Pre-printed ‘Do Not Attempt Resuscitation’ forms improvedocumentation?

Nick Castle a,b,*, Robert Owen b, Gary Kenward a, N. Ineson a,c

a Nurse Consultant in Emergency Care, Frimley Park hospital, Portsmouth Road, Camberley, Surrey, UKb Department of Emergency Medical Care and Rescue, Durban Institute of Technology, Durban, South Africa

c The Royal Defence Medical College, Birmingham, UK

Received 5 March 2003; received in revised form 10 April 2003; accepted 24 April 2003

Resuscitation 59 (2003) 89�/95

www.elsevier.com/locate/resuscitation

Abstract

Objective: Do not-attempt-resuscitate orders are fundamental for allowing patients to die peacefully without inappropriate

resuscitation attempts. Once the decision has been made it is imperative to record this information accurately. However, during a

related research projected we noted that documentation was poor and we thought that the introduction of a pre-printed Do Not

Attempt Resuscitation (DNAR) form would improve the documentation process. Design: Two sets of identical research questions

were applied retrospectively, 12-months apart, to notes of adult patients (�/18 years) who had died during a hospital admission

without under-going a resuscitation attempt. Between the first and the second audit, a new resuscitation policy that incorporated a

pre-printed DNAR form was introduced into our hospital. Results: A pre-printed DNAR form improved documentation when

measured against; clarity of DNAR order (P�/0.05), date decision was made/implementation (P�/0.014), presence of clinician’s

signature (P�/0.001), identification of the senior clinician making the decision (P 5/0.001) and justification for the DNAR decision

(P 5/0.001). However, the pre-printed form made little improvement in encouraging patient involvement in the DNAR decision-

making process (P�/0.348). Conclusion: A pre-printed DNAR form can improve documentation significantly but it has little effect

in encouraging patient involvement in the decision-making process.

# 2003 Elsevier Ireland Ltd. All rights reserved.

Keywords: Do Not Attempt Resuscitation (DNAR); Patient involvement; Pre-printed form

Resumo

Objectivos: As ordens para nao tentar (DNAR) reanimar sao indispensaveis para permitir que os doentes morram em paz. Uma

vez instituıda esta determinacao e imprescindıvel regista-la de forma apropriada. Contudo, ao investigar o processo notamos que a

documentacao era pobre e pensamos que a introducao de um formulario ordens para nao tentar (DNAR) pre impresso poderia

melhorar poderia melhorar a documentacao do processo clınico. Desenho: foram analisadas retrospectivamente duas series de

questoes incidindo sobre os processo de �/ 18anos que morreram durante a hospitalizacao e nos quais nao foi tentada a

reanimacao. Entre a primeira e segunda auditoria implantou-se uma nova polıtica de reanimacao no nosso hospital, que incluıa

DNAR pre-impressas. Resultados: As DNAR pre-impressas melhorou a documentacao avaliada por: clareza da DNAR (p 5/

0,005), data em que a decisao / implementacao foi instituıda (p �/ 0,014), existencia de assinatura do clınico senior (p �/ 0,001),

identificacao do medico senior que tomou a decisao (p B/ 0,001) e justificacao para a DNAR (p B/ 0,001): contudo, a DNSAR pre

impressa aumentou pouco a percentagem de doentes envolvidos na decisao (p �/ 0,348). Conclusoes: as ordens de DNAR pre

impressas melhoram significativamente a documentacao mas tem pouca influencia no aumento do envolvimento dos doentes no

processo de decidir.

# 2003 Elsevier Ireland Ltd. All rights reserved.

Palavras chave: Ordens para nao tentar (DNAR); Envolvimento do doente; Formularios pre impressos

* Corresponding author. Tel.: �/44-1276-60-4604.

E-mail address: nic4healths@aol.com (N. Castle).

0300-9572/03/$ - see front matter # 2003 Elsevier Ireland Ltd. All rights reserved.

doi:10.1016/S0300-9572(03)00176-X

Resumen

Objetivos : Las ordenes de no intentar reanimacion son fundamentales para permitir a los pacientes morir pacıficamente sin

intentos inapropiados de reanimacion. Una vez que se ha tomado la decision es imperativo registrar esta informacion con precision.

