LAUSR

OBJECTIVE Certain antiepileptic drugs (AEDs) may be more suitable for elderly patients with epilepsy (EWE) relative to others. However, little is known regarding which antiepileptic drugs (AEDs) are being used to treat EWE in the United States and how it has changed over time. METHODS We performed a serial cross-sectional study evaluating noninstitutionalized US adults aged 65 years or older with a diagnosis of epilepsy using data from the Medical Expenditure Panel Survey (MEPS) from 2004 through 2015. Trends in AEDs used among EWE were examined. Using each AED as a dependent variable, we determined the p-value for the trend by performing a linear regression with the time interval as the explanatory variable. RESULTS There was a weighted total of 399,801 EWE. Between the years 2004-2006 and 2013-2015 use of phenytoin, carbamazepine and phenobarbital decreased from 60.7% to 31.1% (p ≤ 0.001), 13.7 % to 5.22 % (p = 0.03) and 12.5 % to 5.91 % (p = 0.04), respectively. Use of levetiracetam concomitantly increased from 6.70 % to 43.1 % (p ≤ 0.001). Patients with more medical comorbidities as measured by the Charlson Comorbidity Index had higher odds of levetiracetam use (OR = 2.52, 95 % CI = 1.19-5.34) and lower odds of phenytoin use (OR = 0.46, 95 % CI = 0.24-0.88). CONCLUSIONS There have been significant changes in AED prescriptions to EWE between 2004-2015. However, potentially harmful AEDs (e.g. phenytoin, carbamazepine, phenobarbital, primidone and valproate) were still being prescribed to 42.9 % of all patients between 2013-2015. Increased work to educate providers regarding the use of more appropriate AEDs in this population is needed.