PEDIATRICS Vol. 121 No. 1 January 2008, pp. 123-128 (doi:10.1542/10.1542/peds.2007-0919)
ARTICLE |
Medication Administration Variances Before and After Implementation of Computerized Physician Order Entry in a Neonatal Intensive Care Unit
a Department of Pediatrics, University of Washington, Seattle, Washington
b Nursing Research Service
d Department of Pediatrics, Madigan Army Medical Center, Tacoma, Washington
c Department of Family and Child Nursing, University of Washington School of Nursing, Seattle, Washington
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
OBJECTIVE. The goal was to determine whether implementation of a computerized physician order entry system was associated with a decrease in medication administration variances in a NICU.
METHODS. A prospective observational study was conducted. Research nurses recorded details of medication administrations for patients in a NICU during standardized observation periods. Details of each administration were compared with the medication order; a variance was defined as a discrepancy between the order and the medication administration. Rates of variances before and after implementation of computerized physician order entry in the NICU were compared. Specific types of and reasons for variances were also compared.
RESULTS. Data on 526 medication administrations, including 254 during the pre-computerized physician order entry period and 272 after implementation of computerized physician order entry, were collected. Medication variances were detected for 19.8% of administrations during the pre-computerized physician order entry period, compared with 11.6% with computerized physician order entry (rate ratio: 0.53). Overall, administration mistakes, prescribing problems, and pharmacy problems accounted for 74% of medication variances; there were no statistically significant differences in rates for any of these specific reasons before versus after introduction of computerized physician order entry. Administration of a medication at the wrong time accounted for 53.1% of all variances. Variance rates related to giving a drug at the wrong time were significantly lower in the computerized physician order entry period than in the pre-computerized physician order entry period (rates: 6.7% and 9.9%, respectively; rate ratio: 0.53).
CONCLUSIONS. Implementation of computerized physician order entry in a NICU was associated with a significant decrease in the rate of medication administration variances. However, even with the use of computerized physician order entry, variances were noted for >11% of all medication administrations, which suggests that additional methods may be needed to improve neonatal patient safety.
Key Words: computerized physician order entry • medication errors • medical errors • patient safety
Abbreviations: CPOE—computerized physician order entry • MAMC—Madigan Army Medical Center • ADE—adverse drug event • CI—confidence interval • RR—rate ratio
Medication errors are perhaps the most common threat to patient safety. Providing the proper drug therapy to a hospitalized patient involves several steps and multiple individuals; a mistake at any point in the process from ordering of a medication to its administration may lead to a significant error. For multiple reasons, pediatric patients may be at greater risk for medication errors and more vulnerable to their effects.1–3 Kaushal et al4 reported that the rate of potential adverse drug events (ADEs) resulting from medication errors was threefold higher for children than for adult patients.
Computerized physician order entry (CPOE) has been recommended as a method to reduce medication errors in pediatrics.5 It has been estimated that CPOE could conceivably prevent 76% of potential ADEs in children.6 The results of a comprehensive review of studies on the efficacy of CPOE in adult patients indicated that this technology reduces medication errors substantially.7 Data on pediatric patients are mixed. Although most studies showed that the use of CPOE led to fewer medication errors,8–11 one group of investigators reported a statistically significant increase in the mortality rate after implementation of CPOE in a large children's hospital.12
For the current study, we compared the frequency of medication administration variances before and after implementation of a CPOE system in a NICU. A variance was defined as a discrepancy between a medication order and what was actually administered to the patient. To detect variances, research nurses observed directly the administration of medications to neonates in the NICU. Before the study, we postulated that the rate of medication variances would be reduced after implementation of CPOE.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
A prospective, before/after, observational study was conducted. Research nurses made standardized observations of the administration of medications to infants who were inpatients in the Madigan Army Medical Center (MAMC) NICU and recorded the details of the administration. These data were compared with the information in the medication order and medication administration record to determine whether there was a variance between the order and the administration. Study data were collected between August 2004 and April 2006. During the period between August 2004 and June 2005 (pre-CPOE period), medication orders were written by hand in the NICU. CPOE was introduced in the NICU in July 2005, and data on medication orders with CPOE were collected between August 2005 and April 2006.
The MAMC NICU includes 12 level 3 beds and 18 level 2 beds. Physician care is provided by attending neonatologists and pediatric residents. Nursing care is provided predominately by a team of registered nurses who are trained specifically in neonatal care. These nurses are responsible for administering ordered medications.
