Data centered on antibiotics usage and their determinants in African pediatric populations are limited. In order to define the determinants of antibiotics prescriptions (ABPr), we analyzed the data of a birth cohort in Benin. From 2007 to 2009, 538 infants were followed from birth to 18 months in three different health centers. The following determinants were assessed: infants' clinical findings at consultations, mothers' and children's characteristics at birth, and health parameters recorded at scheduled follow-up of general health parameters. Multilevel logistic models were performed for analysis. Among the 4394 consultations, fever represented 53.7 % of consultations, 64.1 % of which were non-malarial fevers. Antibiotics were prescribed during 44.2 % of the consultations and the proportion of ABPr differed significantly among health centers (p < 10(-3)). Nearly 40 % of ABPr were related to children without fever. During the first semester of life, the percentage of ABPr was twice lower than after (27.4 vs. 54.7, p < 10(-3)). Respiratory and enteric symptoms were positively associated with ABPr (p < 10(-3)). Malaria was significantly associated with a lower ABPr after the first semester [odds ratio (OR) = 0.55, 95 % confidence interval (CI) = 0.44-0.67, p < 10(-3)]. No maternal and child at-birth characteristics were associated with ABPr. ABPr was positively associated with a low breastfeeding score (p < 10(-3)). Studies on the rational use of antibiotics in this population should give priority to children more than 6 months of age, without malaria, and with respiratory and/or enteric symptoms. Our data also advocate for studies specifically designed to assess and improve healthcare providers' compliance to guidelines on antibiotics usage.