Abstracts

Rationale: Recent studies in the United States using claims databases found one third of people with newly diagnosed epilepsy were not treated with antiepileptic drugs (AEDs) and had increased risk of epilepsy-related medical events and healthcare utilization. Whether these findings can be observed in other similar socioeconomic settings is unknown. Methods: The study population comprised of people newly diagnosed with epilepsy between 2009 and 2014 and followed till 2015 (Figure). Patients were defined as immediately treated, delayed and untreated if an AED was prescribed within 30 days, after 30 days, and not prescribed by the end of follow up, respectively. Risks of mortality, hospitalization, and new comorbidity were compared between patients who were immediately treated and not treated at diagnosis using Cox regression and Andersen–Gill model for single and repeated events, respectively. All analyses were adjusted for age, sex, and ethnicity. Results: 3,366 (55% male) patients with newly diagnosed epilepsy were included. Median age at diagnosis was 38 years (interquartile range [IQR] 15-64). Median follow-up duration was 3.4 years (IQR 1.8-5.1). 3123 (93%) patients were immediately treated at diagnosis; 125 (3.7%) had delayed treatment (median delay 127 days; IQR 60-374) and 118 (3.5%) were untreated. Patients were of Maori descent (686, 20%) were less likely to start treatment immediately compared to other ethnicity groups (odds ratio=0.73, p=0.043).575 (17%) patients died during follow-up (standard mortality ratio 4.60, 95% confidence interval [CI] 4.24-4.99). Mortality in patients with delayed/no treatment during the untreated period was similar to those immediately treated (hazard ratio [HR]=0.99, p=0.98). However, mortality in the delayed treatment group during the treated period trended higher than the immediately treated group (HR=1.50, p=0.056). Patients of Maori descent had significantly higher mortality than other ethnic groups (HR=1.39, p=0.011) but delayed/no treatment had no moderation effect on mortality in Maori people (HR=1.04, p=0.93).Among patients without pre-existing comorbidity (n=1852), 286 (15%) was hospitalised for new comorbidity during follow up. Compared to immediately treated patients, those who had delayed/no treatment during the untreated period had lower risk of further seizure-related admission (HR=0.54, p=0.004), but had similar risk of overall admissions (HR=0.96, p=0.83). There was higher risk of developing new chronic obstructive pulmonary disease (COPD, HR=3.73, p=0.036) but the overall risk of developing new comorbidity was similar (HR=1.09, p=0.78).Among the 1790 patients without pretreatment comorbidity, the risks of subsequent seizure-related admission (HR=1.14, p=0.64) and overall admissions (HR=1.12, p=0.64) in the delayed treatment group during the treated period were similar to the immediately treated group. The delayed treatment group during the treated period had significantly higher risk of developing new acute myocardial infarction (HR=15.5, p=0.017), but the overall risk of developing new comorbidity was similar (HR=0.98, p=0.96). Conclusions: Based on hospitalized data, the rate of untreated epilepsy appeared to be lower in New Zealand than the United States. Overall, epilepsy patients bear higher excess mortality. Maoris with epilepsy had higher mortality than other ethnic groups. Compared to immediately treated patients, those untreated or deferred patients in the untreated period had higher risk of COPD, and those deferred patients in the post-deferred treated period had higher risk of acute myocardial infarction. However, the overall risks of mortality and comorbidity were similar. Funding: No funding