CDC light traps underestimate the protective efficacy of an indoor spatial repellent against bites from wild Anopheles arabiensis mosquitoes in Tanzania

Table 1 Estimates of protective efficacy of Mosquito Shield^™ against wild Anopheles arabiensis using different techniques

Technique	Outcome measures	Treatment	Total captured^a	William’s mean^b (95% CI)	% PE (95% CI)	IRR (95%CI)	P-value
Feeding	Feeding inhibition	Negative control	43	11.8 (6.1–17.6)	_	1	< 0.0001
Feeding	Feeding inhibition	Mosquito shield^™	7	3.1 (0–6.4)	84% (58–94)	0.16 (0.06–0.42)	< 0.0001
HLC	Landing inhibition	Negative control	1724	38.0 (30.6–47.2)	_	1	< 0.0001
HLC	Landing inhibition	Mosquito shield^™	427	9.6 (7.4–12.3)	77% (64–86)	0.23 (0.14–0.36)	< 0.0001
CDC-LT	Reduction in numbers collected	Negative control	896	19.1 (14.1–25.7)	_	1	0.160
CDC-LT	Reduction in numbers collected	Mosquito shield^™	658	14.1 (10.5–18.9)	30% (0–56)	0.70 (0.44–1.0)	0.160

^aTotal captured refers to total number of blood-fed An. arabiensis collected for Feeding technique; total number of landed An. arabiensis for HLC technique; total number of An. arabiensis collected in CDC-LT for CDC-LT
^bWilliam’s mean = [(geometric mean of (x + δ))-δ, when δ = 1]. PE (protective efficacy) = 1-IRR estimates

ISSN: 1475-2875