Skip to main content

Table 4 Hierarchical multivariable logistic regression for infant feeding mode (0 = any breastfeeding vs. 1 = no breastfeeding) and four levels of factors according to the socioecological model

From: Understanding breastfeeding behaviours: a cross-sectional analysis of associated factors in Ireland, the United Kingdom and Australia

Variablesa, b

B

Wχ2

AOR

95% CI

Step 1 – Microsystem

Nagelkerke R2 = 52.6%

Age

−0.03

1.37

0.97

0.91, 1.02

Income (fourth to highest quintile)

0.10

0.20

1.11

0.70, 1.75

Education level (university)

0.72

9.35*

2.06

1.30, 3.27

Feeding knowledge (correct)

0.30

1.96

1.35

0.89, 2.06

Delivery (vaginal)

0.29

1.55

1.33

0.85, 2.10

Mum fed as baby (breastmilk)

0.81

12.98*

2.26

1.45, 3.51

First food after birth (breastmilk)

1.52

18.75*

4.55

2.29, 9.04

Feeding attitudes mean

−1.80

152.87*

0.17

0.13, 0.22

Step 2 – Mesosystem

Nagelkerke R2 = 56.2%

Support—friends & family

0.001

0.003

1.00

0.97, 1.03

Support—health professionals & groups

−0.007

0.20

0.99

0.96, 1.02

Support—technology

− 0.091

14.65*

0.91

0.87, 0.96

Breastfeeding encouragement by professional (≤ 30mins)

−0.03

0.02

0.97

0.62, 1.53

Step 3 – Exosystem

Nagelkerke R2 = 58.6%

Urbanisation (city)

0.69

8.40*

1.99

1.25, 3.18

Skin to skin contact after birth (yes)

−0.23

0.90

0.79

0.49, 1.28

Feeding intention (breastfeed)

0.75

7.77*

2.12

1.25, 3.59

Had enough info after birth about breastfeeding (yes)

0.45

3.29*

1.57

0.96, 2.57

Step 4 – Macrosystem

Nagelkerke R2 = 59.9%

Goodness of fit χ2 (df) = 634.84 (18)*

Country (0 = Australia vs. 1 = Ireland)

1.21

15.32*

3.35

1.83, 6.13

Country (0 = Australia vs. 1 = UK)

0.98

11.45*

2.68

1.51, 4.73

  1. Abbreviations: B unstandardised regression coefficient, Wχ2 Wald χ2-test, AOR Adjusted Odds ratio, 95% CI 95% CI for adjusted Odds ratio
  2. *p < 0.05
  3. a Values presented are taken from final model
  4. bReference groups (coded as 0) are presented in brackets