Improving communication and palliative care at times of clinical uncertainty in intensive care units

This is the fifth in a series of six articles that aims to encourage nurse leaders to explore how to use research findings to improve practice and services where they work. The series focuses on research that has been funded by the National Institute for Health Research (NIHR) and included in NIHR Dissemination Centre publications. The authors seek to relate the content directly to the Nursing and Midwifery Council’s code of professional standards of practice and behaviour, specifically the themes of practising safely and promoting professionalism and trust

This is the fifth in a series of six articles that aims to encourage nurse leaders to explore how to use research findings to improve practice and services where they work. The series focuses on research that has been funded by the National Institute for Health Research (NIHR) and included in NIHR Dissemination Centre publications. The authors seek to relate the content directly to the Nursing and Midwifery Council’s code of professional standards of practice and behaviour, specifically the themes of practising safely and promoting professionalism and trust

Introduction by Elaine Maxwell, clinical adviser at the National Institute for Health Research Dissemination Centre:

The NIHR Dissemination Centre critically appraises the latest health research from the NIHR and other research organisations to identify the most reliable, relevant and significant findings. We disseminate these findings as actionable, accessible and authoritative information in the form of Signals and Highlights, as well as themed reviews. In 2015, we looked at the evidence on best practice in caring for people who are dying in our themed review, Better Endings.

Elaine Maxwell

While palliative care is often associated with a gradual decline, and hospices have transformed the final weeks of life for many people, there is a significant number of people for whom death is sudden and unexpected. This might be in any setting, inside or outside healthcare services. One area where sudden death occurs is in intensive care units (ICUs). The Intensive Care National Audit and Research Centre (2017) audit of 248 NHS adult critical care units in England, Wales and Northern Ireland in 2015-16 reporting 21,950 deaths in ICUs.

Our review included the findings of one study that looked at the challenges for staff working with patients and their relatives in intensive care when the outcome is uncertain and change is happening rapidly (Higginson et al 2013).

In this article, the lead researcher acknowledges the difficulty of researching best practice in such complex situations and discusses how the team addressed this. A senior clinical nurse who works in intensive care then reflects on how these findings relate to her practice.

Irene Higginson, head of department, head of division and director of the Cicely Saunders Institute of Palliative Care, Policy and Rehabilitation, King’s College London:

Irene Higginson

Why did we focus on intensive care and uncertainty? 

Managing clinical uncertainty is one of the most difficult things most doctors, nurses and other healthcare professionals face. The interfaces between curative, life-extending, chronic and palliative treatments are complex and often blurred, causing distress. It was my co-author, Wendy Prentice, and ICU colleagues who felt it was vital to improve palliative care in this setting.

ICUs are an acknowledged example of where there is prognostic uncertainty. The number of patients cared for in ICUs is increasing worldwide and, while the central goal in intensive care is to preserve or extend life, the nature of illness or trauma means that more than one in five people admitted will die there.

What did we plan and how did we change?

When we planned this study, care pathways were being promoted to improve end of life care. We initially thought we would be adopting a care pathway for ICU, but we did not want to make assumptions, so our first steps in the research methods were simultaneously to review the literature to find out who else had developed interventions for ICU, and to listen to ICU staff, families and, where possible, patients. We did this using open-ended interviews and focus groups, in qualitative research, to find out whether and how participants thought a care pathway would help.

The findings from the two approaches above changed our minds about what we should do. We found that staff felt that end of life pathways focused on patients where dying is already diagnosed, albeit not always well, taking into account the personal needs of the individual. Staff in intensive care and patients’ families felt there was a need to develop a wider range of interventions before this point to improve symptom control, communication and support, whether patients subsequently survived, deteriorated or died.

At the same time, we found out something very exciting. In the US, a group led by Anne Mosenthal had developed a series of prompts to improve care for everyone in an ICU. Their argument was that good communication and assessment would improve care for all.

The method template we followed

The project followed Medical Research Council (MRC) (2006) guidance for the development and evaluation of complex interventions. This landmark guidance gives a roadmap for how to develop and evaluate complex interventions.

A complex intervention is any treatment or service with several interacting components. This is important because, for example, how people behave, what training they need or how different components of the intervention interact will influence what happens. The MRC guidance is the gold standard that helps to work through these components step by step.

Based on the results so far, we modelled the intervention and tried to develop theory about how it might work. Based on the evidence we gave them, the clinical teams met and over time came up with a tool and plans of how we should train people. It took longer than I thought and at one point I thought we might never get there. They came up with some great principles based on our research. The tool needed to be short, rather than a long checklist, and it needed to support better understanding of individual patients and their families. We then put the training in place and staff started to use the tool.

