Early-onset neonatal infection (EONI), occur within 7 days of birth. They are most often due to Streptococcus B (GBS) and are associated with heavy and costly morbidity and mortality. The strategy combining antenatal detection (PV9) of GBS colonization and intrapartum antibiotic therapy has led to a spectacular decrease in the number of GBS EONI's that have become rare (0.3/1000 births). Current detection is based on the culture of a vaginal swab taken between 35 and 38 SA. Because the positive predictive value of PV9 compared to a culture on the day of delivery is 60%, two problems persist: i) 20% of women and newborns are sometimes unnecessarily exposed to antibiotics with known short-term and long-term harmful effects; ii) more than half of newborns developing EONI are born to mothers with negative PV9. There is a risk of not treating intrapartum colonization when PV9 is negative, and overtreating an uncolonized PV9-positive woman at the time of delivery. These inappropriate antibiotic therapies generate additional maternal-fetal care, examinations, treatments and hospitalizations with significant costs.

Today, a feasible, rapid, sensitive (90-95%) and specific (95-98%) PCR test (Xpert GBS, CEPHEID) can be used to detect women colonized with GBS at the beginning of labor. A recent study (submitted for publication) including 782 women with risk factors for infection (intrapartum fever or prolonged rupture of membranes) who were subjected to PV9 and intrapartum PCR (IP PCR), identified 19% potential reclassification of GBS status, with a potential saving of 6% intrapartum antibiotic.

We postulate that the replacement of PV9 by the generalized use of GBS intrapartum detection would optimize the indications for intrapartum antiobiotherapy, avoiding (i) unnecessary and deleterious care consumption in the absence of intrapartum GBS colonization, and (ii) avoidable EONIs occurring in the absence of intrapartum antiobiotherapy when GBS colonization has not been diagnosed.

We propose to conduct a cost-consequence study because the criteria for clinically relevant judgments do not allow for cost-effectiveness or cost-utility analysis. Indeed, the intrapartum PCR strategy has consequences for both mother and child and these consequences cannot be aggregated.

Thus, cost-consequence analysis based on criteria validated by clinicians and the literature seemed to us to be the most pragmatic approach and the most likely to help public decision making.

The objective of this work is therefore to carry out a cost-consequence analysis comparing the intrapartum antibiotic prophylaxis strategy based on intrapartum GBS colonization screening by PCR, with the current strategy based on antenatal screening by culture between 35 and 38 SA.

• Inclusion visit : The patient is included in the "GBS culture" group or in the "GBS PCR" group determined by the randomization of the hospital in which she gives birth.

A sample to perform the GBS PCR will be taken or not depending on the strategy in place in the center at the time of inclusion. The patient's management will be adapted according to the test result. The administration of intrapartum antibiotic therapy will be decided by the clinician in charge of the patient based on the result of antenatal GBS screening (PCR or culture) and taking into account the risk factors for EONI. In the GBS PCR group, the clinician will not have access to the result of the antenatal detection, even if the result is forgotten or technically impossible.

• M1 follow-up visit (+/-5 days) by phone The study coordinator will follow up the patient by phone.

The data collected are as below:

• Neonatal infection occurring within the first 6 days of life (this diagnosis requires hospital treatment by intravenous route, and will be verified on the hospitalization report),
• Endometritis in the first month after birth,
• Maternal and/or newborn antibiotic therapy (which one? how long? in which indication?)
• Hospitalization of the mother and/or newborn (which hospital, which department, how long? for what reason).
• Hospitalization report.

Inclusion Criteria:

• Patients admitted to the delivery room from 35SA upwards
• Planned vaginal delivery
• Patients who agreed to participate in the study and gave oral consent
• Patient affiliated to a social security system

Exclusion Criteria:

• Complete dilatation (imminent delivery)
• Scheduled caesarean
• Term < 35 SA
• Death in utero
• Medical termination of pregnancy
• Does not speak French
• Opposition to participating in the study

women of childbearing age