Before CPOE, duplicate-copy medication order sheets were kept on a clipboard at the patient's bedside. On each prescribing occasion, the physician handwrote and signed the intended new order and placed the clipboard in a visible location or handed it to the nurse. The nurse transcribed the new order into a computerized charting system, which transferred the order automatically to the medication administration record. The paper copy of the order was then sent via fax machine to the inpatient pharmacy, where it was filled. When the order was filled, the medication was transported back to the NICU.
The Essentris CPOE software system (Clinicomp, San Diego, CA) was used in the NICU at MAMC. This CPOE system had been implemented in most other units at MAMC before its introduction in the NICU. However, CPOE display formats and defaults were configured specifically for use in the NICU. The system included rules and alerts on the order entry screens to prevent errors such as adverse interactions and duplications. Drop-down pick lists were included to simplify drug, dose, and route selection. Providers received warnings when invalid orders were entered or when improper doses of commonly used NICU medications were ordered. In addition, order sets specific for common NICU conditions were provided. Each patient's electronic record included a continuous log of all actions taken for every medication order. Once an electronic medication order was entered, the order entry system automatically populated the information into the electronic medication administration record and other appropriate flowsheets. During the data collection period, the system had inbound pharmacy capability but no outbound interface. The pharmacy received the inpatient medication order from the order entry system; the order was then reentered manually into another computer system for processing.
During the study, research nurses visited the NICU to conduct observations of medication administrations. Observations were conducted during 3-hour periods and occurred during day, evening, and night nursing shifts. Observation periods occurred on different days of the week and at different times of the day, to yield data on a representative sample of medication administrations. One research nurse was present for each observation period.
Before the beginning of each observation period, the research nurse reviewed the medication orders and the medication administration record of each patient in the NICU. The research nurse noted all medications that were due to be administered during the observation period and recorded the details of each ordered medication on a study form. Data also were collected on new medication orders that occurred during the observation period. All oral, intravenous, and topical medications were eligible for inclusion in the study. Data also were collected on total parenteral nutrition solutions. Orders for orally administered fluids and nutrition were excluded, as were orders for blood products.
During an observation period, NICU nurses who had agreed to participate in the study notified the research nurse when they were preparing a medication for administration. Data also were recorded on intravenous medication administrations that had been started before the observation period but were ongoing during the period. Data were not collected for medications administered by nurses who had not agreed to participate in the project.
For each medication order, the following information was recorded: patient research record number, name of drug, dose, route of administration, and time of administration. The same information was collected for each medication administration. In addition, the research nurse noted any problems with the administration and indicated reasons for any detected variances between the medication order and administration. Possible reasons for variances that were listed on the study form included the following: medication not available, order clarification required, medical emergency, care required for another patient, and miscommunication between nurses or with physicians or the pharmacy. The research nurse also recorded reasons for variances that were not listed on the form and provided other narrative details.
The unit of analysis for the study was the medication order, including orders to "hold" or to discontinue a medication that were written during an observation period. Final classification of whether a medication was administered as ordered was based on criteria developed by Barker et al.13 Administration of a medication was considered at variance with the order if administration was omitted, if the drug was given at the wrong time, if an incorrect dose was administered, if the drug was given through the wrong route, or if a drug was administered without an order. A medication was classified as given at the wrong time if it was administered 
60 minutes before or after the time ordered. Medication administrations that were >60 minutes late and were not given before the end of the observation period were considered omitted. Medication doses that differed by >10% from the dose ordered were classified as wrong doses. Medications that were given despite a lack of orders regarding the route of administration were categorized as administrations through the wrong route.
Through review of the additional data recorded by the research nurses, medication administration variances were further classified according to the reason for the variance. Reasons included prescribing problems, pharmacy problems, intentional violations, administration mistakes, intravenous access issues, and the nurse being busy with an acute problem involving another patient. Administrations that were late because of the time needed to clarify an unclear or incorrect order were categorized as prescribing problems. Administrations that were given despite the lack of a specified route of administration were also considered prescribing problems. Pharmacy problems leading to medication administration variances included delays in delivering an ordered drug or sending the wrong drug or the wrong form of a drug, leading to late administration. Intentional violations occurred when a nurse administered a medication at the wrong time on purpose, usually because of an infant's feeding schedule.
