A journey into the unknown: An ethnographic examination of drug-resistant epilepsy treatment and management in the United States.

Patients often recognize unmet needs that can improve patient-provider experiences in disease treatment management. These needs are rarely captured and may be hard to quantify in difficult-to-treat disease states such as drug-resistant epilepsy (DRE). To further understand challenges living with and managing DRE, a team of medical anthropologists conducted ethnographic field assessments with patients to qualitatively understand their experience with DRE across the United States. In addition, healthcare provider assessments were conducted in community clinics and Comprehensive Epilepsy Centers to further uncover patient-provider treatment gaps. We identified four distinct stages of the treatment and management journey defined by patients' perceived control over their epilepsy: Gripped in the Panic Zone, Diligently Tracking to Plan, Riding a Rollercoaster in the Dark, and Reframing Priorities to Redefine Treatment Success. We found that patients sought resources to streamline communication with their care team, enhanced education on treatment options beyond medications, and long-term resources to protect against a decline in control over managing their epilepsy once drug-resistant. Likewise, treatment management optimization strategies are provided to improve current DRE standard of care with respect to identified patient-provider gaps. These include the use of digital disease management tools, standardizing neuropsychiatrists into patients' initial care team, and introducing surgical and non-pharmacological treatment options upon epilepsy and DRE diagnoses, respectively. This ethnographic study uncovers numerous patient-provider gaps, thereby presenting a conceptual framework to advance DRE treatment. Further Incentivization from professional societies and healthcare systems to support standardization of the treatment optimization strategies provided herein into clinical practice is needed.

Education Research: NeuroBytes: A New Rapid, High-Yield e-Learning Platform for Continuing Professional Development in Neurology.

OBJECTIVE: To determine whether NeuroBytes is a helpful e-Learning tool in neurology through usage, viewer type, estimated time and cost of development, and postcourse survey responses. BACKGROUND: A sustainable Continuing Professional Development (CPD) system is vital in neurology due to the field's expanding therapeutic options and vulnerable patient populations. In an effort to offer concise, evidence-based updates to a wide range of neurology professionals, the American Academy of Neurology (AAN) launched NeuroBytes in 2018. NeuroBytes are brief (<5 minutes) videos that provide high-yield updates to AAN members. METHODS: NeuroBytes was beta tested from August 2018 to December 2018 and launched for pilot circulation from January 2019 to April 2019. Usage was assessed by quantifying course enrollment and completion rates; feasibility by cost and time required to design and release a module; appeal by user satisfaction; and effect by self-reported change in practice. RESULTS: A total of 5,130 NeuroBytes enrollments (1,026 ± 551/mo) occurred from January 11, 2019, to May 28, 2019, with a median of 588 enrollments per module (interquartile range, 194-922) and 37% course completion. The majority of viewers were neurologists (54%), neurologists in training (26%), and students (8%). NeuroBytes took 59 hours to develop at an estimated $77.94/h. Of the 1,895 users who completed the survey, 82% were "extremely" or "very likely" to recommend NeuroBytes to a colleague and 60% agreed that the depth of educational content was "just right." CONCLUSIONS: NeuroBytes is a user-friendly, easily accessible CPD product that delivers concise updates to a broad range of neurology practitioners and trainees. Future efforts will explore models where NeuroBytes combines with other CPD programs to affect quality of training and clinical practice.

Automated spectrographic seizure detection using convolutional neural networks.

PURPOSE: Non-convulsive seizures are common in critically ill patients, and delays in diagnosis contribute to increased morbidity and mortality. Many intensive care units employ continuous EEG (cEEG) for seizure monitoring. Although cEEG is continuously recorded, it is often reviewed intermittently, which may delay seizure diagnosis and treatment. This may be mitigated with automated seizure detection. In this study, we develop and evaluate convolutional neural networks (CNN) to automate seizure detection on EEG spectrograms. METHODS: Adult EEGs (12 patients, 12 EEGs, 33 seizures) from New-York Presbyterian Hospital (NYP) and pediatric EEGs (22 patients, 130 EEGs, 177 seizures) from Children's Hospital Boston (CHB) were converted into spectrograms. To simulate a telemetry display, seizure and non-seizure events on spectrograms were sequentially sampled as images across a detection window (26,380 total images). Four CNN models of increasing complexity (number of layers) were trained, cross-validated, and tested on CHB and NYP spectrographic images. All CNNs were based on the VGG-net architecture, with adjustments to alleviate overfitting. RESULTS: For spectrographically visible seizures, two CNN models (containing 4 and 7 convolution layers) achieved >90% seizure detection sensitivity and specificity on the CHB test set and >90% sensitivity and 75-80% specificity on the NYP test set. The one CNN model (10 convolution layers) did not converge during training; while another CNN (2 convolution layers) performed poorly (60% sensitivity and 32% specificity) on the NYP test set. CONCLUSIONS: Seizure detection on EEG spectrograms with CNN models is feasible with sensitivity and specificity potentially suitable for clinical use.

Beyond neuropathic pain: gabapentin use in cancer pain and perioperative pain.

BACKGROUND: Gabapentin (GBP), originally an antiepileptic drug, is more commonly used in the treatment of neuropathic pain. In recent years, GBP has been used as an adjunct or primary therapy in non-neuropathic pain, most commonly for the treatment of perioperative and cancer pain. OBJECTIVES: The aim of this study was to conduct a clinical evidence literature review of GBP's use in perioperative pain and cancer pain. METHODS: Using PUBMED and OVID Medline databases, keyword searches for surgery and cancer in reference to GBP and pain were carried out. Nonblinded studies and case reports that did not present a unique finding were excluded. Studies that focused only on neuropathic pain were also excluded. RESULTS: An initial 142 references focusing on GBP's use in surgical pain and cancer pain were identified. Of these, 48 studies were quality of evidence at a level of II-2 or higher. DISCUSSION: Although efficacy varies, multiple well-designed clinical trials have demonstrated reduced pain and analgesic use with otolaryngology, orthopedic, mastectomy, and abdominal/pelvic surgical perioperative use of GBP, whereas there is limited or no efficacy for cardiothoracic surgery. Cancer pain studies have had greater design variability, often nonblinded, with pain benefit being mild to moderate, and more efficacious with partial neuropathic pain quality. Overall, GBP seems to have significant benefit in neuropathic and non-neuropathic pain associated with the perioperative period and cancer. Considering its favorable side effect profile, GBP represents a beneficial pain adjunctive therapy, beyond neuropathic symptoms.