PP19 Opioid Poisoning Deaths: A National Picture

Abstract

Introduction:

The factors associated with opioid poisoning death are poorly understood. We performed a retrospective autopsy study of decedents (a term used for people who are deceased) of opioid poisoning in Wales in 2015. Using anonymized linked data, we describe demographic characteristics, patterns of emergency service utilization, and clinical presentation prior to death.

Methods:

Decedents of opioid poisoning in Wales in 2015 were identified from the Office of National Statistics (ONS) mortality dataset. Records were linked with the Emergency Department Dataset (EDDS) by the National Welsh Informatics Service (NWIS); and held in the Secure Anonymized Information Linkage (SAIL) databank. The data were accessed and analyzed in the SAIL gateway.

Results:

Age at death ranged from eighteen to seventy-eight years, with a mean age of forty-two years. Average male age was forty-one years and average female age was forty-four and a half years. Seventy-three percent of decedents were men (n = 228/312). Eight-seven percent of decedents (n = 281/312) attended the emergency department in the three years prior to death. In total 2081 attendances were made, forty-one percent of which involved conveyance by ambulance. Attendances per individual ranged from one to 114, with over half of decedents attending more than three times. Diagnostic codes were mostly missing or non-specific, with only seven and a half percent of attendances representing eighty-two decedents, coded as drug related. Treatment codes were also mostly missing or non-specific, with sixteen percent of attendances representing 148 attendees attributed a treatment code. Thirty-nine percent of attendances (n = 822) ended in treatment and discharge, whilst twenty-seven percent (n = 562) led to hospital admission.

Conclusions:

Matching previously published data, we found that fatal opioid poisoning is preceded by a period of high emergency health service utilization. On average decedents were in their fifth decade and more likely to be male than female. Attendances varied widely, with men less likely to attend than women.