A qualitative study of decision-making about the implantation of cardioverter defibrillators and deactivation during end-of-life care

Article history

The research reported in this issue of the journal was funded by the HS&DR programme or one of its preceding programmes as project number 11/2004/29. The contractual start date was in May 2013. The final report began editorial review in December 2015 and was accepted for publication in July 2016. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HS&DR editors and production house have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the final report document. However, they do not accept liability for damages or losses arising from material published in this report.

Declared competing interests of authors

none

Copyright statement

© Queen’s Printer and Controller of HMSO 2016. This work was produced by Standing et al. under the terms of a commissioning contract issued by the Secretary of State for Health. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

Overview of the research

This research study used qualitative methods to investigate patients’, family members’ and professionals’ views about, and experiences of, implantable cardioverter defibrillator (ICD) implantation and deactivation. This chapter presents a synopsis of the study context, rationale, aims and objectives, including an overview of the project team, patient and public involvement (PPI), ethics approval and the subsequent structure of the report.

Context and rationale for the research

In the UK there are 100,000 sudden cardiac deaths per year,1 80% of which are attributable to ventricular tachyarrhythmia. 2 ICDs are recommended for patients at high risk of sudden cardiac death [primary prevention (PP)] and for survivors of cardiac arrest [secondary prevention (SP)]. 2,3 All ICDs combine both a shock function (to treat fast heart rhythms) with a pacing function (to treat slow heart rhythms). In some cases, the pacing function may be very sophisticated and can provide so-called cardiac resynchronisation therapy (CRT) for the treatment of heart failure. CRT itself may be provided by a pacemaker [cardiac resynchronisation therapy with pacemaker (CRT-P)] or in conjunction with an ICD [cardiac resynchronisation therapy with defibrillator (CRT-D)]. The great majority of ICDs are provided for PP for patients with chronic heart failure. They increase life expectancy but may be associated with adverse effects, including unnecessary or inappropriate shocks, device complications, increased hospitalisation and anxiety and depression. 4 Consequently, decision-making about an ICD for individuals should consider the benefit (averting sudden death) against possible future harms, including adverse effects and the potential need for deactivation towards the end of life.

Implantable cardioverter defibrillator implantation rates in the UK vary geographically and are increasing, but the extent to which patients are engaged in decision-making, and what their information and support needs are, is unclear. 5,6 Shared decision-making (SDM), whereby doctors and patients make joint decisions informed by best evidence and aligned with patient values and preferences,7 is associated with improvements in patient involvement, knowledge and accurate risk perception, leading to informed values-based choices and increased treatment adherence. 8,9 There is a dearth of information relating to the role that potential ICD patients can and want to have in the decision-making process and what potential recipients need to know before they decide to have ICD therapy. 5 It is unclear how possible benefits and harms are communicated to patients during the clinical encounter and how patients make sense of this information. A 2014 review10 identified that patients faced with ICD-related decisions often misunderstood the functionality of ICDs, or overestimated their benefit, and the authors recommended a SDM approach to achieve improved patient outcomes.

At the end of life, shocks from ICDs may cause unnecessary pain and suffering to the patient and distress for carers and family members. 11,12 In these circumstances, deactivation may be the most appropriate management. However, there is a paucity of evidence to reflect if, how and when deactivation conversations take place in the UK.

There are clear gaps in the existing body of research with regard to the information needs and preferences of potential ICD recipients with respect to both implantation and deactivation.

The outputs of this study help to address these gaps by providing data on current decision-making while also establishing the information needs, values and preferences of ICD recipients, carers/family members and clinicians, so that decision-making involving patients can be effectively supported. Most patients are not aware of the possibility of ICD deactivation when they have the device implanted and, for those who are, it is unclear when, how and from whom [e.g. general practitioner (GP), cardiologist, heart failure nurse, palliative care clinician] they receive this information. 6,13,14 Little is known about how patients make sense of information about ICDs and what patients’ information needs and preferences are with respect to implantation and deactivation. 5,6,15,16 By documenting current decision-making throughout the care pathway, and by exploring patients’, family members’ and clinicians’ views of decision-making, we have addressed these knowledge gaps to help determine how to better support SDM about ICD implantation and deactivation.

Aims and objectives

Project team

The project was led by Professor Richard Thomson, expert in SDM, and Professor Catherine Exley, medical sociologist and qualitative researcher, in close co-ordination with Dr Kerry Joyce, Senior Research Associate, who was a co-investigator and responsible for the day-to-day project management until September 2014. Dr Joyce’s role was taken over by Holly Standing, Research Assistant, from August 2014, with additional support from Dr Darren Flynn, Senior Research Associate and Practitioner Health Psychologist, from September 2015. The core project team met, on average, once a month to discuss the day-to-day practicalities of running the project, recruitment and data analysis.

The Project Advisory Group consisted of members of the project team and Professor Julian Hughes (Honorary Professor of Philosophy of Ageing and Consultant in Old Age Psychiatry), Dr Stephen Lord and Dr Janet McComb (consultant cardiologists), Dr Paul Paes (Consultant in Palliative Medicine), Mr Tom Bryden, Mr Tom Twedell, Mr Paul Cuskin and Mr Steve Whitely (patient representatives), Mrs Trudie Lobban, Member of the Most Excellent Order of the British Empire (MBE) (founder and trustee of the Arrhythmia Alliance: The Heart Rhythm Charity) and Dr Daniel Matlock (Consultant in Palliative Medicine). The Project Advisory Group provided feedback and advice, and reviewed progress on all aspects of the study.

Patient and public involvement

Active, sustained and non-tokenistic engagement with patients is a crucial component of our research and echoes our commitment to conducting research that supports patients in adopting a more proactive role in decision-making. Public and patient engagement was initiated prior to the development of the outline application and was led by one of the co-investigators, Dr Kerry Joyce. Mrs Trudie Lobban MBE, Founder and Trustee of Arrhythmia Alliance, was contacted at the pre-application stage. Trudie enhanced the focus of the project on patients’ needs by encouraging us to concentrate on both implantation and deactivation decisions, explaining that deactivation is often a key issue that patients and relatives would like to discuss in advance. In this way, deactivation would be addressed in advance rather than leaving it until close to the end of life, when it is often the family/carers who are faced with making the decision rather than the patient themselves, causing additional distress at an extremely emotional time. This benefit of early PPI was used as an example of good practice by INVOLVE, a national advisory group ‘to support active public involvement in NHS, public health and social care research’. 17 Trudie subsequently joined the research team as both a co-applicant and a member of our study advisory group. Trudie provided service user input into all stages, including the development of patient/carer information resources, data analysis, report writing, interpretation and dissemination of study findings. Two patient/carer representatives (with an interest in end-of-life issues) joined the initial project advisory group. The Arrhythmia Alliance and the North of England Cardiovascular Network also reviewed the research proposal and their comments were incorporated, specifically with respect to strengthening the focus on decision-making across the care pathway, including more than one secondary care centre and emphasising patient benefit.

