were exposed to a range of insecticide concentrations to establish a dose–response curve and assess variation around model predictions. In addition, 10 replicates of 20–25 mosquitoes were exposed to a fixed dose with intermediate mortality to assess the degree of variation in mortality. Results The topical application bioassay exhibited the lowest amount of variation in the dose–response data, followed by the WHO tube test. The CDC bottle bioassay had the highest level of variation. In the fixed-dose experiment, a higher variance was similarly found for the CDC bottle bioassay compared with the WHO tube test and topical application bioassay. Conclusion These data suggest that the CDC bottle bioassay has the lowest power and the topical application bioassay the highest power to differentiate between resistant and susceptible populations and assess changes over time and between populations. This observation has significant implications for the interpretation of surveillance results from different assays. Ultimately, it will be important to discuss optimal insecticide resistance surveillance tools in terms of the surveillance objective, practicality in the field, and accuracy of the tool to reach that objective. Graphical Abstract