The aim of the project is to develop a new method of analysis of qualitative datasets based on a new AI model  trained on data from the most recent RCoA National Audit Project (NAP7 which studied the risk factors for and outcomes of, perioperative cardiac arrest). The model was developed and piloted on this data.

The Patient Safety Research Collaboration (PSRC) maternity incident reporting practices project follows on from the previous NAP7 project. The aim of the maternity analysis project is to identify differences in University College London Hospital (UCLH) maternity patient safety reporting practices, gaps in reporting and areas for improvement that will lead to set of recommendations for the improvement of reporting practices at UCLH. A mixed-methods design was adopted to enable a comprehensive exploration of incident reporting practices in maternity services, integrating both contextual and experiential dimensions. Quantitative mapping of patient safety activities and large-scale qualitative analysis of 400 incident reports submitted via the InPhase system over a five-month period (June 20th to November 11th 2024), using AI-supported text analysis tools (Carlena and Infranodus) provided an understanding of the local maternity service and incident landscape. This was complemented by semi-structured interviews with 14 maternity clinicians, offering insight into staff perceptions, behaviours, and barriers to reporting.  The research team includes clinicians from maternity and a PPI team of women with lived experience of maternity services.

Lead Investigator