Ethics approval

Ethics approval for this study was obtained on 26 April 2013 from the National Research Ethics Service Committee North East-Sunderland (reference: 13/NE/0105). Two ethics amendments were subsequently given a favourable opinion, on 14 November 2013 (to amend slightly the strategy for recruiting patients to the study and to be able to include patients eligible for a defibrillator in combination with CRT) and 12 May 2014 (a change to the method of approaching bereaved relatives and widening of the eligibility criteria to 12–18 months post bereavement). All potential participants were provided with information about the study and there was an opportunity to address questions to the researchers prior to participation. Individuals were informed that their participation was entirely voluntary and that they were free to withdraw at any time without reason. All personal identifying information was removed to protect confidentiality.

Structure of the report

Chapter 2 presents the background literature on the implantation and deactivation of ICDs, the rationale for the current study and the study aim and objectives. Chapter 3 details the methods used to address the study objectives, which consisted of observation of consultations, in-depth interviews and interactive workshops. An overview of the context and the participants recruited within each of the three phases is presented in Chapter 4. The findings of the interviews and workshops with reference to implantation issues are presented in Chapter 5. Chapter 6 discusses the findings of interviews and workshops with reference to deactivation issues. Chapter 7 presents a summary of the study findings, along with a discussion in relation to the previous literature, the implications of findings for supporting better SDM about ICDs, the strengths and limitations of the current study and future directions for research.

Rationale

Aim and objectives

Overview of study design

To understand fully the views and experiences of those involved in decision-making about ICD implantation and deactivation, and how such discussions are enacted in practice, requires the use of qualitative methods. In this study, we used a combination of observations of consultations, in-depth individual interviews and interactive group workshops. The study was not longitudinal in nature, that is, we did not follow a particular cohort of patients throughout the care pathway. To reflect the diversity and range of patients’ experiences, we recruited people before and after ICD implantation, as well as people who declined ICD, people considering prospective deactivation and bereaved relatives. We also adhered to the consolidated criteria for reporting qualitative research (COREQ) guidelines. 44

Patients can be referred for an ICD for either PP of sudden cardiac death or as SP after they have experienced a cardiac arrest. PP patients are likely to have more time to discuss and reflect on their decisions about an ICD than those patients who are referred for SP. With this in mind, and because of the different nature of the decision-making process (after a cardiac event), we focused pre-implantation interviews on patients referred for PP. However, to achieve sufficient numbers and to understand fully the nature of decision-making for the two different groups, we did not exclude patients with SP ICDs. We also included patients eligible for, or living with, CRT: patients eligible for a CRT device can also choose to have CRT alone or CRT-D. The decision is presented to patients as the option of having something to improve symptoms or the option of having something to improve symptoms as well as something to increase the chances of living longer. It is important, therefore, that we captured this group of patients in our sample to understand the full range of options available and the nuances of the decision-making process. Moreover, CRT-D is a common option for ICD patients, and excluding this group of patients would limit how representative the study sample was in terms of the wider population of defibrillator patients.

Data collection and analysis for this study followed the principles of the constant comparative method. 45 This is an iterative process whereby data collection and analysis occurred concurrently, with earlier interviews informing subsequent ones and continuing until no new information emerged. The initial phases used observations and interviews to explore lay and professional views and experiences of decision-making about ICD and CRT-D implantation and deactivation. Observations were conducted at one specialist centre, where ICDs are implanted, and at two district general hospitals (DGHs), with ICDs implanted at one of these. We selected these study sites because they provide different pathways to ICD implantation and, therefore, may provide different insights into decision-making between implanting and non-implanting clinicians, and the importance of different organisational approaches, such as the role of the specialist nurse. Our rationale was to understand patient/clinician interactions, information sharing and decision-making in implanting centres and to explore what happens earlier in the pathway by including two DGHs offering contrasting approaches to patient referral, one of which was cardiologist led and the other of which was nurse led. In-depth interviews were conducted with patients, family members and clinicians at the above sites. We also recruited from two additional sites (one in tertiary care and the other in secondary care) to conduct clinician interviews. For the deactivation conversations, patients and family members were recruited through cardiac physiologists, cardiologists or palliative care physicians directly involved in the patients’ care.

In the final stage of the study, these data were used in the interactive group workshops with patients/family members and clinicians to validate the findings of the initial phases and to provide a vehicle to explore their ideas and views about how the findings could be used to support better SDM about ICD implantation and deactivation, as well as to inform advance care planning for ICD recipients.

Phase 1: observations

In order to understand decision-making across the care pathway, non-participant observation46 was conducted by two researchers (KJ and HS) to familiarise themselves with the clinical environment and patient pathways. Specifically, we were interested in learning more about:

1. the nature of ICD consultations (how ICDs are portrayed within the clinical encounter, the focus and duration of interactions, who interacts with whom, who initiates/ends interactions, how the physical environment might constrain/promote the delivery of good-quality decision-making)

2. the nature of decision-making interactions (how information is shared and by whom, how risk communication is enacted and by whom, how the topic of possible future deactivation is approached and by whom)

3. the patient’s journey through the care pathway (what happens and when, key decision points regarding ICDs and how are these negotiated, the stage of the pathway at which risks, benefits and consequences – e.g. possible future deactivation – are communicated to patients).

Patients received a letter about the study with their outpatient appointment letter explaining that a researcher might be present at his/her clinic appointment to observe the consultation. This letter included clear details of how to opt out of the study (see Appendix 1). The letter of invitation included an opt-out reply slip which patients could return to the receptionist/member of clinical team on arrival to enable the opportunity to opt out without having to approach the issue face to face. The consultant cardiologist further relayed the information at the beginning of the consultation and offered a final chance to opt out. Patients had the opportunity to reflect on their decision and discuss their participation in the study with family members/significant others before deciding whether or not to participate (there was usually a period of several weeks between appointment letter and actual appointment). We chose an opt-out process of consent for the observations for three reasons: (1) we wished to avoid overburdening patients at a time when they are already receiving large volumes of information about ICDs; (2) clinicians would be seeking written consent for device implantation so asking for written consent to observe the consultation may result in confusion over what the patient is actually consenting to; and (3) obtaining written consent would lengthen the consultation and place additional demands on the consultant cardiologist.

Experienced qualitative researchers (KJ and HS) observed consultations with patients being considered for ICD. Initially, it was hoped that we would also be able to observe consultations that focused on deactivation, but this was not possible because of the nature and timing of these discussions. It is unknown how many deactivation conversations took place during the study period. We were made aware of very few deactivation conversations, and in some cases the researchers were not notified in time to permit attendance and observation of these consultations.

