Outcomes of adult patients discharged at scene by emergency medical services

Background: The outcomes of patients who call an ambulance but are discharged at scene reflect the safety and quality of emergency medical service (EMS) care. While previous studies have examined the outcomes of patients discharged at scene, none have specifically focused on paramedic-initiated discharge. This study aims to describe the outcomes of adult patients discharged at scene by paramedics and identify factors associated with 72-hour outcomes. Methods: This was a retrospective data linkage study on consecutive adult EMS patients discharged at scene by paramedics in Victoria, Australia, between 1 January 2015 and 30 June 2019. Multivariable logistic regression was used to investigate factors associated with EMS recontact, ED presentation, hospital admission and serious adverse events (death, cardiac arrest, category 1 triage or intensive care unit admission) within 72 hours of the initial emergency call. Results: There were 375 758 cases of adults discharged at scene following EMS attendance, of which 222 571 (59.2%) were paramedic-initiated decisions. Of these, 6.8% recontacted EMS, 5.0% presented to ED, 2.4% were admitted to hospital and 0.3% had a serious adverse event in the following 72 hours. The odds of EMS recontact were increased in cases related to mental health (adjusted OR (AOR) 1.41 (95% CI 1.33 to 1.49)), among low-income government concession holders (AOR 1.61 (95% CI 1.55 to 1.67)) and in areas of low socioeconomic advantage (AOR 1.19 (95% CI 1.13 to 1.25)). The odds of hospital admission were increased in cases related to infection (AOR 3.14 (95% CI 2.80 to 3.52)) and pain (AOR 1.93 (95% CI 1.75 to 2.14)). The strongest driver of serious adverse events was an abnormal vital sign (AOR 4.81 (95% CI 3.87 to 5.98)). Conclusion: The occurrence of hospital admission and adverse events is rare in those discharged at scene, suggesting generally safe decision-making. However, increased attention to elderly, multimorbid patients or patients with infection and pain is recommended, as is further research examining the use of tools to aid paramedic recognition of potential for deterioration.