Abstract
One intervention recommended by the World Health Organization (WHO) to control the spread of dengue fever is a network of lethal ovitraps to capture dengue-carrying Aedes mosquitoes. This study introduces a framework to perform a cost-effectiveness analysis of this intervention that incorporates the substantial operational costs entailed in servicing the network, a factor that has been largely ignored in previous analyses. A vector-host compartmental model was used to model the population dynamics of mosquitoes and to estimate the disability-adjusted life years (DALYs) averted from this intervention. Operational costs were estimated through finding the minimum number of workers that had to be employed to travel to and service all the traps at a specified interval of time. We conducted numerical studies in Singapore and Bangkok to demonstrate how our framework can be used both retrospectively to analyze an existing ovitrap network, and prospectively to assess the cost-effectiveness of setting up a hypothetical ovitrap network. In Singapore, the existing 68,000-trap network was cost-effective over a wide range of service intervals, and the status quo interval of 14 days was near-optimal. In Bangkok, a more heterogeneous setting, three additional findings emerged: allocating traps based on population density was consistently cost-effective; district-specific service intervals improved cost-effectiveness by approximately 30% relative to common single interval; and only the 18 densest districts (of 50) could be made individually cost-effective, surfacing a tradeoff between aggregate cost-effectiveness and equitable coverage. These results illustrate the value of incorporating operational considerations into cost-effectiveness analyses of distributed public-health interventions.