Data used in this study.

Background
Early initiation of breastfeeding within one hour of birth can be predicted by socio-demographic and clinical characteristics of women.The relative influence of many socio-demographic and clinical characteristics on early initiation of breastfeeding has not been established in Sri Lanka. Identifying the significant predictors of early initiation of breastfeeding in Sri Lankan women may influence the Sri Lankan health system to review and renew policies and practices to support women who are at risk of delayed initiation of breastfeeding.
Methods
A cross-sectional survey was conducted with women who had a live baby across selected hospitals in Sri Lanka. Women who were 12 hours post-birth and still admitted to the hospital were invited to participate. Data were collected by interviewing participants and extracting additional clinical information from their medical records. The contribution of socio-demographic and clinical characteristics to explaining variations in early initiation of breastfeeding was estimated using binary logistic regression analysis with simultaneous adjustment.
Results
The rate of early initiation of breastfeeding was 64.5% (n = 195). Participants with a body mass index under 18.5 kg/m2 (OR: 4.17; 95% CI: 1.23–14.05) were more likely to practice early initiation of breastfeeding, and less likely if they gave birth by elective caesarean section (OR: 0.27; 95% CI: 0.14–0.51), were administered antibiotics (OR: 0.4; 95% CI: 1.05–4.73) or had a baby with respiratory distress (OR: 0.08, 95% CI: 0.01–0.57).
Conclusion
Current intrapartum care practices associated with elective caesarean and antibiotic administration, and immediate care of babies with respiratory distress, should be critically reviewed to understand the mechanisms underlying their negative impact on breastfeeding initiation in Sri Lanka. Compliance with breastfeeding support care should be monitored to provide equal care for women,  minimising discrepancies in early initiation of breastfeeding associated with clinical circumstances.