Treatment patterns of antiepileptic drugs and economic outcomes in patients with potential refractory epilepsy in the Texas Medicaid Program

The purpose of the study was to characterize and compare demographic and clinical characteristics, treatment patterns (i.e., medication adherence, persistence, addition, and switching), and healthcare utilization and cost (i.e., all-cause and epilepsy-related) associated with refractory or non-refractory epilepsy. The study used Texas Medicaid claims data from 09/01/07-12/31/13. Prescription and medical service claims of eligible patients analyzed over a 30-month study period comprised of a 6-month pre-period (baseline) and a 24-month follow-up period (annual increments). Patients eligible for the study: 1) were between 18-62 years of age, 2) had a prescription claim for an antiepileptic drug (AED) during the identification period (03/01/08-12/31/11) with no baseline use of an AED and no prophylactic use of an AED at follow-up, and 3) had evidence of epilepsy diagnosis during the study period. Additionally, patients had to be continuously enrolled in Texas Medicaid with no dual eligibility for Medicare and Medicaid. The index date for both the cohorts was the date of the first AED claim. Dependent variables included: treatment patterns, healthcare utilization and cost. The primary independent variable was group (i.e., refractory vs. non-refractory epilepsy). Based on clinical expert opinion and the literature, patients were categorized as “refractory” (i.e., three or more AEDs, excluding diazepam, in the identification period) and “non-refractory” (i.e., less than three AEDs in the identification period). The covariates included age, gender, race/ethnicity, type of epilepsy, type of index AED, baseline CCI, number of psychiatric comorbidities and presence of non-psychiatric comorbidities at follow-up, baseline pill burden, presence of baseline all-cause inpatient visits, baseline number of all-cause outpatient visits, and baseline all-cause total cost. Using a retrospective matched-cohort design, patients in the refractory cohort were matched 1:1 to patients in the non-refractory cohort using propensity scoring. The matched cohorts were compared for treatment patterns and healthcare utilization and costs using multivariate conditional regression models and non-parametric methods. Of the 10,599 eligible patients, 2,789 (26.3%) patients in the refractory cohort were matched 1:1 to patients in the non-refractory cohort for a total of 5,596 patients. Mean (± SD) age of the patients in the matched cohort was 38.0 (± 13.1) years, and the cohort was comprised of a higher proportion of females (56.0%), Caucasians (41.9%), patients with other convulsions (77.2%), and those with claims for sodium channel blockers (35.4%). A higher proportion of patients with refractory epilepsy were initiated on combination AEDs (26.5% vs. 10.7%), followed by GABA analogues (12.0% vs. 10.2%), and calcium channel action agents (7.7% vs. 3.4%) compared to patients with non-refractory epilepsy. During the second year of follow-up, patients with refractory epilepsy had a higher mean (± SD) (2.1 [± 1.5] vs. 1.8 [± 1.4]) number of psychiatric comorbidities, and a higher proportion (51.3% vs. 41.4%) of patients had one or more non-psychiatric comorbidities compared to patients with non-refractory epilepsy. Regarding treatment patterns, compared to patients with non-refractory epilepsy, patients with refractory epilepsy were 3.6 times (OR=3.553; 95% CI=3.060-4.125; p<0.0001) more likely to adhere to AEDs and had a 34.7% (HR=0.653; 95% CI=0.608-0.702; p<0.0001) lower hazard rate of discontinuation of AEDs during the second-year of follow-up, after controlling for covariates. Among those patients on two or more AEDs, patients with refractory epilepsy were 3.7 times (OR=3.723; 95% CI=2.902-4.776; p<0.0001) more likely to add an alternative AED and 3.6 times (OR=3.591; 95% CI=3.010-4.284; p<0.0001) more likely to switch to an alternative AED during the first-year of follow-up, after controlling for covariates. Regarding healthcare utilization and costs during the second year of follow-up, compared to patients with non-refractory epilepsy, patients with refractory epilepsy had a significantly higher number of all-cause outpatient visits (p<0.0001) and pharmacy claims (p<0.0001), higher epilepsy-related inpatient hospital and emergency department (ED) visits (p<0.0001), outpatient visits (p<0.0001), and pharmacy (p<0.0001) claims, after controlling for covariates. Consequently, these patients incurred higher costs for all-cause outpatient visits (p=0.0190) and pharmacy claims (p<0.0001), and higher costs for epilepsy-related inpatient hospital and ED visits (p<0.0001), outpatient visits (p<0.0001), and pharmacy (p<0.0001) claims, after controlling for covariates. Although a majority of the estimates were higher than the second year of follow-up, a similar trend in results was observed during the first-year of follow-up. In conclusion, findings from this study provide evidence for the dynamic patterns of AED use in clinical practice and provide current estimates of the resource utilization and costs associated with Texas Medicaid patients with refractory epilepsy. Management of epilepsy extends beyond the control of seizures and encompasses improvement in overall burden of the disease. As the costs in the second year were lower than in the first year, timely identification and early treatment optimization may help prevent long-term clinical and economic consequences associated with refractory epilepsy.