. 2015 May 1;34:6. doi: 10.1186/s41043-015-0008-y

Table 3.

Predictors of adequate diet diversity (DDS ≥ 4) among children 6-23 months ¹

	Kenya		Uganda		Tanzania		All countries
	aOR	95% CI	aOR	95% CI	aOR	95% CI	aOR	95% CI
Age category, (ref., 6 - 8 mo)
9 - 11 mo	2.57^*	1.47, 4.49	3.99^*	1.81, 8.77	2.25^*	1.40, 3.62	2.61^*	1.87, 3.65
12 - 17 mo	2.65^*	1.67, 4.19	2.74^*	1.31, 5.73	2.64^*	1.75, 3.98	2.75^*	2.07, 3.63
18 - 24 mo	4.00^*	2.24, 7.15	3.45^*	1.63, 7.30	3.19^*	2.06, 4.94	3.61^*	2.63, 4.96
Birth order	0.95	0.87, 1.02	_	_	0.99	0.94, 1.04	0.99	0.95, 1.02
Breast-feeding, (ref., no)	0.95	0.66, 1.38	0.38^*	0.22, 0.65	0.92	0.66, 1.38	0.85	0.67, 1.07
Cough, (ref., no)	_	_	_	_	1.26	0.98, 1.62	_	_
Education (ref., preschool/none)
Primary school	1.59	0.84, 3.00	1.59	0.76, 3.31	1.08	0.77, 1.52	1.22	0.92, 1.62
At least secondary school	1.81	0.90, 3.66	2.10	0.87, 5.08	1.33	0.77, 2.30	1.54^*	1.10, 2.18
Working mother (ref., no)	1.47^*	1.03, 2.12	_	_	_	_	_	_
Wealth Index (ref., first quintile)
Second quintile	1.72^*	1.07, 2.75	0.64	0.32, 1.25	1.34	0.94, 1.91	1.37^*	1.06, 1.76
Third quintile	1.64	0.98, 2.75	1.13	0.62, 2.05	2.02^*	1.40, 2.91	1.81^*	1.38, 2.37
Fourth quintile	2.43^*	1.46, 4.01	1.44	0.72, 2.89	2.47^*	1.61, 3.78	2.38^*	1.78, 3.18
Fifth quintile	1.89^*	1.05, 3.42	3.05^*	1.37, 6.83	4.24^*	2.58, 6.95	3.27^*	2.28, 4.67
Household size (mean)^†	0.98	0.92, 1.04	_	_	_	_	0.99	0.95, 1.03
Rural residence (ref., urban)	0.57	0.31, 1.04	1.04	0.51, 2.11	1.10	0.77, 1.56	0.88	0.65, 1.19
At least 4 PNC visits (ref., no)	2.05^*	1.48, 2.85	_	_	1.06	0.85, 1.31	1.25^*	1.05, 1.50
Immunized (ref., no)	0.96	0.69, 1.32	1.21	0.79, 1.85	0.98	0.78, 1.23	1.01	0.85, 1.20
Received vitamin A (ref., no)	_	_	_	_	1.43^*	1.11, 1.84	1.12	0.92, 1.35
Modern contraceptives (ref., no)	1.31	0.92, 1.86	_	_	1.40^*	1.10, 1.79	1.35^*	1.12, 1.62
Min. feeding frequency (ref., no)	2.54^*	1.80, 3.58	3.13^*	2.06, 4.77	1.51^*	1.21, 1.89	1.98^*	1.67, 2.34

aOR: adjusted odds ratio; SE: standard error; mo: month; min: minimum; PNC: prenatal care.

^*aOR estimates were statistically significant (p < 0.05), ^†OR values computed for one additional household member.