Abstracts

Rationale: More than one-third of epilepsy patients are unable to achieve adequate seizure control with their current antiepileptic drug (AED) regimen and can suffer from breakthrough seizures. The difference in health care resource utilization for patients with breakthrough seizures vs. patients without breakthrough seizures is not well understood.Methods: A retrospective case/control analysis was conducted using U.S. administrative claims from the IMS PharMetrics database. We identified adults with 1 diagnosis for epilepsy (ICD-9-CM: 345.x) between January 1, 2006 and March 31, 2011. Seizures were defined as an ER or inpatient admission with a primary diagnosis of epilepsy. Cases (patients) were identified as having a breakthrough seizure (i.e., index event) if they experienced a seizure while adherent to AED therapy (defined as MPR 0.8) and no history of seizure in the 6-months pre-index (indicating a period of seizure control). Controls had no seizures in the entire study period and met the AED adherence requirements; an outpatient epilepsy claim was randomly selected as their index date. All patients were required to have 1 year pre- and 6 months post-index continuous enrollment. Cases/controls were matched 1:1 using propensity scoring on characteristics including age, gender, payer type, pre-index comorbidities, and pre-index health care costs. All-cause and epilepsy-related (ICD-9-CM 345.x [epilepsy] or 780.3 [convulsions]) resource utilization for cases/controls were compared in the 6-months post index; statistical comparisons were conducted on the median for continuous variables. Results: 20,166 epilepsy patients were eligible for analysis (5,729 cases; 14,437 controls); the final matched sample included 5,279 cases and 5,279 controls. Unmatched cases had a mean of 1.4 seizures during the 6-month follow-up period; the majority (74.8%) had evidence of a single breakthrough seizure (17.9% had evidence of 2 seizures; 7.3% had evidence of 3 seizures). In the 6-month follow-up period, cases experienced significantly higher rates of all-cause hospitalizations and ER visits compared to matched controls (60.5% vs. 9.1%, p<0.0001 and 64.9% vs. 19.2%, (p<0.0001) with at least 1, respectively). Significantly more cases experienced at least 1 epilepsy-related hospitalization and epilepsy-related ER visit compared to controls (59.2% vs. 4.6%, p<0.0001 and 57.4% vs. 6.0%, p<0.0001, respectively). Cases had more all-cause and epilepsy-related outpatient visits per person compared to controls (mean 8.6 vs. 7.6 [median 6.0 vs. 5.0, p<0.0001] and mean 3.2 vs. 1.9 visits [median 2.0 vs. 1.0, p<0.0001], respectively). All-cause resource use increased as the number of breakthrough seizures increased (see Table 1).Conclusions: In this study, AED-adherent epilepsy patients with breakthrough seizures had substantially higher use of ER and inpatient care compared to matched controls, and utilization increased with each subsequent seizure. Better seizure control could potentially reduce the need for costly hospital and ER visits.