We planned two things in tandem: first the implementation of the training and the tool, and second the evaluation of their effects. We tested these side by side and implemented the tool, which we called the Psychosocial Assessment and Communication Evaluation (PACE). PACE was designed for everyone entering ICU to create a more person-centred approach. We did not want just to measure how the tool was completed. In any evaluation, the outcome, that is something that relates to the patients or their families, is the most important not just the process of what happens. Because people in ICU are very ill and cannot complete questionnaires, we decided to ask their families about their experiences using a questionnaire about intensive care quality that had been tested and validated already.

The results took us by surprise. We had thought our PACE tool might affect communication, but we also found improvements in symptom control and trust, which we had not anticipated. Consequently we refined our theory about how PACE worked. We were also pleased with many of the other features of PACE: it was short, did not take long to complete and in general staff thought it was helpful.

Rebecca Harper, senior staff nurse working in intensive care at University College London Hospitals NHS Foundation Trust:

Rebecca Harper

Intensive care nursing presents a number of challenges, not least that, while the intention is to save lives, a large proportion of deaths in hospitals occur in ICUs. However, the hardest for all involved to deal with is the period between full and active treatment in the face of critical illness and the acknowledgement that withdrawal of care is the kindest and most ethical route to follow.

The PACE tool was developed more than five years ago, when pathways for those diagnosed as end of life were still trusted as gold standard. Tools for care outside palliative care settings have not been developed, let alone to address the uncertain times between active treatment and a formal recognition that death is imminent.

The findings of this evaluation are positive in that, not only was implementation feasible and successful in communicating several important patient facts, but relatives reported a greater sense of trust in the teams providing care. Some reported that symptom control, notably agitation and confusion, which are prevalent in patients in ICU, was better even in the face of the slightly higher illness severity of the intervention group.

What was the previous received wisdom in ICU?

There is an argument that, given its complexity and in the face of time-sensitive and changeable scenarios, ICUs have tended towards paternalism, so it is helpful to have a communication aid that reaches out specifically to patients and their families. One theme from the study that particularly stood out for me was that relatives appreciated having their perspectives sought but wanted decision making to remain medically led, in turn challenging the clinician-held convention that relatives would demand greater control of treatment direction, should these types of conversations become commonplace.

It is also great to see another tool in the arsenal to ensure psychosocial aspects of care so that patient preferences can be given more consideration. I would be interested to know if there was any difference in the management of symptoms, bearing in mind that the associated symptoms of dying are also those routinely seen and treated in critical illness, or if better communication simply helped families to understand what care was being given and why.

What gives this study credibility?

Although it was only trialled in two units in one hospital, the ICUs cover a range of patient populations and a number of specialties providing small but varied situations in which to test the PACE tool. The strong feedback relationship between the researchers and the clinicians developing the tool lend it greater credibility; successful change is that which considers specific local needs and challenges. The team also developed its evaluation plan in tandem with implementation and exceeded its target family questionnaire response rate, which adds further credence to the development and evaluation process.


This article presents independent research commissioned by the National Institute for Health Research (NIHR) under the Research for Patient Benefit (RfPB) programme (Grant Reference Number PB-PG-0107-12109).

The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health and Social Care.

The sponsor of the study had no role in study design, data analysis, data interpretation or writing of the report.

How will I change my practice now?

The tool is both practically useful in relation to the information required and in encouraging the team to consider a patient’s preferences. The team is thus able to tailor care to the individual, an approach valuable not just at the end of life but that combats the necessary machine- and process-heavy methods of intensive care more generally. One difficulty the authors discussed was that not all ICU staff had been exposed to palliative care training. Although not an intention of the tool or study, I would support the involvement of palliative care teams in the period of clinical uncertainty as a way of improving symptom relief.

What else do we need to know?

I would like to know more about the barriers to completing the patient preferences section of the PACE tool. Humanising ICU care is implicated in preventing ICU delirium, which can occur not just at the end of life, and better uptake of the tool could be reasonably hypothesised to improve patient, family and staff satisfaction. I also would find it interesting to see, from the observations of care, if symptom relief was indeed better in the intervention arm or if this was a perception brought about through the improved communication aided by the PACE tool. And, as stated in the paper, I would support evaluation of the feasibility and effect of the tool, or a version adapted to an individual unit’s requirements in a number of different units to provide more diverse data.


  • Higginson I, Koffman J, Hopkins P et al (2013) Development and evaluation of the feasibility and effects on staff, patients, and families of a new tool, the Psychosocial Assessment and Communication Evaluation (PACE), to improve communication and palliative care in intensive care and during clinical uncertainty. BMC Medicine. 11, 1, 213. 
  • Medical Research Council (2006) Developing and Evaluating Complex Interventions: New Guidance. 
  • Intensive Care National Audit and Research Centre (2017) Key Statistics from the Case Mix Programme: 1 April 2015 to 31 March 2016. ICNARC, London.

Read the other articles in this series

This article is for subscribers only