Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

A review of nurses' attitudes towards DNAR decisions

  • Comment

Abstract Palmer, R. (2007) Review of nurses’ attitudes towards DNAR decisions.

BACKGROUND There is a great deal of controversy regarding decision-making and do not attempt resuscitation (DNAR) orders. This can be the cause of much anxiety for nurses.

AIM: This literature review investigated the current knowledge of nurses’ attitudes towards and experiences of DNAR decisions.

METHOD:An integrated literature review was undertaken using several critiquing strategies to ensure systematic analysis provide a rigorous process for synthesis of the findings.

RESULTS: The majority of nurses wanted to be part of the DNAR decision-making process but, in practice, this was not always the case. A high proportion was in favour of patient and next of kin involvement, but it appears that in practice they were rarely involved.

DISCUSSION: Although nurses perceived their role as one of patient advocate, few felt they were able to act in this way with regard to resuscitation decisions. This is in conflict with the NMC (2004) code of professional conduct.

CONCLUSION: There is a need for open dialogue about the issues relating to resuscitation decisions, and clear guidelines should be introduced. The rationale for the decision should be made clear and be understood by all involved.

The challenging nature of ethical and legal issues in the use of DNAR orders confronts nurses with some of the most complex dilemmas encountered in clinical practice (Schultz, 1997), and can be the cause of much anxiety for nurses (Simpson (1994).

The British Medical Association et al (2001) resuscitation guidelines advise practitioners on resuscitation issues, stressing that policies should exist in all establishments where resuscitation attempts are made. The guidelines, which encourage all NHS trusts to develop their own local policies, were originally produced in 1993, were revised following requests for detailed guidance on legal and ethical issues from healthcare professionals (Jevon, 2001).

An integrated literature review was undertaken to explore nurses’ attitudes towards and experiences of DNAR decisions. As the research papers reviewed involved small-scale quantitative studies, the review was a quantitative overview. Several critiquing strategies provided a framework which ensured systematic analysis of research papers, and provided a rigorous process for synthesis of findings.


CPR has become established practice in hospitals (Costello, 2002). Ideally, it should only be attempted if there is a good chance of survival, but Tunstall-Pedoe et al (1992) claimed that only 15% of those receiving CPR on general wards survive to discharge, while 50% fail to respond to the initial CPR attempt. According to Peberdy et al (2003), rates of survival following cardiac arrest in hospital are poor in all age groups - fewer than 20% of adult patients survive to discharge. Blackall (1987) suggested CPR is a desperate technique that works infrequently, and virtually never in some types of patients. Despite this, for all in-hospital cardiac arrests, CPR is performed unless a DNAR order exists (Hayward, 1999).

Some patients, especially those with chronic diseases, do not benefit from CPR, so procedures were developed to withhold resuscitative measures which became known as DNAR orders (Kerridge et al, 1998). These indicate that no resuscitation measures are to be taken if a patient has a cardiopulmonary arrest.

BMA et al (2001) emphasised that it is essential to identify those patients for whom CPR is inappropriate. However, even with clear evidence that cardiac arrest and death are likely, ward staff infrequently make decisions about the patient’s resuscitation status (Resuscitation Council UK, 2005).

Jevon (2001) stated that many patients have been successfully resuscitated and go on to have a good quality of life. However, bringing dead patients back to life can prolong the dying process and deny them a dignified and peaceful death (McDermott, 2002). This is reiterated by Carver and Mackinnon (1994), who suggested that unsuccessful resuscitation not only prolongs the dying process but also does not enhance the dignity hoped for when death occurs.

According to McDermott (2002), discussions regarding resuscitation status are invariably initiated by nurses, probably because nurses spend more time with patients and are more likely to observe any deterioration in their condition.

  • Although nurses regularly initiate resuscitation attempts, they are often not included in decision-making relating to resuscitation status (Thibault-Prevost et al, 2000). Historically, they have left resuscitation decisions to medical staff (Jevon, 1999). However, nurses have the responsibility to act as patient advocates, where able (NMC, 2004), and while the final decision lies with the doctor, resuscitation status should be a multidisciplinary decision (BMA, 2001).

