Patient Outcomes Following Prehospital Naloxone Administration in San Francisco

Naloxone (NLX) is available to laypersons and an effective public health intervention. In 2024, San Francisco, CA recorded 638 opioid-related deaths, showing the opioid crisis and need to expand bystander NLX access. As NLX is often administered before EMS arrival, understanding prehospital use patterns and transport dispositions is critical. This study quantifies the effects of these interventions on the EMS system.

The Toxicology Investigators Consortium (ToxIC) Real-World Examination of Naloxone for Drug Overdose Reversal (RENDOR) project is an ongoing multisite observational study where EMS providers collect prehospital data on opioid overdose cases, including demographics, scene type, NLX administrator/dose/route, transport disposition, and provision of leave-behind NLX. Patients were grouped by administrator: Pre-EMS only (family/friend/stranger, police (PD), non-transport fire, and community/social services (C/SS) staff), EMS only, and Pre-EMS + EMS. EMS encounters from 6/2024-12/2025 were included.

Among San Francisco cases (n = 350), 79.1% (n = 277) were male, and the mean age differed by group (p = 0.010): EMS-only recipients were older (45.5 y) than pre-EMS-only (40 y) and pre-EMS + EMS (41.7 y). Most overdoses were on streets/sidewalks (n = 264, 75.4%), with pre-EMS responders giving NLX 73% (n = 193) of these cases. EMS-only responses were more common in private residences (n = 24, 54.5%;p = 0.001). IN NLX was predominant among pre-EMS responders (n = 184, 93.9%) compared to IN + IM/IV/IO (n = 4, 2.0%). EMS used IN in 44.2% (n = 46), similar to IM/IV/IO use (n = 38, 36.5%), and more often than IN + IM/IV/IO (n = 19, 18.3%). Among pre-EMS administrators, bystanders accounted for 87.8% (n = 216), with smaller contributions from PD (n = 10, 4.1%) and C/SS providers (n = 20, 8.1%). Most patients were transported (n = 253, 72.3%), while 25.1% (n = 88) refused/eloped. On-scene deaths were rare (n = 6, 1.7%). No significant difference in transport outcomes was not found for patients receiving either pre-EMS or EMS-only NLX.

In San Francisco, NLX is often administered by pre-EMS responders in public settings and among younger patients, while EMS use is more common in private residences. Most administrations are by bystanders, highlighting the critical role of layperson NLX use and supporting public health interventions that expand bystander distribution in urban settings.