The incidence of neonatal group B streptococcus (GBS) related to early onset sepsis (EOS) has decreased dramatically with antenatal culture screening and intrapartum antibiotic prophylaxis (IAP) implementation during the 1990s. However, while 30 % of women are currently exposed to intrapartum antibiotics, 52 to 82 % of newborns with GBS-EOS were born to mothers with negative antenatal GBS culture screening who did not receive IAP, because of possible intermittent maternal GBS colonization. Intrapartum GBS screening by Polymerase Chain Reaction (PCR) has been suggested as a suitable tool to improve and replace the usual antenatal GBS culture screening between 35 and 38 weeks of amenorrhea for optimizing the IAP indications, but its cost is often cited as an obstacle. Using a cluster crossover randomized controlled trial, we aim i) to conduct a cost-consequence analysis of two screening strategies and ii) to compare newborn and maternal morbi-mortality related to GBS-EOS and antibiotic exposure rate according to an intrapartum screening strategy versus usual antenatal culture screening (DEPIST2P study).