How can end of life care excellence be normalized in hospitals? Lessons from a qualitative framework study

Abstract

Background: There is a pressing need to improve end-of-life care in acute settings. This requires meeting the learning needs of all acute care healthcare professionals to develop broader clinical expertise and bring about positive change. The UK experience with the Liverpool Care of the Dying Pathway (LCP), also demonstrates a greater focus on implementation processes and daily working practices is necessary. Methods: This qualitative study, informed by Normalisation Process Theory (NPT), investigates how a tool for end-of-life care was embedded in a large Australian teaching hospital. The study identified contextual barriers and facilitators captured in real time, as the 'Clinical Guidelines for Dying Patients' (CgDp) were implemented. A purposive sample of 28 acute ward (allied health 7 [including occupational therapist, pharmacists, physiotherapist, psychologist, speech pathologist], nursing 10, medical 8) and palliative care (medical 2, nursing 1) staff participated. Interviews (n = 18) and focus groups (n = 2), were audio-recorded and transcribed verbatim. Data were analysed using an a priori framework of NPT constructs; coherence, cognitive participation, collective action and reflexive monitoring. Results: The CgDp afforded staff support, but the reality of the clinical process was invariably perceived as more complex than the guidelines suggested. The CgDp 'made sense' to nursing and medical staff, but, because allied health staff were not ward-based, they were not as engaged (coherence). Implementation was challenged by competing concerns in the acute setting where most patients required a different care approach (cognitive participation). The CgDp is designed to start when a patient is dying, yet staff found it difficult to diagnose dying. Staff were concerned that they lacked ready access to experts (collective action) to support this. Participants believed using CgDp improved patient care, but there was an absence of participation in real time monitoring or quality improvement activity. Conclusions: We propose a model, which addresses the risks and barriers identified, to guide implementation of end-of-life care tools in acute settings. The model promotes interprofessional and interdisciplinary working and learning strategies to develop capabilities for embedding end of life (EOL) care excellence whilst guided by experienced palliative care teams. Further research is needed to determine if this model can be prospectively applied to positively influence EOL practices.