Authors :
Presenting Author: Samuel Terman, MD, MS – University of Michigan
Alvin Ong, PharmD – Xenon Pharmaceuticals Inc.
Tigwa Davis, PhD – Inovalon
Anthony Yu, PharmD, MS – Inovalon
Yecheng Huang, MS – Inovalon
CJ Park, PharmD – Xenon Pharmaceuticals Inc.
Daniel Thornton, MBA – Xenon Pharmaceuticals Inc.
Rationale:
There is a gap in the literature regarding outcomes related to depression among patients with epilepsy. This study examined treatment patterns and outcomes among patients newly diagnosed with epilepsy, with and without a diagnosis of depression.
Methods:
This retrospective cohort study used 100% Medicare Fee-for-Service claims and the Inovalon MORE2 closed claims database (01/01/2016–12/31/2023) to identify patients with incident epilepsy (01/01/2017–12/31/2019), defined by ≥2 outpatient (OP) or ≥1 inpatient (IP) claims with ICD-10 codes (G40.0X, G40.1X, G40.2X, G40.5X, G40.80X, G40.89, G40.9X). Patients were stratified based on the presence of depression, defined as ≥2 OP or ≥1 IP claims with ICD-10 codes for depression (F32-F33) during the baseline period. The index date was the initiation of the first epilepsy line of therapy (LOT1), defined as therapy lasting ≥30 days. Line of therapy (LOTs) were defined using daily drug arrays and ended at discontinuation (≥60-day gap), switch (new agent with < 30 days overlap), augmentation (new agent with ≥30 days overlap), or partial drop (discontinuation of ≥1 agent for ≥60 days). Treatment failure was defined by any of these criteria. Up to four LOTs were captured per patient.
Results:
A total of 90,738 patients met inclusion criteria, with 21,388 (24%) classified into the depression cohort and 69,350 into the no depression cohort. The median age was 59 (Q1, 43; Q3 71) years for the depression group and 58 (Q1, 39; Q3, 72) years for the no depression group. A higher proportion of females were observed in the depression group (62% vs. 53%). Comorbidity burden was higher among patients with depression, reflected by a higher mean Deyo Charlson Comorbidity Index (CCI) score (3.5 vs. 2.5) and a larger proportion with CCI ≥4 (41% vs. 27%). The depression group also exhibited higher rates of psychiatric co-comorbidities, including anxiety (65% vs. 24%), sleep disorders (42% vs. 21%), psychosis (20% vs. 8%), and bipolar disorder (18% vs. 8%), as well as increased prevalence of chronic cardiometabolic and systemic diseases such as COPD, diabetes, renal disease, heart failure, and peripheral vascular disease. In terms of treatment duration, patients with depression remained on LOT1 for a shorter time (median 145 days vs. 173 days). Durations across subsequent LOTs (LOTs 2–4) were similar between cohorts, with modest declines over time. Logistic regression analysis, adjusted for age, sex, race/ethnicity, and payer, demonstrated significantly higher odds of treatment failure (LOT termination) among patients with depression (OR 1.41; 95% CI 1.34–1.48) compared to those without depression.
Conclusions:
Patients with epilepsy and depression exhibit a higher burden of psychiatric and systemic comorbidities, suggesting a more complex clinical experience. Depression was associated with shorter duration of initial therapy and a higher risk of treatment failure, highlighting the need for individualized treatment approaches in this subgroup.
Funding:
Xenon Pharmaceuticals Inc.