Table 4.

Posterior median and 95 % Bayesian credible intervals (BCI) of the geostatistical model based on environmental/climatic, malaria intervention coverage and SES predictors

	Geostatistical model 1	Geostatistical model 2	Geostatistical model 3
Variables	OR (95 % BCI)	OR (95 % BCI)	OR (95 % BCI)
LSTN	0.78 (0.69, 0.88)	0.81 (0.72, 0.90)	0.82 (0.72, 0.93)
Rice field proximity	0.49 (0.27, 0.81)	0.56 (0.34, 0.98)	0.73 (0.43, 0.99)
No	1.00	1.00	1.00
Yes	0.49 (0.27, 0.81)	0.56 (0.34, 0.98)	0.73 (0.43, 0.99)
Area type
Urban	1	1	1
Rural	3.91 (2.88, 5.22)	2.50 (1.89, 3.35)	2.36 (1.79, 3.13)
SES
Most poor		1	1
Very poor		0.76 (0.61, 0.95)	0.77 (0.61, 0.96)
Poor		0.90 (0.71, 1.14)	0.92 (0.73, 1.16)
Less poor		0.70 (0.55, 0.89)	0.71 (0.56, 0.90)
Least poor		0.32 (0.24, 0.43)	0.33 (0.25, 0.45)
Age (years)
0–1		1	1
1–2		1.56 (1.23, 1.56)	1.21 (0.95, 1.55)
2–3		1.72 (1.35, 2.21)	1.71(1.33, 2.19)
3–4		1.88 (1.48, 2.41)	1.87 (1.47, 2.39)
4–5		2.08 (1.62, 2.67)	2.07 (1.61, 2.65)
Mother’s education
No education		1	1
Primary		1.08 (0.89, 1.31)	0.85 (0.71, 1.03)
Secondary		1.06 (0.74, 1.54)	0.88 (0.61, 1.27)
Higher		1.44 (0.29, 6.49)	0.42 (0.05, 2.13)
Case management (ACT)	–	1.45 (0.49, 4.21)	0.13 (−1.49, 1.67)
U5 sleep under net (ITNsU5)	–	1.66 (0.89, 3.08)	0.25 (−0.37, 0.90)
House Spray (IRS)	–	1.14 (0.17, 7.23)	0.11 (−1.75, 1.70)
Variances
Gaussian process	0.87 (0.66, 1.15)	0.6 (0.44, 0.80)	0.55(0.39, 073)
Spatially varying ITNsU5			0.57 (0.38, 0.88)b
Spatially varying IRS			0.76 (0.42, 1.75)b
Spatially varying ACT			0.75 (0.42, 1.84)b
Range (km)a	10.61 (2.39, 20.82)	3.00 (0.40, 10.00)	2.70 (0.44, 9.75)

Model 1 includes only climatic factors

Model 2 includes climatic factors, age, SES, intervention measures

Model 3 has spatially varying covariates for IRS, ACT and ITNsU5

^aMinimum distance in kilometer at which the spatial correlation is lower than 5 %

^bPosterior median