Patients whose epilepsy was uncontrolled had more injuries, spent more time in the hospital, received more prescriptions, and incurred greater health care resource usage and costs than did patients whose epilepsy was under control in a retrospective, longitudinal matched-cohort study.

Additionally, uncontrolled epilepsy in private insurance patients incurred nearly $2,900 more in work time lost for disability and sick leave, comprising about one-fifth of these patients’ total direct health care costs.

Lead author Ranjani Manjunath of GlaxoSmithKline and her associates analyzed public and private insurance claims to determine how health care resource utilization and costs and epilepsy-related injuries differed between patients with uncontrolled epilepsy and those with well-controlled epilepsy. The findings were reported online Oct. 17 (Neurology 2012;79:1908-16).

The researchers defined patients with uncontrolled epilepsy as those who had at least two consecutive changes in their antiepileptic drug (AED) regimen (at least a month apart) and at least one subsequent hospitalization or emergency department visit within the next year. Patients taking AEDs but who had no changes in their therapy as well as no hospitalizations or ED visits were defined as having well-controlled epilepsy.

A total of 3,454 Medicaid patients and 602 private insurance patients with uncontrolled epilepsy were matched 1:1 with well-controlled epileptic patients for the study. Propensity score matching was used to reduce sample selection bias, and adjusted risk ratios for outcomes took into account age, sex, state or region, baseline AED use, use of other drugs known to increase seizure risk, baseline costs, and psychiatric conditions or epilepsy-related comorbidities (Alzheimer’s disease, brain tumor, meningitis, migraine, and stroke). A lower percentage of uncontrolled epilepsy patients were using AEDs at baseline: 48.1% of uncontrolled Medicaid patients, compared with 52.6% of well-controlled Medicaid patients (P less than .001), and 40% of uncontrolled private patients, compared with 45.5% of well-controlled private patients (P less than .05).

The public patient records were pulled from 110,312 Medicaid patient claims from Florida, Iowa, Kansas, Missouri, and New Jersey, with varying year ranges within the time period of 1997-2009. The private claims came from 36,529 patients with private insurance through their self-insured employers, and were gathered from Ingenix Employer records from 1999 to 2008.

The researchers selected only adult patients who had a prescription for an AED and who had been diagnosed with epilepsy or diagnosed with two nonfebrile seizures more than a month apart. All patients had been enrolled in their insurance plan for at least a year, and baseline included the 180 days before the patient’s first AED prescription. Patients were tracked until death, the end of continuous enrollment, or the end of the period for which data were available.

The study found that Medicaid and private patients with uncontrolled epilepsy had similarly higher incidence rate ratios (incidence rate defined as events divided by patient-years) of injury with 1.9-2.2 times more fractures and head injuries. Medicaid patients had 2.45 times more car accident injuries and 10 times more status epilepticus episodes.

Compared with patients who had well-controlled epilepsy, Medicaid patients had 1.94 times more AED prescriptions, 1.47 times more non-AED prescriptions, 6.65 times more hospitalizations, 7.72 times more days in the hospital, 3.67 times more ED visits, 1.66 times more outpatient services, and 3.09 times more neurologist visits. Likewise, private patients received 1.75 times more AED prescriptions, 1.34 times more non-AED prescriptions, 5.37 times more hospitalizations, 7.27 times more days in the hospital, 5.05 times more ED visits, 1.41 times more outpatient services, and 2.28 times more neurologist visits. (All P values were less than .05.)

Overall, Medicaid patients with uncontrolled epilepsy incurred an average $12,258 more in costs than Medicaid patients with well-controlled epilepsy, and uncontrolled epileptic private patients’ total extra cost was $14,582. Most of these costs came from hospitalization; outpatient services and prescription drugs comprised the next largest share. Private patients also missed 2.5 times more workdays, including 61% more sick days.

The study was limited by the conservative definition of epilepsy, the potential for unobservable confounders between cohorts, and the possibility that more uncontrolled patients existed who were not included because they did not require hospitalization or an ED visit.

The study was funded by GlaxoSmithKline, where Dr. Manjunath and another author are full-time employees. Five others are full-time employees of Analysis Group, which conducted the study with GSK’s grants, and another consults for Analysis Group.