Sin embargo, durante una investigacion relacionada proyectada, notamos que la documentacion era pobre y pensamos que la

introduccion de un formulario pre impreso de ‘No Intentar Reanimacion’ (DNAR) mejorarıa el proceso de documentacion. Diseno :

Se aplicaron retrospectivamente dos conjuntos de preguntas de investigacion identicos, separados 12 meses, de las notas de pacientes

adultos (�/18 anos) quienes murieron durante su estadıa hospitalaria sin ser objeto de intento de reanimacion. Entre la primera y

segunda evaluacion, se introdujo una nueva polıtica de reanimacion que introducıa un formulario de DNAR pre impreso.

Resultados : Un formulario pre impreso de DNAR mejoro la documentacion al compararlo en: claridad de la orden de DNAR (P

�/ 0.05), fecha en que se tomo la decision / implementacion (P �/ 0.014), presencia de la firma del clınico ( P �/ 0.001),

identificacion del clınico mayor que toma la decision (PB0.001) y justificacion de la decision de DNAR (PB0.001). Sin embargo, el

formulario pre impreso mejoro muy poco en lo que concierne involucrar al paciente en el proceso de toma de decision de DNAR (P

�/ 0.348).Conclusion : Un formulario pre impreso de DNAR puede mejorar significativamente la documentacion pero tiene poco

efecto en fomentar el involucrar al paciente en el proceso de toma de decision.

# 2003 Elsevier Ireland Ltd. All rights reserved.

N. Castle et al. / Resuscitation 59 (2003) 89�/9590

Palabras clave: No intente reanimacion (NIR); Involucrar al paciente; Formulario pre impreso

1. Introduction

The decision to withhold cardio-pulmonary resuscita-

tion is fraught with difficulties and in the United

Kingdom (UK) has become a political issue following

a number of high profile Do Not Attempt Resuscitation

(DNAR) decisions [1�/3]. A number of influential

groups, primarily representing the elderly [4], have

become involved demanding that DNAR decision and

the process surrounding how these decisions are made

be open to greater scrutiny [5]. This is supported by

formal instruction from the Chief Medical Officer in the

Department of Health in the UK [6].

Increasing political pressure, supported by advice

from the Chief Medical Officer, resulted in an edict

from the Department of Health outlining a nationwide

DNAR policy [7]. All hospitals in the UK were

subsequently instructed to review their DNAR policy,

and to assist in this process, the British Medical

Association (BMA), the Royal College of Nursing

(RCN) and the Resuscitation Council UK (RC(UK))

up-dated their previously published guidance documents

[8].

Frimley Park Hospital had revised it’s DNAR policy

in-line with best recommended practice and, although

the decision making process was considered robust,

concerns remained about the quality of documentation.

It had been noted previously during an ongoing

research project into the prevention of cardiac arrest

that only 11.5% of patient who died underwent a

resuscitation attempt, equating to a DNAR rate of

88.5% during 1999 [9].

Our DNAR rate of 88.5% is in-line with other

published DNAR rates of between 75 and 90% [10,11]

but is much higher than the rate identified by Layson et

al. [12]. Steen [13] quotes the work by Layson et al. [12]

to highlight the generally low DNAR rate in some

countries/institutions.

Furthermore our reported cardiac arrest rate per 1000

admissions of 3.3% is well within the recently reported

range by Weerasinghe et al. (1.65/1000�/4.75/1000) based

on a review of all in-patient cardiac arrests for the

Australian health district of New South Wales [14].

During research into the prevention of cardiac arrestan expert panel [9] noted that a small number of patients

who had been resuscitated should have been considered

for a DNAR order (unpublished data) [15]. It, therefore,

appears that in general our DNAR decisions, when

made, were appropriate although a small number of

other patients required further consideration of their

suitability for cardio-pulmonary resuscitation.

The principle objective of this paper was to note thequality of DNAR documentation in the presence of a

well-established DNAR policy. Other papers have

identified that poor documentation of DNAR decisions

can be problematic [16]. Concerns about the general

standard of medical record keeping have also been

widely reported [17�/20]. Substandard record keeping

can lead directly to medico-legal action [20,21] or make

decisions made in good faith more difficult to defend.

