Evaluation of community-based vector surveillance system for routine entomological monitoring under low malaria vector densities and high bednet coverage in western Kenya

Abstract

Abstract Introduction. Entomological surveillance is traditionally conducted by supervised teams of trained technicians. However, it is expensive and limiting in the number of sites visited. Surveillance through community-based collectors (CBC) may be more cost-effective and sustainable for longitudinal entomological monitoring. This study evaluated the efficiency of CBCs in monitoring mosquito densities compared to quality-assured sampling conducted by experienced entomology technicians. Methodology. Entomological surveillance employing CBCs was conducted in eighteen clusters of villages in western Kenya using indoor and outdoor CDC light traps and indoor Prokopack aspiration. Sixty houses in each cluster were enrolled and sampled once every month. Collected mosquitoes were initially identified to the genus level by CBCs, preserved in 70% ethanol and transferred to the laboratory every two weeks. Parallel, collections by experienced entomology field technicians were conducted monthly by indoor and outdoor CDC light traps and indoor Prokopack aspiration and served as a quality assurance of the CBCs. Results. Per collection, the CBCs collected 80% fewer An. gambiae s.l. [RR=0.2; (95%CI: 0.14 – 0.27)] and An. coustani [RR=0.2; (95%CI: 0.06-0.53)] and 90% fewer An. funestus [RR=0.1; (95%CI: 0.08-0.19)] by CDC light traps compared to the quality assured (QA) entomology teams. Significant positive correlations were however observed between the monthly collections by CBCs and QA teams for both An. gambiae and An. funestus. In paired identifications of pooled mosquitoes, the CBCs identified 4.3 times as many Anophelescompared to experienced technicians. The cost per person-night was lower in the community-based sampling at $9.1 compared to $89.3 by QA per collection effort. Conclusion. Unsupervised community-based mosquito surveillance collected substantially fewer mosquitoes per trap-night compared to quality-assured collection by experienced field teams, while consistently overestimating the number of Anophelesmosquitoes during identification. However, the numbers collected were significantly correlated between the CBCs and the QA teams suggesting that trends observed by CBCs and QA teams were similar. Further studies are needed to evaluate whether adopting low-cost, devolved supervision with spot checks, coupled with remedial training of the CBCs, can improve community-based collections to be considered a cost-effective alternative to surveillance conducted by experienced entomological technicians.