Our observations helped to understand how conversations about ICD implantation and subsequent deactivation are enacted. Our approach avoided imposing structure on the observations and brief field notes were written up immediately after the observation period. 47 Prior to beginning data collection, a pilot observation was conducted opportunistically, after which the researcher discussed the consultation with the consultant cardiologist in order to shape an observation grid and coding frame as a guide to their note-taking around the key components of a typical ICD consultation (see Appendix 2). This included, but was not limited to, discussions of diagnosis and checking understanding; communication of the risk of sudden cardiac death; presentation of the ICD, with risks and benefits information; and the impact of the ICD on quality of life and, in some cases, the capacity of the ICD to change the mode of death. It should be noted that this is not a deep ethnographic study and the purpose of the observations was to provide context for subsequent interviews. Consequently, the observations enabled us to understand how decision-making about ICD implantation and deactivation was enacted in different settings and the nature of ICD consultations and decision-making interactions (what is said and what remains unsaid), including the patient’s journey through the care pathway. Notes were taken by hand either during or immediately after the consultation in order to minimise the presence of a researcher within the clinic environment. Observations at each site took place prior to beginning recruitment for phase 2, thus providing local context and helping to inform the purposive sampling strategy and interview guides for in-depth interviews with patients, relatives and clinicians.

Phase 2: in-depth interviews

Interviews were conducted by KJ and HS with patients (aged > 18 years), partners/relatives (aged > 18 years) and clinicians to explore understandings and experiences of, and the factors that influence, decisions about (1) ICD implantation, and (2) deactivation towards the end of life. We excluded patients and family members who lacked capacity to participate in the study and patients and family members who were unable to speak English. Interviews were conducted at a time and place convenient to the interviewee. The majority of interviews were conducted face to face, but a small group of people preferred to be interviewed by telephone. We ensured that patient and family member interviews were conducted during times when support from the clinical team was available in the event that any of the interviewees experienced undue distress. We also made participants aware of sources of support and further information where necessary, including Arrhythmia Alliance, Patient Advice and Liaison Service, and Cruse bereavement support (appropriate information was included in the participant information sheet and was reiterated at the beginning of the interviews).

One-to-one, in-depth interviews are the method of choice where the issue under investigation requires detailed exploration of sensitive issues. In-depth interviews are particularly instructive where the current evidence base is limited,47 as they enable space and flexibility for the interviewee to shape the issues discussed, rather than the direction being predetermined by the researchers’ interests or hypotheses. The interviews investigated how decision-making about ICDs is currently undertaken and if, and at what stage, device deactivation is addressed; patients’ and clinicians’ information and support needs regarding ICDs in general and regarding deactivation in particular; patients’, carers’ and clinicians’ views about how best to support decision-making about the implantation and deactivation of ICDs; and individual and organisational facilitators and barriers to discussions about implantation and timely decision-making about deactivation.

Purposive sampling was used to capture a broad range of views and experiences and, as data collection progressed, we deliberately sought out individuals who had different views or experiences to challenge our analysis. During recruitment we sought to interview people about both implantation and deactivation decisions and purposively sampled so that the views and experiences of the following groups of patients and/or family members were included.

With regard to implantation decisions:

• before implantation – patients referred for an ICD from secondary care

• before implantation – patients recruited from implanting centres who elected to have a device implanted

• after implantation – ICD patients who had a device implanted (we purposively sampled to capture a range of experiences from those recently implanted to others who have lived with the device for up to 12 months)

• declined ICD – patients who were referred for an ICD but who declined [at either secondary (declined referral) or tertiary (declined device) stages of the pathway].

With regard to deactivation decisions:

• patients who were prospectively considering device deactivation

• bereaved family members (4–18 months post bereavement).

Where possible, we sought to interview prospectively patients who were considering deactivation via tertiary care (e.g. advanced cancer patients). However, discussions about deactivation often take place only a few days (or even hours) before death,13 which presented practical and ethical issues with regard to interviewing patients and relatives. Where deactivation was close to the end of life, and we were unable to speak to patients, we sought to interview family members of patients 4–18 months post bereavement.

Our aim was to continue with interviews until no new information emerged (theoretical saturation). The interview schedule focused on understandings of and feelings about ICDs; experiences of decision-making about implantation; if, where, when and how ICD deactivation was discussed; availability and appropriateness of information; preferences for information and decision support; and exploration of patient pathways. In addition, we included some specific prompts to tap into the values and ethical viewpoints that underpinned peoples’ decisions. Interviews were audio-recorded with written consent, transcribed verbatim and anonymised. Permission was sought from participants to retain contact details for the purposes of inviting them to participate in workshops in phase 3.

Recruitment strategy for interviews

A summary of the recruitment process is shown in Figure 1.

FIGURE 1.

Recruitment process for patient interviews.

We used a slightly different approach to recruit bereaved relatives: a letter of invitation, together with an information sheet, was sent by post to bereaved relatives identified by the physiologists at the tertiary centre, which holds a register of all those people who died with an active ICD in situ and all those people who had their device deactivated before death. We included both groups of patients in our approach (i.e. those who had their device deactivated before death and those whose device was deactivated afterwards). If interested in participating, the individual was asked to return the accompanying consent to contact form. In the first instance, we contacted bereaved relatives 4–6 months post bereavement; however, we were faced with low levels of participation so the eligibility criterion was widened to 12–18 months post bereavement. It was also made clear in the invitation that relatives could choose to be interviewed with a friend or family member if preferred. Equally, if those people contacted did not feel that they were the most appropriate person to complete the interview, they were invited to pass the information on to whoever was most closely involved in the care of their loved one. The letter about the research was accompanied by a cover letter from a cardiologist member of the study team (see Appendix 3). The purpose of the covering letter was to reassure potential participants that the researchers had no access to the clinical data of their loved one and to emphasise that there was no obligation to take part in the research. The consultant’s contact details were also included in the letter to provide a point of contact in the instance that someone encountered distress from our recruitment approach.

Phase 3: interactive group workshops with clinicians and patients/relatives

Two interactive group workshops (2 hours in duration), involving Microsoft PowerPoint (Microsoft Corporation, Redmond, WA, USA) presentations, small group work and plenary discussions, were conducted with (1) clinicians and (2) patients/relatives. These were aimed at validating the findings of the initial phases and as a vehicle to explore their ideas and views about how the findings could be used to support better SDM about ICD implantation and deactivation, as well as informing advance care planning for ICD recipients.

Workshops were audio-recorded and were facilitated by two researchers (DF and HS). Field notes were taken during workshops, which were audio-recorded.

Data collection and analysis strategy for workshops

The clinician workshop took place within a tertiary care centre. Following an overview of the aims, objectives and methods, including a brief overview of SDM, attendees were presented with a summary of the key findings from phases 1 and 2 and were invited to comment on whether or not they resonated with their views and experiences in a group plenary. This was followed by group work, in which attendees were allocated to one of two small groups to consider the following issues, with reference to the generic question ‘How can patients and their relatives be better supported to make informed values-based decisions about ICD implantation/deactivation in partnership with clinicians?’:

• Who should discuss the pros and cons of ICDs with patients/relatives?

• When should a discussion about the pros and cons of ICDs with patients/relatives take place?

• What information should be provided to patients/relatives?

• How could specific barriers to SDM be overcome?

The workshop concluded with key issues identified from each small group being discussed in a plenary session. A copy of the clinician workshop slides can be found in Appendix 4.