2. Methods

2.1. Population and location

Frimley Park Hospital is a medium size hospital with

700 in-patient beds including an eight-bedded coronary

care unit, nine-bedded intensive care/high dependencyunit, and an emergency room that treated 75 684

patients during 2001. The hospital provided in-patient

care for 53 146 patients as well as care for 217 178

outpatients during 2001. The hospitals activity, size and

catchment area has changed little between the two audit

dates.

2.2. Audit process

This was a retrospective audit involving the medicalnotes of all patients who died at the hospital during a 1-

week period in July 2001 with the process being repeated

for the same period 1-year later. All patients over the

age of 18 years were included unless they had experi-

enced a period of active resuscitation.

Audit questions were agreed with both the hospital’s

risk management and audit departments and ratified by

the director of nursing (Table 1). Ethical committeeapproval was not sought as this was a retrospective

audit of the DNAR documentation process.

The aim of the first audit was to identify a baseline

standard of documentation and to highlight areas for

improvement. Soon after the first audit a pre-printed

DNAR form was introduced along with educational

support for staff. The second audit compared the

standard of documentation before and after the pre-printed DNAR form and education was introduced.

A one-year gap was left between audits to allow for

the newly introduced pre-printed DNAR form to

become part of established hospital practice. This was

based upon anecdotal experience that newly introduced

policies require a period of regular practice before being

assessed.

2.3. Statistical analysis

The data was coded and analysed with the StatisticalPackage for the Social Sciences (SPSS) version 11.0.

Proportions were compared using Pearson’s Chi-

squared (x2) test for independence and where categories

contained less than five Fishers Exact test was applied.

A ‘P ’ value of less than 0.05 was considered statistically

significant. Confidence intervals could not be generated

because the data collected was nominal (categorical) in

nature, producing a Yes/No answer that does notsupport the application of confidence intervals.

2.4. Documentation process

The previously established method of recording

DNAR decisions was usually via a coded entry (e.g.not for 333, the hospital emergency number) being

recorded on the inside cover of the patient’s notes or

occasionally within the main-body of the patient’s

clinical notes. The DNAR decision was also recorded

occasionally within the nursing notes either as a hand-

written entry or using a pre-printed highly visible

adhesive sticker.

The DNAR decision-making process was limited todoctors with full General Medical Council registration

but any team member could complete the DNAR form

as long as the name of the senior clinician making the

decision was clearly identifiable.

Following the initial audit, the hospital’s medical

director instructed that the pre-printed DNAR form be

introduced and the previously established documenta-

tion process should stop.

3. Education process

As part of the introduction of our pre-printed DNAR

form a briefing document was produced and added to

our junior doctor’s induction package. The process

surrounding DNAR decisions was also included in

qualified nursing staff basic life support training.All admitting consultants and each ward/department

manager received a copy of both the up-dated DNAR

policy and the pre-printed form. In addition, a number

of copies of the DNAR policy (and pre-printed form)

were placed in the hospital library and the postgraduate

medical education centre as well as being published on

the hospitals intra-net.

All senior hospital staff were given ample opportunityto provide feedback to the medical director with regard

to the new DNAR policy and the associated change in

the documentation process. No changes were put

forward by the consultants or other bodies within the

hospital, and therefore, the new policy was ratified by

both the executive board and the consultant staff

committee and was instigated from August 2001.

Our approach placed great emphasis on ward-basednursing staff to ensure that accurate documentation was

undertaken, particularly in the immediate period follow-

ing junior doctor rotation.

4. Results

There were 17 patients identified in the first audit(July 2001) and 20 patients in the second (July 2002).

Following introduction of the pre-printed form, we

noted a statically significant improvement in the follow-

Table 1

Research questions

(1) Was the DNAR decision clearly recorded in the patient’s notes

without the use of code?

(2) Was the date that the DNAR decision was made clearly recorded?

(3) Was the DNAR order signed?

(4) Was the clinician recording the DNAR orders clearly visible by

name and by grade (e.g. was name printed)?

(5) Was a reason for the DNAR decision given?

(6) Was the patient involved in the decision making process?

N. Castle et al. / Resuscitation 59 (2003) 89�/95 91

ing key areas: clarity of DNAR statement (P�/0.05), the

date the decision was made/implemented (P�/0.014),

the presence of a clinician’s signature (P�/0.001),

identification of the senior clinician making the decision(P 5/0.001) and justification for the DNAR decision

(P 5/0.001). The sixth key area, involvement of the

patient in the decision making process, demonstrated a

slight but non-statistically significant improvement (P�/

0.348).

