Drug use during pregnancy in Sweden – assessed by the Prescribed Drug Register and the Medical Birth Register

Abstract

Olof Stephansson, Fredrik Granath, Tobias Svensson, Bengt Haglund, Anders Ekbom, Helle KielerClinical Epidemiology Unit and Centre for Pharmacoepidemiology, Department of Medicine, Karolinska University Hospital and Institutet, Stockholm, SwedenPurpose: The purpose of this research is to study drug use during pregnancy in Sweden and agreement between use according to antenatal medical records and dispensed drugs from a pharmacy database.Patients and methods: From the Swedish Medical Birth Register (MBR), we established a population-based cohort of 102,995 women who gave birth in 2007. Using the unique personal registration number, information on dispensed drugs from the Prescribed Drug Register (PDR) was obtained prior to, during, and after the pregnancies and compared with MBR information on drug use from standardized antenatal medical records.Results: According to the PDR, 57.6% of the 102,995 women filled a prescription with at least one drug during pregnancy and 50.9% during the lactating period (until 3 months after delivery). The most dispensed drugs during pregnancy were B-lactam antibacterials and penicillins. Agreement between drugs recorded in antenatal medical records and dispensed drugs was highest for drugs used for chronic conditions. The agreement was particularly high for thyroid therapy (85.3%), anti-intestinal inflammatory drugs (80.3%), antiepileptics (69.2%), immunosuppressants (67.4%), and insulin (63.8%). Agreement for drugs used for occasional use was generally lower, ranging between 42.5% for antihistamines and 0.8% for gynecological anti-infectives.Conclusions: A large proportion of women filled a prescription during pregnancy or the lactating period. Agreement between drug use in medical antenatal records and register information from a national pharmacy database was high for drugs used for chronic conditions but low for occasional use. For occasionally used drugs, medical record and register-based data may provide incomplete exposure information because of nonreporting or noncompliance.Keywords: drug utilization, pregnancy, lactation, pharmacoepidemiology