The patient/relative workshop was convened in a community setting. The structure was similar to the clinician workshop, commencing with an overview of the study aims, objectives and methods, including a brief overview of SDM, followed by a summary of the key findings from phases 1 and 2 and a subsequent opportunity to comment on/discuss them within a group plenary session. This was followed by small group work that also focused on eliciting their ideas on how the findings could be used to support better SDM about ICD implantation and deactivation. A structured small group exercise with reference to a summary of preference-sensitive decision points and opportunities for discussions with patients/relatives across the generic heart failure pathway (Figure 2) was used to elicit their ideas on the following questions:

• What information about options (ICD, CRT-D or doing nothing/active monitoring) should be provided to patients/relatives?

• Who should discuss the pros and cons of the options (ICD, CRT-D or doing nothing/active monitoring) with patients and their relatives?

• When should a discussion with patients and their relatives about deactivation of ICDs/CRT-Ds take place?

FIGURE 2.

Decision points and opportunities for discussions with patients/relatives about ICDs.

A copy of the patient/relative workshop slides and structured exercise can be found in Appendix 4.

Field notes and transcriptions of audio-recordings from the workshops were subjected to functional content analysis (use of a priori and emergency coding with the specific function of identifying generic categories of information) for discussion with the wider research team to establish whether or not the descriptions of our findings (presented in Chapters 5 and 6) accorded with the understandings of the participants, and to inform recommendations on how to support better SDM about ICDs.

Phase 1: observations

A total of 38 consultations involving clinicians and patients being considered for both ICD implantation and deactivation were observed across three sites (July 2013 to January 2015). Consultations ranged in length from around 10 to 30 minutes.

• Site 1: a tertiary care (implanting) centre – 29 observations of two different cardiologists consulting with their patients.

• Site 2: DGH, nurse-led clinics – three observations of two heart failure nurses consulting with patients.

• Site 3: DGH – six observations of one cardiologist consulting with patients.

Owing to a lesser frequency of potential ICD candidates seen in secondary care clinics, and the difficulties in managing to observe a consultation (owing to some patients showing improvement in the repeat echo and discussion of the possibility of an ICD or CRT-D no longer being appropriate), lower numbers of observations were conducted in secondary care than initially anticipated.

Phase 2: in-depth interviews

A total of 80 interviews were undertaken with patients/relatives and clinicians (Table 1).

Fifty-one interviews were conducted with patients and/or relatives in total; of these, seven interviews were conducted with bereaved relatives. Forty-four interviews were undertaken with patients: 33 men and 11 women, aged between 47 and 85 years [mean 68.4 years, standard deviation 10.3 years]. The majority of patients (n = 34) had been offered an ICD for PP and 10 people had received one for SP. Eleven patients had a diagnosis of cardiomyopathy and the remaining patients (n = 33) were diagnosed with ischaemic heart disease. Twenty-seven interviews were conducted with patients alone and 17 were conducted with patients along with a family member (usually a spouse) present. The largest proportion of patients was recruited via tertiary care following implantation of an ICD, or they were recruited in tertiary and secondary care contexts while considering ICD therapy. We were able to interview eight patients who declined an ICD (one of whom accepted CRT but not the ICD), but only two people who had decided in advance to have their ICDs deactivated were interviewed. Of the seven bereaved relative interviews, four were with spouses, in each case the widow of the deceased. A further three group interviews were conducted. These included two interviews with the spouse and daughter of the deceased and one interview with the son and daughter-in-law of the deceased. Our sampling strategy was successful in recruiting the range of medical/clinical specialties involved in the treatment and care of ICD patients across the pathway.

Phase 3: interactive group workshops

A total of 11 clinicians participated in the group workshop from the following specialties: two physiologists; two implanting cardiologists; two consultant cardiologists; one palliative care consultant; one clinical psychologist; two heart failure nurses; and a public health registrar working with the Cardiovascular Clinical Research Network. The patient/relative workshop was attended by 12 participants: five patients, six relatives and a patient representative from a charitable organisation.

Information provision to patients/relatives about the rationale for implantable cardioverter defibrillator therapy

As discussed in Chapter 1, the pathway to receiving an ICD is complex as there are a number of potential routes by which patients may come to be offered a device. The patient’s pathway is dependent upon several factors, including if the device is being offered for PP or SP. Furthermore, the pathway for PP patients may also vary depending on if the patient’s heart failure is being managed in the community or in a secondary care setting. Here, we outline some of the key information provided to patients when deciding whether or not to have an ICD implanted.

The purpose of the pre-implantation consultations was to establish the patient’s need for an ICD. The interviews and workshops indicated that there was variability in the timing of the initial ICD discussion. In particular, patients who were being managed for heart failure in a secondary care setting sometimes had an initial discussion about the device before accepting a referral to an implanting centre (and some patients declined the device at this point). However, it was evident from the observations, interviews and workshops that the majority of patients had limited knowledge about the device at the time of the implanting centre consultation. This may be due to one or more possibilities, most probably a combination of patients being unable to recall fully or to understand information provided on ICDs prior to referral to the implanting centre or limited discussion about ICDs before referral.

Importance of time and checking on patient understanding

In both individual interviews and the workshop, clinicians expressed concerns about patients’ abilities to retain the information given to them. Although they said that they endeavoured to provide the patient with as much information as they needed, clinicians often believed that much of this was forgotten soon after the consultation. However, having time to reflect was an important factor in patients’ decision-making processes:

It took us 2 or 3 month you [know], ’cause I didn’t even read the leaflet. I just tossed it to one side when I first got it . . . [partner’s name] read it first and she says, ‘You want to read this.’ ‘Aye, I’ll read it tomorrow, aye, I’ll read it tomorrow.’ And I eventually did read it, you [know]. But I thought to myself, ‘I’ll read it in my time.’ You [know]?

Alex, 53 years old, PP, pre implant

For patients being offered an ICD for PP, being afforded time to consider the information, discuss with appropriate others and return to clinicians with any additional questions was perceived as extremely beneficial.

Dr Elm had prepped me, and I’d been able to talk to [name of GP] about it in advance, I didn’t, incidentally, get on the web . . . So, I mean, I think where I am placed, and particularly having cardiologist [secondary care] and the rehab nurses, and . . . I’ve talked a little bit about it to [first name], the GP. I think I’ve had lots of chances to talk about it.

Bob, 67 years old, PP, post implant, ICD

However, SP patients, for clinical reasons (i.e. to minimise the risk of an early recurrent episode) were often given little time to consider their decision. Nonetheless, as SP patients have also experienced the trauma of a recent cardiac event, this may further limit their ability to process information given to them about the ICD.

Perhaps there needs to be a checking of understanding. Perhaps that’s necessary. Because I’m sure I probably seemed awake and alert and I very quickly realised that, you know, I, I was in hospital – You know, I woke up and thought, you know, ‘I’m in hospital’ . . . But obviously what I now realise was traumatic for my body obviously did have these effects that lasted longer than possibly it was obvious that it had done.

