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


This study suggests that CPOE may prevent errors from reaching and harming patients, but it does not provide evidence that facilities with CPOE have fewer medication errors than facilities without CPOE because different institutions appeared to have different levels of underreporting of medication errors under the voluntary reporting system. It is apparent that Medmarx data cannot be used to assess the potential benefits of CPOE previously reported[2,11,12] or compare rates of medication errors between providers.

Our analysis suggests substantial variation among voluntarily participating facilities in the diligence with which they reported medication errors. Such variation in reporting diligence explains our failure to show the expected benefits of the CPOE systems. It should be aware that, as the number of participating facilities increases, Medmarx may have enough power to detect statistically significant differences in medication errors between CPOE and non-CPOE facilities. But such statistical differences should be understood in the context of the data collection process and should not be seen as evidence for comparing safety performance between reporting providers.

Consistent with findings of Koppel et al.,[13] our quantitative and qualitative analyses indicate that CPOE can lead to medication errors not only due to faulty computer interface design but also due to common human errors such as typing errors. Our analysis suggests that the CPOE systems currently implemented often lack adequate decision support to intercept drug interactions and contraindications and that the systems are often not linked with other systems such as medication administration systems. The development and implementation of CPOE systems should include not only optimal system design but also features that prevent common human errors.

This study sheds light on the value and the proper use of voluntary medical error reporting, which has been a focus of patient safety efforts in the past five years.[14,15] This study demonstrates that a voluntary medical error-reporting system offers valuable insights into medical errors associated with specific technologies or drugs. The reported cases reveal where errors occur, their causes, their consequences, and so on. The textual descriptions of errors provide further lessons about how the errors occur and how to prevent them.

This study also reveals important limitations of a voluntary medical error-reporting system in patient safety research. It does not support estimation of prevalence. Because the reporting is voluntary, some events may not be reported. In addition, determining the denominator is difficult, if possible at all. Consequently, voluntarily reported data do not support comparative analysis of institutions on overall safety performance. It is apparent from the Medmarx data that CPOE and non-CPOE facilities reported errors very differently, as indicated by large variation in the number of errors reported by the participating facilities. Many factors, such as leadership and resource commitment, staffing, legal concerns, and experience with the reporting system, affect how completely an institution reports errors voluntarily. A clear understanding of the levels of underreporting by participating organizations and, ideally, commitment of voluntary participants to report all or selected categories of medical errors would substantially increase the value of the data in terms of identifying best practices for reducing medical errors.


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: