OBJECTIVES:Much attention has been given to prehospital blood use for trauma, but evidence is lacking for non-traumatic hemorrhage. Therefore, we sought to evaluate the epidemiology of patients receiving prehospital whole blood for nontraumatic hemorrhage. METHODS:We conducted a retrospective observational study of patients receiving prehospital blood transfusions for non-traumatic hemorrhagic shock within two fire-based emergency medical services (EMS) systems, between February 20, 2024, and April 23, 2026. Patients were included if blood products were administered for a non-traumatic cause of hemorrhagic shock. Exclusion criteria included cardiac arrest and blood transfusion prior to EMS arrival. Data collected from electronic patient care records included demographics, scene type, blood product administered, operational time intervals, and physiologic parameters. Shock index was calculated at patient-level by dividing heart rate (HR) by systolic blood pressure (SBP) at initial EMS assessment, immediately prior to transfusion, and emergency department (ED) arrival. RESULTS:Of 163 blood transfusions, 50 (30.7%) were administered for non-traumatic hemorrhagic shock. After exclusions, 47 patients were included. Mean age was 66 years, and 55% were male. Gastrointestinal bleeding (GI) was the most common source (49%), followed by unspecified hemorrhagic shock (43%) and obstetric/gynecologic hemorrhage (4.3%). Most encounters occurred in private residences (57%) or rehabilitation/skilled nursing facilities (32%). Mean heart rate decreased from 102 at initial EMS assessment to 96 on ED arrival, while mean systolic blood pressure increased from 83 mmHg to 106 mmHg. Mean shock index decreased from 1.34 to 1.01. CONCLUSIONS:Our two-system observational study found just under a third of patients received blood for non-traumatic reasons, with the majority being for GI bleeding. Further work is needed to better understand the impact of blood for medical patients on patient outcomes.