Drivers of Epilepsy-related Cost of Care in an Integrated Healthcare System (P6.280)

Objective: Determine the factors associated with epilepsy-related care costs

Background: Drivers of epilepsy-related care costs have not been systematically studied

Design/Methods: We reviewed electronic medical records and cost data of epilepsy-treated patients between 2013–2016 from Partners Healthcare System’s Enterprise Data Warehouse. Total cost of care for epilepsy (TCC-E) was defined as sum of costs associated with all epilepsy-specific encounters (inpatient, outpatient, procedure, laboratory and hospital pharmacy) over 365 days. Changes in antiepileptic drug (AED) therapy, including additions/substitutions were assessed. Linear regression was used to model TCC-E; logistic regression modelled the likelihood of being in the top 5% and 1% of TCC-E. All models adjusted for demographic factors, insurance type, and neurologist access

Results: Data were available from 979 incident and 4392 prevalent cases. The mean and median TCC-E was $4768 and $602, respectively. TCC-E for incident cases was 104% higher than prevalent cases (p<0.001). Compared to focal seizures, total costs were 88% (p<0.001) higher for mixed seizures. Every additional Charlson Comorbidity Index (CCI) point was associated with 38% (p<0.001) higher, and psychiatric comorbidity with 159% (p<0.001) higher TCC-E. After controlling for AED generation(old vs new) and number of AEDs, calcium channel blockers (Pregabalin, Gabapentin, Ethosuximide) and SV2A binding agents (Levetiracetam) were associated with 64% and 98% higher TCC-E (p<0.001) compared to sodium channel blockers (Carbamazepine, Phenytoin, Fosphenytoin, Oxcarbazepine, Lacosamide, Lamotrigine). TCC-E was 88% higher for every additional AED change event; substitution and addition events were associated with 127% and 83% higher TCC-E (p<0.001). Incident status, comorbidities, and medication changes were significantly (p<0.05) associated with odds of being in top 5% and top 1% of TCC-E.

Conclusions: Epilepsy-related factors, comorbidities, and AED-related factors are significant predictors of TCC-E. Study limitations include cross-sectional design and non-inclusion of outpatient pharmacy costs

Study Supported by: UCB Pharma

Disclosure: Dr. Kakarmath has nothing to disclose. Dr. Agarwal has nothing to disclose. Dr. Golas has nothing to disclose. Dr. Felsted has nothing to disclose. Dr. Dye has received personal compensation for consulting, serving on a scientific advisory board, speaking, or other activities with Yes. I am currently employed by UCB, Inc. since Jan 3, 2017 and formerly with Humana’s Comprehensive Health Insights for over 4 years. Dr. Dye has received personal compensation in an editorial capacity for Yes. I received a one-time compensation from the Journal of Managed Care Pharmacy for editorial review of manuscripts for a special edition. Dr. Fishman has received personal compensation for consulting, serving on a scientific advisory board, speaking, or other activities with Yes, I am an employee of UCB, Inc. Dr. Levey has received personal compensation for consulting, serving on a scientific advisory board, speaking, or other activities with Yes, I am an employee of UCB, Inc. Dr. Agboola has nothing to disclose.