Reckmann, Margaret H; Westbrook, Johanna I; Koh, Yvonne; Lo, Connie; Day, Richard O

Table 1

Details of Studies Evaluating the Impact of CPOE on Prescribing Error Rates in Pediatric Hospital Populations

Study (Reference)	Study Design	Study Setting	Study Population	CPOE System	Method of Error Detection	Prescribing Error RatePre-CPOE	Prescribing Error RatePost-CPOE	Outcome—Change in Prescribing Error Rate	Other Outcomes Measured
Potts292004	Prospective cohort study prepost CPOE implementation 2 + 2 mo	Tertiary care academic Children's Hospital (Nashville, United States) 20 bed pediatric critical care unit (PCCU)	13,828 medication orders (n = 514 Patients)	Home gtrn (some CDSS)	Chart review by designated clinical pharmacist	2662 errors in 6803 orders (39.1%)	110 errors in 7025 orders (1.6%)	95.9% relative reduction (39.1 Versus 1.6%, p < 0.001) in all types of errors associated with medication ordering	Potential ADEs reduced by 40.9% (2.2 versus 1.3%, p < 0.001) PEs reduced by 99.4% (30.1 versus 0.2%, p < 0.001) rule violations (RVs) reduced by 97.9% (6.8 versus 0.1%, p < 0.001)
Cordero242004	Retrospective cohort study prepost CPOE 6 + 6 mo	University Medical Center neonatal intensive care unit (NICU) (Columbus, OH, United States)	111 very low birth weight (VLBW) infants born consecutively within 6 mo before and 100 VLBW infants born within 6 mo after implementation of CPOE	Commercial extensively customized to NICU needs (some CDSS)	Full time NICU dedicated clinical pharmacist pre-data: obtained retrospectively from medical records Post-data: obtained electronically from computerized patient record	At time of admission: 14 of 105 (13%) gentamicin dosages deviated > 10% from recommended dose At time of suspected late-onset sepsis: 2 of 31 (6%) gentamicin dosages incorrect	At time of admission: no medication errors in 92 gentamicin dosages At time of suspected late-onset sepsis: no errors in 28 gentamicin dosages	Error rate in gentamicin dosage reduced to zero
Kim262006	Prospective prepost CPOE study 241 + 296 d	Academic Medical Center (Baltimore, United States) pediatric oncology	1259 pre-CPOE paper pediatric chemotherapy orders (n = 176 patients); 1116 post-CPOE pediatric orders (n = 167 patients)	Commercial (unknown CDSS)	Two-phase daily audit of chemotherapy orders	4.0% incorrect order format (treatment plan), 2.3% incorrect order format (order); 1.1% order and treatment plan nonmatch; 5.8% incorrect calculation	2.6% incorrect order format (treatment plan), 0.06% incorrect order format (order); 6.0% order and treatment plan nonmatch; 0.54% incorrect calculation	4 of 6 high importance steps in the daily chemotherapy ordering process were less likely to be incorrect after CPOE deployment	Daily audit of successful completion of specific steps in treatment plan (confusing audit data analysis results)
Vaidya302006	Simulated test environment, crossover study (handwritten and CPOE)	University Medical Center (Baltimore, United States) pediatric ICU	9 common IV infusions, 234 orders generated by 26 volunteer physicians	Home gtrn (some CDSS)		170 of 234 (73%) continuous IV infusion orders contained ≥ 1 prescribing errors	10 prescribing errors for 234 continuous IV infusion orders (4.3%)	94.1% relative reduction (73 versus 4.3%, p = 0.0001) in prescribing errors	Handwritten orders: 25% of errors were judged to be “high risk” (→ potential serious ADEs) each prescriber made at least one error CPOE orders: no “high risk” errors

ADE = adverse drug event; CDSS = clinical decision support systems; CPOE = computerized provider order entry; ICU = intensive care unit; IV = intravenous.

Does Computerized Provider Order Entry Reduce Prescribing Errors for Hospital Inpatients? A Systematic Review

This Article

Access policy for JAMIA