Operationalising FRAM in Healthcare: A critical reflection on practice

Resilience Engineering principles are becoming increasingly popular in healthcare to improve patient safety. FRAM is the best-known Resilience Engineering method with several examples of its application in healthcare available. However, the guidance on how to apply FRAM leaves gaps, and this can be a potential barrier to its adoption and potentially lead to misuse and disappointing results. The article provides a self-reflective analysis of FRAM use cases to provide further methodological guidance for successful application of FRAM to improve patient safety. Five FRAM use cases in a range of healthcare settings are described in a structured way including critical reflection by the original authors of those studies. Individual reflections are synthesised through group discussion to identify lessons for the operationalisation of FRAM in healthcare. Four themes are developed: (1) core characteristics of a FRAM study, (2) flexibility regarding the underlying epistemological paradigm, (3) diversity with respect to the development of interventions, and (4) model complexity. FRAM is a systems analysis method that offers considerable flexibility to accommodate different epistemological positions, ranging from realism to phenomenology. We refer to these as computational FRAM and reflexive FRAM, respectively. Prac-titioners need to be clear about their analysis aims and their analysis position. Further guidance is needed to support practitioners to tell a convincing and meaningful system story through the lens of FRAM.

Automated summary

Resilience Engineering principles are gaining popularity in healthcare and FRAM is the most well-known method used. However, there is a lack of guidance on its application, which may hinder its adoption and lead to misuse. This article provides self-reflective analysis of FRAM use cases and identifies lessons for successful application in healthcare. Four themes are developed, including core characteristics, flexibility in epistemological paradigm, diversity in interventions, and model complexity.