Objective To assess the impact of separation of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) polymerase chain reaction (PCR)-positive mother-newborn dyads on breastfeeding outcomes. Study design This observational longitudinal cohort study of mothers with SARS-CoV-2 PCR-and their infants at 3 NYU Langone Health hospitals was conducted between March 25, 2020, and May 30, 2020. Mothers were surveyed by telephone regarding predelivery feeding plans, in-hospital feeding, and home feeding of their neonates. Any change prompted an additional question to determine whether this change was due to coronavirus disease-2019 (COVID-19). Results Of the 160 mother-newborn dyads, 103 mothers were reached by telephone, and 85 consented to participate. There was no significant difference in the predelivery feeding plan between the separated and unseparated dyads (P = .268). Higher rates of breastfeeding were observed in the unseparated dyads compared with the separated dyads both in the hospital (P < .001) and at home (P = .012). Only 2 mothers in each group reported expressed breast milk as the hospital feeding source (5.6% of unseparated vs 4.1% of separated). COVID-19 was more commonly cited as the reason for change in the separated group (49.0% vs 16.7%; P < .001). When the dyads were further stratified by symptom status into 4 groups-asymptomatic separated, asymptomatic unseparated, symptomatic separated, and symptomatic unseparated-the results remained unchanged. Conclusions In the setting of COVID-19, separation of mother-newborn dyads impacts breastfeeding outcomes, with lower rates of breastfeeding both during hospitalization and at home following discharge compared with unseparated mothers and infants. No evidence of vertical transmission was observed; 1 case of postnatal transmission occurred from an unmasked symptomatic mother who held her infant at birth.