Table 2. The instantaneous association between village repellent distribution and Plasmodium spp. infection detected by RDT and PCR.

Factor	RDT			PCR
	All species* (n = 32,194)			All species (n = 13,157)			P. falciparum† (n = 13,068)			P. vivax† (n = 13,068)
	AOR	95% CI	p-Value	AOR	95% CI	p-Value	ARRR	95% CI	p-Value	ARRR	95% CI	p-Value
Intervention
No repellent	Ref.	—	—	Ref.	—	—	Ref.	—	—	Ref.	—	—
Repellent	0.25	0.004, 15.2	0.512	0.82	0.62, 1.09	0.180	0.67	0.47, 0.95	0.026	1.41	0.80, 2.47	0.233
Time (month)	0.87	0.72, 0.97	0.013	0.97	0.89, 1.07	0.582	1.02	0.95, 1.11	0.523	0.95	0.87, 1.04	0.247
Season
Cool	Ref.	—	—	Ref.	—	—	Ref.	—	—	Ref.	—	—
Hot	3.64	0.89, 14.8	0.072	1.10	0.35, 3.44	0.871	0.81	0.27, 2.40	0.700	11.2	2.23, 56.2	0.003
Rainy	3.15	0.89, 11.1	0.074	1.17	0.45, 3.02	0.747	0.46	0.17, 1.21	0.115	20.5	4.84, 86.6	<0.001

Instantaneous treatment effect comparisons: AORs and ARRRs, 95% CIs, and p-values from generalised linear mixed modelling (GLMM) using generalised structural equation modelling (GSEM).

*All-species estimates from crossed random effects generalised (logit) linear mixed model with random effects for cross-sectional (month) and village-specific heterogeneity in infection, and village-specific heterogeneity in effect of repellent distribution.

^†Species-specific (ARRR) estimates from crossed random effects generalised (multinomial) linear mixed model with random effects for cross-sectional (month) and village-specific heterogeneity in infection. No infection was the reference group for the outcome.

AOR, adjusted odds ratio; ARRR, adjusted relative risk ratio; PCR, polymerase chain reaction; RDT, rapid diagnostic test.