Study population and data
EUROmediCAT is a network dedicated to the study of medication safety in pregnancy which includes EUROCAT population-based CA registries collecting data on medication exposure during pregnancy. Currently 20 EUROCAT registries from 14 countries participate in EUROmediCAT with annual surveillance covering approximately 753,000 births per year throughout Europe (www.euromedicat.eu)32.
The registries collect and send anonymised individual case data to the EUROmediCAT Central database. The standard data on each registration are described in EUROCAT Guide 1.4.31 One syndrome and up to eight malformations are coded using International Classification of Diseases, Ninth Revision (ICD-9) or Tenth Revision (ICD-10) codes with British Pediatric Association (BPA)/RCPCH one-digit extensions. Registrations with only anomalies on the EUROCAT list of minor anomalies are excluded leaving registrations with one or more major malformations. This list of minor anomalies includes also anomalies of non-congenital origin including pyloric stenosis, and Patent Ductus Arteriosus in preterm births. All registrations are classified to EUROCAT subgroups according to their ICD9/10-RCPCH codes, as set out in EUROCAT Guide 1.4.1 The sources used to obtain data on CA vary across registries and include maternity, neonatal, and paediatric records; fetal medicine, cytogenetic, pathology, and medical genetics records; paediatric cardiology services; and hospital discharge and child health records.33 The majority of the registries record anomalies diagnosed up to at least one year of age.
Medications taken in the first trimester of pregnancy are coded according to the Anatomic Therapeutic Chemical (ATC) classification. Maternal disease before and during pregnancy are recorded using ICD codes. Most registries (84%) collect prospective data on medication exposure during the first trimester, mainly from maternity and other medical files, and some collect retrospective data from interviews with women after delivery or other sources, either alone (17%) or in combination with prospective sources (28%)34. All participating registries reviewed antibiotic exposures to confirm that they were first trimester exposures.
All registries recording information on antibiotic exposures (regardless of timing), with data over the period of this study (1995-2012), were eligible to participate in the study. Eighteen of the 23 (at start of study) eligible registries agreed to participate in the study and gave permission for their data to be extracted.