Abstracts

Rationale: The extent to which recommendations for epilepsy monitoring unit (EMU) VEEG reporting are followed in the United States (US) has not previously been described. Here, we comprehensively review National Association of Epilepsy Centers (NAEC) EMU VEEG reports for the presence of recommended data elements and quality indicators, as an initial step in defining elements relevant to center accreditation. Methods: The BIDMC Committee on Clinical Investigations determined that this project does not constitute human subjects research. Existing guidelines and recommendations (Table 1) for VEEG report documentation were reviewed for recommended data elements; elements were reconciled and categorized as 1) clinical care element; 2) standardized terminology; or 3) quality variable. 777 EMU VEEG reports from the year 2017 from 166 of the 237 NAEC accredited epilepsy centers were analyzed for the presence or absence of each of these variables. Reports were written for routine clinical purposes; at the time of submission, NAEC did not recommend or require adherence to specific report guidelines. Reports were deidentified; all patient identifiers were redacted by originating centers according to the HIPAA Privacy Rule prior to submission. Sites were identified only by level (3 or 4), population served (adult, pediatric, or both), and geographic region (Table 2). Chi-square tests were used to determine if there were statistically significant differences between NAEC level 3 and 4 centers, adult and pediatric VEEG reports, and data element types. Results: There was considerable variation in VEEG reporting practices throughout the US. 60.1% of NAEC centers with reports appeared to have a center-wide standard template in use, and 64.9% of centers used cumulative summary VEEG reports (the remainder used daily 24-hour reports). For clinical care elements, reporting was high (e.g. nearly all reports contained proper patient and study identifiers, 86.% described the presence or absence of interictal epileptiform activity, 95.9% of reports with seizures described clinical semiology, and 95.5% described the ictal EEG). For standardized terminology, 51.6% of reports with seizures used International League Against Epilepsy classification-compatible seizure terminology, and 74.8% of reports with epileptiform activity or rhythmic/periodic patterns used ACNS critical care-compatible terminology. For potential quality indicators, there was much more variability: 9.8% of reports stated whether a seizure event was typical of home events, and 19.3% of reports described polygraphic channels during seizure events. There were some differences between adult and pediatric reports (e.g. pediatric reports more likely to include treatment or follow-up recommendations, 72.0% vs 33.3%, p = 0.0008). Conclusions: There is significant variability in EMU VEEG reporting practices in the US. Reports include most elements recommended for clinical care, but utilize inconsistent formats. Inclusion of recently-developed standardized terminology is moderate. As expected, the highest variability is found in inclusion of potential quality indicators in these clinically-oriented reports. The work involved for centers to include this small number of additional elements would likely be modest and feasible. This project will inform efforts to create a uniform template for VEEG reports that centers can utilize to locally assess and benchmark the utility, safety, quality, and outcomes of VEEG, and that could facilitate multicenter quality improvement and health outcomes research. Funding: PCORnet, PCORI, 02/01/2018-6/30/2020 Epilepsy Learning Health System