The incidence of variance of medication administrations in the pre-CPOE study period was compared with that in the CPOE period with the use of negative binomial regression analysis. Estimates of rate ratios (RRs) for which the 95% confidence interval (CI) did not include 1.0 were considered indicative of statistically significant differences. The date of each observation was included in the model as a random-effects term. To account for differences in patient volume, the patient census on each observation day was included in the regression model. Finally, to adjust for acuity level, the average patient workload for each observation day was also included in the regression model. The average patient workload was calculated by dividing the sum of the values for required nursing care hours for all patients in the unit by the NICU census. The required nursing care hours in the MAMC NICU were determined by using a Department of Defense system (the Workload Management System for Nursing). With this system, the nursing staff estimates the number of care hours required to provide care for each patient on the unit, for each shift.
In addition to the comparison of total rates of medication administration variances, similar analyses comparing different types of and reasons for variances during the pre-CPOE and CPOE periods were conducted. Finally, subanalyses were conducted after exclusion of data for observations made in August 2005, shortly after the introduction of CPOE in the NICU.
In an observational study of medication administrations to adult patients, a 19% medication error rate was reported.13 On the basis of this estimate, data on 
200 administrations from the pre-CPOE period and 200 administrations via CPOE were needed for an 80% chance of detecting a 10-percentage point difference in variance rates with the 2 systems.
The study was approved by the MAMC institutional review board. Written informed consent was obtained from nurses who were observed administering medications. Of the 42 neonatal nurses at MAMC, 33 (79%) consented to participate. Among the 9 who chose not to participate, 4 were leaving the unit soon and 3 were either on-call or part-time nurses who were not on duty during an observation period. No identifying data on nurses were collected as part of the study. In situations in which a research nurse observed that a significant medication error was about to occur (eg, the research nurse observed that an incorrect dose of medication was about to be administered to a patient), the nurse who was to administer the medication was notified, so that the error could be prevented. However, these events were counted as variances for the study analyses.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Research nurses observed medication administrations on 53 separate days during the study. A comparison of observation sessions conducted in the pre-CPOE and CPOE periods is shown in Table 1. Data were collected on 526 administrations, including 254 during the pre-CPOE period and 272 medications ordered using CPOE. Observations were made for 48 different medications. The 10 most commonly observed medication administrations are listed in Table 2. Overall, these drugs accounted for similar proportions of the total observed medication administrations during the pre-CPOE and CPOE periods (72.4% and 72.0%, respectively; P = .92); however, there were differences for some individual drugs. Caffeine and sodium chloride each accounted for greater proportions of observed medication administrations during the pre-CPOE period (P = .019 and P = .018, respectively), and gentamicin and ampicillin each constituted greater proportions of observed administrations after the implementation of CPOE (P < .001 and P = .013, respectively).
|
|
Because of incomplete data, no determination of variance could be made for 5 of the 526 medication administrations. A variance was noted for 81 (15.5%) of remaining 521 administrations. The rate of variance was significantly lower during the period when CPOE was used for medication orders, compared with the pre-CPOE period (variance rates: 11.6% and 19.8%, respectively; RR: 0.53; 95% CI: 0.33–0.84). During the initial month of data collection after institution of CPOE, a variance was noted for 9 (26.5%) of 34 observed medication administrations. This variance rate was significantly higher than that observed during the rest of the CPOE period (RR: 2.67; 95% CI: 1.22–5.82). When data obtained during this "rollout period" were excluded, the difference in variance rates between the CPOE and pre-CPOE periods was more pronounced (rates: 9.4% and 19.8%, respectively; RR: 0.42; 95% CI: 0.25–0.71).
The reasons for medication administration variances during the pre-CPOE and CPOE periods are summarized in Table 3. As can be seen in Table 3, the overall rates of variance for administration mistakes, pharmacy problems, and prescribing problems were similar. These 3 reasons accounted for 74% of all variances observed. Except for pharmacy problems, the RRs comparing variance rates for the pre-CPOE and CPOE periods were <1.0. However, the 95% CI for each RR estimate included 1.0, which indicated that there were no statistically significant differences.
|
The frequencies of different types of medication administration variances are displayed in Table 4. As can be seen in Table 4, variances related to administration of drugs via the wrong route and at the wrong time were both significantly lower after the implementation of CPOE. Administration of medications at the wrong time accounted for 53.1% of all observed variances. Of the 43 administrations given at the wrong time, 5 were given too early. Among the 38 medication administrations that were late, 7 (18.4%) were attributable to administration mistakes, 5 (13.2%) were attributable to prescribing problems, 12 (31.6%) were attributable to pharmacy problems, 8 (21.1%) were intentional violations, 5 (13.2%) were attributable to intravenous access issues, and 1 (2.6%) occurred because the nurse was busy with another patient with emergency needs.
