Web-based decision support system for personalized selection of antiepileptic drugs in adolescents and adults: an external validation study
This article was originally published here
Eur J Neurol. November 5, 2021. doi: 10.1111 / en.15168. Online ahead of print.
Antiepileptic drugs (ASMs) should be tailored to individual characteristics, including type of seizure, age, gender, co-morbidities, co-drugs, drug allergies, and reproductive potential. We previously developed a web-based algorithm for patient-tailored ASM selection to help healthcare professionals prescribe medication using a decision support app ( https://epipick.org). In this validation study, we used an independent dataset to assess whether the ASMs recommended by the algorithm are associated with better outcomes than the ASMs considered less desirable by the algorithm. Four hundred and twenty-five consecutive patients with newly diagnosed epilepsy were followed for at least one year after starting an ASM chosen by their doctor. Patient characteristics were fed into the algorithm, blinded to physician’s ASM choices and results. The algorithm recommended ASMs, categorized into hierarchical groups, with Group 1 ASMs labeled as the best option for this patient. We assessed retention rates, seizure-free rates, and adverse effects leading to discontinuation of treatment. Survival analysis contrasted results between patients who received preferred drugs and those who received lower tier drugs. Propensity score matching corrected for possible imbalances between groups. ASMs ranked by the algorithm as the best options had a higher retention rate (79.4% vs. 67.2%; p = 0.005), a higher rate of freedom from seizure (76.0% vs. 61, 6%; p = 0.002) and a lower dropout rate due to adverse effects (12.0% vs. 29.2%; p
PMID: 34741372 | DOI: 10.1111 / en.15168