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


Data Source

This study used Medmarx data from 2003. Medmarx is an anonymous, Internet-accessible reporting system for voluntarily participating hospitals and health care systems to report medication errors in a standard format (www.medmarx.com). USP analyzes the reports regularly, identifies patterns, makes recommendations to improve medication safety, and shares information with participating organizations. This system has been in place since August 1998 and has experienced substantial increases in the number of error reports.[16]

Medmarx uses the National Coordinating Council for Medication Error Reporting and Prevention (NCC MERP) definition of medication errors.[17] NCC MERP defines a medication error as any preventable event that may cause or lead to inappropriate medication use or patient harm while the medication is in the control of the health care professional, patient, or consumer. Such events may be related to professional practice, health care products, procedures, and systems including prescribing, order communication, product labeling, packaging, nomenclature, compounding, dispensing, distribution, administration, education, monitoring, and use. The information Medmarx collects includes whether the error occurred in an in-patient or outpatient setting; type, severity, and cause of error; location; staff and products involved; contributing factors; and corrective actions taken.[16] Medication errors are classified according to the NCC MERP index for categorizing medication errors, which classifies medication errors into four levels and nine severity categories: potential for error (category A), actual error without harm (categories B, C, and D), actual error with harm (categories E, F, G, and H), and actual error that may have contributed to or resulted in a patient's death (category I).[17] Harm is defined as impairment of the physical, emotional, or psychological function or structure of the body or pain. Since May 2003 reporting facilities have been asked to determine whether the error reported was caused by CPOE.[16]

Facilities that subscribe to the Medmarx program complete a facility profile form that provides descriptive information about the facility, such as bed size, type of facility, ownership, and monthly dose dispensed.[16] One question on the current facility profile asks facilities to report if the organization has CPOE and, if so, if CPOE is used throughout the organization or only in limited areas. The facility profile does not identify the vendor, the sophistication of clinical support, and how long the facility has used CPOE.


Three analyses were conducted. The first analysis compared the number of medication errors reported by facilities that had full CPOE implementation with the number reported by those that had no CPOE. The number of errors per 100,000 doses was calculated based on the monthly total dosages reported on the facility profile, and the rates were compared between CPOE and non-CPOE facilities. Facilities that reported partial CPOE implementation were excluded because the extent of CPOE implementation could vary substantially and could not be determined. Category A errors (potential errors) were excluded from the analysis because of concern that different facilities might have different understandings about the definition of category A errors.

The second analysis examined characteristics of errors reportedly caused by CPOE. This analysis was restricted to May-December 2003 Medmarx data, during which time facilities were asked to report whether the reported errors were caused by CPOE, and included errors in categories B-I, as well as category A, to explore circumstances in which CPOE caused errors. The third analysis was qualitative, entailing reading of the descriptions of the errors caused by CPOE, synthesizing them, and grouping them.

T tests were used to determine between-group differences, and χ2 statistics were used to determine differences in distributions of types and causes of errors. P values less than 0.05 were considered statistically significant.


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.