Nursing and Physician Perception of Inappropriate Intensive Care - Protocol


Nursing and physician perception of inappropriate intensive care


Qualitative Protocol Development Tool


  • This protocol has regard for the HRA guidance and order of content


A multi-point survey to investigate the incidence and outcome of patients who doctors or nurses feel are unlikely to leave hospital alive from intensive care and the rate of agreement between doctors and the bedside nurse.


Nursing and physician perception of inappropriate intensive care



Version 3.3 (02/12/2018)




IRAS Number: 248372
SPONSORS Number: 201810






The undersigned confirm that the following protocol has been agreed and accepted and that the Chief Investigator agrees to conduct the study in compliance with the approved protocol and will adhere to the principles outlined in the Declaration of Helsinki, the Sponsor’s SOPs, and other regulatory requirement.

I agree to ensure that the confidential information contained in this document will not be used for any other purpose other than the evaluation or conduct of the investigation without the prior written consent of the Sponsor

I also confirm that I will make the findings of the study publically available through publication or other dissemination tools without any unnecessary delay and that an honest accurate and transparent account of the study will be given; and that any discrepancies from the study as planned in this protocol will be explained.


For and on behalf of the Study Sponsor:


Date: ……/……/……
Name (please print):


Position: …………………………………………………………………………………………  




Chief Investigator:

Signature: ………………………………………………………………………………………… Date: ……/……/……
Name: (please print):






Chief Investigator Dr Richard Innes

Consultant Intensive Care Medicine

Musgrove Park Hospital, Parkfield Drive, Taunton TA1 5DA


Study Co-ordinator Dr Deborah A Webster

Registrar Intensive Care Medicine and Anaesthetics

Derriford Hospital, Derriford Road, Plymouth, PL68DH


Sponsor Taunton & Somerset NHS Foundation Trust, Musgrove Park Hospital, Parkfield Drive, Taunton TA1 5DA

01823 333444

Joint-sponsor(s)/co-sponsor(s) N/A
Funder(s) N/A
Key Protocol Contributors Richard Innes and Deborah Webster (see above)
Committees South West Anaesthetic Research Matrix

Contact : Deborah Webster (Chair)




Study Title A multi-point survey to investigate the incidence and outcome of patients who doctors or nurses feel are unlikely to leave hospital alive from intensive care and the rate of agreement between doctors and the bedside nurse.
Internal ref. no. (or short title) Nursing and physician perception of inappropriate intensive care
Study Design Multicentred, survey based study investigating the incidence of patients thought to be receiving inappropriate levels of intensive care in the South West of England.
Study Participants Intensive Care nurses and doctors in trusts throughout the South west of England
Planned Size of Sample (if applicable) Doctors and nurses caring for 900 patients
Follow up duration (if applicable) 6 months
Planned Study Period 8 weeks of initial data collection at each site with follow up of patient outcomes at 6 months
Research Question/Aim(s)


Define incidence (frequency and percentage) of patients who are felt to be receiving inappropriate intensive care in the South West of England.

Assess agreement between nursing staff and doctors in whether patients are receiving appropriate care.

Review 6 month outcome data (survival to discharge, survival at 6 months and level dependence at 6 months) in patients who were felt to be receiving inappropriate levels of care to contextualise staff perception.





(Names and contact details of ALL organisations providing funding and/or support in kind for this study)

South West Anaesthesia Research Matrix

Chair: Deborah Webster, Intensive Care registrar, Derriford Hospital

Non-financial support: trainee research network, assist with recruitment and data capture in trusts throughout the Peninsula deanery
Severn Trainee Anaesthetic Research

Chair: Katie Samuel, Anaesthetic Registrar, North Bristol NHS Trust

Non-financial support: trainee research network, assist with recruitment and data capture in trusts throughout Severn deanery




Sponsor (Taunton & Somerset NHS Foundation Trust): Overall responsibility for the initiation and management of the study and for indemnity. No direct role in decisions made about study design, conduct, data analysis or interpretation, manuscript production or dissemination of results.

Funder: N/A



Study Steering Groups

  • Patient & Public Involvement Group
    • Patients are not study participants in this study and as such Patient and Public Involvement (PPI) Groups have not been utilised in study design. We anticipate that given the study subject matter, it may be beneficial to involve PPI groups at the stage of write up and dissemination of results in order to ensure that results are reported sensitively.
  • SWARM and STAR
    • These are two well established trainee led research groups in the Peninsula and Severn deaneries. SWARM and STAR involvement in the study will mean that staff across the region can be surveyed improving the quality of the data captured.



