No causal relationship between transdermal scopolamine and seizures: methodologic lessons for pharmacoepidemiology.

Because of case reports suggesting that use of transdermal scopolamine might be associated with the subsequent development of seizures, a retrospective cohort study was performed with computerized Medicaid claims data. Patients receiving transdermal scopolamine were compared with patients receiving diphenhydramine, meclizine, prochlorperazine, and promethazine. A four-fold increased risk of seizures after transdermal scopolamine use was observed in the claims data. However, this was not supported by the primary medical records. All patients who had seizures after using transdermal scopolamine either had seizures before receiving the drug as well or did not really suffer from seizures. The original finding appeared to be the result of the use of transdermal scopolamine for "dizziness, rule out seizures"; the ICD-9-CM coding system does not include "rule out" diagnoses. Thus these data do not confirm the existence of an association between seizures and the use of transdermal scopolamine. In addition, this study demonstrates the usefulness of pharmacoepidemiology studies in documenting drug safety and the importance of obtaining primary medical records when performing pharmacoepidemiologic studies with claims data.