Background: Breastfeeding benefits both mother and infant. U.S. breastfeeding initiation rates are steadily increasing, but duration rates still fall short of recommendations. Postnatal depression has a significant negative impact on breastfeeding duration. Studies show grandmothers may play an important role in infant feeding decisions. Objective: Evaluate the relationship of maternal postnatal depression and social influence (SI) of infant’s grandmothers on breastfeeding duration (weeks). Methods: Data from the Infants Feeding Practices Study II (2005-2007) were analyzed using Cox regression (survival analysis). SI was measured as each grandmother’s opinion of infant feeding weighted by the importance of her opinion to the participant (range=-4.00, 4.00). Possible postnatal depression was indicated by a score of at least 10 on the Edinburgh Postnatal Depression Scale. Participants were excluded if they reported not having a mother or mother-in-law. Results: Most women in this study (N=1,846; N=429 censored), were White (87.5%), married (78.7%), educated beyond high school (83.0%), employed (53.4%), had breastfed previous children (94.0%) and did not have a score indicating postnatal depression (76.4%). Average breastfeeding duration was 21.00 (SD=17.35; Range=1-65) maternal grandmother SI score was 1.30 (SD=1.78), and paternal grandmother SI score was 0.80 (SD=1.45). In bivariate analysis, scores indicative of postnatal depression (HR=1.16, 95%CI: 1.02-1.33), maternal grandmother’s SI (HR=0.87, 95% CI: 0.85-0.90) and paternal grandmother’s SI (HR=0.91, 95% CI: 0.87-0.95) were respectively associated with an increased and decreased hazard of stopping breastfeeding. However, in the final model including demographic covariates along with both SI and depression variables, only maternal grandmother’s SI remained significantly associated with breastfeeding duration (HR=0.88, 95% CI:0.84-0.92). Conclusions: Pro-breastfeeding SI of maternal grandmothers is positively associated with longer breastfeeding duration and remains significant when controlling for postnatal depression. Future research and interventions should leverage the SI from maternal grandmothers by including them in the conversation about breastfeeding practice; thus, providing a cross-generational impact.