Abstracts

Rationale: Adverse drug reaction profiles of antiepileptic drugs (AEDs) are as important as efficacy of AEDs. Adverse drug reactions (ADRs) are reported by clinical trials before drugs come to market or voluntary drug adverse event reporting systems. However, it is a challenge to predict adverse reaction rates observed in clinical practice. The Observational Health Data Sciences and Informatics (OHDSI) collaboration provides a Common Data Model (CDM) for standardizing electronic health record data into a common format to support large-scale observational studies. The objective of the study is to 1) develop an OHDSI CDM towards building a pharmacovigilance detection platform for AEDs, and 2) evaluate the feasibility and utility of OHDSI CDM for ADR monitoring in a single institution. Methods: Electronic health records required by the CDM were extracted and transformed into OHDSI CDM structure. Patients who have a diagnosis of epilepsy and exposed to AED (valproate, topiramate, oxcarbazepine, levetiracetam, lamotrigine) monotherapy for at least 6 months were included. We evaluated 4 domains of ADRs related to antiepileptic drugs: 1) hematological alteration, 2) hyponatremia, 3) abnormality in liver function test, 4) subclinical hypothyroidism. Results: Our dataset contained 1344 patients. Thrombocytopenia occurred in 14.1% of valproate monotherapy group. Mild and severe hyponatremia were reported in 9.1% and 0.4% of oxcarbazepine monotherapy group, respectively. Mild and severe degree of elevation in liver enzyme were reported in 18.4% and 0.9% of valproate monotherapy group, respectively. Of 550 patients, 134 patients (24.4%) had subclinical hypothyroidism. Conclusions: Our observational study from a single institution demonstrates that ADRs of antiepileptic drugs using the OHDSI CDM are feasible. Standardized OHDSI CDM platform can be used as a valuable tool for monitoring nationwide antiepileptic drug safety surveillance. Funding: Funded by Seoul National University Bundang Hospital