373 - Identifying Fatal Pediatric Poisonings: A Comparison of Child Death Review, Poison Center, and Death Certificate Data

Poster Number: 373
Publication Number: 373.1642

Background: Poisonings are a recognized cause of pediatric mortality in the United States (US). The development of effective poisoning prevention efforts relies on complete and accurate data. Three data sources used for poisoning surveillance and research—Child Death Reviews (CDR), Poison Center (PC) calls, and death certificates—employ different methodologies and data collection instruments, and often provide disparate estimates of fatal poisonings.

Objective: Our objectives were: (1) to compare the demographic and substance characteristics of fatal child poisonings described in death certificate, PC, and CDR data, and (2) characterize the number of fatalities captured by PC and CDR data compared to death certificate data by year.

Design/Methods: We acquired CDR data from the National Fatality Review-Case Reporting System (NFR-CRS), PC call data from the National Poison Data System (NPDS), and death certificate data from the Centers for Disease Control Wide-ranging Online Data for Epidemiologic Research (WONDER). Our study was limited to poisoning deaths in the US among children 0-19 years old between 2005-2020. We analyzed substances associated with death and shared variables between these databases using descriptive and chi-squared statistics. As the number of states contributing to the NFR-CRS varies by year, we present both state-limited and national-level comparisons when describing deaths captured by NPDS and NFR-CRS, using WONDER as a reference standard.

Results: During the study period, 3,938 poisoning fatalities involving children aged 0-19 were reported to NFR-CRS, 2,321 to NPDS, and 16,679 to WONDER (Table 1). A greater proportion of NFR-CRS (22.4%) and NPDS fatalities (25.8%) involved children ≤4 years of age compared to WONDER (7.6%). Using WONDER as the reference standard, NFR-CRS captured 71.1% of fatalities in infants < 1, 62.9% in children ages 1-14, and 17.2% in those 15-19. NPDS captured 30.9% of fatalities in infants < 1, 51.1% in children ages 1-14, and 8.9% in those 15-19. As a percentage of WONDER fatalities, the annual number of NFR-CRS fatalities ranged from 13.4% to 28.1% across the study period; the annual number of NPDS fatalities ranged from 10.6% to 20.5% (Table 2). Opioids were the most common substance in NFR-CRS and WONDER but not NPDS (Figure 1).

Conclusion(s): The demographic, substance characteristics, and number of deaths captured differ between the NFR-CRS, NPDS, and WONDER databases. Optimizing the capture of poisoning fatalities by surveillance systems is critical to informing the development of effective poisoning prevention strategies.