Table 3. Risk differences of delays in two or more domains and delays in early development domains, using on-site observations (total, n = 1957; intervention, n = 1037; control, n = 920). All outcomes are analysed at cluster-level.

	Intervention (clusters, n = 16)	Control (clusters, n = 16)	Crude intervention-control difference (95% CI); P-value	Adjusted intervention-control difference (95% CI); P-value
	Proportion outcome (95% CI)		Crude intervention-control difference (95% CI); P-value	Adjusted intervention-control difference (95% CI); P-value
Observation outcome
Two or more delays	0.15 (0.12 to 0.18)	0.35 (0.26 to 0.45)	-0.20 (-0.30 to -0.10); 0.001	-0.20 (-0.30 to -0.10); 0.001
Communication	0.20 (0.15 to 0.23)	0.29 (0.21 to 0.37)	-0.10 (-0.18 to -0.01); 0.032	-0.09 (-0.18 to 0.00); 0.042
Gross motor	0.12 (0.05 to 0.17)	0.25 (0.16 to 0.33)	-0.13 (-0.23 to -0.02); 0.015	-0.13 (-0.23 to -0.02); 0.015
Fine motor	0.09 (0.06 to 0.12)	0.24 (0.13 to 0.34)	-0.14 (-0.25 to -0.04); 0.010	- 0.14 (-0.24 to -0.04); 0.009
Problem-solving	0.06 (0.05 to 0.07)	0.13 (0.09 to 0.16)	-0.06 (-0.10 to -0.02); 0.003	-0.01 (-0.07 to -0.05); 0.003
Personal-social	0.22 (0.17 to 0.26)	0.38 (0.28 to 0.48)	-0.16 (-0.27 to -0.05); 0.005	−0.16 (-0.26 to -0.05); 0.004

The analysis is adjusted for clustering and controlled for several covariates (sex of the child, child weight at registration, child height at registration, child age at registration and endline, maternal age, maternal education, number of additional children the mother has, number of additional children aged <5 years, and family structure) by logistic regression analysis.
CI = confidence interval.