Key themes

The following key themes and sub-themes were identified from the literature review and are discussed below:

  • Ethical issues:
    - Age bias;
    - Cessation of treatment;
    - Quality of life issues;
    - Informed consent.
  • Involvement in DNAR decision-making process:
    Nurse involvement;
    - Patient involvement;
    - Next of kin/family involvement.
  • Documentation.

Ethical issues

According to Jevon (2001) the decision not to resuscitate is complex, with many ethical, legal and moral implications.

In a British study by Bass (2003), the entire study sample thought human rights and patient choice were important ethical issues. The general consensus of opinion was that no one else should make decisions if patients are capable of doing so themselves. Other ethical issues highlighted included patients not being informed that they had DNAR status, and ensuring death with dignity. Most of the reports reviewed mentioned ethical principles.

Law and ethics provide a framework by which decisions about withdrawing or withholding treatment may be made. The Human Rights Act (1998) incorporates several articles that are relevant to resuscitation decisions. Under the act, it is essential that decisions relating to resuscitation must be compatible with human rights. As do not resuscitate orders infringe the right to life, healthcare professionals should be able to justify these decisions. There are four ethical principles that summarise moral responsibility to others: autonomy (respect for individual choice and confidentiality); beneficence (doing good); non-maleficence (avoiding harm); and justice (distribution of resources) (Levack, 2002). Although these principles can be used to guide DNAR decision-making the application of DNAR orders involves ethical principles that sometimes conflict: autonomy (patient’s right to decide treatment) and beneficence (doctors should provide treatment that does good) (Cherniack, 2002).

Age bias

Hakim et al (1996) highlighted an age bias in the DNAR decision-making process. It was found that patients over 75 years received DNAR decisions more quickly than younger patients, regardless of prognosis.

DNAR orders have also been placed on older people’s medical notes without them having been discussed with the patient or their relatives (Age Concern, 2000). This controversy prompted Ebrahim (2000) to suggest that such orders written without consulting the patient constituted unethical and inadequate care.

Some 78% of nurses in the US study of burn nurses by Purvis et al (1998) agreed that age should not be used as a criterion for a DNAR decision-making. However, in another study, when asked whether age of patient influenced DNAR decision-making, 41% said it did (Thibault-Prevost et al, 2000).

Cessation of treatment

According to Henneman et al (1994), a key concern of the DNAR order is whether a patient will still receive other treatments because, while a DNAR order is often understood to refer to withholding CPR, the legal definition allows for other medical interventions, such as ventilatory support and invasive monitoring.

Withdrawing or withholding medical treatment is common for patients with DNAR orders. It is therefore imperative that policies are in place to ensure the relevant legal and ethical issues are addressed. Such policies facilitate standardisation of decisions and reduce controversy relating to DNAR status (Jezewski, 1994).

Only 46% of nurses in the study by Purvis et al (1998) thought that all medical treatment except CPR should continue. While 33% approved of discontinuing mechanical ventilation, over 60% of respondents opposed discontinuing intravenous fluids, tube feeding and medications. In contrast, 72% of participants in the Canadian study by Thibault-Prevost et al (2000) thought a DNAR order should discourage initiating aggressive treatments. It was also reported that 60% stated that withdrawal of therapy occurred after the DNAR order was written, and that 65% deemed admission to critical care units inappropriate. Nurses in this survey did not associate DNAR status with no care: 98% felt that nursing care should be maintained until patients die.

Quality of life

Most of the nurses in the Bass (2003) study felt DNAR decisions depended on how patients perceived their quality of life, and whether risks of resuscitation outweighed the benefits. Some stated that the way the multidisciplinary team felt about the patient’s quality of life should be taken into account.

