Objective: An enhanced dose range checking (DRC) system was developed to evaluate prescription error rates in the pediatric intensive care unit and the pediatric cardiovascular intensive care unit.
Methods: An enhanced DRC system incorporating “soft” and “hard” alerts was designed and implemented. Practitioner responses to alerts for patients admitted to the pediatric intensive care unit and the pediatric cardiovascular intensive care unit were retrospectively reviewed.
Results: Alert rates increased from 0.3% to 3.4% after “go-live” (P < 0.001). Before go-live, all alerts were soft alerts. In the period after go-live, 68% of alerts were soft alerts and 32% were hard alerts. Before go-live, providers reduced doses only 1 time for every 10 dose alerts. After implementation of the enhanced computerized physician order entry system, the practitioners responded to soft alerts by reducing doses to more appropriate levels in 24.7% of orders (70/283), compared with 10% (3/30) before go-live (P = 0.0701). The practitioners deleted orders in 9.5% of cases (27/283) after implementation of the enhanced DRC system, as compared with no cancelled orders before go-live (P = 0.0774). Medication orders that triggered a soft alert were submitted unmodified in 65.7% (186/283) as compared with 90% (27/30) of orders before go-live (P = 0.0067). After go-live, 28.7% of hard alerts resulted in a reduced dose, 64% resulted in a cancelled order, and 7.4% were submitted as written.
Conclusions: Before go-live, alerts were often clinically irrelevant. After go-live, there was a statistically significant decrease in orders that were submitted unmodified and an increase in the number of orders that were reduced or cancelled.
From *The University of Arizona College of Medicine–Phoenix; and †Department of Pediatric Critical Care, Phoenix Children’s Hospital, Phoenix, Arizona.
Correspondence: Paul Bakerman, MD, Cardiovascular Intensive Care Unit, Phoenix Children’s Hospital, 1919 E. Thomas Rd, Phoenix, AZ 85016 (e-mail: email@example.com).
The authors disclose no conflict of interest.