Automating the Human Factors Analysis and Classification System (HFACS): An Initial Investigation Based on Error Reports in Radiation Oncology

This study constitutes an evaluation of two datasets of error reports from the University of North Carolina School of Medicine's Department of Radiation Oncology. These errors were reported in accord with the Human Factors Analysis and Classification System (HFACS), using HFACS Level 1 and HFACS Level 2 codes. Keywords from an initial dataset of 58 reports, a list of HFACS theory keywords, and a list of related thesaurus words were used to develop a dictionary of signal words. These words are related to the HFACS Level 1 code for "Condition of Operator" and the HFACS Level 2 code for "Inattention-Distraction." These words were evaluated for relevance to the "Inattention-Distraction" category, both at face value, and in the context of "Inattention-Distraction" reports from the initial dataset of 58 reports. The signal words were then evaluated a second time in the context of a second dataset of 3459 reports, for confirmation of contextual relevance. The findings suggest that while more data is needed for future research, findings could lead to the development of an automated and more user-friendly HFACS reporting system at UNC School of Medicine's Department of Radiation Oncology.Master of Science in Information Scienc