Applying the human factors analysis and classification system to critical incident reports in anaesthesiology.

BACKGROUND: The Human Factors Analysis and Classification System (HFACS) was developed as a practical taxonomy to investigate and analyse the human contribution to accidents and incidents. Based on Reason's "Swiss Cheese Model", it considers individual, environmental, leadership and organizational contributing factors in four hierarchical levels. The aim of this study was to assess the applicability of a modified HFACS taxonomy to incident reports from a large, anonymous critical incident database with the goal of gaining valuable insight into underlying, more systemic conditions and recurring schemes that might add important information for future incident avoidance.
METHODS: We analysed 50 reports from an anonymous, anaesthesiologic, single-centre Critical Incident Reporting System using a modified HFACS-CIRS taxonomy. The 19 HFACS categories were further subdivided into a total of 117 nanocodes representing specific behaviours or preconditions for incident development.
RESULTS: On an individual level, the most frequent contributions were decision errors, attributed to inadequate risk assessment or critical-thinking failure. Communication and Coordination, mostly due to inadequate or ineffective communication, was contributory in two-thirds of reports. Half of the reports showed contributory complex interactions in a sociotechnical environment. Ratability scores were noticeably lower for categories evaluating leadership and organizational influences, necessitating careful interpretation.
CONCLUSIONS: We applied the HFACS taxonomy to the analysis of CIRS reports in anaesthesiology. This constitutes a structured approach that, especially when applied to a large data set, might help guide future mitigation and intervention strategies to reduce critical incidents and improve patient safety. Improved, more structured reporting templates could further optimize systematic analysis.