Using Life-Saving Interventions to Determine Optimal Vital Sign Ranges among Adults Encountered by Emergency Medical Services.

Background: Vital signs are an essential component of the prehospital assessment of patients encountered in an emergency response system and during mass-casualty disaster events. Limited data exist to define meaningful vital sign ranges to predict need for advanced care.

Study objectives: The aim of this study was to identify vital sign ranges that were maximally predictive of requiring a life-saving intervention (LSI) among adults cared for by Emergency Medical Services (EMS).

Methods: A retrospective study of adult prehospital encounters that resulted in hospital transport by an Advanced Life Support (ALS) provider in the 2022 National EMS Information System (NEMSIS) dataset was performed. The outcome was performance of an LSI, a composite measure incorporating critical airway, medication, and procedural interventions, categorized into eleven groups: tachydysrhythmia, cardiac arrest, airway, seizure/sedation, toxicologic, bradycardia, airway foreign body removal, vasoactive medication, hemorrhage control, needle decompression, and hypoglycemia. Cut point selection was performed in a training partition (75%) to identify ranges for heart rate (HR), respiratory rate (RR), systolic blood pressure (SBP), oxygen saturation, and Glasgow Coma Scale (GCS) by using an approach intended to prioritize specificity, keeping sensitivity constrained to at least 25%.

Results: Of 18,259,766 included encounters (median age 63 years; 51.8% male), 6.3% had at least one LSI, with the most common being airway interventions (2.2%). Optimal ranges for vital signs included 47-129 beats/minute for HR, 8-30 breaths/minute for RR, 96-180mmHg for SBP, >93% for oxygen saturation, and >13 for GCS. In the test partition, an abnormal vital sign had a sensitivity of 75.1%, specificity of 66.6%, and positive predictive value (PPV) of 12.5%. A multivariable model encompassing all vital signs demonstrated an area under the receiver operator characteristic curve (AUROC) of 0.78 (95% confidence interval [CI], 0.78-0.78). Vital signs were of greater accuracy (AUROC) in identifying encounters needing airway management (0.85), needle decompression (0.84), and tachydysrhythmia (0.84) and were lower for hemorrhage control (0.52), hypoglycemia management (0.68), and foreign body removal (0.69).

Conclusion: Optimal ranges for adult vital signs in the prehospital setting were statistically derived. These may be useful in prehospital protocols and medical alert systems or may be incorporated within prediction models to identify those with critical illness and/or injury for patients with out-of-hospital emergencies.