Abstracts

Rationale: Patients’ first-person descriptions of their own seizures is often used clinically but may not agree with seizure descriptions by an external observer. The agreement between patients’ self-report of their seizures and those of observers present during the events has not been previously evaluated. In particular, impaired cognitive function during and following seizures may be difficult for patients to describe accurately, because these impairments themselves could interfere with reliable reporting. Assessing patient-observer agreement on seizure descriptions can provide important insight into which aspects of seizures are more reliably reported in a clinical setting.  Methods: 404 patients newly diagnosed with focal epilepsy aged 12-60 were enrolled in the multicenter Human Epilepsy Project (HEP). Coordinators administered DISCOVER (Diagnostic Interview for Seizure Classification Outside of Video EEG Recording) questionnaires to patients and external observers (if available) for each patient-reported seizure type. A total of 285 seizure types were collected in 183 patients in which both the patient and observer completed 12 DISCOVER questions which were retrospectively analyzed. Inter-rater reliability between patients and observers was evaluated using the Cohen’s kappa coefficient. Results: The highest agreement was found for the question asking if the patient remembers what happens during the seizure. Other questions, including those describing decreased ability to respond, glassy stare, purposeless hand movements, talking nonsense, speaking repetitive phrases, the patient finding themselves in a different position, having mouth movements, drooling, eyes closed throughout, both sides of the body stiff, or falling to the ground showed only moderate agreement at best. Seizure characteristics that typically occur during seizures with more severe impairment (glassy stare, mouth movements, aimless hand movements, talking nonsense, speaking repetitive phrases, drooling) were endorsed more frequently by observers than patients, whereas characteristics that might be experienced directly by patients such as falling and finding themselves in a different position were more commonly reported by patients than by observers.  Conclusions: We found a high agreement between patients and observers when asked if the patients remember what happens during seizures. These finding suggest that clinicians can obtain relatively reliable information about degree of impairment during seizures by asking patients or observers if the patient remembers what happens during seizures. This question is not always asked in routine clinical practice but perhaps should be. In contrast only moderate agreement was found between patients and observers for other behavioral descriptions of the seizures. This suggests that taking a history from both patients and observers can provide independent and important information about behavior during seizures.  Funding: The Epilepsy Study Consortium (ESCI) is a non-profit organization dedicated to accelerating the development of new therapies in epilepsy to improve patient care. The funding provided to ESCI to support HEP comes from industry, philanthropy and foundations (UCB Pharma, Eisai, Pfizer, Lundbeck, Sunovion, The Andrews Foundation, The Vogelstein Foundation, Finding A Cure for Epilepsy and Seizures (FACES), Friends of Faces and others).