Direct breastfeeding during the first week after NICU discharge among preterm infants: prevalence and associated factors

Background Direct breastfeeding (DBF) offers nutritional and immunological benefits and supports the coordination of sucking, swallowing, and breathing in preterm infants, yet many preterm infants do not feed directly at the breast around the time of neonatal intensive care unit (NICU) discharge. This study aimed to describe the proportion of preterm infants who fed directly at the breast during the first 7 days after NICU discharge and to identify factors associated with this outcome. Methods This was a multicenter prospective cohort study conducted in the NICUs of two tertiary hospitals in China. Using consecutive convenience sampling, we enrolled 421 preterm infant-parent units between July 2025 and July 2026. During hospitalization, mothers independently completed a general information questionnaire, the Breastfeeding Self-Efficacy Scale-Short Form (BSES-SF), the Edinburgh Postnatal Depression Scale (EPDS), and the Self-Rating Anxiety Scale (SAS), while fathers completed the Father's Breastfeeding Self-Efficacy Scale-Short Form (FBSES-SF). Feeding status was assessed at the 1-week post-discharge follow-up and classified into three mutually exclusive categories (exclusive DBF, partial DBF, and no DBF), with exclusive and partial DBF combined into a derived binary outcome, any DBF. Binary logistic regression was used to identify factors associated with any DBF, with study center retained as a fixed effect. Results Of the 421 preterm infants, 215 (51.07%) had any DBF during the first 7 days after NICU discharge. Of these, 76 (18.05%) were exclusively directly breastfed, and 139 (33.02%) were partially directly breastfed; the remaining 206 (48.93%) had no DBF. In the adjusted binary logistic regression model ( n = 404; 213 with any DBF and 191 with no DBF), the following variables were associated with any DBF (adjusted odds ratio [aOR], 95% confidence interval [CI]): oropharyngeal colostrum administration, ≥1 vs. 0 sessions (aOR 5.58, 95% CI 2.48–12.58); severity of illness (overall P < 0.001), grade 3 vs. grade 1 (aOR 0.09, 95% CI 0.03–0.24) and grade ≥4 vs. grade 1 (aOR 0.02, 95% CI 0.01–0.07); DBF during hospitalization, ≥1 vs. 0 sessions (aOR 5.41, 95% CI 2.34–12.50); maternal educational level, bachelor's degree or above vs. junior college or below (aOR 4.78, 95% CI 1.97–11.58); monthly household income (overall P = 0.008), 10,000–19,999 CNY vs. < 10,000 CNY (aOR 4.16, 95% CI 1.54–11.19) and ≥20,000 CNY vs. < 10,000 CNY (aOR 4.35, 95% CI 1.30–14.54); anxiety score, per 1-point increase (aOR 0.90, 95% CI 0.85–0.96); and paternal breastfeeding self-efficacy, per 1-point increase (aOR 1.07, 95% CI 1.04–1.10). Severity of illness grade 2 vs. grade 1 and study center did not reach statistical significance in the adjusted model. Conclusions About half of the preterm infants in this cohort fed directly at the breast during the first 7 days after NICU discharge. In this observational study, receipt of oropharyngeal colostrum administration, DBF during hospitalization, higher maternal educational level, higher monthly household income, and higher paternal breastfeeding self-efficacy were associated with higher odds of any DBF, whereas greater severity of illness and higher maternal anxiety score were associated with lower odds of any DBF. These associations do not establish causal effects and require confirmation in interventional studies before they are used to guide care.