The protocol has been designed by lead trainee (Deborah Webster), Chief Investigator (Richard Innes) in conjunction with Professor Karen Forbes (Department of Palliative Care Medicine, University of Bristol) and Dr Lucy Selman (Department of Social Medicine, University of Bristol).

The sponsor has played no role in development of the study design or study conduct. They will provide overall responsibility for initiation and management of the study and for indemnity cover.

There is no funder attached to this study – local trainee research networks will provide logistical support free of charge.

Patient and Public Initiative groups have not been involved in study design but their advice will be sought prior to publication of results.


KEY WORDS: Futile care, Inappropriate care, Excessive care





Gantt Chart
                           Month Dec 18 Jan 19 Feb Mar Apr May      Jun      Jul        Aug       Sept       Oct      Nov       Dec    Jan 20
Finalise protocol  


IRAS/R&D approval  


Pilot questionnaire  


R&D approval all sites  


Main patient recruitment  


Six-month follow-up  


Data entry and validation  

         ———–                                           ———

Analysis and writing up  





A multi-point survey to investigate the incidence and outcome of patients who doctors or nurses feel are unlikely to leave hospital alive from intensive care and the rate of agreement between doctors and the bedside nurse.


1          BACKGROUND

Inappropriate care refers to the situation where care is disproportionate to the patient’s prognosis or wishes. Decisions about escalation of a patient’s treatment are often complex and influenced by many factors including patient comorbidities, patient wishes, stage of acceptance of patient and relatives, multidisciplinary team advice and often a degree of conjecture. It may take time for all members of the multidisciplinary team, the patient and relatives to come to terms with the fact that ongoing care is not going to lead to clinical benefit.

Inappropriate or futile care within intensive care is a sensitive but important subject particularly as health care resources are increasingly scarce. Indeed, it has been identified by the UK Critical Care Research Forum as one of their top priorities. Despite this there is very little international data on its precise incidence and no UK data.

In the largest European study (APPROPRICUS), a one-day point incidence survey conducted in 2010,  27% of doctors and nurses felt that at least one patient was receiving inappropriate care (1). 89% of the time this was “too much care” and in 38% of survey responses staff felt “other patients would benefit more”. They also found a strong association between perception of inappropriate care and intent to leave job. This was a pan – European study but did not include data from the UK.

Huynh et al (2) investigated the incidence of futile care within 5 Intensive care units in a single tertiary academic teaching hospital in California during a 3-month period in early 2012. They found 11% of patients were deemed to have received “futile” care, and a further 8.6% received care that was “probably futile”. Of the 123 patients who received futile care, 104 had died by six months, 16 were discharged to other hospitals or long-term care facilities and 3 returned home for palliative care. Anstey et al (2013) using a similar methodology to the APPROPRICUS study surveyed staff across multiple ICU’s within California and found that 38% of surveyed doctors and nurses felt that at least one patient currently treated on their intensive care unit was receiving care that was inappropriate (3). In 93% of cases this referred to care being “too much” and in 75% of cases care was disproportionate to prognosis or patient wishes.

By definition, inappropriate care is not beneficial to the patient. It is costly in financial terms and is not a just allocation of resource (4).There is increasing evidence that lengthy, drawn out deaths on intensive care result in complex grieving amongst relatives and so arguably delayed decisions to withdraw care or failure to limit treatment escalation causes harm. Delivering inappropriate care is also one of the leading causes of conflict for clinicians working within intensive care leading to moral distress, burnout, depression, job dissatisfaction and poor staff retention (5). Hence intensive care medicine has one of the highest rates of burn out within health care (6).

No study to date has looked at the incidence of futile or inappropriate care in UK intensive care units. There are financial, health care structure and cultural differences which may translate to significant differences in our perception of inappropriate care delivery by comparison to Europe or the United States. We also do not know whether perception of inappropriate care will vary between nursing staff and doctors, or indeed between intensive care units throughout the South West of England.


2          RATIONALE

No study to date has looked at the incidence of futile or inappropriate care in UK intensive care units. There are financial, health care structural and cultural differences which may translate to significant differences in our perception of inappropriate care delivery by comparison to Europe or the United States. We also do not know whether perception of inappropriate care will vary between nursing staff and doctors, or indeed between intensive care units throughout the South West of England.

