Inter-Rater Agreement on the EEG terminology, Brief potentially Ictal Rhythmic Discharges (B(I)RDs)
Abstract number :
2.028
Submission category :
3. Neurophysiology / 3C. Other Clinical EEG
Year :
2019
Submission ID :
2421478
Source :
www.aesnet.org
Presentation date :
12/8/2019 4:04:48 PM
Published date :
Nov 25, 2019, 12:14 PM
Authors :
Ji Yeoun Yoo, Mount Sinai Hospital; Bonnie Wong, Mount Sinai Hospital; Lara Marcuse, Mount Sinai Hospital; Madeline Fields, Mount Sinai Hospital; James Young, Mount Sinai Hospital; Nathalie Jette, Mount Sinai Hospital
Rationale: This study aimed to investigate the inter-rater agreement (IRA) on the proposed definition and categories of B(I)RDs among EEG experts at multiple centers in North America. Methods: EEG samples used in this study were collected from the Mount Sinai Health System EEG database of adult patients (age 18 years or older) who received long term (> 12 hours) EEG recordings. The questionnaire included 10 scenarios with 18 questions, and the respondents were asked to answer which pattern the particular EEG demonstrated among a list of multiple choice options including (a) normal or benign variants, (b) B(I)RDs, (c) polyspikes, (d) seizure, (e) other. Participants were allowed to type a suggested name for the pattern if they selected other. If B(I)RDs was selected as the answer, participants were then asked to select 2 categories of B(I)RDs: definite vs possible, with vs. without evolution. The web-based questionnaire was sent to all members of the Critical Care EEG Monitoring Research Consortium (CCEMRC). Gwet's (kappa) statistic was calculated for the IRA. Results: 42 members of CCEMRC participated in the survey. For the patterns representing B(I)RDs, the percent correct response was 61%, and the kappa for agreement 0.44 [95% Confidence Interval 0.27-0.62] indicating moderate agreement with this term. For the category of evolution, the percent correct response was 83% for evolution, 86% for no evolution, and the kappa was 0.54 [95% Confidence Interval 0.34-0.74] indicating moderate agreement. For the category of definite vs possible, the percent correct response was 70% and 85% respectively, and the kappa was 0.17 [Confidence interval -0.012-0.35] showing slight agreement. For the questions with < 60% correct response for BIRDs, answers varied among polyspikes, seizures, or other (specifically; Paroxysmal Fast Activity (PFA)). Conclusions: The IRA was moderate for the identification of B(I)RDs and the concept of evolution. IRA was slight for the definite vs. possible categories of B(I)RDs. We expect the IRA to improve with strict definition and standards of reporting of B(I)RDs, polyspikes, seizures and PFA terminology. Funding: No funding
Neurophysiology