Related Experiment Video
Updated: May 7, 2025
Inverse Probability of Treatment Weighting Propensity Score using the Military Health System Data Repository and National Death Index
Published on: January 8, 2020
Identifying fatal poisonings using child fatality review, poison centre and death certificate data in the USA
Christopher E Gaw1,2, Allison E Curry3,4, Kevin C Osterhoudt3,4
1Division of Emergency Medicine, Nationwide Children's Hospital, Columbus, OH, USA christopher.gaw@nationwidechildrens.org.
Background:
Three sources used for poisoning surveillance-child fatality reviews (CFRs), poison centre (PC) calls and death certificates-employ disparate data methodologies. Our study objectives were to (1) characterise the number of fatalities captured by CFRs and PC data compared with death certificates by age and (2) compare demographic and substance characteristics of fatalities captured by the three sources.
Methods:
We acquired CFR data from the National Fatality Review-Case Reporting System (NFR-CRS), PC calls from the National Poison Data System (NPDS) and death certificate data from Centers for Disease Control and Prevention's Wide-ranging Online Data for Epidemiologic Research (WONDER) on poisoning fatalities among children 0-17 years old between 2005 and 2020.
Results:
A total of 6376 poisoning fatalities among children 0-17 years were reported to WONDER, 3460 to NFR-CRS and 1622 to NPDS. Using WONDER as the reference standard, NFR-CRS captured 71.1% of fatalities among infants, and 68.0% among children 1-4 years. NPDS captured 30.9% of fatalities among infants and 59.3% among children 1-4 years. Children≤4 years represented a greater proportion of fatalities in NFR-CRS (25.5%) and NPDS (37.0%) than WONDER (19.9%). NFR-CRS had a slightly higher proportion of fatalities involving Black or African American race (16.8%) compared with WONDER (14.4%). Opioids were the most common fatal substances associated with NFR-CRS and WONDER.
Conclusions:
Fatality counts, as well as demographic and substance characteristics of those fatalities, differ between poisoning databases used by investigators and health agencies. Reliable death classification can improve data quality. Optimising poisoning fatality capture is critical for informing effective prevention strategies.