Isobel, 67 years old, SP, post implant

These data suggest the need for checks to ensure that patients have retained the necessary information about the device. Such checks of understanding would be useful throughout the pathway between the initial implanting consultation and the decision, as well as beyond any implantation.

Fostering patients’ and their relatives’ expectations about living with the device

There was variation in the extent to which patients were aware of the therapeutic purpose of the ICD or the negative impacts of living with an ICD. One important aspect of this was the understanding that the ICD is a risk modifier that offers potential survival benefits; however, a simple ICD does not confer any symptom improvement to the recipient (some devices such as a CRT-D, a specialist form of pacemaker, can offer symptom improvement). Erroneous expectations of symptom improvements and quality-of-life benefits are likely to influence patients’ decision-making.

I try to make a very clear distinction between feeling better and living longer. And I try to make it quite clear, if it’s a straight ICD, that this is not about feeling better, this is about prolonging life expectancy, so that their anticipation is not that they will feel better as a result of that process.

Dr Thyme, non-implanting cardiologist

Such confusion about the likelihood of symptom improvement was often related to a conflation of ICDs with other devices with which the patient was more familiar, such as pacemakers. As such, a major aspect of the information giving process is to distinguish between these devices.

I don’t think they realise what that box is actually going to do for them. I think a lot of people think, ‘I’m getting a pacemaker’. That’s what they, because their friends have got one and sometimes I say, ‘But yours is a very special device, yours has got other properties’, and they’ll say, ‘oh have I got the sort of, the deluxe model’.

Nurse Mint, heart failure nurse

As pacemakers are more widely utilised in the treatment of heart failure and have been around for several decades, many patients being offered an ICD possess some prior knowledge of these devices. Many patients have heard of pacemakers and often know someone, such as a friend or family member, who has received this type of device, whereas relatively few patients have prior knowledge of an ICD. Clinicians have to ensure that patients understand the differences between these devices and that they are aware of which device they are being offered.

We batter that [calling the ICD a pacemaker] out of them. Because there is a strong disincentive to call it a pacemaker, because it causes so much confusion. So any time a defibrillator patient says they’ve got a pacemaker in we say, you know ‘it’s like a pacemaker, but don’t use that word because it causes confusion with everybody’.

Ms Forsythia, physiologist

In addition to the potential for unrealistic expectations about improvements to symptoms, if patients erroneously believe that they are receiving a pacemaker, this may leave them potentially unprepared for the defibrillating function of the device. Despite clinicians’ reported attempts to try and prevent patients from confusing ICDs with pacemakers, it was evident that there were still some patients who possessed a degree of confusion, several of whom referred to the ICD as a pacemaker throughout their interview:

She [cardiologist] said she had sent a letter to [implanting centre S01] for putting my name forward for a pacemaker.

Jim, 80 years old, PP, pre implant

Furthermore, one bereaved relative expressed no knowledge of the fact that her husband had a defibrillator, believing that he had only a pacemaker.

The key difference between ICDs and pacemakers is the ICD’s capacity to deliver a defibrillation. Defibrillation is the main function by which the device mediates the patient’s risk of sudden cardiac death. In addition to the inability of a simple ICD to offer symptom benefit, patients should be made aware prior to implantation that the device has the potential for adverse impacts on quality of life. The experience of defibrillation itself has the potential to be distressing to patients.

Your ICD will not improve your quality of life and we know that in some patients it makes their quality of life worse, particularly those who are unhappy with inappropriate shocks or those who get very obsessed with their heart function and their device.

Dr Rosemary, implanting cardiologist

Although the defibrillation, or shocking function, is the key mechanism by which the device may extend a patient’s life, it is also this function that has the potential to be psychologically distressing. Indeed, several patients were interviewed who had experienced psychological sequelae following the experience of shocks.

Variations were apparent in the manner in which clinicians discussed the defibrillations; many employed humorous comparisons:

Well, a lot, a lot of people tend to say it’s like a, being kicked in the chest by a camel, but, or, I mean I don’t know how they know that’s what it feels like. But it’s, it’s, it’s a bit of a kinda, I think it’s a bit of a lighter way to tell them, so that they know it’s going to cause a bit of discomfort, but kinda not, not quite making light of it.

Mr Broom, physiologist

As with anthropomorphising the device, comparing the shocking function of the device to being kicked in the chest by a horse or camel may be useful, as this is an image that might be easily recalled by some patients. Another common method employed by clinicians to illustrate the shocking function of the device was to hit the desk very hard and suddenly. These methods appeared to be successful in instilling in patients the appropriate expectations of the defibrillation. Discussion of the shocking function and how this may be experienced appear to be an essential element of the information given to patients pre implantation.

Impact of the implantable cardioverter defibrillator on physical appearance

Another important consideration in decision-making is the potential impact of the ICD on the appearance of a patient’s body.

A patient I’ve already met but, mentioned but who is considering having an ICD fitted and she is very body conscious. She’s a young woman and she did talk about, ‘well what will it look like? Well, what’s that going to be like, having this thing that’s visible?’ . . . So that was certainly a concern for her.

Dr Aspen, psychologist

However, our data suggest that this was not an isolated incident. For example, Dan felt self-conscious about the physical appearance of the ICD:

I think one of the big subconscious things for me, in, for a personal point of view, is how protruding it is, you know, when I, when I take me [sic] shirt off, you know, it’s quite evident that it’s there. I thought in my mind’s eye at the time, that, where you know, that it would be a lot more you know, less sightly to see. You know, it, they are a fairly big thing, if you like. I am, I am a fairly slight person anyways and I think, for me, consciously, if I was away on holiday and was having to take me top off anywhere, subconsciously that’d be on my mind.

Dan, 48 years old, PP, post implant, ICD

It was evident that not all participants in this study felt fully informed about the size of the device prior to implantation. Some, such as Phil (68 years old, SP, post implant, ICD), discovered, following implantation, that the device was much larger than they had expected:

They said it was about the size of a box of matches . . . They should of [sic] said, what was it, Swan Vestas [Republic Technologies (UK) Ltd, High Wycombe, UK]!

A big box of matches? So it was bigger than you expected it to be?

And a damn site more painful.

Implantation scars, and the way in which the device can protrude, can leave some people feeling self-conscious about their changed bodies, and others reported experiencing ongoing pain or discomfort:

Yeah, it hurts when I lie on me [sic] left side and I find that very strange. It’s as though, where’s it [going] to [fall] out. And actually sometimes when you, you go over the scar it hurts, it’s very strange . . . But when I lie on my left side and I’m like in a really comfortable position, oops, no it wakes me up. It’s as though it’s squashing together. And it’s, you can see it. Now you know where they go in . . . It’s a neat scar, but it’s got a lump . . . Well I was told, because, I’m not exactly Twiggy, you wouldn’t see a thing which just shows that’s wrong.

Molly, 60 years old, PP, post implant, CRT-D

Those patients, and their families, who were able to see the ICD before the procedure reported that this was useful. Furthermore, this was something that was reinforced at the ICD support group.

