Generic analysis method to learn from serious adverse events in Dutch hospitals: a human factors perspective.

Abstract

Hospitals in various countries such as the Netherlands investigate and analyse serious adverse events (SAEs) to learn from previous events and attempt to prevent recurrence. However, current methods for SAE analysis do not address the complexity of healthcare and investigations typically focus on single events on the hospital level. This hampers hospitals in their ambition to learn from SAEs. Integrating human factors thinking and using a holistic and more consistent method could improve learning from SAEs. This study aims to develop a novel generic analysis method (GAM) to: (1) facilitate a holistic event analysis using a human factors perspective and (2) ease aggregate analysis of events across hospitals. Multiple steps of carefully evaluating, testing and continuously refining prototypes of the method were performed. Various Dutch stakeholders in the field of patient safety were involved in each step. Theoretical experts were consulted, and the prototype was pretested using information-rich SAE reports from Dutch hospitals. Expert panels, engaging quality and safety experts and medical specialists from various hospitals were consulted for face and content validity evaluation. User test sessions concluded the development of the method. The final version of the GAM consists of a framework and affiliated questionnaire. GAM combines elements of three methods for SAE analysis currently practised by Dutch hospitals. It is structured according to the Systems Engineering Initiative for Patient Safety model, which incorporates a human factors perspective into the analysis. These eases aggregated analysis of SAEs across hospitals and helps to consider the complexity of healthcare work systems. The GAM is a valuable new tool for hospitals to learn from SAEs. The method can facilitate a holistic aggregate analysis of SAEs across hospitals using a human factors perspective, and is now ready for further extensive testing.