Severity of medication administration errors detected by a bar-code medication administration system

From Clinfowiki
Jump to: navigation, search

This article examines the potential impact of medication errors collected from bar-coded medication administration systems from six community hospitals. Reports released by the institute of medicine between 1999 and 2001 brought the subject of medication errors to the forefront of quality improvement agendas.

Six types of errors were measured and rated by a panel of three pharmacists, two registered nurses, and one physician. The categories included four types of dose or administering-related errors and two types of order-related errors. The panel was presented with each error situation and asked to rate them from a scale of 0-10 for their potential to cause harm in a given situation. A total of 945 scenarios were reviewed, including 579 prevented errors and 366 that showed discrepancies after the medication was administered. 91% of the errors detected by the system were rated as having minimal potential for harm in a typical scenario. Only 1% of total errors were rated as having a potential for severe harm, including life-threatening or lasting effects.

Evolving bar-code medication administration systems could have these results built into decision-support systems for clinicians administering medications. This could help reduce the effect of alert fatigue, where clinicians don’t pay attention to error-related messages that are frequently bypassed. Detected errors with a more severe potential for harm could trigger different alerts that cannot be bypassed with a single click or keystroke. Errors of higher severity could also be tracked more closely for prevention purposes.

The context of the medication errors was not included in the study. Each error was summarized in the form of what medication was ordered and the discrepancy that the system detected. The category of errors related to lack of orders in the system was rated the most harmful by the review panel, which 50% of scenarios rated as having a potential to cause moderate or severe harm. The context of this error, however, was not presented to the reviewers. One weakness of this approach was that medications for which there was no order in the system would not be analyzed for contraindications for certain conditions or drug interactions. With this in mind, there is a possibility that these types of errors were more harmful than the study found.

Although the prevention of medication errors is important in improving the quality of medicine, we may be devoting more resources to their study due to the fact that they are easy to quantify and measure. Studies have suggested the most serious errors are related to diagnosis, and many are not detected using our current methods of data collection.