The Value and Limitations of Pregnancy Drug Registries
Janet D. Cragan, Centers for Disease Control and Prevention, Food and Drug Administration, Atlanta, GA 30341
Interest in establishing pregnancy registries to improve post-marketing surveillance for prescription drugs used by women of reproductive age has increased among pharmaceutical manufacturers. Established registries include those for acyclovir/ valacyclovir, antiretroviral drugs, antiepileptic drugs, sumatriptan/naratriptan, and bupropion. These registries prospectively record voluntary reports by health-care practitioners of patients exposed prenatally to the drug of interest. The presence of major defects is then obtained from the same practitioners after delivery, and defect rates are compared with those reported for the general public. The Acyclovir Pregnancy Registry is the largest, begun by Glaxo Wellcome in 1984. As of July 31, 1998, the proportion of births with defects in the registry was 28/1045 (2.7%, 95%CI:1.8-3.9%). No increased risk or unique pattern of defects was identified. Strengths of the registry approach include its prospective nature, which minimizes selection bias, and its statistical power to detect a defect resulting from a specific drug exposure. Weaknesses include the lack of power due to small sample sizes, lack of detail about exposures and outcomes, and lack of a defined comparison group. Additional issues include loss to follow-up and data confidentiality.