Table 4.

Scores of currently breast-feeding, meal frequency, dietary diversity, composite feeding and hand washing for all five visits among non-participants and CPNP participants in intervention area, Habro and Melka Bello districts, Ethiopia (2012–2014)^*

	Score	Non-participants†			CPNP participants†			Difference in score between non-participants and
	range	n	Mean	sd	n	Mean	sd	CPNP participants‡	95 % CI	P §
Currently breast-feeding score	0–7	338	6·20	0·97	232	6·16	0·88	−0·11	−0·28, 0·05	0·18
Meal frequency score	0–15	338	7·67	2·26	232	8·36	2·24	0·43	0·07, 0·78	0·02
Dietary diversity score	0–10	338	4·20	1·94	232	5·38	2·61	0·06	−0·26, 0·38	0·73
Composite feeding score_1║	0–30	338	18·08	2·98	232	19·89	3·65	0·49	0·01, 0·96	0·046
Composite feeding score_2║	0–25	338	11·88	3·03	232	13·73	3·51	0·49	0·01, 0·96	0·046
Hand washing score	0–6	413	3·55	1·88	261	2·64	1·87	0·07	−0·18, 0·32	0·60

CPNP, community-based participatory nutrition promotion.

^* n 570 for scores of currently breast-feeding, meal frequency, dietary diversity, composite feeding_1 and composite feeding_2; n 674 for hand washing score.

^†Non-participants, mothers who never attended CPNP session in the intervention area; CPNP participants, mothers who ever attended CPNP session in the intervention area.

^‡Differences (95 % confidence intervals) in scores between non-participants and CPNP participants were derived from contrasts of each score using mixed-effect linear regression analysis accounting for clustering effect, father’s education level, household food insecurity and district location.

^§ P values were obtained using mixed-effect linear regression analysis accounting for clustering effect, father’s education level, household food insecurity and district location

^║Composite feeding score_1 was constructed based on scores of breast-feeding, dietary diversity and meal frequency; composite feeding score_2 was constructed based on scores of dietary diversity and meal frequency.