And she, she did show the size and they did explain how they put them in. She did, she explained everything . . . they even passed one around. ’Cause they passed the one, they passed a pacemaker around as well to show you the difference. But do you know, they’re getting smaller all the time ’cause his is quite round and large, and it’s evident, you can see it. But I’ve seen some on the telly and they’re not, they’re not very big.

Wife of Burt, 74 years old, SP, post implant, ICD

Being able to see and touch the device appeared to facilitate patients’ development of realistic expectations regarding the impact of the device on their bodies.

Many people are involved in implantation decisions

At its core, SDM involves more than two actors. The findings of this study indicated that decision-making around ICD implantation, consistent with other preference-sensitive decisions,45 often involves multiple actors, including the patient and implanting clinician dyad but also family members, significant others and other clinicians.

Timing of deactivation discussions

Our data provide evidence of different views of clinicians and patients/family members about whether or not patients should be informed about the deactivation of ICDs prior to implantation. There appear to be several potential opportunities across the patient pathway when these discussions could be held and these are discussed in relation to their respective benefits.

Both patients and clinicians were asked about their opinions of when was the best, or an appropriate, time to discuss deactivation. In particular, we elicited views about the appropriateness of advance discussions where the issue of deactivation is introduced pre implantation. It was apparent from our data that there was a lack of consensus between patients and clinicians, and among clinicians, regarding the appropriateness of advance deactivation conversations. One argument for the inclusion of deactivation in pre-implantation discussions was the suggestion that this information could influence how the patient feels about the device and may therefore be an important factor to inform their decision-making about implantation.

It’s a tricky one. But I think that would be the ideal place [for deactivation discussions], would be before [implantation]. ‘Cause once you’ve got it, if you then decide that, ‘Well I don’t really want that because of X, Y and Z’ . . . then it seems to have ‘Had I known . . .’ decision. I think if it’s going to be talked about it should be talked about beforehand, definitely.

Mr Jasmine, physiologist

Indeed, patient interviews suggested that some patients, such as Fred (70 years old, SP, deactivated ICD patient), regretted the decision to have the device implanted when they were faced with deactivation:

I think if there was more discussion about them and about the whys and the wherefores.

So back at the time, before you have it implanted, kind of, more look to the future and what might happen?

Definitely . . . I think at the time when I had the heart attack, and before the ICD, if I’d have known what was coming, I wouldn’t have had it . . . I definitely wouldn’t have had it, no.

And is that because of the personal impact on you or is that, for you, or are you back thinking about the costs again there?

I think it would be both. I think I am practical – if nothing’s going to help, nothing’s going to help, and that’s what it is about. And what’s the point in clutching at straws?

Fred’s misgivings about the device took into account both personal and broader health service (economic) factors. Overall, he felt that the decision to have an ICD was a mistake. Furthermore, Fred’s account indicates that his view is not specific to the ICD, but rather reflects his feelings about medical interventions more broadly; he had also decided to discontinue enzyme replacement therapy, as he no longer perceived any benefit from it. Fred’s account indicates that he might have benefited from greater discussion of the future and the potential outcomes of the ICD at the time of implantation, as these factors were important in his decision-making.

A number of clinicians, in particular palliative care physicians, drew comparisons between heart failure and oncology services, and the manner in which medical interventions are presented:

It’s not exactly a parallel situation, but it’s not dissimilar to chemotherapy. I don’t give chemotherapy, but obviously I work closely with oncologists who do and many of those I’ve seen having the discussions and many of the discussions I’ve heard about subsequently have explained that, ‘This treatment may have beneficial effects but we’d need to review the effectiveness with repeat scans after a certain interval, and there may be a time when it’s no longer appropriate.’ I think anything that you do to somebody needs an exit strategy as well as an entry strategy.

Dr Basil, palliative care clinician

Dr Basil highlights how chemotherapy is not presented as something that will be used indefinitely for the remainder of the patient’s life, but rather as something that will be routinely reviewed for its appropriateness and clinical effectiveness. He indicates that such an approach would be appropriate for ICD patients, as it would highlight that there may come a point when the device is no longer appropriate, enabling patients to develop realistic expectations regarding the long-term outcomes of the device.

Furthermore, advance discussions about deactivation can provide patients, and their families, with an awareness that deactivation is a possibility in the future.

And that sounds like, you know, not putting the patient first, but if you can have an easier conversation [about deactivation] with the patient then the patient will feel more comfortable having that conversation. And so it’s not just to make it easier to have the conversation for us, it’s to make it easier for the patient to have the conversation as well, and all the family. And it’s just a simple sentence, ‘Your device will be switched off at some point, you can’t live forever’, you know.

Nurse Cypress, heart failure nurse specialist

Several clinicians spoke of advance discussions in terms of ‘sowing a seed’ by introducing the potential need for deactivation early so that it can be built upon in subsequent consultations. As Nurse Cypress suggests, the advance discussions do not need to be in-depth, unless specifically desired by the patient, but rather should present an opportunity to start to set expectations and understandings about the future. As such, when the time comes that deactivation is a consideration, this will be a concept with which the patient, and their families, are familiar.

Indeed, some clinicians recounted experiences where advance discussions had successfully facilitated the end-of-life pathway:

I think the family certainly were aware that the ICD needed to be deactivated at some point. So, I think the cardiologist had had a conversation previously, not to say, ‘We need to deactivate it now’, but to say ‘there will be some point that we need to deactivate this’ because the gentleman has been becoming more frail over the preceding months. So it wasn’t a complete shock for the family and they recognised that that was what was happening, that the patient was dying and they wanted their dad to be as comfortable as possible.

Dr Buckthorn, palliative care consultant

The findings suggest that early advance discussions about deactivation can be useful and may facilitate a realisation in patients and family members that a patient is approaching the end of his or her life. Advance conversations about deactivation and other end-of-life issues allow for planning, whereby the patient may be moved away from an active treatment plan to a more palliative approach. Clinicians indicated that, in their experiences, difficulties in the deactivation process occurred when there had been a failure to engage with the patient about future possibilities and appropriate planning had not taken place. Failure to plan for the end of life often resulted in deactivation decisions occurring in the last days or hours of the patient’s life, or not at all, and deactivation decisions being made quickly.

[If] they’ve got some other problem like cancer and are reaching the end of their life, and we have to make decisions quite quickly, unless it’s been thought through already, about their ICD.

Dr Mulberry, palliative care consultant

It seems that there is often a failure to recognise when a patient is approaching the end of life. In these instances, where deactivation often occurs out of necessity because the patient is close to death, the patient often cannot be engaged in the decision, and there were even some instances recounted of occasions on which deactivation did not take place until post mortem.

In spite of participants acknowledging the benefits of advance discussions about deactivation, our data suggest a lack of agreement regarding the timing of these conversations. For example, there appeared to be variation in the extent to which clinicians chose to engage in these discussions at the time of implantation; some gave only cursory attention to these issues:

Well again, not in any great detail perhaps sadly. But so it’s often a throwaway line along the lines of, ‘If it’s, if it’s shocking you when it shouldn’t or you’re dying of something else then the, there is the option to switch it off.’ And I’ll be honest, not everyone has that discussion.

