Effects of Paramedic Tenure on the Accuracy of Prehospital Stroke Identification

Introduction: Stroke is the fifth-leading cause of death in the United States and the No. 1 cause of disability, affecting approximately 795,000 people per year. Reliable identification of stroke in the field by prehospital personnel expedites delivery of acute stroke therapy. While the National Institutes of Health (NIH) scale is comprehensive, it is difficult to perform in the prehospital environment, and abbreviated scales are much more common. Provider experience, however, may factor into the correct identification of stroke.

Objective: To determine if years of prehospital experience affect the accuracy of stroke identification.

Methods: A systematic review was conducted of data from a suburban EMS agency answering approximately 34,000 calls per year with access to a Level 1 trauma center and a certified stroke center. All patients determined to be having a stroke by paramedics between January 1, 2013 and December 31, 2017 were included in the analysis. There were no exclusions based on age, sex, race, or comorbidities. ICD-10 diagnostic codes were collected for each patient. Multivariate logistic regression with robust standard errors was utilized.

Results: A total of 621 patients were included in the data analysis. Of the patients identified by the paramedic as having a stroke, 51% were subsequently admitted to the hospital for stroke, while 49% were not determined to be suffering from an acute stroke. For each additional year of experience, the paramedic was 6.5% more likely to correctly identify stroke (OR 1.065, p=0.112). It was also found that paramedics were more likely to correctly diagnose stroke in male patients (OR 1.020, p=0.045) and less likely to correctly diagnose stroke in female patients for each additional year of experience (OR 0.982, p=0.046). 

Conclusions: This study may suggest that paramedics with more tenure are more accurate at recognizing stroke in the prehospital setting. Additionally, a disparity appears to exist between male and female patients. Patients included in this study appear to have been overtriaged by paramedics. While there is an acceptable level of overtriage related to stroke, it should be noted that this may lead to alarm fatigue and overuse of scarce resources.