Abstracts

Rationale: To assess if early treatment in newly diagnosed epilepsy is associated with lower health burden than no treatment in a high-income setting.

Methods: We performed linkage analysis of patients with newly diagnosed epilepsy between 1999 and 2016 with statewide healthcare databases to extract hospital admission, ambulatory psychiatric care, and mortality data from 1970 to 2019. Data were compared between untreated and treated patients, with the latter divided into those treated immediately after epilepsy diagnosis and those treated later, after subsequent seizures (Fig. 1). Analyses were performed at up to 5 years post-diagnosis, adjusted for seizure type, age at onset, sex, baseline comorbidity, and relevant follow-up times.

Results: 603 patients with newly diagnosed epilepsy were studied. 422 (70%) were treated immediately after diagnosis, 110 (18%) received delayed treatment, and 71 (12%) were not treated during clinical follow-up (median 6.8 years, interquartile range [IQR] 4.0–10.0) (Table 1).

Incidence of seizure-related admissions or emergency department presentations in the first year post-diagnosis was higher in the immediate treatment group (318 per 1000 person-years [PYs]) than the untreated (29 per 1000 PYs; incidence rate ratio [IRR]=10.4; 95% confidence interval [CI] 5.54–19.6; p< 0.001) or delayed treatment groups (151 per 1000 PYs; IRR=2.13; 95% CI: 1.20–3.78; p=0.010).

Of 453 (75%) patients without a physical comorbidity in the 5 years pre-diagnosis, no difference in risk of developing new comorbidities was found between immediate treatment and untreated groups (IRR=0.82; 95% CI: 0.35–1.89; p=0.63).

In the 5 years pre-diagnosis, 66 (11%) patients attended psychiatric clinics and 16 (2.7%) were diagnosed with psychiatric comorbidities excluding substance abuse and intellectual disability. Among 537 (89%) patients without a pre-diagnosis psychiatric clinic attendance, 74 (12%) attended psychiatric clinics within 5 years of diagnosis, and psychiatric comorbidities were diagnosed in 19 patients (3.2%). No difference in risk of developing a new psychiatric comorbidity was found between treatment groups (Cramér's V=0.07; p=0.32).

A total of 70 (11.6%) patients died during 5 years’ follow-up, 7 (10%) had epilepsy recorded as cause of death. There were no differences in overall mortality between immediate treatment and delayed treatment (hazard ratio [HR]=1.18; 95% CI: 0.55–2.48; p=0.69) or untreated groups (HR=1.59; 95% CI: 0.48–5.25; p=0.45). Compared to the general population, standardized mortality ratio was 3.61 (95% CI: 2.80–4.66) in the immediate treatment, 4.21 (95% CI: 2.11–8.43) in the delayed treatment, and 3.99 (95% CI: 1.29–12.4) in the untreated groups.

Conclusions: Patients with newly diagnosed epilepsy who were untreated or received delayed treatment did not have increased risks of hospitalization, comorbidity, or mortality versus patients treated immediately. Patients with newly diagnosed epilepsy who require immediate treatment have a high rate of seizure-related hospital admissions and further research to better understand the reasons for this is warranted.

Funding: Please list any funding that was received in support of this abstract.: UCB Pharma-funded.