Abstract / Summary
Implementing vector control interventions against infectious diseases is operationally challenging, with delivery and population uptake taking considerable time within a community. Studies have suggested that the slow mass drug administration (MDA) of the endectocide, ivermectin, may have contributed to the lower epidemiological impact against malaria observed in recent cluster-randomised controlled trials, although evidence supporting this hypothesis is lacking. It is unclear whether interventions which have short but highly effective durations of entomological impact are more impactful than tools which kill the same number of mosquitoes but over a longer period. Transmission dynamics mathematical models can be used to explore the trade-offs between product profiles and investigate how different implementation periods could influence disease control. First, we use a simple vector-borne disease compartmental model to explore the epidemiological impact of killing the same number of mosquitoes over 10, 30 or 90 days. We find that for different mosquito-killing periods, the total number of cases averted is similar, although the observable impact may depend on when efficacy is assessed. Secondly, we adapted an existing malaria transmission model and simulated the fast and slow deployment of an effective endectocide MDA campaign in different seasonal settings. In perennial and seasonal settings and for a given coverage, slow MDA deployment averts more clinical cases of malaria than a fast synchronised distribution (completed in a single day, although the difference was modest). The benefits of the intervention depend on how well-timed a deployment is relative to the peak of the transmission season, with the efficacy of an ivermectin-like endectocide MDA being up to approximately 42% lower if started late. The work indicates that the time taken to complete MDA campaigns is unlikely to drive major differences in the effectiveness of short-lived endectocides and showcases how transmission dynamics models can be used to support epidemiological trial interpretation.