Research has shown that human error in anaesthesia is a major contributor to critical incident in anaesthesia, what is unclear is how this occurs and what impact these incidents have on patients. The purpose of this thesis was to examine human error in anaesthesia using Reason’s (1990) framework of error and Swiss Cheese Model to identify the stages of anaesthesia in which errors occur, the frequency and severity of those errors, and the role of fatigue, stress, and usability in human error in anaesthesia. A two stage process was used to collect information on human error in critical incident. The first stage consisted of a task analysis and Flanagan’s (1954) critical incident analysis which allowed for the collection of information on the anaesthetic process, this information led to suggestions for a questionnaire to be used to collect data which could be quantitatively analyzed. In the second stage, a questionnaire was constructed and taken online by participants sent invitations from their respective District Health Board in New Zealand. A small sample size of data was acquired (n=12 responses) of which 8 were complete and used. Results were analyzed by a critical incident analysis. Human error was found to occur during all stages of anaesthesia with induction representing the most incidents and emergence the least. Incidents involving human error were found to be rare, occurring between once per yearly quarter to a few times per lifetime. Most incidents were found to be near misses, and almost a quarter of incidents were found to be of harm, of these only one was judged to be of moderate harm. Fatigue and stress were found to be associated with half of incidents, and equipment design was found to only be associated with a few incidents. This investigation is effective in highlighting examples of modern critical incidents of anaesthesia. These results indicate that fatigue and stress possibly play large roles in contributing to human error in anaesthesia and may be good areas for future research.