Incident analysis is a structured method of investigation to determine the underlying cause of adverse events in healthcare so that recommendations for safety enhancements can be implemented to improve patient safety. Despite significant investment in incident analysis for over 18 years, Australian hospital reports suggest that the rate of adverse events is not decreasing. In Queensland (QLD) public hospitals, the human error and patient safety (HEAPS) method is the most widely used form of incident analysis. HEAPS teams consist of members from different health professions and status levels and include those who have been involved in the adverse event. Since healthcare is an intergroup context, this makes incident analysis an intergroup environment. The first aim of this thesis was to identify the challenges to incident analysis from the perspective of the healthcare executives, policy makers, and clinicians involved in the process. The second aim was to use intergroup theories to understand how professional identity and status level of those involved impacts the analysis effectiveness. Methods Four qualitative studies were undertaken. Data were analysed using a thematic content analysis approach. A data-driven approach was used to understand participants' perspectives of challenges to analysis effectiveness. A theory-driven approach invoking social identity theory (SIT) and communication accommodation theory (CAT) was used to investigate how intergroup factors impact. In Study 1, 11 senior healthcare executives were interviewed to provide an organisational and regulatory context and in Study 2, three Patient Safety Officers, who conduct analysis meetings, were interviewed. In Study 3, three HEAPS analysis meetings in a Queensland hospital were observed and communication categorised according to CAT. Finally, to triangulate the findings, in Study 4 six participants from the observed meetings were interviewed. Results Study One: Power, status and professional identity were categorised as barriers to analysis effectiveness: impacting open disclosure; feedback to stakeholders; and dissemination of learnings. iii Organisational capacity to remedy error was perceived to conflict with organisational requirements to meet key performance indicators around financial performance and capital expenditure. The current model of accreditation and a decentralised model of error management were perceived to adversely impact the quality of the analysis process, service improvements and the sharing of learnings. The capacity of clinical governance committees to enact change was perceived to be hampered by status, organisational agendas and cost concerns. Stakeholders' limited understanding of human-factors principles resulted in a persons'-approach to incident analysis. Study Two: PSOs were found to view analysis as an intergroup process. They used communication strategies to reduce intergroup differences and build a superordinate team identity. From a CAT perspective, interpersonal control, interpretability, emotional expression a...