Previous Abstract | Next Abstract
Printable Version
October 26, 2005
9:00:00 AM - 11:00:00 AM
Hall C4
Failure in Context: Linking Observed Behavior to Cognition, Tasks and Adverse Events
Mark E. Nunnally, M.D., Yuval Bitan, Ph.D., Chris Nemeth, Ph.D., Michael O'Connor, M.D., Richard I. Cook, M.D.
Anesthesia and Critical Care, University of Chicago, Chicago, Illinois, United States
Introduction: Implementation of new technologies such as infusion devices poses unique challenges in the clinical setting [1]. Practitioners have difficulty programming infusion devices [2], but the causes of their difficulty is not well understood by either practitioners or their leadership. Users program devices to accomplish specific tasks, but imbedded in this process is a deeper level of cognition, in which goals, perceptions and beliefs translate into action. Video recording is a method that enables researchers to document and analyze observable behavior. Such recordings are potentially valuable sources of information, although this requires a way to link recorded behavior to cognitive theory, task categories and real clinical events. Computerized video presentation enables investigators to organize and categorize videotaped behaviors in a way that permits probing the ways people relate to and operate devices.

Methods: Forty clinicians were videotaped performing simple tasks on an infusion device. A reviewer selected examples of programming behaviors in 15- to 60-second long video segments. The researchers sorted video segments according to programming activities such as data entry, and cognitive activities such as decision-making. They sorted reports from an incident reporting database, looking for specific examples where observed behaviors may have played a role. This organization method formed the basis of a comparison tool to show the multiple links between data and concept.

Results: The researchers assembled traits into a table with hypertext links to video clips. These serve as examples of user programming activity. Pull-down text in the table explains the use context, the particular situation, and defines terms. By clicking on specific icons, users view related case reports. The multiple displayed relationships show how problems might manifest. The table and video segments will be available via the Internet and enable clinicians to understand the nature of programming complex electronic devices in the clinical setting.

Discussion: Research techniques that evaluate cognition and task performance provide insight into how failures arise [3]. Video recording captures clinician behavior in a form that can be made available for review and analysis. Human factors knowledge related to cognition sheds light on the larger issues related to clinicians' interactions with technology. Display tools that link concepts, recorded behaviors and actual events translate the clinician programming activity data into interpretive lessons. These insights about the clinical use of complex devices allow manufacturers, practitioners, and others who are interested in patient safety to develop guidelines for better design and training.

[1] Cook RI, Woods DD. Implications of Automation Surprises in Aviation for the Future of Total Intravenous Anesthesia (TIVA). J Clin Anesthesia (1996) 8, 29S-37S

[2] Nunnally M, Nemeth C, Brunetti V, Cook RI. Lost in Menuspace: User Interactions with Complex Medical Devices. Special issue on Studies in Healthcare Technical Work. IEEE Transactions on Systems, Man and Cybernetics-Part A. (2004) 34:736-742

[3] Hollnagel E, Woods DD. Cognitive Systems Engineering: New wine in new bottles. Int J Human-Computer Studies (1999) 51, 339-356

Anesthesiology 2005; 103: A1296