|
As shown in Table 4, all of the administrations via the wrong route occurred during the pre-CPOE period. Nine of the 10 variances of this type were attributable to a medication being given without a route of administration being specified in the order, a problem addressed directly by CPOE. When data for this specific type of problem were excluded from the analysis, differences in the rates of variances between the CPOE and pre-CPOE periods were of only marginal significance (RR: 0.62; 95% CI: 0.37–1.04; P = .07). When data from the CPOE rollout period were not included, the rate of variance after implementation of CPOE was significantly less than that during the pre-CPOE period even with exclusion of this one type of problem (RR: 0.49; 95% CI: 0.28–0.86).
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The results of this study indicate that the introduction of CPOE in a NICU was associated with a significant reduction in medication administration variances. After the initial rollout period, when the variance rate might be expected to increase temporarily,14 there was a more than twofold reduction in the rate of administration variances when medications were ordered via CPOE, compared with the pre-CPOE period. Rather than an improvement in a specific part of the process of ordering, dispensing, and administering a drug, the reduction in variances observed with CPOE seemed to be the result of an overall general improvement. The increased efficiency associated with the use of CPOE led to significantly fewer medications being given at the wrong time, in addition to eliminating the problem of drugs being administered without an order for a specified route.
Our measured rate of medication administration variances of 19.8% during the pre-CPOE period was remarkably similar to the 19% error rate that Barker et al13 found in their study of medication administrations in selected hospitals and skilled nursing facilities in Colorado and Georgia. Those investigators also performed direct observation to detect administration errors. As in our study, Barker et al13 found that administration of medications at the wrong time was the most common type of error. Although the prescribing problems, pharmacy problems, and mistakes in administration that we found should be categorized as errors, classification of variances related to intravenous access problems, intentional violations, and the nurse being busy with a patient with an acute condition was more difficult.
Intentional violations, which usually occurred when a nurse delayed giving an orally administered medication to coincide with an infant's feeding schedule, are a particularly difficult type of variance to categorize. In observational studies of commercial airline pilots during actual flights, a similar type of variance, known as conscious "noncompliance" with standing operating procedures, was the most common type of "error" detected and occurred on >50% of observed flights. However, only 2% of those errors were considered "consequential" by the observer.15 Similarly, virtually all of the intentional violations detected in our study were trivial in nature and were the result of a highly trained professional exercising his or her judgment regarding proper care.
Our main finding, that variances between medication orders and administrations were reduced after implementation of CPOE, is consistent with other quantitative evaluations of this technology in preventing medication errors in pediatrics. King et al8 compared medication error rates and ADEs in 2 wards in a tertiary pediatric hospital in Canada; CPOE was used in one of the wards. Data on errors were collected via passive reporting systems. At baseline, error and ADE rates were similar. After implementation of CPOE, however, the error rate in the ward using this technology was significantly lower than that in the control ward (RR: 0.6; 95% CI: 0.48–0.74). No difference in the rates of ADEs was noted. Potts et al9 reviewed medication orders in a pediatric critical care unit before and after the implementation of CPOE. Those investigators found that the rate of potential ADEs decreased by 40.9% with the use of CPOE. In a study of NICU patients, Cordero et al11 assessed the efficacy of CPOE in reducing medication errors by focusing on 2 specific medication administration issues, namely, the time from ordering to administration of caffeine loading doses and the accuracy of gentamicin dosing. Those investigators found a significant decrease in the time to caffeine loading with CPOE, compared with the pre-CPOE period. Before the implementation of CPOE, 16 (11.7%) of 136 orders for gentamicin were 10% greater or 10% less than the recommended dose. With CPOE, 0 of 128 evaluated gentamicin doses were incorrect.
