Investigating outcomes associated with medication use during pregnancy: a review of methodological challenges and observational study designs.

In the absence of randomised controlled trials, knowledge of outcomes associated with medication use during pregnancy is dependent on observational studies. Numerous observational study designs exist, with the decision on which is most appropriate depending on a number of factors, including the exposure and outcome under investigation and knowledge of key methodological issues. This review provides an overview of the key methodological issues involved in undertaking observational studies to investigate medication use during pregnancy, including selection bias, exposure and outcome classification, information bias, confounding and statistical analysis. This review also discusses observational study types used to investigate outcomes associated with medication use during pregnancy and summarises their relative strengths and weaknesses. Knowledge of the strengths, weaknesses and methodological issues associated with observational studies can assist clinicians in making assessments about the validity and applicability of results presented in order to provide the best recommendations to patients.