Abstracts

Rationale: Hospitalizations for epilepsy can be influenced by multiple factors. We examined the effect on outcomes of drug-therapy related factors amenable to extraction from healthcare administrative databases. Methods: We analyzed 2 large US databases (SDI [open data set] and MarketScan [closed data set]) with healthcare claims between Nov 2009 and Sept 2011 for patients ages ≥12 years with epilepsy who had 3 years of continuous prescription data (no gap >180 days) including an antiepileptic drug (AED) prescription in the previous year. Comparisons between patients on old (eg, phenytoin, carbamazepine or valproate) and new-generation AEDs and patients with or without a drug-therapy change, were made for the number of therapy days on treatment between hospitalizations. Prescriber's specialty, patient proximity to a member center of the National Association of Epilepsy Centers (NAEC), and payer formulary inclusiveness were evaluated for impact on AED use. NAEC proximity and formulary inclusiveness were sourced from other data. Results: 17,743 claims were used in the SDI analysis which, due to its database design, focused on patients with prior hospitalizations and 144,753 in the MarketScan analysis which was prior hospitalization agnostic. In analyses, epilepsy-related hospital visits were less frequent if new-generation rather than older AEDs were prescribed. Within the SDI data (Table), the risk reduction with newer AEDs was similar for men and women and benefit increased with patient age. Patients with comorbidities were more likely to have hospitalizations; still, all patients exhibited risk reduction from using newer AEDs. A drug therapy change (add-on/switch) reduced the relative risk for hospitalizations by 18.4% (p ≤.01); greater benefit was obtained when switching from old- to new-generation AEDs (26.7% risk reduction vs 10.3% reduction when switching from new- to old-generation AED). The rate of use of new AEDs was highly correlated with treatment by a neurologist, proximity to a NAEC, and formulary inclusiveness (new-generation AEDs with preferred status or no formulary restrictions). MarketScan analysis was confirmatory of the SDI analysis. Conclusions: Utilization of new generation AEDs and drug-therapy changes lead to improved patient outcomes as assessed by the surrogate measure of hospital encounter frequency. Greater use of new generation AEDs is influenced by access to a neurologist, formulary coverage and proximity to centers of treatment excellence. Funded by UCB Inc.