Understanding safety differently: developing a model of intravenous insulin infusion use in hospital inpatients

Introduction: Variable rate intravenous insulin infusions (VRIIIs) are very effective in managing hyperglycaemia in hospital inpatients when used correctly. However, if used incorrectly, they can cause harm or even death. Unlike traditional safety approaches that focus on ensuring adherence to predefined standards and identifying errors to prevent their recurrence, Resilient Health Care (RHC) shifts the focus to understanding variability in work, including both successes and failures. RHC achieves this by comparing Work as Imagined (WAI) (what people expect themselves and others to do) with Work as Done (WAD) (what people actually do) with the aim of realigning WAI with WAD in order to reflect reality. Aim: To develop a model of VRIII use in adult inpatients informed by the RHC framework. Methods: This mixed methods study was conducted in three phases at a Vascular Surgery Unit in a tertiary teaching hospital in England, UK. Phase I: Exploring WAI. Analysis of all documents related to VRIII use, and focus groups with users of VRIII. Documents and transcripts were analysed using inductive/deductive analysis. Phase II: Exploring WAD. Video Reflexive Ethnography methodology and quantitative data were used. The qualitative data were analysed using thematic analysis and the quantitative data were used to judge the outcomes of the observed tasks. Phase III: Model development. Two separate hierarchical task analyses (HTAs) were produced, one for each phase, to identify similarities and differences and to inform the development of the model of VRIII use. Results: WAI was perceived as a complex process that involved iterative tasks of producing hospital-specific guidelines and a multi-pronged approach to ensuring implementation of and adherence to guidelines. WAD was accomplished by using standardised practices (adhering to VRIII guidelines) and context-dependent adaptations (assigning blood glucose monitoring to other colleagues during busy shifts). Reflexive meetings revealed that a lack of knowledge in the selection of intravenous fluids and monitoring of blood glucose tasks was the main challenge faced while using VRIII. Suggestions from healthcare practitioners designed to overcome this challenge focused mainly on the need for face-to-face, VRIII-focused training that is tailored to their need. The comparison between WAI and WAD HTAs highlighted that most of the tasks in both HTAs were mostly aligned. The comparison also assisted in developing a model for the use of VRIII. The model showed that standardisation (WAI) and flexibility (context�dependent adaptations) must complement each other to ensure the delivery of patient care while using VRIII. The model also highlighted two key points to be considered when monitoring and continuously reviewing everyday clinical work and its resultant outcomes: 1) Understanding the permanence status of adaptations is a crucial step in differentiating between work that has long-term and short-term success. 2) Using monitoring tools such as checklists, clinical audits and VRE, corresponding to the type of task being performed is fundamental in reviewing everyday work and designing and implementing effective interventions that help in improving patient care while using VRIIIs. Conclusion: This research explored safety and complexity in the use of VRIIIs by integrating innovative methodologies inspired by the RHC construct of comparing WAI with WAD and by the VRE principles of exnovation, care, reflexivity and collaboration. The methods employed and findings of this thesis could help researchers as well as healthcare practitioners to better understand the use of VRIIIs in situ and move beyond describing concepts to providing practical recommendations, based on practical data obtained from a Vascular Surgery Unit, for improving safety and patient care delivery while using VRIII