This study will look at outcome data in patients whom doctors and/or nursing staff perceive as receiving inappropriate levels of care. This will help to contextualise how accurately doctors and nurses predict poor outcome from critical illness. This information will be useful for several reasons. If high levels of inappropriate care are being delivered and patients unanimously have poor outcome then we need to improve our shared decision making. If large differences exist between nursing staff and doctors perceptions of inappropriate care then this needs to be further investigated given the links to job dissatisfaction and burnout. If patients perceived to be receiving inappropriate care make good recoveries by 6 months, then staff wellbeing and retention may benefit from enhanced feedback regarding patient outcome.



This study will be conducted using a survey questionnaire. The questionnaire will be like that used by Huynh et al in order that an incidence of inappropriate care can be defined (2). Other studies have used questionnaires that result in ambiguous incidence statistics (e.g. % respondents who perceived at least one patient received inappropriate care). As per Huynh et al, the questionnaire will ask why care is inappropriate with five options selected by a study group, but also an option to insert free text (2).

The American study made reference to “futile” care. Our questionnaire will be adapted to use less emotive wording on advice of Professor Karen Forbes (Palliative care Bristol) and Dr Lucy Selman (Social medicine, Bristol University). The questionnaire will be piloted prior to commencing data collection to ensure the questionnaire detects cases where care is felt to be excessive or disproportionate to patient prognosis.



Primary Aim

To define the incidence of inappropriate care provision on intensive care units in the South West of England according to nursing staff, junior doctors and consultants who are caring directly for intensive care patients.


Secondary Aims

Assess agreement between nursing staff and doctors in perception of inappropriate care

Assess patient outcome in cases that healthcare staff feel care is inappropriate



4.1       Objectives

1) Define incidence of patients who are perceived to be receiving inappropriate care in the South West of England by conducting a multicentred questionnaire based survey.


2) To elucidate the reason for the perception of inappropriate care (through semi structured questionnaire options)


3) To assess the level of agreement between doctors and nursing staff perception of inappropriate care

4) To assess patient outcome at 6 months in cases where care has been perceived to be inappropriate in terms of

  • Death in hospital
  • Death within 6 months
  • Discharge to same level of independence



4.2       Outcome

1) Incidence of perceived inappropriate care in intensive care in South West of England

2) Define the key reasons care is perceived to be inappropriate

3) Define level of agreement between doctors and nurses perceptions of appropriateness of care

4) Define how well staff perceptions of inappropriate care relate to patient outcome at 6 months.



Study registration and Good Clinical Practice

The Project lead (Deborah Webster) will identify site leads in each participating hospital in the South West Deanery (Peninsula and Severn) through the South West Anaesthesia Research Matrix and Severn Trainee Anaesthetic Research groups. The SWARM website will have access to recruitment posters, the study protocol and survey questionnaires. Site leads will be responsible for gaining local Research and Development approval prior to start of data collection and for recruiting researchers to assist with data collection. All site leads and researchers will complete online Good Clinical Practice training prior to data collection.


Data collection

This is a multicentred survey that will be run in parallel over an eight-week period in all general, neurosurgical and cardiothoracic intensive cares throughout the South West. The participants are the staff who are caring for patients requiring organ support in these units. All eligible staff will be approached. Participation is voluntary.

Site leads can select one day from each Monday-Sunday week block to conduct the survey. The same day each week is not to be used in case this introduces bias. Data collection should occur during the day shift (as it is impractical to capture medical staff out of hours).

Data collectors should identify which patient fulfil inclusion and exclusion criteria at 8am on the day of the survey and record the patients hospital number, date of birth and name and bed number on the master sheet alongside the patients identifier code as below.


Site number. Study day. Bed space number


Bedside nurse survey

Bedside nurses will complete a questionnaire on the patients they are directly caring for (generally one or two patients). One survey sheet will be completed for each patient under the nurse’s care who fulfils inclusion/exclusion criteria. The survey will be labelled with the patient’s code and no other identifiers.

Nursing team leaders and doctor surveys

Nurse team leaders (nurse in charge of the shift) and all doctors will complete the survey on all eligible patients under their care. These surveys will ask questions regarding multiple patients. Enter the site number and study day at the top of the questionnaire and fill in the bed spaces requiring organ support so that it is clear to respondents which patients they should be answering the survey about. No patient identifiers should be entered on these survey sheets.

All surveys will be distributed in envelopes to reduce the risk of questionnaires causing distress by being seen by patients and relatives.

Patients admitted after 8am of the study day will be excluded.

