Potential Benefits and Problems With Computerized Prescriber Order Entry

Analysis of a Voluntary Medication Error-Reporting Database

Chunliu Zhan; Rodney W. Hicks; Christopher M. Blanchette; Margaret A. Keyes; Diane D. Cousins


Am J Health Syst Pharm. 2006;63(4):353-358. 

In This Article


During 2003, 570 facilities submitted 235,164 medication error reports to Medmarx. Among these 570 facilities, 120 reported that they had CPOE in all clinical areas, 81 facilities reported having CPOE in some clinical areas, and 339 did not have CPOE. Table 1 presents the selected characteristics of the facilities with full CPOE and those without CPOE. Most CPOE facilities were federally owned, and most non-CPOE facilities were general community hospitals.

Table 2 presents the number of medication errors reported by severity category for CPOE facilities and non-CPOE facilities. Most of the errors reported were from inpatient settings. Less than 2% of the errors reported resulted in harm to patients (categories E-I), and 20 of them resulted in patient deaths (category I). Table 2 indicates that a smaller percentage of errors reported by CPOE facilities reached patients compared with non-CPOE facilities (categories C-I, 41.78% versus 65.85% in inpatient errors, 39.66% versus 57.55% in outpatient errors, p < 0.01) or harmed patients (categories E-I, 1.37% versus 2.23% in inpatient errors, 0.69% versus 2.18% in outpatient errors, p < 0.01).

Table 3 compares medication errors reported by CPOE facilities and non-CPOE facilities. The number of events reported varied widely from facility to facility, ranging from 1 to 3450 inpatient medication errors, and such variation resulted in the large standard deviations shown in Table 3 . With the exception of total outpatient errors, there were no significant differences between CPOE facilities and non-CPOE facilities in mean number of errors reported, errors per 100,000 doses dispensed, and harmful errors.

Table 4 summarizes the characteristics of the 7029 CPOE-related medication error records. Most CPOE-related errors were category A errors (i.e., circumstances or events with the capacity to cause errors). About 4.7% of CPOE-related errors reached patients (categories C-E), and about 0.1% inflicted temporary harm to patients (category E). No patient suffered serious injuries due to CPOE-related errors (categories F-I).

Of the 6679 errors for which types of errors were identified, more than half were dosing errors, and most often these were wrong doses (79%), followed by wrong dosage forms (13%) and extra doses (8%). Prescriptions using CPOE were reportedly assigned to the wrong patients 231 times.

Of the 5457 errors for which one or more causes of errors were identified, knowledge deficiency was most frequently cited. A substantial number of errors involved computer entry, such as the use of abbreviations and dose calculations. There were errors reported, not shown in the table, when the system failed to flag drug-drug interactions, drug-allergy contraindications, and drug-disease contraindications.

Of the 1025 medication errors for which at least one contributing factor was identified, distractions were most often cited, followed by staff inexperience and heavy workload.

Table 5 lists a few examples of the textual descriptions of the CPOE-related errors. Some of the examples point to the faulty computer interface in the current CPOE systems. Many examples point to CPOE design failures, especially lack of connection with other parallel systems such as those for medication dispensing and administration. There were also examples pointing to the inadequacy of decision support in the CPOE systems used. Human errors could occur in interactions with the computer, such as picking the wrong items on a multiple-choice list on the computer screen and failing to differentiate look-alike patient names, such as John Smith and John Smith, Jr.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: