Patient identification (ID) errors occurring during the medication administration process can be fatal. The aim of this study is to determine whether differences in nurses' behaviors and visual scanning patterns during the medication administration process influence their capacities to identify patient ID errors. Nurse participants (n = 20) administered medications to 3 patients in a simulated clinical setting, with 1 patient having an embedded ID error. Error-identifying nurses tended to complete more process steps in a similar amount of time than non-error-identifying nurses and tended to scan information across artifacts (e.g., ID band, patient chart, medication label) rather than fixating on several pieces of information on a single artifact before fixating on another artifact. Non-error-indentifying nurses tended to increase their durations of off-topic conversations-a type of process interruption-over the course of the trials; the difference between groups was significant in the trial with the embedded ID error. Error-identifying nurses tended to have their most fixations in a row on the patient's chart, whereas non-error-identifying nurses did not tend to have a single artifact on which they consistently fixated. Finally, error-identifying nurses tended to have predictable eye fixation sequences across artifacts, whereas non-error-identifying nurses tended to have seemingly random eye fixation sequences. This finding has implications for nurse training and the design of tools and technologies that support nurses as they complete the medication administration process.
(c) 2011 APA, all rights reserved.