S90. Changing EEG ordering behavior based on modification of an electronic medical record ordering platform: Preliminary data. (May 2018)

• Record Type: Journal Article
• Authors: Hanrahan, Brian; Lee, James; Jackowiak, Eric; Su, Xiaowei; Paolini, Stephanie; Bagić, Anto; Urban, Alexandra
• Is Part Of: Clinical neurophysiology. Volume 129(2018)Supplement 1
• Page Start: e175
• Page End: e176
• Publication Date: 2018-05
• Subjects: Neurophysiology -- Periodicals; Electroencephalography -- Periodicals; Electromyography -- Periodicals; Neurology -- Periodicals; 612.8
• DOI: 10.1016/j.clinph.2018.04.450
• Languages: English
• ISSNs: 1388-2457

Abstract

Abstract : Introduction: We observed that stat-EEGs are often requested for convenience rather than medical necessity leading to prolonged turn around times. Limited EEG technician availability, equipment availability and lack of clinical acumen are also hypothesized as additional contributors to the misuse of stat-EEG. At our institution, stat-EEGs represented 43% of all inpatient EEG studies between 9/2015 and 8/2016. We sought to reduce potential misuse of EEG by incorporating a prescreening process into the EEG order set. It was our hypothesis that a required prescreening method would guide ordering physicians to the proper EEG study type (Stat vs. Routine) and likely reduce the proportion of spot EEGs ordered as stat. Methods: The EMR ordering process for inpatient EEGs was updated with the pre-screening process in early 10/2017. If an urgent EEG is wanted by an ordering provider, the pre-screening process requires the provider to select from a list of indications for a stat EEG. These indications include: Coma or mental status change, Patient fails to wake up after a seizure, Suspicion of status epilepticus and Patient pharmacologically paralyzed. Post-EMR change data collection was started on 10/22/2017 had its rate of stat-EEGs calculated. This variable was compared to 1774 spot EEGs (routine and stat) ordered from 1/1/2017 to 10/21/2017. A subgroup analysis between daytime (7A-7P) and nighttime (7P-7A) studies was also performed. Continuous EEG studies were excluded from the analysis. Results: 117 post-EMR change studies (61 STAT and 68 Routine) have so far been collected. There was a 7% decrease in the rate of stat-EEGs in the post-EMR change group (47.2%) when compared to the pre-EMR change group (54.6%). This did not reach statistical significance ( p = 0.12). With subgroup analysis, there was an 11% decrease and 4% decrease in the rate of STAT EEGs during the day and night respectively. The difference in daytime rate of stat-EEGs reached near statistical significance ( p = 0.06). Conclusion: Preliminary data of this ongoing QI project suggests that including a pre-screening process within an EMR EEG order set may alter ordering behavior. While there was a 7% decrease in the rate of STAT EEGs ordered it did not reach statistical significance due to a small post-EMR change group. Statistical significance might be reached with continuing data collection.