Abstract
Errors related to the administration of medication delivery are less common than those that occur during the ordering phase.2 However, traditional medication delivery systems have fewer safeguards to intercept administration errors, and serious consequences may result.
In a study by Leape et al,3 5% of medication errors identified were related to parenteral drug delivery. To address the problem of infusion errors, preprogrammed, customizable, computerized systems are being developed. ALARIS Guardrails software is designed to apply user-defined maximum/minimum infusion rate limits for selected medications (eg, those that are frequently used or have a narrow therapeutic index).
To gain a better understanding of this new technology, a observational, prospective study was conducted. The study's goals were to document how often smart infusiondevices alerted users to potential errors, to quantify specific alert types, to specify the medications involved in alerts, and to detect trends associated the devices. User satisfaction with the devices was also measured. This information was used to assess the value of smart-infusion software in the prevention of infusion-device programming errors. Data representing all device alerts that occurred on three special care units over an 8-month period (May 20, 2002 to January 13, 2003) was downloaded from the infusion devices. Seven categories were used to describe each alert. One-hundred fifty-seven alerts resulted in the discontinuation of the initial entry and subsequent reprogramming of the device to a different rate of infusion. A model was developed to assess the rate at which errors occurred and to assess the possible harm potential of events that may have been avoided by the alerts. A potentially life-threatening dose was initially programmed in 17 different events analyzed. The number of reprogramming events was small compared with the total number of doses administered. However, the outcome from any one of those errors could have been significant. Therefore, the device may prove valuable even in settings that have achieved a low rate of infusion-related errors.
Get full access to this article
View all access options for this article.
