The objective of this was to compare prescription error rates before and after introduction of computerized provider order entry on a pediatric emergency department. Error assessment was limited to errors with potential to cause life-threatening injury, failure of therapy, or an adverse drug effect. 29.6% of prescriptions generated an alert, with an almost 90% false-positive rate. 7,268 medication orders before and 7,292 after were compared, showing a significant reduction in the errors per 100 prescriptions (10.4 before vs. 7.3 after; absolute risk reduction = 3.1, 95% confidence interval [CI] = 2.2 to 4.0)Sethuraman 2015 (Link) | PubMed 25998704 | Author Search
A retrospective study of weight-based dosing errors (>10% variance) committed by residents at either the beginning or end of their training. Error rates were not significantly different between the beginning and end of the academic year. Error rates (16% of prescriptions were out of dosing bounds) were similar to previously reported reports. Most of the errors were of no consequence to the patient and were errors only in the sense that they departed form published dose ranges.Pacheco 2012 (The Journal of Emergency Medicine) | PubMed 22464610 | Author Search
Review of prescriptions written in an academic, suburban, tertiary care pediatric emergency department. Sixteen percent of prescriptions had an error, but over 60% of these were considered insignificant. About 1 in 6 of the erroneous prescriptions had a significant error. Two of the 60 identified erroneous prescriptions were considered to be serious errors, where only one dose of albuterol was prescribed instead of a scheduled regimen. The majority of significant errors were found to be due to incorrect dosing, which suggests that users may not be using the decision support provided. Pediatr Emerg Care. 2015;31:368-372.Nelson 2015 (Link) | PubMed 25931343 | Author Search
International Journal of Medical Informatics
A retrospective analysis of inpatient medication orders and dosing alerts at a large, academic, pediatric medical center. Pediatric orders had higher odds for an alert compared to adult orders. The alert rate was highest for immunosuppressive agents (> 50% alert rate), medications for neonates (6.7%), orders for rectal administration (9.5%), and orders for continuous infusion (7.9%). Orders for albuterol continuous aerosol and heparin injection solution had near-100% alert rates. Opioid alerts had among the lowest alert rates, at 7.4%. There was also a temporal analysis that indicated that January wasStultz 2015 (Link) | PubMed 25466381 | Author Search
Compared rates of errors between handwritten and computerized orders in a pediatric intensive are unit. CPOE resulted in errors (of omission in 22% of cases), but at a far lower rate than in handwritten orders (66%). Drug dosing error rates were similar for both methods (21%) and largely involved under-dosing. (Intensive Care Medicine)Maat 2014 (Link) | PubMed 24352486 | Author Search
Maat B, Rademaker CM, Oostveen MI, Krediet TG, Egberts TC, Bollen CW
JPEN J Parenter Enteral Nutr 2013 Jan;37(1):85-91
BACKGROUND: Prescribing glucose requires complex calculations because glucose is present in parenteral and enteral nutrition and drug vehicles, making it error prone and contributing to the burden of prescribing errors.
OBJECTIVE: Evaluation of the impact of a computerized physician order entry (CPOE) system with clinical decision support (CDS) for glucose control in neonatal intensive care patients (NICU) focusing on hypo- and hyperglycemic episodes and prescribing time efficiency.
METHODS: An interrupted time-series design to examine the effect of CPOE on hypo- and hyperglycemias and a crossover simulation study to examine the influence of CPOE on prescribing time efficiency. NICU patients at risk for glucose imbalance hospitalized at the University Medical Center Utrecht during 2001-2007 were selected. The risks of hypo- and hyperglycemias were expressed as incidences per 100 patient days in consecutive 3-month intervals during 3 years before and after CPOE implementation. To assess prescribing time efficiency, time needed to calculate glucose intake with and without CPOE was measured.
RESULTS: No significant difference was found between pre- and post-CPOE mean incidences of hypo- and hyperglycemias per 100 hospital days of neonates at risk in every 3-month period (hypoglycemias, 4.0 [95% confidence interval, 3.2-4.8] pre-CPOE and 3.1 [2.7-3.5] post-CPOE, P = .88; hyperglycemias, 6.0 [4.3-7.7] pre-CPOE and 5.0 [3.7-6.3] post-CPOE, P = .75). CPOE led to a significant time reduction of 16% (1.3 [0.3-2.3] minutes) for simple and 60% (8.6 [5.1-12.1] minutes) for complex calculations.
CONCLUSIONS: CPOE including a special CDS tool preserved accuracy for calculation and control of glucose intake and increased prescribing time efficiency.
Policy from the American Academy of Pediatrics related to prescribing. This article does not break down the functions in a way that would allow testing for the presence of any of the functionality, but does speak in general terms about the importance of weight-based dosing, the need for EHR integration, and appropriate literacy levels in patient education materials.[su_cite_pediatrics url_fragment = ‘131/4/824’ author = ‘Del Beccaro’ year = ‘2013’] | PubMed 23530170 | Author Search
A technical report from the American Academy of Pediatrics that reviews what is known about computerized medication prescribing in children, and current public policy initiatives. Includes a short list of requirements for prescribing systems, including weight-based dosing, the ability to identify compounding pharmacies, and messaging standards that include weight with prescriptions.[su_cite_pediatrics url_fragment = ‘131/4/e1350’ author = ‘Johnson’ year = ‘2013’] | PubMed 23530183 | Author Search
A 2007 technical report on electronic prescribing systems and their limitations and potential benefits to the pediatrician in the ambulatory setting. Written in an era when much e-prescribing was done in separate systems dedicated to that purpose, rather than from within an integrated EMR, which is the usual case today.[su_cite_pediatrics url_fragment = ‘119/6/e1413’ author = ‘Gerstle’ year = ‘2007’] | PubMed 17545368 | Author Search