The national burden of E-code-identified adverse drug events among hospitalized children using a national discharge database.

PURPOSE: The purpose of this study was to provide a national-level assessment of pediatric adverse drug events (ADEs), including types, frequencies, and burdens.
METHODS: Discharge data were obtained from the 2006 Kids' Inpatient Database. ADEs were identified by the International Classification of Diseases, Ninth Revision, Clinical Modification and supplemental E codes as adverse effects (AEs), accidental poisonings (APs), or those involving neuropathy, dermatitis, and contact dermatitis. For ADEs occurring in the hospital, visits were matched, by all patient refined diagnostic-related group, age, and gender, to one control visit without an ADE code. Burden was measured as excess length of stay and excess cost relative to the control. Using regression analysis, we obtained estimates on the effects of over 100 predictors on excess length of stay and excess cost of cases relative to the control.
RESULTS: Out of 7,558,812 hospital discharges in 2006, there were 84,510 ADEs identified during 69,620 visits (0.9% of the total number of discharges); 55,285 (79.4%) visits involved an AE; and 13,630 (19.6%) involved an AP; 12,151 (17.5%) were characterized by an ADE (usually an AP) at admission. The national pediatric ADE burden was estimated at 104,230 days with direct costs of $252.9 million. The most common AEs occurred with antineoplastic and immunosuppressive drugs (20.4%) and adrenal corticosteroids (12.5%). The most common APs involved aromatic analgesics (13.7%), cardiovascular drugs (9.5%), antidepressants (8.6%), and benzodiazepine tranquilizers (8.0%).
CONCLUSION: By identifying specific ADEs that occur most often and/or have the highest burden, physicians and hospital administrators can better target their strategies for reducing pediatric medication-related harm.