BACKGROUND: Previous studies have used emergency medical services (EMS) naloxone administration and EMS overdose impression as surrogate markers to identify opioid-associated (OA) out-of-hospital cardiac arrests (OA-OHCA). We evaluated the accuracy of these two surrogate markers using medical examiner post-mortem toxicology data in OHCA patients who died. METHODS: We conducted a retrospective cohort study of all adult (≥18) non-traumatic OHCA patients treated by Milwaukee County EMS between January 1st, 2015 to December 31st, 2016. EMS naloxone administration and EMS impression of cardiac arrest etiology were used to classify expired patients into 2 groups: (1a) Received ≥1 Naloxone (1b) No Naloxone, and (2a) EMS-impression overdose, (2b) Not EMS-impression overdose. Milwaukee County Medical Examiner (MCME) opioid toxicology data was used as the reference standard and matched to EMS-treated OHCA patients that expired to determine the sensitivity (SN), specificity (SP), positive predictive value (PPV), and the negative predictive value (NPV) of both surrogate markers. RESULTS: 1,654 OHCA cases were included
112 (6.8%) received ≥1 Naloxone and 1,542 (93.2%) received No Naloxone
60 (3.6%) were EMS-impression overdose and 1,594 (96.4%) were not EMS-impression overdose. Of the 525 opioid-associated deaths in the MCME Drug-related Death Database, 150 (28.6%) were OA-OHCA cases treated by EMS. The SN, SP, PPV, and NPV for EMS naloxone administration was 39%, 96%, 52%, and 94% respectively. The SN, SP, PPV, and NPV for EMS-impression overdose was 27%, 99%, 68%, and 93% respectively. CONCLUSION: EMS naloxone administration and EMS- impression overdose had limited sensitivity for identifying OA-OHCA in expired patients in this large urban EMS system. Prehospital and public health researchers should identify improved methods for accurately classifying this OHCA subpopulation.