It is noteworthy that relative involvement in the

decision making process surrounding DNAR has re-

mained unchanged (P�/1.0) between the two auditsdespite improved documentation and the lack of any

formal legal role for relatives with regards DNAR

within the UK [6,22]. The role of relatives is limited to

stating what a patient would wish and this is clearly

highlighted within both our local as well as the national

DNAR policy [8].

5. Changes to pre-printed DNAR form

Following a notes review and discussion with both

junior doctors and ward nurses, it was decided to changethe wording on the pre-printed DNAR form with

regards to communication with relatives. The aim was

to change the process from a passive process of passing

information to an active process of informing relatives

of the medical decision that had been made. This

involved changing the statement of ‘discuss with next-

of-kin/significant other’ to ‘inform next-of-kin/signifi-

cant other’.This decision was aimed at emphasising the role of the

clinician in making the DNAR order and to de-

emphasise the role of relatives. This change was

instigated following reports from senior nurses that

medical staff were delaying recording appropriate

DNAR decisions until it had been discussed with

relatives.

6. Discussion

As can be seen by a direct comparison of the twodifferent sets of data (Fig. 1) significant improvements

where made in five-out-of-the-six key areas following

the introduction of the pre-printed DNAR form sup-

ported by education. However, no statistically signifi-

cant improvement was noted with regards patient

involvement nor was any significant change noted with

regards the role of the next-of-kin.

The keys to the successful implementation of the newpolicy and the associated pre-printed DNAR form were

the senior ward-based nursing staff, who provided

continued support to medical staff and the hospital

consultant body who retained overall responsibility for

decisions made in their name [6�/8].

An area where the ‘pre-printed form’ made no

statistically significant improvement was the involve-ment of patients within the decision-making process as

out-lined in the BMA/RCN/RC(UK) DNAR guidelines

[8] and the European Human Rights act [23]. This lack

of communication had been the catalyst for the nation-

wide edict from the Department of Health [7] following

a number of high-profile cases in the media [1,3,4].

We noted that only 1 out of 16 patients were involved

in the decision-making process prior to the introductionof a pre-printed DNAR form and that this increased to

only 4 out-of 20 patients following the introduction of

the pre-printed form (P�/0.348).

Our experiences regarding the lack of patient involve-

ment in decision-making is not unique, having been

highlighted by other authors [24]. Smith et al. noted that

less than 50% of doctors in his study were aware of the

patient’s wishes regarding DNAR decisions. The mainreasons given for a lack of formal doctor-patient

discussion was the time such discussions required [24].

Fukaura and colleagues noted that in Japan only 5%

of patients were involved in DNAR decisions and that

typically a patient surrogate (usually a family member)

was consulted regardless of the competency of the

individual patient [25]. The data presented by Fukaura

et al. serves to highlight the different cultural issuessurrounding DNAR decision-making and patient in-

volvement [25].

The role of surrogates appears to differ from country

to country with Europe placing little store in their role

(outside of stating what a patient may have wanted)

through to formal protection in law in other countries

such as Japan and some American states [25�/27]. It is

noteworthy that under English law, relatives have noauthority to give consent but are only able to imply what

the patient may have wanted [6,22]. It would seem that

in practice relatives believe they have a greater say and

as we have found during our audit, attending medical

staff continue to empower relatives.

The process of involving patients within the decision

making process is open to debate. Ebrahim suggests free

and open discussion with all patients and implies that alack of patient involvement is unethical [28]. However,

other clinicians highlight that this may be a cruel and

inhuman process leading to inappropriate resuscitation

attempts [29,30].

The reasons for non-involvement of patients within

the decision making process was not assessed during our

audit. A number of suggestions have been proffered by

clinicians, including a fear of upsetting patients [30,31],stereotyping [32], patients being too sick or unable to

comprehend and time constraints [24]. Despite this it

would appear that patients do wish to be involved in

DNAR decisions [33] and clinical practice needs to

N. Castle et al. / Resuscitation 59 (2003) 89�/9592

change to reflect this [34]. Many patients and relatives

have exaggerated expectations regarding the success of

resuscitation efforts [35,36]. Once they have been given a

balanced explanation they are often able to make

informed decisions [37]. It is also important that all

parties realise that DNAR decisions do not necessarily

mean the withdrawal of active treatment and that many

patients designated DNAR survive to leave hospital.

