Javascript must be enabled for the correct page display

Alarm Hazards on the Pediatric Intensive Care Unit

Brinks, K. (2015) Alarm Hazards on the Pediatric Intensive Care Unit. Master's Thesis / Essay, Artificial Intelligence.

[img]
Preview
Text
AI-MHMC-2015-K.BRINKS.pdf - Published Version

Download (1MB) | Preview
[img] Text
toestemming.pdf - Other
Restricted to Backend only

Download (711kB)

Abstract

Medical machines alert nurses on a pediatric intensive care unit (PICU) to changes in a patient’s status or treatment by producing alarms. The current organization of medical alarms causes hazards that can be detrimental to a patient’s care. An excessive number of alarms, lack of alarm integration into central points, or inappropriate alarm boundary configuration are examples of these hazards. This study addressed alarm hazards on the University Medical Center Groningen’s (UMCG) PICU by answering three questions: how can alarm hazards in the ICU environment be analyzed, what are the most hazardous alarm risks found on the UMCG’s PICU, and how can the most prominent alarm hazard be addressed? The study used a cognitive work analysis to find all relevant aspects in the work of critical care nurses regarding alarms. The analysis provided a formalization of the nurses’ work domain, tasks, and strategies. A consecutive risk analysis used the obtained framework to find the most hazardous alarm risks. This risk analysis was based on a combination of the SHERPA and HFMEA methods, and obtained its information from experienced PICU nurses. Results showed the most prominent hazards to be the setting of alarm boundaries for patient monitors and IV pumps, the malfunctioning of pagers, and a lack of knowledge to identify the cause of alarms. The latter was addressed by building a knowledge-based system prototype that supports decision making and training by suggesting diagnoses for patient monitor alarm combinations. The system was built using knowledge and rules elicited from the most experienced PICU nurses. The validity and effectiveness of the system was confirmed during expert reviews. Although the resulting system is only a prototype, the application of knowledge-based systems to reduce alarm hazards looks very promising. The full research provided a theoretically grounded approach for improving alarm hazards in critical care that can be used as a paradigm by researchers, ICU managers, and medical device manufacturers interested in improving alarms.

Item Type: Thesis (Master's Thesis / Essay)
Degree programme: Artificial Intelligence
Thesis type: Master's Thesis / Essay
Language: English
Date Deposited: 15 Feb 2018 08:07
Last Modified: 15 Feb 2018 08:07
URI: http://fse.studenttheses.ub.rug.nl/id/eprint/13273

Actions (login required)

View Item View Item