Abstract 92: Documentation Discrepancies of Time-Dependent Critical Events in Out of Hospital Cardiac Arrest
Background: The timing of and interval between prehospital interventions and patient-related events during out of-hospital cardiac arrest (OHCA) resuscitation are important for system design and patient outcome. Since most prehospital research relies on data abstraction from the medical record, it is imperative to assure the accuracy and precision of time-stamped events in prehospital documentation. Austere environments and/or limited human resources may impair the accuracy and precision of this documentation. OHCA is a time-sensitive disease that necessitates rapid interventions; even small errors in timing documentation may introduce unrecognized bias into analysis.
Objective: Determine the accuracy and precision of EMS documentation of time-stamped events during OHCA.
Hypothesis: Times of critical events (CE) in OHCA documented by EMS are inaccurate when compared to audio documentation of the same cases.
Methods: We reviewed prehospital documentation of OHCA resuscitation from two regional EMS agencies. We abstracted the documented times of CE in the prehospital medical record, as well as synchronized, time-stamped audio recordings from the monitor/defibrillator corresponding to each case. CE included IV/IO access, first epinephrine administration, successful airway placement, and return of spontaneous circulation. We compared the difference in minutes between medical record and audio sources, rounding down to the whole minute.
Results: A total of 411 patients suffered OOHCA during the study period. Of these, 192 patients had both documentation times and audio times for at least one CE. A total of 422 CEs were documented. A median absolute difference of 2 minutes (IQR 1-4 minutes) was found for all CE. 136 (32.8%) of the cardiac arrest cases had 2 or more CE documented. There was a median discrepancy of 4.5 minutes (Range 0-27, IQR 1-4.5) in the interval between CE in the same case. Within case error shows that the differences are random, and not simply related to synchronization of timing devices.
Conclusion: Documentation may introduce an error of 2 minutes in the times of CE and 4.5 minutes in the estimated intervals between CE. These errors are meaningful for studies of time dependent diseases like OHCA.
- © 2013 by American Heart Association, Inc.