‘Is the patient completely alert?’ – accuracy of emergency medical dispatcher determination of patient conscious state

Abstract

Introduction During emergency ambulance calls, one of the key issues assessed is the patient’s level of consciousness. An altered conscious state can be indicative of a need for a high priority response; however, the reliability of the resulting triage depends on how accurately alertness can be ascertained over the phone. This study investigated the accuracy of emergency medical dispatcher (EMD) determination of conscious state in emergency ambulance calls in Perth, Western Australia. Methods The study compared EMD determination of patient alertness based on the Medical Priority Dispatch System (MPDS), with conscious state as recorded by paramedics on arrival, for all emergency ambulance calls in a 1-year period in metropolitan Perth. Diagnostic accuracy was reported across the whole system and stratified by MPDS chief complaint. Results There were 109,678 calls included for analysis. In terms of identifying patients as not alert, the overall positive predictive value was 6.62% and negative predictive value was 99.93%, with 10 times as many patients dispatched as not alert than found to be not alert at scene. Sensitivity was only 69.94%. There was significant variation in accuracy between chief complaints. Conclusion The study found high levels of inaccuracy between dispatch identification of not-alert patients, and what paramedics found on scene. While not-alert dispatch was 10 times more common than patients being determined not-alert on scene, only 70% of not-alert patients on scene were classified as such during dispatch. Further research is suggested into the factors that affect the accuracy of EMD determination of patient conscious state.