Doctors and nurses must complete the questionnaires during their shift, without conferring and away from the patient’s bedside. Completed questionnaires will be placed in a centrally located box. The survey does not collect staff identifiers other than clinical role and grade, so responses are anonymous. Completion of the survey is voluntary. Consent is assumed if staff complete the survey.


Data storage

Data will be collated and anonymised using a combination or Microsoft Excel software (Microsoft Corporation, Redmond, WA, USA) and the REDCap (Research Electronic Data Capture) tool hosted at the University of Bristol. REDCap is a secure web-based application designed to support data capture for research studies. The study codes will be uploaded alongside respondents’ survey responses. Patient identifiable information will be stored locally on paper master sheets only.

Paper questionnaires and master sheets will be stored securely in box files in Research offices which can only be entered by clinical and research staff. It is the site leads responsibility to ensure that a locked room which can only be accessed by clinical and research staff is available and that data is stored there appropriately. Once the data from the master sheets and questionnaires has been uploaded and any follow up of outcomes has occurred at 6 months. Paper documentation will be moved to secure storage in Research and Development archives. It will be kept here for 5 years as per local Research and Development policy.


Follow up

At six months after data capture, Redcap will prompt researchers to complete follow up data on any patients who were perceived to have received inappropriate care. In most cases this will be done through the hospital clinical IT system.

We are measuring the following simple outcome measures for patients deemed to have received inappropriate care

1) Alive/dead at hospital discharge

2) Alive/dead at 6 months

3) for those alive at hospital discharge – discharge destination (hospital, home, nursing home, residential care)

4) For those alive at 6 months – discharge destination (as above)

If it is unclear from hospital records, then the patients’ GP should be contacted in writing or by telephone to gain this information. Patients will not be contacted.

No patient identifiers will be stored on central database, outcome measures will be uploaded to REDCap.


Data Analysis

Data will be uploaded from REDCap to a statistical programme (SPSS) for analysis.



6          STUDY SETTING

This is a multicentred survey that will be run in parallel over an eight week period in all general, neurosurgical and cardiothoracic intensive cares throughout the South West. The participants are the staff who are caring for patients requiring organ support in these units. All eligible staff will be approached. Participation is voluntary.



7.1       Eligibility Criteria


7.1.1    Inclusion criteria

  • All nursing staff and doctors who are caring for
    • adult patients
    • requiring one or more of the following organ supports
      • Respiratory (invasive ventilation or new non-invasive ventilation requirement)
      • Cardiovascular (inotropes/vasopressors/mechanical devices)
      • Renal replacement therapy
    • in general, cardiothoracic or neurosurgical intensive care


7.1.2    Exclusion criteria

  • Staff caring for
    • Patients age <18years.
    • Paediatric intensive care patients
    • Patients requiring only high flow nasal oxygen therapy
    • Patients who are receiving end of life or palliative treatment


7.2       Sampling


There is no UK data on incidence of inappropriate care within critical care as judged by health care professionals although internationally it is estimated to be around 10%.


7.2.1    Size of sample

This initial study should provide data on perceived appropriateness of care of around 900 patients.


7.2.2    Sampling technique

We will collect data on one day a week over an 8-week period. Data collection will not occur on the same day of each week in case this introduces bias. Dates for data collection will be organised locally at each site.


7.3       Recruitment

We are asking the opinion of health care professionals who are directly looking after patients on the study day.


7.3.1 Sample identification


Participants will be identified as the nurse or doctor directly caring for the patient. They will be identified by the local site lead.


7.3.2 Consent

This is a self-completion questionnaire. Posters explaining the study will be placed in staff communal areas and there will be a short introductory paragraph accompanying the survey explaining the purpose of this study, how information will be used and stored and how the findings will be made available to them.

Completion of the questionnaire will be used to indicate consent on behalf of the participant (doctor or nurse).









The data collected, whilst of a sensitive nature, is anonymous (participants will not be required to identify themselves). There will be no feedback at the time of collection to healthcare staff regarding perceived cases of inappropriate case, so the conduct of the study will not alter clinical decision making. Discussions should already be taking place between nursing staff and doctors regarding whether care is appropriate. In order to ensure patients can be followed up to ascertain their outcome, it is necessary to link the patient with the questionnaire but this data will only be kept locally on the master sheets. The central data base will not collect any patient identifiers.


8.1       Assessment and management of risk

This study has no safe guarding implications


8.2        Research Ethics Committee (REC) and other Regulatory review & reports

  • Before the start of the study, a favourable opinion will be sought from a REC for the study protocol, and other relevant documents.


