Skip to main content

Table 7 Results of logistic regression analysis predicting infants’ breastfeeding at 3 months of corrected age

From: A prospective cohort study on lactation status and breastfeeding challenges in mothers giving birth to preterm infants

Variables

All infants

Twins

Singleton

 

aOR1

95%CI

p*

aOR1

95%CI

p*

aOR1

95%CI

p*

Twin

0.27

0.09, 0.86

0.03

–

–

–

–

–

–

Gestational age

      

0.71

0.44, 1.12

0.14

Exclusive breastfeeding at discharge

–

–

–

7.29

1.74, 30.56

0.007

–

–

–

Own mother’s milk≥50% at 2 weeks post-discharge

8.20

2.6, −25.50

< 0.001

–

–

–

32.66

3.00, 355.25

0.004

  1. Note. Breastfeeding at 3 months of corrected age was included as dependent variable. (1) In all infants, a backward LR variable selection method was used to identify significant variables among cesarean section, maternal age, twin, GA, exclusive breastfeeding at discharge, own mother’s milk ≥50% at 2 weeks post-discharge, and direct feeding at 2 weeks post-discharge. The inclusion and removal of variables in models were based on p values of 0.05 and 0.10, respectively. Twin and own mother’s milk ≥50% at 2 weeks post-discharge entered the model. The model yielded statistical significance (χ2 = 19.31, df = 2, p < 0.001), and was checked by Hosmer-Lemeshow test for goodness-of-fit (p = 0.52). (2) In twins, a backward LR variable selection method was used to identify significant variables among cesarean section, maternal age, GA, exclusive breastfeeding at discharge, own mother’s milk ≥50% at 2 weeks post-discharge, and direct feeding at 2 weeks post-discharge. However, the Hosmer-Lemeshow test χ2 = 0.00. (3) In singleton infants, a backward LR variable selection method was used to identify significant variables among cesarean section, maternal age, GA, exclusive breastfeeding at discharge, own mother’s milk ≥50% at 2 weeks post-discharge, and direct feeding at 2 weeks post-discharge. The inclusion and removal of variables in models were based on p values of 0.05 and 0.10, respectively. Gestational age and own mother’s milk ≥50% at 2 weeks post-discharge entered the model. The model yielded statistical significance (χ2 = 14.11, df = 2, p = 0.001), and was checked by Hosmer-Lemeshow test for goodness-of-fit (p = 0.47)
  2. * p-value is based on the results of a multivariate binary logistic regression analysis. aOR = adjusted Odds Ratio. CI = Confidence Interval