Association of antiepileptic drug nonadherence with risk of seizures in adults with epilepsy

Abstract

This study evaluated the potential effect of antiepileptic drug (AED) nonadherence on the risk of subsequent seizure. Retrospective insurance claims from the United States were analyzed. Inclusion criteria were: age 21–64 years, diagnosis of epilepsy or nonfebrile convulsions, ⩾2 AED prescriptions, and insurance enrollment for ⩾6 months pre- and ⩾60 days post-AED initiation. Seizure was defined as a hospital or emergency admission associated with epilepsy or nonfebrile convulsions. Observation began 7 days post-drug initiation, ending with the first of the following: seizure, insurance disenrollment, or 365 days post-drug initiation. Adherence was measured using the medication possession ratio (MPR), with MPR <0.8 defining nonadherence. Seizure risk was assessed using an extended Cox proportional hazards model. Of 18,073 subjects identified, 2467 (14%) had ⩾1 seizure. Mean follow-up was 133 days among subjects with event and 305 days for patients without event. Seizure risk was 21% higher among nonadherers (hazard ratio = 1.205, P = 0.0002) than adherers.

Introduction

Epilepsy carries heightened morbidity and increased risk of mortality compared with the general population [1], [2], [3]. Lack of seizure control is believed to be a primary driver in the underlying increased morbidity and mortality in this population [4]. Uncontrolled seizures have been associated with physical injuries such as fractures, head injuries, and burns; psychosocial comorbid conditions, such as depression and anxiety; reduced quality of life; and sudden death in epilepsy [5], [6]. Prolonged seizures also can lead to neuronal death and cognitive impairment.

There are many disease-based and treatment-related factors that can contribute to poor seizure control among patients with epilepsy, including age at disease onset, seizure etiology, seizure type, epilepsy syndrome, comorbid conditions, and nonadherence to antiepileptic drugs (AEDs). Because nonadherence may be a modifiable factor in seizure control, there has been research to understand its role. An early study found that 31% of seizures may be attributable to medication nonadherence [7]. Jones and colleagues [8] showed a correlation between nonadherence and increased seizure frequency, and an earlier study [9] suggested that 45% of patients reported a seizure after missing a dose of medication.

Medication nonadherence, particularly when associated with increased seizures, can further amplify the morbidity and mortality associated with epilepsy. Nonadherence can be evaluated in a variety of ways, including electronic monitoring caps, patterns of prescription refills observed from pharmacy claims data, and patient-reported survey data. As a result, estimates of medication nonadherence in epilepsy have ranged broadly from approximately 30% to 60% [8], [10], [11], [12]. In a recent claims database study, AED nonadherence was estimated at 39% and associated with increased emergency department visits and hospitalizations, as well as increased total health care costs in adults with epilepsy [13]. In another recent study conducted in Medicaid enrollees with epilepsy, nonadherence was associated with a greater than threefold increased risk of mortality compared with adherence [14].

Previous literature has signaled an association between nonadherence and poor seizure control. However, this association has not been widely investigated using large-scale observational data from real-world practice settings nor approached from a temporal perspective. To address these research gaps, we conducted a retrospective observational study to measure the temporal association between nonadherence to AEDs and the risk of seizure in a large cohort of adult patients diagnosed with epilepsy. Knowledge of this relationship is valuable in aiding clinicians in identifying time periods of patient nonadherence and risk of seizures when managing patients with epilepsy.