Research In Action
Research In Action
Breadcrumb

Pediatric poisonings are a leading cause of unintentional injury death in the United States (US), and child poisoning fatalities have increased in recent years. Though the drivers of this trend are multifactorial, the opioid epidemic has been identified as a major contributor. In response, there have been increasing calls among investigators and advocates for child-focused research and policy efforts.
To prevent child poisonings, we need complete and accurate data to inform the development of prevention and harm mitigation interventions. Currently, there are multiple sources that collect data on fatal child poisonings, including child death reviews, poison center calls, and death certificates. Each of these data sources has strengths and limitations.
A Better Understanding of Data Sources
To help investigators and policymakers better understand these data sources, we conducted a methodologic study comparing the characteristics of three databases used in poison research:
- The National Fatality Review-Case Reporting System (NFR-CRS; child death reviews)
- The National Poison Data System (NPDS; poison center calls)
- The Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research (WONDER; death certificates).
We examined poisoning fatalities, as defined by the respective databases, from 2005-2020 among children at birth through age 17. Our findings were recently published in Injury Prevention, and here is what’s key to know:
Demographic and Substance Characteristics
- NFR-CRS (26%) and NPDS (37%) described more fatalities among children ≤4 years of age compared to WONDER (20%).
- The proportion of fatalities involving Black or African Americans described by NFR-CRS (16.8%) was slightly higher than that of WONDER (14.4%).
- Opioids were the most common substance associated with death in WONDER and NFR-CRS, but in NPDS they were the fourth most common substance.
Fatality Comparison
In our study, we used WONDER, which is considered a comprehensive death record, as a reference standard to compare fatality capture within the NFR-CRS and NPDS. By doing so, we noted:
- WONDER described approximately two times as many poisoning fatalities than NFR-CRS and four times as many than NPDS.
- Compared to WONDER, NFR-CRS fatality capture was high among infants <1 year old (71%), children 1-4 years (68%), and children 5-9 years (72%).
Key Takeaways
Our study highlights how these databases differ in demographic and substance characteristics, as well as the relative numbers of fatalities reported. Some key findings:
- Researchers and policymakers should be aware of the differences and limitations associated with these data sources.
- WONDER describes a large number of fatal poisonings, but NFR-CRS and NPDS contain a greater proportion of pediatric fatalities involving young children.
- The unique variables contained in NFR-CRS and NPDS may help answer research questions that WONDER cannot.
- No data source is perfect, and there are multiple opportunities to improve the accuracy and completeness of poison fatality databases in the US.