Thibault-Prevost et al (2000) asked critical care nurses’ when they believed DNAR orders should be written: 71% said this should happen when quality of life after CPR would be poor, 66% said when present quality of life is unacceptable and 61% said when CPR was considered futile. However, Mello and Jenkinson (1998) warned that futility-related decisions ignore patients’ autonomy, while the term futility is avoided in the BMA et al (2001) guidelines, as it is viewed as paternalistic and could be used in place of the real issues and rationale behind the DNAR decision.

Informed consent

Informed consent can be seen as the foundation on which patient empowerment is based and therefore should be central to decision-making. It is a legal requirement in healthcare practice as well as a moral obligation, and should be sought for all resuscitation decisions (Thomas, 1997).

Although CPR is a medical treatment and requires patient consent, this is not always obtained. All nurses in the Bass (2003) study who thought patients should be included in resuscitation decisions felt that informed consent was important before any decision could be made. Curiously, not all respondents felt they could raise the subject with patients themselves. In practice however, respondents reported that medical staff usually made the decision; this does not reflect the nurses’ desire for informed consent, patient choice or empowerment.

Manias (1998) suggested that the incongruence between nurses’ attitudes about who should make decisions and their experiences of who actually makes those decisions may lead to ethical dilemmas because patient autonomy has not been supported.

Involvement in the DNAR decision

Miles and Burke (1996) suggested that nursing staff normally instigate DNAR discussions, although consultants would have overall responsibility. Giles and Moule (2004) asked nurses who should be involved in the initial DNAR discussion: 97% agreed that nurses should be involved, while 99% said the doctor should instigate such discussions. Almost all reported it was important to assess the best interests of the patient before such discussion. It was also reported that patients and their relatives were only involved if they had initiated conversation around the subject.

The majority of nurses in the study by Bass (2003) felt the decision should be a multidisciplinary one. Non-medical members of the multidisciplinary team build up different and sometimes closer relationships with patients, they are therefore able to act as advocate or give support during resuscitation discussion. Harris and Davies (2007) conducted an audit into DNAR documentation and found 84% of DNAR decisions were discussed with relatives,100% were documented in nursing notes and 90% were made by seniorteam members.

Nurse involvement

Jevon (1999) identified the nurse’s role in resuscitation decisions as one of advocacy and accountability. Nurses are often not included in DNAR decision-making, even though they are most likely to initiate resuscitation attempts, and have close contact with the patients and their families. This is in contrast with their code of professional practice (NMC, 2004), which states that nurses should act as a patient’s advocate.

The majority of nurses in the Purvis et al (1998) study wanted to be part of the DNAR decision, which they felt should include multidisciplinary team members, the patient and their relatives.

Giles and Moule (2004) found that nurses perceived their views to be valuable, and felt they should be involved in DNAR discussions for both competent (85%) and incompetent patients (90%). However, only 33% reported they were involved for competent patients and 40% for incompetent patients.

When asked who should be involved in the DNAR decision, 81% of respondents in the Thibault-Prevost et al (2000) study said nurses, but 63% claimed nurses were infrequently involved. Most respondents agreed that patients (97%) and family (95%) should participate in the decision, and almost all (99.5%) agreed that the doctor should be involved.

However, only 40% of nurses surveyed by Giles and Moule (2004) felt competent patients were encouraged to be involved in the decision.

In order to make appropriate decisions, nurses should be involved early in the DNAR decision-making process (De Gendt et al, 2007).

Patient involvement

The BMA guidelines (2001) state that patients should be involved in discussions about their resuscitation status, and decisions should be based on the following aspects relating to the patients: overall chance of success; views and wishes of competent patients; and patients’ human rights relating to the right to life while being free from degrading treatment.

Some nurses in the Bass (2003) study felt there was a greater need than ever for patients to be involved in any treatment decision, mainly because of the risk of litigation. However, they reported that in practice very few patients were involved and that it was mainly a medical decision. The majority of respondents felt their role was to act as an advocate for the patient, someone to support them and uphold their choice, while some felt their role included protecting patients from futile resuscitation attempts.

