Table 1.

Negative binomial regression model comparing mosquitoes counts in villages treated with three different Bti-interventions during non-intervention (2013 baseline) and intervention (2014) periods for female Anopheles captured indoors and outdoors, analysed together and separately. The random effect was integrated at village level

Bti interventions	N	Incidence-rate ratio (95% CI)	p	% reduction (95% CI)
All (N = 4031)
Untreated control	2973	–
Guided treatment	409	0.393 (0.340–0.455)	< 0.001	60.7% (54.5%–66.0%)
Full treatment	649	0.304 (0.258–0.359)	< 0.001	69.6% (64.1%–74.2%)
Indoors (N = 2063)
Untreated control	1522	–
Guided treatment	208	0.384 (0.324–0.455)	< 0.001	61.6% (54.5%–67.6%)
Full treatment	333	0.318 (0.274–0.369)	< 0.001	68.2% (63.1%–72.6%)
Outdoors (N = 1968)
Untreated control	1451	–
Guided treatment	201	0.403 (0.345–0.471)	< 0.001	59.7% (52.9%–65.5%)
Full treatment	316	0.285 (0.222–0.367)	< 0.001	71.5% (63.3%–77.8%)