Dr Oak, implanting cardiologist

Others indicate that the issue of deactivation was not part of their routine implantation discussion at all:

Generally I don’t because of the . . . at that point in time they’re not really . . . it doesn’t seem quite the right time to have that conversation. I mean, it’s certainly something I would think about. But I can’t think that I’ve ever had a ‘we’re going to put this in and think about deactivating’ conversation. I mean it might be something that perhaps we should think about. Certainly, I don’t as a matter of routine and I can’t think . . . I’ve ever had a conversation, ‘well, we’re going to put this in but we could turn it off at some point if you want’ conversation. . . . it’s slightly counterintuitive to think about end-of-life care at the time you are implanting the device.

Dr Rosemary, implanting cardiologist

Furthermore, it was apparent that some clinicians were opposed to the idea of including deactivation in the pre-implantation discussions, believing that too much focus on this idea at the time of implantation could be potentially harmful:

Yeah. The biggest decision is, is death. That’s the biggest decision, and we [physiologists] have to disagree with everybody else, that death cannot be mentioned when you’re putting a defibrillator in too strongly. How could you? How could you say to a patient, ‘We’re going to give you this device that’s going to save your life, but when you’re going to die, think about this conversation, now.’ It just doesn’t work. I think you’d be, you’d be giving mixed messages. So we have in the waiting area, our sole means of communication is, there is an Arrhythmia Alliance leaflet up that says, ‘Dying’. Hopefully, that, that is enough for a family member, sitting waiting for dad or mum to come out. They’ll look and go, ‘Oh I hadn’t thought of that.’

Ms Forsythia, physiologist

Ms Forsythia suggests that raising the issue of deactivation at the time of implantation has the potential to cause confusion and distress, giving patients mixed messages. Rather than engage in potentially uncomfortable conversations about death and the end of life, it seems that it is hoped that the literature available in the waiting room might be read and help patients and/or their families come to the realisation themselves that it is something that they may need to consider.

In addition to concerns about the appropriateness of including a discussion of deactivation at the time of implantation, clinicians also expressed concerns about the ability of patients to retain this information if it was given earlier in the pathway:

It’s no different to other aspects of medicine where if you’re able to offer somebody a cure to a problem, then that is the thing that they most need and most want, and whether they’re in that position to be able to take on board any of those things and really embrace it, I’m not so sure about . . . I think it would be a mistake to try and move everything upstream and say that when we put the ICDs in we need to be having much more upfront discussions about end of life ’cos I just don’t think it’s realistic for patients.

Dr Mulberry, palliative care consultant

However, an alternative approach was suggested, whereby deactivation forms part of an ongoing conversation that is introduced early on, potentially at implantation, and is then followed up throughout the patient pathway:

. . . [deactivation] should be acknowledged at the time of insertion, so that when they come for checks that they’re not just coming for checks of how frequently it’s fired and how they’re doing. Discuss the relative burdens and risks of having it and how they feel about continuing to have it.

Dr Buckthorn, palliative care consultant

Trigger points for discussions about deactivation

In addition to implantation, it was suggested that the issue of deactivation should be returned to throughout the patient pathway, to ensure that patients are informed and involved in these decisions. One suggested means of insuring that deactivation conversations take place in a timely manner was to identify trigger points for the discussions and to build these into the care of ICD patients.

It would be good to make sure that there was a trigger there to just be saying ‘Has this patient changed significantly? Is the ICD still appropriate? Should we be initiating a discussion about it?’ would be useful.

Dr Mulberry, palliative care clinician

I suppose it can be kind of key points in someone’s illness as well and whether things like hospital admissions for whatever reason, something that’s a necessity to a change and care environment, means that something has potentially changed, erm, and almost a kind of near-miss.

Dr Echinacea, palliative care clinician

These trigger points could occur at different stages throughout the patient pathway to ensure that these conversations were conducted with all relevant individuals. These may include the development of another life-limiting condition, a deterioration in the patient’s heart failure, recurrent hospital admissions or listing on a palliative care register. The production of do not attempt resuscitation (DNAR) forms or advanced directives were suggested to be the final trigger point by which to indicate that deactivation conversations should have occurred. It seemed from the interviews that this was not current practice and a recognition that the ICD needed to be deactivated often did not occur until very late in the end-of-life pathway, if at all:

And unfortunately what does happen is that it takes the patient to be really on their death bed and then they’ve got no, they’re not cognitive, well they’re not aware, they haven’t got the cognition that things are going to be switched off and but that their, it’s the families that suffer the trauma as well, you know.

Nurse Cypress, heart failure nurse

Failure to engage in deactivation conversations can cause distress for both the patient and their family if the device needs to be deactivated when the patient is in the last moments of life.

Finally, a need for deactivation to be an ongoing consideration in the care of ICD patients was identified:

[It] isn’t a one-off it’s an ongoing dialogue and an ongoing process; and not something that you can – I would imagine, it’s not something that you can do in a ten minute outpatient appointment.

Dr Echinacea, palliative care consultant

It was suggested that, rather than there being one time point for deactivation discussions, the decision-making process should take place over time:

I think that the discussion is not a one off, it’s going to be part of a series of consultations, which perhaps take place with their, their usual specialist at a more frequent interval than they would have been used to having a review in the past.

Dr Basil, palliative care consultant

As Dr Basil indicates, the decision-making process may be an escalation of the reviews that the patient is used to having. The ideal pathway would include an introduction to the idea of deactivation at the time of implantation, which is revisited across the pathway, as and when necessary, and conversations may become more frequent as the patient’s condition deteriorates.

Actors involved in deactivation decisions

A key theme that emerged from the data was the lack of clarity regarding whose responsibility it is to instigate deactivation conversations and engage patients in the decision-making process. There are several key groups of clinicians who may be appropriate to discuss the deactivation decision. These include cardiologists, palliative care clinicians, physiologists, heart failure nurses and primary care clinicians. The following sections explore participants’ accounts of each of these.

Benefits of advance deactivation discussions for patients and their families

During the interviews with patients, it was evident that advance discussions had the potential to position the idea of deactivation as something positive:

Well there are the end-of-life considerations. I’ve read somewhere something . . . And I thought, ‘Oh yes . . . well I’m aware that, that this could happen. I think, yes, if, if I was seriously ill and near the end of life I wouldn’t want the thing shocking me.

Isobel, 67 years old, SP, post implant

For Isobel, being informed about deactivation, and that this may at some point be an issue that she needs to consider, was not a frightening or distressing concept; rather, she recognised that the shocking function of the device would be inappropriate as she approached end of life and it was not something that she wished to experience. As such, the potential to deactivate the device in advance of this was viewed positively.

Furthermore, awareness of deactivation and having advance conversations about deactivation afforded some patients a sense of control over their situation.

I went and I looked and found out. It gave me a bit of feeling of, sort of, control. You haven’t really got great control or any autonomy I suppose. But it gave me a sense of that, really. And it was as if I’d left no stone unturned and my eyes are wide open about what I was going through.

