RESUMEN

PURPOSE: To examine the rates and determinants of breastfeeding initiation (BFI) amongst women with epilepsy (WWE) and women without epilepsy (WWoE) in Manitoba, Canada. METHODS: We conducted a retrospective cohort study using province-wide health databases from 1995 to 2019. Annual BFI rates for WWE and WWoE were examined. Multivariable logistic regression models were used to quantify the association between maternal and infant characteristics and BFI in both groups. RESULTS: During the study period, 1,331 pregnant WWE and 357,334 WWoE were examined. Among WWE, 70.9 % initiated breastfeeding compared to 81.8 % among WWoE. We observed a significant small increase in yearly trends of BFI in both WWE (ß=0.45, p = 0.008) and WWoE (ß=0.23, p < 0.001). In WWE, BFI was associated with caesarean delivery (aOR=0.72,95 % CI: 0.53-0.97), chronic pain (aOR=0.67,95 % Cl: 0.46-0.97), lower income (aOR=0.34,95 % Cl: 0.26-0.44), and gestational age (aOR= 1.09,95 % CI:1.01-1.18). In WWoE, BFI was associated with chronic pain (aOR=0.83,95 % Cl: 0.80-0.86), lower income (aOR=0.45, 95 %CI:0.44-0.46), mood and anxiety disorder (aOR=0.84,95 % CI:0.81-0.86), and gestational age (aOR=1.13,95 % Cl:1.12-1.14). The use of any ASM (aOR=0.66,95 % Cl:0.51-0.85), new generation (aOR=0.86,95 % Cl: 0.62-1.20), polytherapy (aOR=0.46,95 % Cl: 0.31-0.69) and gabapentin (aOR=0.49,95 % Cl: 0.17-1.24) reduced the likelihood of BFI among WWE. CONCLUSION: BFI was approximately 10 % lower in WWE compared to WWoE. Determinants such as low income, ASM use, and comorbidities were significant contributors to a reduced BFI in both groups. Targeted counselling for WWE on breastfeeding benefits is essential. Further research is needed to investigate breastfeeding continuation in WWE.