As well as changing the wording on our DNAR pre-

printed form to emphasise the role of the patient and to

downplay the role of relatives/surrogates it was pro-

posed that a patient information leaflet be introduced.

Fig. 1.

N. Castle et al. / Resuscitation 59 (2003) 89�/95 93

The BMA and the (RC(UK)) have produced a sample

leaflet to assist with this process [38].

7. Limitations of this study

The small numbers of patients enrolled may affect

the ability of this audit to generalise but it does give

insight into the process. Despite this, statically signifi-

cance was seen in five out-of-the six key audit questions.

The small patients numbers were due to the audit being

limited to a 1-week period. This was partly to ensure

that medical staff did not become aware of the audit,

and therefore, change how they documented DNARdecisions.

No attempt was made to see if the DNAR decision

was appropriate as this was not the purpose of this

audit. Neither did we note if there was an increase in the

number of DNAR orders made although this will be the

subject of ongoing audit and clinical education.

8. Conclusion

DNAR orders are an established part of medical care

as they facilitate the natural process of dying without

aggressive medical intervention. This is particularly

important when we consider the poor outcome asso-

ciated with resuscitation that is further compounded by

the costs associated with futile resuscitation attempts.

Once the decision has been made, it is imperative thatboth the decision, and any associated rationale, is

carefully and accurately recorded so as to minimise

errors. We have found that a structured DNAR form

Fig. 1 (Continued)

N. Castle et al. / Resuscitation 59 (2003) 89�/9594

will aid the documentation process and improve com-

munication between health care professionals but it had

little impact with regard to increasing patient involve-

ment in DNAR decisions.

References

[1] Cancer patient’s fury at doctor who wrote her off on hospital’s

death Ward’, The Guardian, 13 April 2000.

[2] Munro R. A heart-stopping moment. Nurs Times 2000;96:12.

[3] Communication surrounding a decision not to resuscitate a

patient-W258/89-90. In: Report of the Health Service Commis-

sioner, 1990/1991. London: HMSO, 1991. p. 12.

[4] Age Concern England. Turning your back on us*/older people

and the NHS. London: Age Concern, 2000.

[5] The beginning of the end to the scandal of ‘Not for resuscitation’.

Age concern England. Astral House, 1268 London Road,

London, 2001.

[6] Chief Medical Officer. Letter to all consultants on resuscitation

policy. London: Department of Health, 1991 (PL/CM0(91)22).

[7] NHS Executive Resuscitation policy (HSC 2000/028) September

2000 (England). Scottish Executive Health Department Resusci-

tation policy (HDL 2000/22). Edinburgh Scottish Executive,

November 2000.

[8] British Medical Association, Resuscitation Council (UK), Royal

College of Nursing. Decisions relating to cardiopulmonary

resuscitation: a joint statement from the British Medical Associa-

tion, the Resuscitation Council (UK) and the Royal College of

Nursing. J Med Ethics 2001;27:310�/6.

[9] Hodgetts JT, Kenward G, Vlackonikolis I, Payne S, Castle N,

Crouch R, Ineson N, Shaikn L. Incidence, location and reasons

for avoidable Cardiac arrest in a district general hospital.

Resuscitation 2002;54:115�/23.

[10] Bedell SE, Pelle D, Maher PL, Cleary PD. Do not resuscitate

orders for critically ill patients in the hospital: how are they used

and what is their impact. J Am Med Assoc 1986;256:233�/7.

[11] Skogvoll E, Isern E, Sangolt GK, Gisvold SE. In-hospital

cardiopulmonary resuscitation. 5 years incidence and survival

according to the Utstein template. Acta Anaesthesiol Scand

1999;43:177�/84.

[12] Layson RT, Adelman HM, Wallach PM, Pfeifer MP, Johnston S,

McNutt RA. Discussion about the use of life-sustaining treat-

ments: a literature review of physicians’ and patients’ attitudes

and practices. End of Life Study Group. J Clin Ethics

1994;5:195�/203.