Our project was substantially different from the other studies on CPOE for children in 2 significant ways. First, we focused exclusively on medication administrations, because this represents the final point at which a medication variance (or error) can occur or can be prevented before it reaches the patient. Regardless of whether CPOE is used to order a medication or the order is handwritten, the systems in place to detect serious medication errors are robust. Most prescribing errors are detected and fixed by pharmacists and/or nurses, and most dispensing errors are caught by the nurse administering the medication.4,16 For example, in the seminal study by Kaushal et al,4 616 medication order errors were detected, but only 5 (0.8%) of those resulted in preventable ADEs. By comparing administration variances that occurred before and after the implementation of CPOE, we were able to account for the inherent robustness of the system in our evaluations. Conversely, even if an order for a medication is written and dispensed perfectly, the patient will receive the proper drug only if it is administered correctly. Finally, almost 50% of the variances we detected involved late administration of medications. Although some of this type of variance was related directly to administration mistakes, more commonly the cause of late administration of a medication involved the pharmacy not dispensing the medication correctly or in a timely manner. Incorrect or unclear physician orders also led to late administrations. Therefore, monitoring of timeliness allowed assessment of multiple phases of the process of medication administration.
The other unique feature of our assessment of CPOE for children was the use of direct observation to detect medication administration variances. One of the main reasons for using this method is that direct observation provides more complete assessment of medication errors, compared with other commonly used techniques. In a previous study of 2556 medication doses, 300 errors were detected through direct observation, compared with 17 via chart review; only 1 error was reported via incident reports.17 Therefore, the use of direct observation is the most-rigorous method for comparing variances in administration of medications ordered with CPOE or with handwritten orders. This technique also allows unbiased comparison of these 2 methods for ordering drugs. CPOE systems are designed specifically to reduce certain types of errors, and it is likely that those specific errors will be reduced. Unfortunately, implementation of new technology leads frequently to new and unforeseen errors. When a CPOE system was implemented at the Children's Hospital of Pittsburgh, there was an unexpected increase in the mortality rate.12 Some of this increase might have been related to unforeseen delays in ordering medications with the new system for recently admitted, critically ill children. There is a growing awareness of medication errors that are related specifically to CPOE.18–21 In any comparison of rates of medication errors (or variances), it is crucial that surveillance for these unanticipated events be as rigorous as that for expected errors. Direct observation allows such an evaluation.
There are weaknesses in the design of our study that limit the interpretation of the results. By focusing on medication administrations, we might have missed some significant errors in drug ordering. Direct observation is a labor-intensive method. Although we attempted to collect data in a representative manner, the frequency and timing of observation sessions were limited by the availability of the research nurses. Because of this limitation, observations were conducted in 2 months (May and June) during the pre-CPOE period for which no comparable data were collected after the implementation of CPOE. It is possible that the NICU nurses changed their processes for medication administrations when they were being observed, thus skewing our results. Although this might change the proportions of variances detected, it is doubtful that any change in nurse performance related to being observed would bias the comparison between the pre-CPOE and CPOE periods. The sample size determination for the study was based on having sufficient power to detect an overall change in variance rates between the pre-CPOE and CPOE periods. With this sample size, however, the power to detect differences in specific types of variances or reasons for variances was limited. Finally, as with any study using a before/after comparison, differences other than the implementation of CPOE might have influenced the results. In our analysis, we attempted to control for changes in NICU census and acuity level; however, factors such as turnover in nursing, physician, and pharmacy personnel might have altered the results.
Despite these limitations, we are confident that our main finding, that is, that the use of CPOE was associated with a reduction in medication variances in the NICU at MAMC, is valid. However, with the use of a rigorous method, a variance between medication orders and administration was detected 10% of the time for drugs ordered with CPOE. Although many of these variances were likely trivial in nature, this finding highlights the complexities of providing hospitalized patients with the proper dose of the proper medication at the proper time, and it suggests that other interventions, in addition to CPOE, are needed to solve the problem of medication errors.
|
ACKNOWLEDGMENTS |
|---|
This study was funded by a grant from the Agency for Healthcare Research and Quality.
|
FOOTNOTES |
|---|
Accepted Jul 2, 2007.
Address correspondence to James A. Taylor, MD, Department of Pediatrics, University of Washington, Box 354920, Seattle, WA 98195. E-mail: uncjat{at}u.washington.edu
The authors have indicated they have no financial relationships relevant to this article to disclose.
The views expressed in the article are those of the authors and do not reflect the official policy of the Department of the Army, the Department of Defense, or the US Government. The investigators have adhered to the policies for protection of human subjects prescribed in 45 CFR 46.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
- Koren G, Barzilay Z, Modan M. Errors in computing drug doses. Can Med Assoc J. 1983;129 :721 –723[Abstract]
- Rowe C, Koren T, Koren G. Errors by paediatric residents in calculating drug doses.
