We determined the effect of an early discharge program on the relationship between PCV and ER use in the first 3 months of life. Methods: Normal newborns, 997 (Early discharge (ED)) and 1171 controls (standard discharge (C)) formed the cohort. Only patients who selected a clinic for PC were included in this analysis. (ED=670 C=730). Computer and manual searches were done at specific community based clinics (CBC infants=799) and a single Children's Hospital based clinic (HBC infants=601) to determine primary care and ER use. Data were analysed using bivariable (T test, chi-sq) and multivariable analysis (Logistic regression). Maternal variables included age, race, prenatal care use and parity. Infant variables included birth weight and gestational age. Results: 35% of C and 34% ED had an ER visit during the 1st 3 months. 77% of C and 79% of ED had a PCV at the stated site. 65% of C and 69% of ED visited their primary care site prior to a visit to the ER (P=0.15, Chi-sq). The mean age (±SD) at first PCV was 20±15 days in C and 16±11 days in ED infants (P<.0001 T-test). Multiparous women, younger women and nonwhite race were significantly related to ER use. Prior primary care visit (OR=0.464, CI 0.407, 0.528 (P<.0001) was strongly predictive of ER use. Those who had a prior primary care visit had a mean age of first ER visit of 47.8 days compared to 22.8 days for those who did not(P<.0001). Early discharge did not increase the risk of ER use (OR= 1.04 CI 0.916, 1.18 P=.76) or change the relationship between prior PCV and ER use.Conclusion: There is a significant relationship between PCV and ER use in the first 3 months of life. Infants who had a prior PC visit visited the ER later. Earlier discharge can be safely achieved without increasing the use of ER service, but this requires effective coordination between the hospital and outpatient primary care clinics.