Table 2.

Quality assessment of trials on complementary food with or without education

	Quality Assessment					Summary of Findings
				Directness		No of events
No of studies	Design	Limitations	Consistency	Generalizability to population of interest	Generalizability to intervention of interest	Intervention	Control	RR or SMD (95% CI)
Height gain: Moderate outcome-specific quality
4 studies[17,24-26]	RCT	Random effect model was used because of heterogeneity	2 studies suggest benefit	To food insecure population		257	255	SMD 0.34(-0.09, 0.78) Food insecure population
Height for age: Moderate outcome-specific quality
7 studies[17,22,24-26,29,30]	RCT +non RCT	Random effect model was used because of heterogeneity	2 studies suggest benefit	To food insecure population		704	948	SMD 0.39 (0.05, 0.73) Food insecure population
Stunting: Moderate outcome-specific quality
7 studies[17,22,24-26,29,30]	RCT +non RCT	Random effect model was used because of heterogeneity		To food insecure population		704	948	RR 0.33 (0.11, 1.00) Food insecure population
Weight gain: Moderate outcome-specific quality
4 studies[17,24-26]	RCT	Random effect model was used because of heterogeneity	1 study suggest benefit	To food insecure population		247	255	SMD 0.43 (-0.42, 1.27) Food insecure population
Weight-for-age: Moderate outcome-specific quality
3 studies[22,25,30]	RCT+ non RCTs	Random effect model was used because of heterogeneity	1 study suggest benefit	To food insecure population		162	156	SMD 0.26 (0.04, 0.48) Food insecure population
Underweight: Moderate outcome-specific quality
1 study[30]	Non RCT			Only one study and to food insecure population		170	149	RR 0.35 (0.16, 0.77) Food insecure population