Next of kin/family involvement

Next of kin involvement was supported by 83% of nurses in Giles and Moule’s (2004) study, while 78% reported that this happened in discussions relating to incompetent patients. Only 55% felt this was the case for competent patients.

The Australian study by Manias (1998) gained similar results, with respondents placing high importance on patient involvement. Nurses overwhelmingly acknowledged the importance placed by society on patient autonomy, but said this was not evident in the clinical setting. In their experience patients, next of kin and nurses did not play predominant roles in influencing patients’ DNAR status. The nurses reported feeling frustrated at their lack of input in DNAR decisions and at the lack of involvement of patients and their families.

The study showed that 91% of respondents felt they should be involved in decisions relating to DNAR status, 98% felt patients should be involved and 71% placed importance on the next of kin being involved. Strikingly, only 1.4% believed that it was a collaborative decision involving patient, next of kin and nursing staff.

Manias (1998) suggested that further research is needed into why nurses, patients and family do not contribute to the DNAR decision. It is also recommended that nurses enhance their role as patient advocate, thereby enhancing patient autonomy.


According to Mello and Jenkinson (1998), DNAR status should be recorded immediately in the patient’s medical notes along with a clear rationale for the decision. Ineffective communication and documentation can lead to CPR being performed or withheld inappropriately (Jevon, 2001).

Pennels (2001) stated that if nothing has been documented in a patient’s note, resuscitation should be carried out. According to Harris and Davies (2007), in order to improve the clarity of documentation, specific forms for recording DNAR decisions should be used. All nurses in the Giles and Moule (2004) study agreed that resuscitation status should be recorded in the patient’s notes and 91% said this actually occurs in practice. While 94% agreed that resuscitation status should be made clear at every nursing handover, only 72% said that this actually happens.

Miles and Burke (1996) reported that although DNAR decisions are recorded in the medical notes, rarely was a full explanation given. This reflects similar findings by Giles and Moule (2004), whose respondents identified varying practices regarding the documentation of the rationale for DNAR decisions.

Only 55% of the respondents from the Thibault-Prevost et al (2000) study stated DNAR decisions were well documented, while 70% observed doctors’ hesitancy in writing them. Indeed, nurses in the Bass (2003) study felt their accountability included prompting medical staff to complete the correct documentation once a DNAR decision had been made.

A high percentage (96%) of nurses in the Thibault-Prevost et al (2000) study stated that nurses should be involved in ensuring organisations had a DNAR policy, while 99% thought that doctors should be involved. When asked if DNAR policy facilitated nursing practice, 75% agreed. However, only 59% knew that they had a local DNAR policy. Manias (1998) showed similar results - only 45% of nurses surveyed were aware of a DNAR policy.

Thibault-Prevost et al (2000) recommended documentation of a comprehensive patient treatment plan, educational programmes and increased dialogue to facilitate understanding of issues relating to DNAR orders. Purvis et al (1998) also argued that an open dialogue is essential, and suggested there was a need for acceptable guidelines for all team members.


There is incongruence between nurses’ attitudes about who should make the DNAR decision, and nurses’ experience of who actually makes that decision.

The majority of nurses in the reviewed studies wanted to be part of the DNAR decision-making process, but this was not always the case in practice. A high proportion were in favour of patient and next of kin involvement, but it appears that in practice they were rarely involved.

Although nurses perceived their role as patient advocate, few felt they were able to act in this way with regard to resuscitation decisions (Bass, 2003). This is in conflict with the NMC (2004) code of professional conduct, which stipulates that nurses must act as patient advocate where possible. Acting as the patient advocate, nurses can collaborate with medical colleagues when DNAR decisions are made, thereby enhancing patient autonomy.

The literature review has highlighted the need for open dialogue about the issues relating to resuscitation decisions, and that clear guidelines should be introduced. Furthermore, the rationale for decisions should be made clear and be understood by all involved.

  • Comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Links may be included in your comments but HTML is not permitted.

Related Jobs