Abstracts

This is a Late Breaking abstract

Rationale: Dravet syndrome (DS) is the most common genetic epilepsy, characterized by treatment-resistant seizures and disease-specific clinical features that begin within the first year of life. However, the median age of diagnosis is 4.2 years, and there remains a diagnostic lag for the majority of individuals. Here, we aimed to delineate the disease course of individuals with an early versus delayed DS diagnosis through large-scale analysis of healthcare claims data.

Methods: Mapping ICD-10 codes to clinical concepts and analyzing claims data including hospital admissions and medication prescriptions, we characterized the clinical landscape of individuals with Early (before 48 months old) vs. Delayed (after 48 months old) Dravet diagnoses (mean age at diagnosis = 24 vs. 83 months, respectively). We assessed differences in the longitudinal phenotypic profiles across both cohorts using a unique clinical dataset derived from claims data. Diagnosis was defined as the first ICD-10 code for DS, unless earlier indicators of diagnosis preceded the availability of the DS ICD-10 code, i.e., prior genetic test, discontinued use of sodium channel blocker medication, or Z15.89 claim (Genetic susceptibility to other disease).

Results: Clinical histories were reconstructed across 770 individuals with DS, including 357 Early and 413 Delayed diagnoses. Significant differences (p < 0.01) in the first ICD-10 code for Dravet symptoms were identified for seizure- and epilepsy-focused claims (G40), febrile seizures, status epilepticus, and developmental delay. Individuals with Early diagnosis were referred to a neurologist faster (14 vs. 45 months) and initiated anti-seizure medications (ASM) earlier (21 vs. 53 months) than the Delayed cohort. Hospitalization with a relevant ICD-10 code occurred earlier (inpatient encounter at 12 vs. 49 months, ER visit at 14 vs. 45 months, p< .001 for all 8 comparisons, Figure 1). While sodium channel blockers were rarely used in both cohorts, discontinuation was associated with timing of a DS diagnosis, suggesting that exacerbating seizures can alert providers to the possibility of DS when these drugs are used. Phenotypically, individuals with Delayed diagnosis experienced higher rates of developmental delay, speech issues, conduct disorder, and ADHD at the time of diagnosis (p < .001-.01). Those with Early diagnosis had higher rates of epilepsy/seizure symptoms, but also respiratory issues and tachycardia (Figure 2).