Last month, NICE published new draft guidance on the care of dying adults. The extensive media coverage that followed reflects the context in which it was developed: the guidance serves partly to fill the gap left by the Liverpool Care Pathway for the Care of the Dying (LCP), which was phased out in a storm of controversy following the 2013 Neuberger review. The well intentioned aim of the LCP was to identify the essential elements of good care of the dying from hospices, and transform them into a series of prompts to guide professionals in hospital and community settings. Given that the new draft guidance is almost 300 pages long, one can understand why the Liverpool team thought that condensing this information into a few pages would be helpful. However, the LCP ultimately failed, not because the essential elements were wrong, but because of the way that the tool was used and implemented. How can we avoid repeating this mistake?

This week we have published a paper in BMJ Open that aims to help us understand more about what went wrong with the implementation of the LCP. We made use of qualitative data that had been collected in 2009 as part of a mixed methods study to develop a tool to improve care of the dying in Intensive Care Units. In the study 25 healthcare professionals (nurses, doctors, social workers, in various specialties) were interviewed about their experiences with using integrated care pathways for the dying (including the LCP and its derivatives). Because the data were collected in 2009 we were able to investigate professional’s views of the potential benefits and harms of integrated care pathways, independent of subsequent press criticisms or the Neuberger review.

We found that most healthcare professionals interviewed described benefits of using integrated care pathways for the dying. However, these benefits related almost exclusively to processes of care, and were experienced by the healthcare professional themselves: integrated care pathways provided clarity about what was to be done, and consistency. Though intended as guides, pathways were often interpreted as protocols, as a set of instructions, and this aspect appeared to be particularly valued especially by the most junior clinicians. Surprisingly, interviewees did not speak of integrated care pathways as directly benefitting patients or their families by helping to ensure better outcomes in death or bereavement. Where patient outcomes were mentioned, it was in the context of harm: some interviewees spoke of concern about the use of integrated care pathways, that such pathways could be dangerous at times, that there was a reliance on process which could lead to patient harm.

With the benefit of hindsight it seems extraordinary, if these views were representative of more widely held opinions, that such a tool became so rapidly and universally accepted. Our data provide insights into how this may have happened. Integrated care pathways for end of life care appeared to have symbolic value for healthcare professionals. They legitimised death as an outcome, provided a positive focus to care, and were used as a signal to herald the change in focus of care from active to palliative treatment, as a non-verbal form of communication. Patchy education and training in palliative care may have created a vacuum that allowed a tool for which there was no strong evidence to become accepted and valued. The strong symbolic value may, in turn, have made it easier for healthcare professionals (as well as institutions and policy makers) to overlook shortcomings in evidence.

This study provides important messages for the successful development and implementation of future tools that aim to guide care of the dying. Firstly, that comprehensive education and training in palliative care is critical. A Royal College of Physicians audit in 2014 found that mandatory training in care of the dying was only required for doctors in 19% of trusts, and for nurses in 28%. Without such training it is unlikely that staff would be able to use any pathway well, or potentially recognise when it is being used poorly. Secondly, our study highlights the importance of grounding any future tool around patient and carer reported outcomes; professionals, particularly the most junior, may need to be re-orientated from processes of care (whether morphine is prescribed) to outcomes (whether pain is relieved). Lastly, the study demonstrates the importance of collecting qualitative data when developing and implementing interventions. Qualitative research can be undervalued (both by health professionals and journal editors) and is often overlooked in favour of quantitative methods of evaluation. Collecting, reporting, and using qualitative data is essential to understand, refine, and improve any future tools that aim to improve care of the dying.

Katherine Sleeman is a clinical lecturer in palliative medicine at King’s College London, Cicely Saunders Institute, Department of Palliative Care Policy and Rehabilitation. She is on Twitter @kesleeman

Competing interests: I declare that I have read and understood the BMJ Group policy on declaration of interests and I have no relevant interests to declare. I have an NIHR funded clinical lectureship in palliative medicine. I split my time between clinical work as a palliative medicine registrar, and academic work at the Cicely Saunders Institute, King’s College London.