Table 5:

Multipredictor model of exclusive breastfeeding, screening status divided by indication (n = 8297)*

VariableOR (95% CI)
< 3 glucose measures, screen not indicatedRef.
Screened (≥ 3 measures), preterm0.23 (0.14–0.38)
Screened (≥ 3 measures), small for GA0.25 (0.14–0.44)
Screened (≥ 3 measures), birthing parent with diabetes0.57 (0.44–0.73)
Screened (≥ 3 measures), large for GA0.72 (0.59–0.87)
Birth year
 2014Ref.
 20150.89 (0.78–1.01)
 20160.66 (0.58–0.76)
 20170.71 (0.62–0.81)
 20180.90 (0.72–1.11)
Antenatal parity ≥ 11.82 (1.64–2.01)
Hypertension or pre-eclampsia0.85 (0.70–1.04)
Obesity0.77 (0.67–0.90)
Smoking by the birthing parent0.80 (0.66–0.96)
Social issues or perinatal inpatient drug screening0.61 (0.40–0.94)
Marital status divorced, single or widowed (v. married or common law)0.66 (0.55–0.80)
Pyrexia0.70 (0.50–1.00)
Induced0.85 (0.77–0.95)
Mode of birth
 Simple vaginalRef.
 Assisted vaginal0.91 (0.76–1.08)
 Elective cesarian0.79 (0.66–0.93)
 Emergency cesarian0.80 (0.67–0.97)
5-minute Apgar score ≥ 81.55 (1.01–2.38)
Brief postnatal CPAP1.32 (0.89–1.96)
Newborn cardiorespiratory issue0.87 (0.59–1.26)
Laceration
 NoneRef.
 First or second degree0.92 (0.80–1.04)
 Third or fourth degree0.81 (0.62–1.05)
Postpartum hemorrhage or anemia0.96 (0.82–1.13)
Birthing parent obstetric complication0.90 (0.82–0.99)
  • Note: CI = confidence interval, CPAP = continuous positive airway pressure, GA = gestational age, OR = odds ratio, Ref. = reference category.

  • * The final step of the model is presented; 2668 newborns were omitted owing to missing data. The logistic regression was significant over null hypothesis (χ2(26) = 499; p < 0.001), and a small improvement over the first model (Table 3) (χ2(4) = 96; p < 0.001) with an Nagelkerke R square = 0.078.