Assessing the quality of patient safety incident investigation reports

Between 4% and 17% of hospital inpatients experience a patient safety incident. Many healthcare organizations undertake analysis and investigation of serious incidents to understand them and prevent future occurrences. Tools have been developed to assess investigation quality as a reflection of a health service's learning and improvement process. However, a broad-scale examination of the quality of investigation reports has not been conducted in Australia. This study aimed to assess the quality of a sample of Australian patient safety incident investigation reports.

A deductive, directed content analysis was conducted to assess the quality of 300 incident investigation reports from 56 Australian health services. Each report was assessed on the extent to which they met predefined quality criteria using the Dutch Health and Youth Care Inspectorate (IGZ) scoring instrument and the United Kingdom's Learning Response Review and Improvement Tool (LRRT).

A detailed and complete description of events was present in just over half of the reports (57%). There was also variable application of systems approaches to incident causation in the reports. Generally, there was evidence that the people affected were engaged with, and the avoidance of blame and counterfactual reasoning was well executed, as was overall writing quality. Areas in need of improvement included identifying contributing factors beyond the staff and local hospital, the use of appropriate scientific literature and using observational insights to enhance understanding of work-as-done-in our sample only 4% used observation techniques to examine healthcare processes.

This study establishes that stronger foundational, evidence-based approaches to incident investigation are necessary. These approaches include using multiple data sources, like observations of real work, and inclusion of people with skills to apply a systems thinking-driven analytical process to effectively identify contributing factors beyond the individual to drive learning and continuous systemic improvement.