[13] Steen PA. Letter to the Editor. Resuscitation 2002;55:349.

[14] Weerasinghe DP, MacIntyre RC, Rubin GL. The epidemiology of

cardiac arrests in a Sydney hospital. Resuscitation 2002;53:53�/62.

[15] Hodgetts, JT, Kenward G, Vlackonikolis I, Payne S, Castle N,

Crouch R, Ineson N, Shaikn L. Unpublished data.

[16] Wynne G, Owen R, Evans T. Haphazard documentation of do

not attempt resuscitation (DNAR) orders. Resuscitation

1998;37:S53.

[17] Chua RV, Cordell WH, Ernsting KL, Bock HC, Nyhuis AW.

Accuracy of bar codes versus handwriting for recording trauma

resuscitation events. Ann Emer Med 1993;22:1545�/50.

[18] Bateman ND, Carney AS, Gibbin KP. An audit of the quality of

operation notes in an otolaryngology unit. J R Coll Surg

Edinburgh 1999;44:94�/5.

[19] Sanghavi R, Shefler A. The ABCs of recording paediatric cardiac

arrests. Resuscitation 2002;55:167�/70.

[20] Lesar T, Briceland L, Stein DS. Factors related to errors in

medication prescribing. J Am Med Assoc 1997;277:312�/7.

[21] Marsh B, Narain J. Patient dies because of GP’s bad handwriting,

Daily Mail 4 February 2003;35.

[22] Dimond B. Legal aspects of nursing 2nd. London: Prentice Hall,

1994.

[23] Humans Rights Act, 1998. The Stationary Office October 2000.

[24] Smith TJ, Desch CE, Hackney MH, Shaw JE. How long does it

take to get a ‘do not resuscitate’ order. J Palliat Care 1997;13:5�/8.

[25] Fukaura A, Tazawa H, Nakajima H, Adachi M. Do-not-

resuscitate orders at a teaching hospital in Japan. New Engl J

Med 1995;333:805�/8.

[26] Haverkate I, van Delden JJM, van Nijen AB, van der Wal G.

Guidelines for the use of do-not-resuscitate orders in Dutch

hospital. Crit Care Med 2000;28:3039�/43.

[27] Mellor M, Jenkinson C. Comparison of medical and nursing

attitudes to resuscitation and patient autonomy between a British

and American teaching hospital. Soc Sci Med 1998;46:415�/24.

[28] Ebrahim S. Do not resuscitate decisions: flogging dead horses or a

dignified death?: resuscitation should not be withheld from elderly

people without discussion. Br Med J 2000;320:1155�/6.

[29] Tunstall-Pedoe H. Do not resuscitate decisions: resuscitation

should not be part of every death. Br Med J 2001;322:102.

[30] Smith G. Do not resuscitate decisions: sound clinical reasons for

withholding cardiopulmonary resuscitation must not be confused

with ageism. Br Med J 2001;322:104.

[31] Matthews E. Can paternalism be modernized. J Med Ethics

1986;12:133�/5.

[32] Thompson BL, Lawson D, Croughan-Minihane M, Cooke M.

Do patients’ ethnic and social factors influence the use of do-not-

attempt-resuscitate orders. Ethn Dis 1999;9:132�/9.

[33] Bruce-Jones P, Roberts H, Bowker L. Resuscitating the elderly:

what do the patients want. J Med Ethics 1996;22:154�/9.

[34] Say when you want to die, patients asked. Daily Mail 24 February

2003;15.

[35] Miller DL, Jahnigen DW, Gorbien MJ. Cardiopulmonary

resuscitation: how useful? Attitudes and knowledge of an elderly

population. Arch Intern Med 1992;152:578�/82.

[36] Murphy GJ, Burrows D, Santilli S. The influence of the

probability of survival on patients’ preferences regarding pul-

monary resuscitation. New Engl J Med 1994;330:545�/9.

[37] Walker RM, Schonwetter RS, Kramer DR. Living wills and

resuscitation preferences in an elderly population. Arch Intern

Med 1995;155:171�/5.

[38] Decisions relating to cardiopulmonary resuscitation: model

information leaflet. London: BMA, 2002. Available on http://

www.resus.org.uk.

N. Castle et al. / Resuscitation 59 (2003) 89�/95 95