Background: In recent years, there has been an increase in the use of automated healthcare databases for drug safety in pregnancy evaluation; their suitability for this purpose needs to be evaluated.Aim: To evaluate the utility of the United Kingdom’s General Practice Research Database (GPRD) to act as an alternative to pregnancy registries, using anticonvulsants as a case study.Methods: Pregnancies in women with epilepsy were identified and first trimester anticonvulsant exposure was determined. Major congenital malformations in the offspring were identified and verified. The risk of major congenital malformations following exposure to a range of anticonvulsants was calculated and compared to those reported by the UK Epilepsy and Pregnancy Register. The ability to identify a known teratogenic association using GPRD data was also assessed. An algorithm was created to identify and classify different types of pregnancy loss in an automated manner.Results: The risks of a pregnancy outcome with a major congenital malformation following first trimester anticonvulsant exposures, were found to be similar in the GPRD to those of the UK Register. The number of exposures to individual products in the GPRD was often small and therefore lacked statistical power. It was, however, possible to identify a known teratogenic association using data from the GPRD. Verification of the algorithm developed to classify pregnancy losses demonstrated that, although not perfect, it would be a beneficial tool when using the GPRD for drug safety in pregnancy research.Conclusion: It is unlikely a single data source or study design will be sufficient for monitoring all aspects of the safety of medicine use during pregnancy. The GPRD has the potential to make a valuable contribution to this field of research and could play an important role in complementing the work of other surveillance systems.
|Date of Award||30 Apr 2012|
|Supervisor||Corinne De Vries (Supervisor) & John Weil (Supervisor)|
- congenital abnormalities