Regulatory Review & Compliance

  • Before any site can enrol patients into the study, the Chief Investigator/Principal Investigator or designee will ensure that appropriate approvals from participating organisations are in place. Specific arrangements on how to gain approval from participating organisations are in place and comply with the relevant guidance.
  • For any amendment to the study, the Chief Investigator or designee, in agreement with the sponsor will submit information to the appropriate body in order for them to issue approval for the amendment. The Chief Investigator or designee will work with sites (R&D departments at NHS sites as well as the study delivery team) so they can put the necessary arrangements in place to implement the amendment to confirm their support for the study as amended.


This is a relatively short survey so we do not envisage any amendments being required.


8.3       Peer review

This study was presented at the South West Anaesthetic Research Matrix and following a competitive process was selected to be taken forward as a SWARM study. This process is led by Dr Gary Minto, Consultant anaesthetist Plymouth in conjunction with the SWARM committee. It has also been reviewed internally within Musgrove Park Hospital Research Department and there have been no adverse comments.


8.4       Patient & Public Involvement

At this stage we have not involved patients or the public. We aim to involve patients when we have the results through our local critical care follow -up clinic. This provides a pool of patients who have survived critical care after a significant illness. They will be given the opportunity to review the results prior to dissemination for their opinions (if they so wish). This will be particularly useful for informing the final publication and presentation of this work.



8.5       Protocol compliance

  • Accidental protocol deviations can happen at any time. They will be documented and reported to the Chief Investigator and Sponsor immediately.

8.6       Data protection and patient confidentiality


The protocol describes the relevant steps to ensure participant and patient confidentiality.




8.7       Indemnity


This study will be covered by NHS indemnity via the sponsor (Musgrove Park Hospital). There are no obvious risks of harm to patients or participants.



8.8       Access to the final study dataset

Dr Richard Innes and Dr Debbie Webster will have access to the final dataset. Site investigators will not have access to the full study dataset whilst the study is underway. Following completion of the study and once all data is collated and anonymised then all local site investigators will have access to the data as the organisers for this study (SWARM) necessitates a collaborative authorship and as such all co-investigators will be responsible for agreeing the validity of the anonymised data.




9.1       Dissemination policy


This data is owned by the South West Anaesthetic Research Matrix (SWARM) and will be published under their logo.

It will be presented at the regional intensive care meetings within the South West and published in an appropriate journal.


9.2       Authorship eligibility guidelines and any intended use of professional writers


This study will be published with joint authorship by the various co-investigators and documentation of the contribution of the various investigators under the auspices of SWARM.

10        REFERENCES


  1. Perceptions of Appropriateness of Care Among European and Israeli Intensive Care Unit Nurses and Physicians. JAMA. 2011; 306(24):2694-270
  2. Thanh N. Huynh et al The Frequency and Cost of Treatment Perceived to Be Futile in Critical. JAMA Intern Med. 2013;173(20):1887-1894.
  3. Matthew H Anstey et al. Perceptions of the appropriateness of care in California adult intensive care units. Critical Care2015.

4 Thanh N. Huynh et al.  The Opportunity cost of Futile Care in the ICU. Crit Care Med. 2014 September; 42(9): 1977–1982

5.Azoulay E et al; Conflicus Study Investigators and for the Ethics Section of the European Society of Intensive Care Medicine. Prevalence and factors of intensive care unit conflicts: the conflicus study. Am J Respir Crit Care Med. 2009;

6.Randall Curtis. Is There an Epidemic of Burnout and Post-traumatic Stress in Critical Care Clinicians? AJRCC. VOL 175: 2007






11.1     Appendix 1- Required documentation

Academic CVs of site leads

Proof of up to date Good Clinical Practice certification for all data collectors

Written evidence of local R&D approval of protocol at each site



11.2     Appendix 2 – Schedule of Procedures (Example)


Procedures Data collection once weekly for 8 week period, follow up data collection at 6 months.
Screening and Baseline 6 Months Analysis and write up
Informed consent Not applicable
Master sheet linking study code to patient identifiers x
Nursing questionnaire x
Physician questionnaire x
Collection follow up data x
PPI involvement for advice regarding dissemination of key results to public (if necessary)  




13.3     Appendix 3 – Amendment History

Amendment No. Protocol version no. Date issued Author(s) of changes Details of changes made
N/A This version of protocol has been reviewed by sponsor