Arch Dis Child. 1998;79
:56
–58
[Abstract/Free Full Text] - Sullivan JE, Buchino JJ. Medication errors in pediatrics: the octopus evading defeat. J Surg Oncol. 2004;88 :182 –188[CrossRef][Web of Science][Medline]
- Kaushal R, Bates DW, Landrigan C, et al. Medication errors and adverse drug events in pediatric inpatients.
JAMA. 2001;285
:2114
–2120
[Abstract/Free Full Text] - Stucky ER. Prevention of medication errors in the pediatric inpatient setting.
Pediatrics. 2003;112
:431
–436
[Abstract/Free Full Text] - Fortescue EB, Kaushal R, Landrigan CP, et al. Prioritizing strategies for preventing medication errors and adverse drug events in pediatric inpatients.
Pediatrics. 2003;111
:722
–729
[Abstract/Free Full Text] - Kaushal R, Shojania KG, Bates DW. Effects of computerized physician order entry and clinical decision support systems on medication safety: a systematic review.
Arch Intern Med. 2003;163
:1409
–1416
[Abstract/Free Full Text] - King WJ, Paice N, Rangrej J, Forestell GJ, Swartz R. The effect of computerized physician order entry on medication errors and adverse drug events in pediatric inpatients.
Pediatrics. 2003;112
:506
–509
[Abstract/Free Full Text] - Potts AL, Barr FE, Gregory DF, Wright L, Patel NR. Computerized physician order entry and medication errors in a pediatric critical care unit.
Pediatrics. 2004;113
:59
–63
[Abstract/Free Full Text] - Upperman JS, Staley P, Friend K, et al. The impact of hospitalwide computerized physician order entry on medical errors in a pediatric hospital. J Pediatr Surg. 2005;40 :57 –59[CrossRef][Web of Science][Medline]
- Cordero L, Kuehn L, Kumar RR, Mekhjian HS. Impact of computerized physician order entry on clinical practice in a newborn intensive care unit. J Perinatol. 2004;24 :88 –93[CrossRef][Medline]
- Han YY, Carcillo JA, Venkataraman ST, et al. Unexpected increased mortality after implementation of a commercially sold computerized physician order entry system.
Pediatrics. 2005;116
:1506
–1512
[Abstract/Free Full Text] - Barker KN, Flynn EA, Pepper GA, Bates DW, Mikeal RL. Medication errors observed in 36 health care facilities.
Arch Intern Med. 2002;162
:1897
–1903
[Abstract/Free Full Text] - Weant KA, Cook AM, Armitstead JA. Medication-error reporting and pharmacy resident experience during implementation of computerized prescriber order entry.
Am J Health Syst Pharm. 2007;64
:526
–530
[Abstract/Free Full Text] - Helmreich RL. On error management: lessons from aviation.
BMJ. 2000;320
:781
–785
[Free Full Text] - Wang JK, Herzog NS, Kaushal R, Park C, Mochizuki C, Weingarten SR. Prevention of pediatric medication errors by hospital pharmacists and the potential benefit of computerized physician order entry. Pediatrics. 2007;119(1) . Available at: www.pediatrics.org/cgi/content/full/119/1/e77
- Flynn EA, Barker KN, Pepper GA, Bates DW, Mikeal RL. Comparison of methods for detecting medication errors in 36 hospitals and skilled-nursing facilities.
Am J Health Syst Pharm. 2002;59
:436
–446
[Abstract/Free Full Text] - Walsh KE, Adams WG, Bauchner H, et al. Medication errors related to computerized order entry for children.
Pediatrics. 2006;118
:1872
–1879
[Abstract/Free Full Text] - Koppel R, Metlay JP, Cohen A, et al. Role of computerized physician order entry systems in facilitating medication errors.
JAMA. 2005;293
:1197
–1203
[Abstract/Free Full Text] - Horsky J, Kuperman GJ, Patel VL. Comprehensive analysis of a medication dosing error related to CPOE.
J Am Med Inform Assoc. 2005;12
:377
–382
[Abstract/Free Full Text] - Campbell EM, Sittig DF, Ash JS, Guappone KP, Dykstra RH. Types of unintended consequences related to computerized provider order entry.
J Am Med Inform Assoc. 2006;13
:547
–556
[Abstract/Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2008 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
