Table 2. Univariate and multivariable effect size estimates of the larviciding intervention on malaria prevalence in Dar es Salaam, 2004–2008 (N = 64,537).

	Univariate		Multivariable
	OR*	95% CrI†	OR*	95% CrI†
Larviciding intervention	0.79	0.66–0.93	0.79	0.66–0.93
Age
Under five years of age	-	-	1.00	-
≥5 and <15 years of age	-	-	0.82	(0.76–0.90)
≥15 and <30 years of age	-	-	0.67	(0.61–0.73)
≥30 and <45 years of age	-	-	0.60	(0.54–0.66)
≥45 and <60 years of age	-	-	0.55	(0.48–0.63)
≥60 years of age	-	-	0.47	(0.40–0.56)
Male sex	-	-	1.08	(1.01–1.15)
Slept outside ward (previous 2 weeks)	-	-	0.90	(0.77–1.04)
Treated for malaria (previous round)	-	-	0.65	(0.56–0.75)
Took malaria drug (previous 2 weeks)	-	-	1.02	(0.90–1.16)
ITN used the night before	-	-	0.93	(0.86–0.99)
House has closed ceiling	-	-	0.93	(0.85–1.01)
House has window screens	-	-	0.90	(0.83–0.98)
Trend for time (AR1 § )	Yes		Yes
Semi-parametric smooth for rainfall	Yes		Yes
Random effects (TCU & Household)	Yes		Yes

Statistically significant results are bolded.

^*OR  =  Odds Ratio.

^†CrI  =  Credible Intervals.

^§AR1  =  First Order Autoregressive Process.