Evidence
Alert

ICU admission decision support tool showed promise but was rarely used

A decision support tool developed to help doctors determine whether patients should be admitted to intensive care showed promise in facilitating patient-clinician communication, but was not often used by doctors, with fewer than 30% using the forms.

Intensive care can deliver lifesaving treatment. It can be invasive and distressing with no guarantee of success. At present, there is little to guide doctors in the decision-making process, and this NIHR-funded study sought to help doctors by developing a support tool.

Those who used the tool were generally positive about it, saying it prompted greater patient involvement and provided reassurance for more uncertain cases, but the majority did not use it. Increase in documentation was a major barrier, and while integration into the existing electronic record might alleviate some of this burden, there is still some doubt as to the long-term feasibility of such a tool.

Why was this study needed?

For eligible patients, prompt admission to the ICU can increase their chance of survival by up to 23%. Yet those that do survive may experience lasting physical and emotional effects, and it is the job of the clinician to carefully weigh up the potential gains and risks of admission in what is often a time-pressured environment.

There are currently no national guidelines to help the decision-making process, and evidence suggests it is influenced by a range of factors, with considerable variation between clinicians. In addition, patients and their families are not always fully informed or consulted.

This study explored current practice in order to create a decision support tool that could be used to help take some of the uncertainty out of the process, thereby improving decisions and, when possible, also informing the discussions with the patient and their family.

What did this study do?

This mixed-methods study used systemic reviews to examine the factors that influence admission to ICU and the experiences of those involved. This information was supplemented by a three-week study of six NHS hospitals to observe current practice and conduct interviews.

In addition, a survey was used to ascertain the most important criteria when deciding whether to admit a patient, including factors such as age, level of ward staffing and illness severity. A total of 303 intensive care unit consultants and 187 critical care outreach nurses completed the hypothetical choice experiment.

An initial draft of the decision support tool was subsequently developed, and feedback was used to modify the tool prior to it being tested in three hospitals for eight weeks. The utility of the tool was assessed via interviews with clinicians and examination of patient records.

What did it find?

  • Use of the decision-support forms was under 30%. Those that used them stated they helped clarify their reasoning and prompted greater patient and family involvement. Reasons for not using the form included the additional work created in filling them in as well as the perception that they undermined professional expertise. This was especially true for referrals from the emergency department which were mostly verbal due to the requirement to immediately stabilise the patient with the help of ICU expertise.
  • The various strands of the study highlighted that doctors often found it difficult to weigh up benefits and harms of ICU admission, with many struggling or failing to communicate their reasoning. Consequently, implicit values sometimes became part of the decision-making process.
  • It was not always possible to identify the final referral decision from the clinical record or important factors that contributed to making that decision.
  • In the implementation feasibility study, the specially designed information leaflets were not given to patients or families. Reasons given were the timing of the decision and the patient being too unwell to be able to take in the information.
  • Out of the eight factors in the choice survey, patient age was found to have the greatest influence upon consultants’ decisions to admit (relative influence 23.9%). This was followed by family views (relative influence 19.9%). The registrar’s assessment of the patient was more influential than the National Early Warning Score. Critical care outreach nurses also deemed patient age to be most influential (relative influence 21.6%), followed by the severity of main comorbidity (relative influence 17.1%) and National Early Warning Score (relative influence 17.4%).

What does current guidance say on this issue?

There is no recent guidance; the study authors state that a 1996 Department of Health guideline is the most relevant guideline on this topic.

This guideline suggests consideration of whether the condition is reversible, whether the patient has any significant co-morbidities and whether the patient has made their wishes clear in advance.

Ultimately it seeks to ensure patients are not admitted when there is little chance of improvement, as such admission may only act to prolong suffering for patients and their families.

What are the implications?

While this study shows that those referring patients to ICU could benefit from greater support, the decision support tool trialled in this study would need some adaptation to fit the time-pressured realities of the users.

The process did seem to help clinicians articulate and communicate their reasoning for admission. Perhaps, as the authors say, if the tool were to be integrated into existing systems the perceived additional workload may be diminished.

Another not insignificant finding is that although clinicians stated they valued patient’s wishes, in some cases there was a lack of patient and family involvement.

Citation and Funding

Bassford C, Griffiths F, Svantesson M et al. Developing an intervention around referral and admissions to intensive care: a mixed-methods study. Health Serv Deliv Res. 2019;7(39).

This project was funded by the NIHR Health Services and Delivery Research Programme (project number 13/10/14). The University of Aberdeen and the Chief Scientist Office of the Scottish Government Health and Social Care Directorates fund the Health Economics Research Unit.

Bibliography

Department of Health. Guidelines on admission to and discharge from intensive care and high dependency units. London: Department of Health, 1996.

NHS website. Intensive care. London: Department of Health and Social Care; updated 2019.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre

Definitions

Template content only

 

Commentaries

Expert commentary

The systematic review element of this study found that a decision to admit a patient to ICU was inconsistently determined by a number of factors including patient age and available bed numbers. This lack of consistency and transparency was further confirmed in a series of choice experiments where clinicians were asked to make admission decisions about hypothetical patients.

The ethnographic part of the study, where researchers would follow clinicians during the decision-making process, provided many valuable anecdotal insights into the reasons for this lack of consistency with both referring and critical care team dynamics playing an important role. A retrospective notes-based survey demonstrated a common lack of documentation of the decision-making process and also highlighted a frequent lack of consultation with either the patients or their families.

The adoption of some of the decision support interventions which provide a rational structure for the decisions, developed by the authors, would improve the quality of admission decision-making and would allow units to better explain their decisions if challenged.

Dr Simon Baudouin, Consultant in Critical Care Medicine, ex-board member of the FICM and Trustee and Board member of ICNARC, The Newcastle upon Tyne Hospitals NHS Foundation Trust

The commentator is a board member of the Intensive Care National Audit and Research Centre

 

Expert commentary

People with severe life-threatening illnesses may need to be admitted to an ICU for life-support treatments. However, there is little information to guide doctors on how to reach a decision in a clear and fair way.

This study provides some much-needed insights into this process. For example, many ICU doctors strive to involve families in decision-making, but this did not often happen in this study. In addition, doctors need to develop strategies to better explain the benefits and harms of treatments.

While the tools developed by the researchers were not readily adopted in the study hospitals, these resources will be a useful starting point for doctors to improve their own decision-making.

Dr Nazir I Lone, Senior Clinical Lecturer in Critical Care and Honorary Consultant in Critical Care, The Royal Infirmary of Edinburgh, University of Edinburgh

The commentator declares no conflicting interests