Homepage > Research abstracts > Improving prediction of diagnosis carried out by ambulance teams in cases of dyspnea and chest pain using hospital and EMS database
Improving prediction of diagnosis carried out by ambulance teams in cases of dyspnea and chest pain using hospital and EMS database
Researchers: Amit Frenkel1, Oren Wacht2, Lior Rokach2, Refael Strugo3, Lena Novack1
- Soroka University Medical Center
- Ben Gurion University of the Negev
- Magen David Adom
Background: Making a correct diagnostic determination is crucial for administering appropriate treatment. The prehospital setting of emergency medical service (EMS) is challenging due to limited diagnostic resources and urgency. Errors occurring this early can have substantial implications for patients’ well-being.
Objectives: We aimed to improve the diagnostic abilities of the EMS teams in two clinical conditions: chest pain and dyspnea.
Method: We conducted a population-based retrospective study among patients brought to the emergency department of Soroka University Medical Center between Jan 2018-Dec 2020 by a Mobile Intensive Care Unit (MICU) operated by Magen David Adom (MDA). Diagnostic decisions made for patients with chest pain and the dyspnea symptoms by the EMS teams were compared to the diagnosis assigned by the hospital team. Discrepancy between the two teams was modelled in a multivariable analysis.
Findings: The study population comprised 13,847 patients arriving at the hospital by MICU, with 9,467 (~73%) cases identified by MDA with any of the pre-specified pre-hospital suspected diagnosis, and 4,380 cases who arrived via MICU but were not diagnosed with the same diagnoses in the pre-hospital setting. The general agreement between the prehospital and hospital diagnoses was relatively high. Judging by validity estimates, paramedics were generally good at ruling out diagnoses (specificity parameters between 78-99%), although frequently struggled to correctly identify certain conditions (sensitivity in the range 16-52%). For each clinical condition we indicated factors increasing chances of discrepant diagnosis.
Conclusions: The prehospital and hospital diagnoses were in good agreement, mostly due to high specificity function.
Recommendations: Our findings emphasize the importance of ongoing feedback for paramedics made possible by connecting the hospital records with the MDA system. This way EMS teams will be informed on the patients they have treated on their way to the hospital and improve their professional judgment
Research number: R/295/2020
Research end date: 06/2024