Development and validation of a tool to aid writing and reviewing of healthcare safety investigation reports: A modified Delphi study

Introduction Healthcare organisations worldwide are expected to investigate or review incidents that unintentionally harm or could cause harm. Investigation findings are presented in a written report, which is arguably a proxy measure for the quality of investigations. The report is an important document to be read and acted upon. The aims of this study were: (1) to design a tool to support the writing and review of healthcare safety investigation reports, and (2) to validate the content of the tool. Methods The study was conducted in two phases. Phase 1: The initial content of the tool was developed using relevant published literature and building consensus with 25 specialist participants from the UK and Australia. Phase 2: Content validity of the tool was assessed for relevance and clarity in two e-Delphi survey rounds with users of the tool. Using a 4-point scale, a median of 3 or 4 and an interquartile range of less than or equal to 1.5 were used to determine consensus. Results Phase 1: A tool containing eight areas of review with accompanying descriptors was developed. For each area of review, a three-point ordinal rating scale along with a comments box for formative self-assessment was included. Phase 2: At the end of the round 1 survey, there was consensus on all but one area of review in the tool. By the end of the round 2 survey, consensus was reached on all areas of review. No additional areas of review were added and none were removed. Some descriptors were amended to improve clarity. Conclusion We co-designed and validated the content of a tool that can be used to inform the quality of safety investigation reports and learning in terms of a systems-based approach. The tool has multiple uses ranging from self-assessment for report writers to facilitating oversight of the quality of healthcare safety investigation reports. Future work could focus on building further evidence of the tool's overall utility.