RT Journal Article
SR Electronic
T1 Feasibility of linking violent death decedents to prior-month emergency department visits in North Carolina, 2019–2020
JF Injury Prevention
JO Inj Prev
FD BMJ Publishing Group Ltd
SP 355
OP 362
DO 10.1136/ip-2022-044821
VO 29
IS 4
A1 Neuroth, Lucas M
A1 Johnson, Lois C
A1 Fliss, Mike Dolan
A1 Waller, Anna E
A1 Harmon, Katherine J
YR 2023
UL http://injuryprevention.bmj.com/content/29/4/355.abstract
AB Objective Linking data between violent death decedents and other sources can provide valuable insight, highlighting opportunities for prevention of violent injury. This study investigated the feasibility of linking North Carolina Violent Death Reporting System (NC-VDRS) records with North Carolina Disease Event Tracking and Epidemiologic Collection Tool (NC DETECT) emergency department (ED) visit data to identify prior-month ED visits among this population.Methods NC-VDRS death records from 2019 through 2020 were linked to NC DETECT ED visit data from December 2018 through 2020 using a probabilistic linkage approach. Linkage variables included date of birth, age, sex, zip code and county of residence, date of event (death/ED visit) and mechanism of injury. Potential linkable ED visits were filtered to those occurring in the month prior to death and manually reviewed for validity. Linked records were compared with the NC-VDRS study population to assess linkage performance and generalisability.Results Among the 4768 violent deaths identified, we linked 1340 NC-VDRS records to at least one ED visit in the month prior to death. A higher proportion of decedents dying in medical facilities (ED/outpatient, hospital inpatient, hospice or nursing/long-term care facility) linked to a prior-month visit (80%) relative to those dying in other locations (12%). When stratified by place of death, linked decedents demographically resembled the overall NC-VDRS study population.Conclusions Though resource intensive, an NC-VDRS-to-NC DETECT linkage was successful in identifying prior-month ED visits among violent death decedents. This linkage should be leveraged to further analyse ED utilisation prior to violent death, expanding the knowledge base surrounding prevention opportunities for violent injuries.No data are available. Data are not available for sharing.