Entomological baseline data collection and power analyses in preparation of a mosquito swarm-killing intervention in south-western Burkina Faso

Abstract

Background: Insecticides are currently the main tools used to reduce the transmission of malaria; therefore, the development of resistance to insecticides in malaria vectors is of major concern for malaria control. The resistance level to pyrethroids is particularly high in the Western region of Burkina Faso and may affect the efficacy of insecticidal bed nets and indoor residual spraying. Adult mosquito swarming and other nocturnal behaviours exhibit spatial and temporal patterns that suggest potential vulnerability to targeted space spraying with effective insecticides. Indeed, targeted space-spraying against adult mosquito swarms has been used to crash mosquito populations and disrupt malaria transmission. Methods: Prior to impact assessment of swarm killing, a baseline data collection was conducted from June to November 2016 in 10 villages divided into two areas in western Burkina Faso. The data considered both ecological and demographic characteristics to monitor the key entomological parameters. Results: The mean number of swarms observed was 35 per village, ranging from 25 to 70 swarms according to the village. Female density in both areas varied significantly as a function of the village and the period of collection. The human biting rate was significantly affected by the period of collection and depended upon whether the collection was carried out indoors or outdoors. Averages of parity rate were high in both areas for all periods of collection, ranging from 60 to 90%. These values ranged from 80 to 100% for inseminated females. Sporozoite rates ranged between 1.6 and 7.2% depending upon the village. The molecular identification of resting and swarming mosquitoes showed the presence of the three major malaria vectors in Burkina Faso, but in different proportions for each village. Conclusions: The distribution of the potential swarm markers and swarms in villages suggested that swarms are clustered across space, making intervention easier. Power simulations showed that the direct sampling of swarms provides the highest statistical power, thereby reducing the number of villages needed for a trial.