Objective: To quantify the clinical and economic burden of uncontrolled epilepsy in patients requiring emergency department (ED) visit or hospitalization.
Methods: Health insurance claims from a 5-state Medicaid database (1997Q1-2009Q2) and 55 self-insured US companies ("employer," 1999Q1 and 2008Q4) were analyzed. Adult patients with epilepsy receiving antiepileptic drugs (AED) were selected. Using a retrospective matched-cohort design, patients were categorized into cohorts of "uncontrolled" (≥ 2 changes in AED therapy, then ≥ 1 epilepsy-related ED visit/hospitalization within 1 year) and "well-controlled" (no AED change, no epilepsy-related ED visit/hospitalization) epilepsy. Matched cohorts were compared for health care resource utilization and costs using multivariate conditional regression models and nonparametric methods.
Results: From 110,312 (Medicaid) and 36,529 (employer) eligible patients, 3,454 and 602 with uncontrolled epilepsy were matched 1:1 to patients with well-controlled epilepsy, respectively. In both populations, uncontrolled epilepsy cohorts presented about 2 times more fractures and head injuries (all p values < 0.0001) and higher health care resource utilization (ranges of adjusted incidence rate ratios [IRRs] [all-cause utilization]: AEDs = 1.8-1.9, non-AEDs = 1.3-1.5, hospitalizations = 5.4-6.7, length of hospital stays = 7.3-7.7, ED visits = 3.7-5.0, outpatient visits = 1.4-1.7, neurologist visits = 2.3-3.1; all p values < 0.0001) than well-controlled groups. Total direct health care costs were higher in patients with uncontrolled epilepsy (adjusted cost difference [95% confidence interval (CI)] Medicaid = $12,258 [$10,482-$14,083]; employer = $14,582 [$12,019-$17,097]) vs well-controlled patients. Privately insured employees with uncontrolled epilepsy lost 2.5 times more work days, with associated indirect costs of $2,857 (95% CI $1,042-$4,581).
Conclusions: Uncontrolled epilepsy in patients requiring ED visit or hospitalization was associated with significantly greater health care resource utilization and increased direct and indirect costs compared to well-controlled epilepsy in both publicly and privately insured settings.