Abstract T P340: The Electronic Medical Record: Are the Times Accurate?
Background: The Electronic Medical Record (EMR) was designed to improve patient outcomes, eliminate unnecessary costs, and provide continuity of care. Intravenous tissue plasminogen activator (IV t-PA) is most effective when administered early. In order to effectively decrease door to needle times, a stroke center must be able to correctly identify the steps within their pathway where there is room for improvement. A reliable way to measure and record time is therefore crucial. In our experience, the EMR has been extremely helpful with data collection; however, there are limitations, particularly with respect to the automatic time stamp for electronic documentation. We sought to determine the accuracy of the EMR time stamp versus real-time paper charting in our assessment of the acute stroke patient.
Methods: 33 patients treated with IV t-PA between July 2013 and July 2014 were included in analysis. Specific time points were documented on a Nursing Flow Sheet by a dedicated stroke nurse who is part of each acute stroke assessment and were compared to the time points documented in the EMR. The average discrepancy between the Nursing Flow Sheet and EMR for each time point was calculated. Arrival time to the Emergency Department (ED), time to CT scanner, lab collection time, and t-PA bolus time were evaluated.
Results: The average discrepancy on arrival time to the ED was 7 minutes. The Nursing Flowsheet documented earlier arrival times in all but 2 cases. The arrival time obtained from the EMR required extrapolation- either from the point of comprehensive triage where vital signs were obtained (n=24) or from the code sheet scanned into the electronic record (n=9). Times documented on the code sheet were only 3 minutes different from the Nursing Flow Sheet versus 9 minutes for those with vital signs charted. The average discrepancy between time to the CT scanner on the Nursing Flow Sheet and the order time in the EMR was 13 minutes. The majority of patients (88%) had imaging completed prior to the order being placed. The discrepancies for lab collection time and t-PA bolus time were smaller (4 minutes and 0 minutes respectively).
Conclusion: Automatic time stamps provided by an EMR may not accurately reflect true time points, making improvement of door to needle times more difficult.
Author Disclosures: E. Lawrence: None. C. Woolford: None. D. Merbach: None. S. Thorpe: None. R.H. Llinas: None. E.B. Marsh: None.
- © 2015 by American Heart Association, Inc.