Mosquito composition and density
As shown in table 1, a total of 1,105 mosquitoes were collected comprising 647 (58.6%) Anopheles sp., 402 (36.4%) Culex sp., 28 (2.5%) Aedes sp. and 2 (0.2%) Coquellitidia sp. The Anophelines comprised of An. gambiae s.l. (95.2%), Anopheles funestus (2.9%), Anopheles brohieri (1.2%), Anopheles paludis (0.5%) and Anopheles ziemanni (0.2%). Of the 647 Anophelines, 154 (23.8%) collected indoors comprising An. gambiae s.l. 149(96.7%), An. funestus 4 (2.6%), and An. ziemanni1(0.6%). 493(76.2%) were collected outdoor, made up of An. gambiae s.l. 467(94.7%), An. funestus 15(3.04%), An. paludis 3(0.6%) and An. brohieri 81.6(%). An. gambiae sensu stricto (s.s.) was the only member of the An. gambiae complex found.
With regards to the proportion of Anophelines collected based on housing status (improved/non-improved), 429 were collected from improved houses, of which 20.04% were indoor and 79.96% outdoor. In the non-improved houses, 218 Anophelines were collected, amongst which 31.2% was indoor and 68.8% outdoor.
Nocturnal activity and biting cycle of anophelines
Overall, the average man biting rate was observed to increase gradually between 6pm and 4am, peaking between 2am and 4am and then slowly declining to 6am (Figure 2). The overall man biting rate for the Anopheles was 0.098 bites per person per night (b/p/n). Anopheles gambiae was the most aggressive species, representing 95.2% of the total number of bites (0.094b/p/n) with peak biting hours between 2am and 4am regardless of the place of bite. Despite the small number collected compared to An. gambiae, the peak biting hours for An. funestus, was also observed at the same period both indoor and outdoor (Fig. 2).
A total of 488 female Anopheles was dissected for parity status with an overall parity rate of 61.3% (Table 1). Segregating by species, the parity rates were 62.4% (290/465), 53.8% (7/13), 100% (1/1) and 12.5% (1/8) for An. gambiae, An. funestus, An. ziemanni and An. brohieri, respectively.
Infection rates and entomological inoculation rates
A total of 615 female Anopheles mosquitoes were processed to ascertain the presence of P. falciparum circumsporozoite protein by CSP-ELISA. Of these, 210 were infected, giving an overall circumsporozoite protein rate of 34.2% (Table 1). Despite The circumsporozoite protein rate for An. gambiae (33.6%), the most abundant species, was lower compared to An. funestus (52.6%). While the lone An. ziemanni and 3 An. paludis collected were infected, none of the An. brohieri captured was infected. The intervention did not reduce indoor sporozoite infection rates of all Anopheles (IE = 1.1). It however reduced relative indoor sporozoite infection rates of An. gambiae by 1.8-fold. The overall average EIR was 0.29 infective bites per person per night (ib/p/n) with An. gambiae and An. funestus contributing to most of the transmission (Table 1).
Effect of house improvement on entomological indices
Effect of house improvement on mosquito density
In the improved homes, the relative number of indoor Anopheles significantly reduced by 1.8-fold (RFI=3.99; RFN=2.21; P=0.001) compared to the unimproved houses. In those improved homes, the relative number of An. gambiae entering houses increased by 1.7-fold (RFI=3.81; RFN= 2.26; P=0.004). Although the number of An. funestus collected indoors was 12-fold lower than the number collected outdoors in the improved houses, this effect did not differ statistically (RFI=12; RFN=1; P=0.07) probably due to the small sample size (Table 2).
Effect of house improvement on mosquito parity status
Table 3 summarizes the effect of the intervention on the number of parous anophelines by species. Improving houses generally led to a reduction in the number of parous anophelines collected indoor by 1.7-fold (RFI=4.48; RFN=2.67; p=0.05). The relative number of parous An. gambiae significantly reduced by 1.8-fold (RFI=4.32, RFN=2.63; p=0.03). The intervention was associated with 1.3-fold reduction in indoor parous rates for An. gambiae and 1.2-fold overall reduction of indoor parous rates (Table 3).
Effect of house improvement on entomological inoculation rate
Table 4 depicts the indoor and outdoor variation in entomological inoculation rates (EIR) in the two groups of houses. It was observed that improving the houses led to a reduction in the number of infective bites received per person per night indoors. A relative reduction of 1.7-fold (RFI=4.84, RFN=2.81) for all Anopheles and 1.6-fold (RFI=4.75; RFN=3.04) for An. gambiae was recorded.
Effect of house improvement on the night biting cycle of the anopheles
As described in Fig. 3, the number of mosquitoes caught indoor in improved houses rose from 8pm to 2 am. A reduction was subsequently observed up till 4am. At this point a second increase started. At 6am, when the catches were stopped, the highest number of mosquitoes collected indoor was found. Indoor and outdoor Anopheles abundance displayed the same hourly variation, with a peak at 4:00am (fig 4) in non-improved houses. In both group of houses, mosquitoes continued biting at 6:00am when the catches were stopped.