Date of Award

3-12-2026

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Applied Science

First Advisor

John John Bush

Abstract

For over thirty years, patients have been visiting the Arkansas Epilepsy Program for diagnosis and treatment for seizures and seizure-like episodes. As part of their clinical evaluation, patients often undergo ambulatory EEG (electroencephalogram) monitoring. This routine process produces valuable data for treating the patient. In this study, over 2000 ambulatory EEG reports from 1998 to 2016 were reviewed. A large database of seizures was created from the reports, with information on 407 patients, 1611 EEG-confirmed seizures, and 1726 patient-reported seizures (IRB Protocol #17-093). The database was used to address two important issues in epilepsy. The first topic of the study was the reliability of patient seizure reporting. The treatment of epilepsy depends on patients to report the frequency and severity of their seizures. Erroneous reporting diminishes the physician's capability to properly diagnose and treat patients. Similarly, clinical trials aiming to approve new epilepsy medications also rely on patient reporting. Unreliable reporting may result in incorrect assessments of the efficacy of experimental medications. Thus, the reliability of patient reporting is critical to diagnosis, treatment, and clinical trials. Based on previous studies,7-13,19,20 it was hypothesized that patient seizure reports are often erroneous. This study aimed to measure how inaccurate patient reporting is and to dissect the phenomenon in greater detail. Using the database, two measurements of patient reporting accuracy were calculated: false positives and missed seizures. False positives occur whenever patients report events that they think are seizures, but without the corresponding changes in the brain’s electrical activity (EEG). For the study, the EEG is the final word on whether a seizure happened or not (with exception of simple focal seizures, which are not reliably detected on EEG). If a patient reports a seizure but it isn’t on the EEG, it is considered a false positive. Alternatively, if a seizure is detected on the EEG but the patient doesn’t report it, it is called a “missed seizure,” the second measurement of patient reporting accuracy. In our study, the average patient missed 60% of seizures and reported 80% false positives. Analysis was also carried out for subpopulations based on age, sex, epilepsy type, seizure type, and seizure origin in the brain. The study concludes that patient reporting is unreliable and should not be used as the sole or primary basis to determine effective treatments for patients or the outcome of clinical trials. Additional techniques and technologies should supplement or replace patient reporting to achieve a more accurate assessment. Due to the large numbers of patients and seizures in this study, there is an unprecedented statistical significance for many of its findings. With respect to previous studies, this study also greatly improved the metrics for assessing reporting accuracy, statistical analysis, and study conditions and procedures – reflecting the patients’ real-life environments. It is clear that patient reporting of seizures is unreliable. Other tools, such as patient-worn devices or EEG monitoring should be used to monitor seizure occurrence. Better reporting allows for better-tailored treatment and for more accurate assessment of medications, including those in clinical trials. The database also contained information about the daily pattern of seizure occurrence. Hence, the study also aimed to identify temporal trends in seizure activity. For this topic, we analyzed the database’s 1611 seizures in 175 patients (the other patients in the database did not experience EEG-detected seizures and were omitted from this analysis). The temporal distribution of seizures was elucidated in different subgroups based on epilepsy type, seizure type, and patient demographics. Generally, the fewest seizures were experienced between 12 AM and 8 AM. The highest number of seizures were experienced between 4 PM and 8 PM by most subgroups. These trends have nothing to do with patient reporting, all of the seizures in this analysis are detected on the EEG with confirmation by the physician. Statistically significant effects on seizure timing were observed in patients grouped according to body mass index (p value = 0.0189), benzodiazepine use (0.00064), the use of antiepileptic medication (0.015) and the site of seizure origin within the brain (2.2e-16). The effect of BMI on seizure timing is a novel finding unexplored by previous works. Regarding benzodiazepines, an unusually high seizure peak at 4-8 PM was discovered, as well as a shift in seizure frequency to later hours in the day associated with antiepileptic medications. The patterns elucidated by this study can be used to optimize patient medication schedules for a more optimal therapeutic effect. This information can also be used by pharmaceutical companies to design drugs with specific time-release profiles, allowing for precise seizure treatment aimed at times of high risk.

Download

Included in

Biostatistics Commons

Share

COinS