Abstracts

Rationale: Epilepsy is diagnosed when an individual has two or more unprovoked seizures more than 24 hours apart. However, previous research has suggested that some individuals have multiple seizures prior to their first recognized or recorded index seizure and there may be substantial delay between the onset of seizures and an epilepsy diagnosis. A longer time between seizure onset and epilepsy diagnosis may lead to delays in treatment and poor outcomes. This study characterized time delays between first and second seizures and epilepsy diagnosis in a large cohort of Medicaid beneficiaries and assessed demographic variables that may be associated with a delayed epilepsy diagnosis. Given known racial disparities in epilepsy care, we were particularly interested in time to diagnosis among racial and ethnic minorities. Methods: Medicaid claims data for 16 geographically and demographically diverse states were analyzed to identify an incident epilepsy diagnosis (ICD-9-CM: 345.XX) in 2012 - 2014 for beneficiaries 18-64 years old. Data for 2010 and 2011 were further included as look-back periods, to identify both the incident seizure diagnosis (ICD-9-CM: 780.3X) as well as subsequent seizure diagnoses prior to the epilepsy diagnosis. Time from first- and second- seizure diagnosis to incident epilepsy diagnosis, as well as number of seizure claims, were calculated with differences between demographic groups evaluated. Results: There were 80,949 individuals with a first epilepsy diagnosis in the study period, a majority of whom were women (53.3%) and White (56.0%). However, individuals of other races/ethnicities were well represented with 22.6% Black, 0.84% American Indian/Alaskan Native (AIAN), 1.1% Asian or Pacific Islander (API), 0.9% Native Hawaiian or Other Pacific Island (NHOPI) and 9.0% Hispanic. The average age at index seizure date was 41.1 (SD: 13.1) years, with a median of 41.7. Overall, the median time between incident seizure and epilepsy diagnosis was 228 days (Range: 1 - 1,821). Differences between racial groups were significant (p < 0.0001) but in the opposite direction hypothesized. Whites had a higher median of 255 compared to Blacks (230), AIAN (179), API (188), and Hispanics (186) but lower when compared to NHOPI (301). For those who had only 1 seizure claim prior to epilepsy (43.1% of the population), the median time between seizure and epilepsy was 77 days (range: 1 - 1,816). In comparison, those who had multiple seizure claims before epilepsy diagnosis, time from their index seizure to epilepsy diagnosis was a median of 434 days (range: 1 - 1,821). Finally, those individuals who were diagnosed with intractable epilepsy at their index epilepsy diagnosis (6.5%) had a shorter median time (150) compared to non-intractable patients (234). Conclusions: There are substantial time gaps between first seizure diagnosis and epilepsy diagnosis for Medicaid beneficiaries. Demographic sub-group analysis did not suggest that racial and ethnic minorities were more likely to experience a delayed epilepsy diagnosis. Understanding the causes and predictors of delay in epilepsy diagnosis is critical to improving diagnosis times and reducing poor health outcomes due to delayed diagnosis and treatment. Additionally, while there are many strengths to claims data, it will be important to assess whether findings are similar or different when using clinical data such as information derived from the electronic health record. Funding: This work was funded by the CDC Special Interest Project 3 U48 DP005030-05S1.