Abstracts

Rationale: Epilepsy is a neurological chronic disorder that requires daily compliance to anti-epileptic drugs (AEDs) to achieve optimal seizure control. Missed doses can lead to severe consequences. The purpose of this study is to study the association of non-compliance to AEDs with emergency room (ER) visits, hospitalization, and mortality among epileptic patients.Methods: A retrospective cohort design using state Medicaid claims data from Florida, Iowa and New Jersey in the period of 01/97-06/06 was employed. Patients aged ≥18 with ≥1 diagnosis of epilepsy and ≥1 AED dispensing were included. Medication possession ratio (MPR) was used to evaluate AED compliance for each quarter during the observation period (i.e., date of first AED dispensing after epilepsy diagnosis to first occurrence of death, last Medicaid eligibility date, defined study end date, or for hospitalizations and ER visits, last AED dispensing date plus days supplied). Quarters with MPR ≥0.8 were marked as compliant and <0.8 as non-compliant. The association of AED compliance with mortality was assessed using a Cox proportional hazards model that included a time-varying covariate for AED compliance. Incidence rates for hospitalizations and ER visits during compliance and non-compliance quarters were calculated and compared using incidence rate ratios and their 95% confidence intervals calculated based on Poisson distribution.Results: A total of 89,033 epileptic patients with at least one filled prescription for an AED were identified with mean (SD) age of 47.0 years and 56.4% female. Together, they contributed 606,185 (63%) compliance and 358,348 (37%) non-compliance quarters, with mean (SD) MPR being 0.96 (0.05) during compliant quarters and 0.36 (0.29) during non-compliant quarters, respectively. There were 374,227 ER visits and 205,399 hospitalizations occurring during compliant quarters, and 252,425 ER visits and 134,134 hospitalizations during non-compliant quarters. Non-compliant patients had a significantly higher incidence of ER visits and hospitalizations when compared to compliant patients. Preliminary unadjusted results also suggest that non-compliance to AEDs may be associated with an increased risk in mortality (hazards ratio=2.03, p<0.001), however multivariate regression analyses to adjust for demographic and disease-specific characteristics are currently ongoing as Phase II of the research.Conclusions: Based on this retrospective univariate analysis of 89,033 Medicaid participants with epilepsy, non-compliance to AEDs is significantly associated with increased risk of hospitalization and ER visit. These findings suggest that improving AED compliance among epilepsy patients may positively impact resource utilization and potentially mortality. Phase II of this research will more fully demonstrate the association between AED non-compliance and mortality and will be available in July 2007. (Study was sponsored by GlaxoSmithKline)