Emergency department use after pediatric pharmaceutical ingestion: comparison of two national databases

Abstract

Background: As no "gold standard" measure exists for the number of children evaluated in emergency departments (EDs) for medication-related injuries, the public health impact is based on estimates. In January 2006 the Morbidity and Mortality Weekly Report published a National Electronic Injury Surveillance System-All Injury Program (NEISS-AIP) estimate on unintentional pediatric medication exposure to children < or =4 years of age that resulted in an ED visit: 53,517 per year for the period 2001-2003. We sought to generate a parallel estimate using American Association of Poison Control Centers' (AAPCC) National Poison Data System (NPDS) to see how these estimates compare.

Methods: To match data reported from the NEISS-AIP, NPDS was searched for the electronic medical records of children < or =4 years of age whose call type was an unintentional exposure to a pharmaceutical that involved presenting to a health care facility (HCF) in 2001-2003.

Results: 178,513 met all of the criteria: 57,100 in 2001; 60,098 in 2002; and 61,315 in 2003. Comparing NEISS-AIP to NPDS: 10% versus 13% were hospitalized; 72% versus 68% were either 1 or 2 years of age; and the substance distribution was acetaminophen (8.1%, 6.8%), cough/cold (7.5%, 9.6%), cardiovascular (7.8%, 11.0%), anticonvulsant (3.6%, 3.2%), and vitamins (4.5%, 3.4%).

Conclusion: These results are close suggesting that the actual number is near these numbers. The NPDS number is greater than NEISS-AIP point estimate but within the 95% confidence interval. As NPDS is an actual count and NEISS-AIP is an extrapolation from a sample, to the extent that every child presenting to an ED following a medication exposure is not reported to a poison center, both databases may underestimate the problem. The NEISS-AIP extrapolation tool may need to be reassessed.