Glenn S. Takata, MD, Wilbert Mason, MD, MPH, Carol Taketomo, PharmD,
Tina Logsdon, MS, and Paul J. Sharek, MD, MPH
Division of General Pediatrics,
Department of Pediatrics,
University of Southern California School of Medicine,
Los Angeles, California
OBJECTIVES: The purposes of this study were to develop a pediatric-focused tool for adverse drug event detection and describe the incidence and characteristics of adverse drug events in children’s hospitals identified by this tool.
METHODS: A pediatric-specific trigger tool for adverse drug event detection was developed and tested. Eighty patients from each site were randomly selected for retrospective chart review. All adverse drug events identified using the trigger tool were evaluated for severity, preventability, ability to mitigate, ability to identify the event earlier, and presence of associated occurrence report. Each trigger and the entire tool were evaluated for positive predictive value.
RESULTS: Review of 960 randomly selected charts from 12 children’s hospitals revealed 2388 triggers (2.49 per patient) and 107 unique adverse drug events. Mean adverse drug event rates were 11.1 per 100 patients, 15.7 per 1000 patient-days, and 1.23 per 1000 medication doses. The positive predictive value of the trigger tool was 3.7%. Twenty-two percent of all adverse drug events were deemed preventable, 17.8% could have been identified earlier, and 16.8% could have been mitigated more effectively. Ninety-seven percent of the identified adverse drug events resulted in mild, temporary harm. Only 3.7% of adverse drug events were identified in existing hospital-based occurrence reports. The most common adverse drug events identified were pruritis and nausea, the most common medication classes causing adverse drug events were opioid analgesics and antibiotics, and the most common stages of the medication management process associated with preventable adverse drug events were monitoring and prescribing/ordering.
[Editorial Commentary: These findings (that only 3.7% of adverse events find their way into hospital error reports) is very alarming, and suggests that previous reports have only documented the tip of the iceberg.]
CONCLUSIONS: Adverse drug event rates in hospitalized children are substantially higher than previously described. Most adverse drug events resulted in temporary harm, and 22% were classified as preventable. Only 3.7% were identified by using traditional voluntary reporting methods. Our pediatric-focused trigger tool is effective at identifying adverse drug events in inpatient pediatric populations.