Gwen, 53 years old, PP, post implant, CRT-D

Throughout her interview it was obvious that Gwen desired to be fully informed about the CRT-D and the potential outcomes of life with the device. At times she described frustration in her encounters with clinicians, where she faced difficulties eliciting the level of information she desired. Gwen expressed how having information about the CRT-D was important as she wished to process the potential outcomes in advance. This could then be retained and returned to at such point in her future at which it needs to be dealt with.

Furthermore, it was evident among some of the bereaved relatives whose partner had been deactivated pre mortem that deactivation of the ICD was viewed positively and was seen as a natural part of the end-of-life process:

You can’t fire, start a car when there’s no petrol in it, yeah? . . . Yeah, I’ve made this up . . . It’s no good firing your starter motor, starting your car, when there’s no petrol in it. So it seems that this thing [ICD] was superfluous.

Janet, bereaved relative of ICD patient, deactivated pre mortem

Janet’s husband had reached a level of deterioration at which she could recognise that he was coming towards the end of life, or, to follow her analogy, ‘out of petrol’, and, as such, she no longer saw the ICD as having a purpose. The device firing in the event of a dangerous heart rhythm would do little to change his overall condition; at this point, the purpose of the device as a protection against sudden cardiac death was futile.

Barriers to timely deactivation decision-making

Themes identified in the context of ICD deactivation included several salient barriers to timely discussions with patients/relatives: reticence of secondary/tertiary care clinicians to discuss end-of-life care, and mechanisms for appropriate identification of patients who would benefit from a discussion about deactivation, as well as ethical issues.

Summary of implantation issues

During the pre-implantation period, patients reported variable levels of knowledge about the severity and/or gravity of their condition and the associated risk of sudden cardiac death. In particular, patients being considered for ICD therapy as a PP strategy were frequently unaware of the clinical rationale for ICD therapy until their consultation with a tertiary (implanting centre) care specialist. This contributed to some patients, and their relatives, experiencing negative reactions such as a sense of shock and disbelief. This appeared to confound their capability to process adequately the information being conveyed to them about the rationale for the ICD and to differentiate ICDs from CRT-Ds and pacemakers (the latter of which was a core issue in relation to patient confusion and unrealistic expectations about the benefits of ICD therapy). Interviews with clinicians affirmed these findings. They acknowledged that communicating the rationale for ICD therapy, and the fact that the simple ICD does not confer any symptom or quality-of-life benefit, was a challenging task. Patients often reported being unaware of the availability of choice, including no active therapy (no ICD) as an option available to them.

Clinicians were observed to use, and reported the use of, metaphors to describe the function of an ICD (e.g. anthropomorphising the device by describing it as a sentry on guard duty against cardiac events) alongside the use of verbal descriptors and numerical risk information during consultations with patients. The use of verbal descriptors (using elastic language such as low, medium and high risk) was influenced by clinicians’ perceptions of their patients’ capability to comprehend probabilistic information.

Issues related to the need to understand the rationale for ICD therapy (i.e. as a risk modifier, as opposed to a risk eliminator, for sudden cardiac death and not as a treatment to improve their symptoms) was also evident from interviews with bereaved relatives of ICD patients.

The primary function of an ICD is to deliver a shock (defibrillation) that has the ‘benefit’ of preventing sudden cardiac death. However, appropriate and, in particular, inappropriate defibrillations (shocks) can be extremely distressing for patients and are likely to be a key contributing factor to the development of mental health problems such as depression and anxiety. Clinicians used a range of metaphors to describe the experience of defibrillation from an ICD, ranging from likening a shock to being ‘kicked in the chest by a camel’ to using physical actions such as ‘hitting a desk with their fists’ to evoke a vivid image in patients’ minds.

Variability in the way in which clinicians related information about the outcomes of implantation on appearance (owing to the protrusion of the implanted device in patients’ chests and post-surgical scars) was also evident. Interviews with patients revealed that they were often not fully informed about these important implications, including potential negative consequences in terms of perceived body image. There was also evidence from patient interviews that information about the surgical procedure to implant an ICD in terms of the complexity and the use of local (as opposed to general) anaesthesia was not always conveyed by clinicians prior to implantation, which was a source of considerable distress to some patients.

Patients, relatives and clinicians acknowledged the importance of providing ‘balanced’ information on the risks and consequences of ICD therapy, including the provision of information on the surgical procedure used to implant the device. However, concerns were raised about the detrimental impact of information overload, and there was a need to identify and fulfil the information needs of individual patients/relatives, as well as to gauge their desire for involvement in decision-making, which patients often felt was suboptimal. Many patients reported experiencing frustration at clinicians’ failures to recognise/address their personal information needs, which would help them to form personal preferences and values in relation to the options available. In many cases, this went beyond the prevention of sudden cardiac death to considerations about their values towards mode of death and the impact of ICDs on their capability to retain a driving licence, which was key to quality of life for many. This was compounded by a lack of time and opportunities for the development of a rapport with patients in consultations; indeed, on many occasions, the implanting cardiologist had not previously had any contact with the patient prior to the implantation consultation.

Clinicians expressed variable access to, use of and perceived value of patient educational materials produced by charities such as the British Heart Foundation and the Arrhythmia Alliance, which could further compound the issues of patient understanding about the rationale for ICD therapy, including an awareness of the associated benefits, risks and consequences of ICD implantation before, during and after any implantation procedure.

This was further complicated by clinicians often reporting that patient retention of information was suboptimal; a common remedial strategy suggested by clinicians was ‘information checks’ during subsequent clinical contacts across the pathway.

Adequate time for patients and their relatives to reflect on information provided by clinicians and the options available was considered an important prerequisite for engagement in effective SDM. Consistent with other contexts, multiple actors were important influences on decision-making about ICD therapy, including expert and trusted clinicians (in particular heart failure nurses with whom many patients in the current study had developed a close and continuing relationship) and close family members.

Summary of deactivation issues

We identified evidence of differing views of clinicians and patients/family members about whether or not patients should be informed about the deactivation of ICDs prior to implantation. However, our data suggest that patients, and their families, preferred to be informed about end-of-life issues in advance. Such discussions could be hypothetical in nature, offering an introduction to the idea that a patient’s condition may deteriorate or that they might be diagnosed with an additional life-limiting condition, at which time the shocking function of the device may no longer be considered appropriate and may diminish the quality of life remaining.

These views were shared by some clinicians who described an initial discussion to ‘plant the seed’ regarding deactivation and to address unrealistic expectations about ICD therapy, but other clinicians intimated that deactivation was not a routine component of the pre-implantation discussions; this was driven by a clinical belief that discussing end-of-life issues alongside a device to prevent cardiac death (which appeared to be an illogical juxtaposition of information on implantation and deactivation) may lead to harm to patients, in particular for patients offered ICD therapy as a SP strategy following a recent cardiac event.

Indeed, there was a lack of consensus among clinicians about when in the pathway a conversation about deactivation should occur. Introducing the idea of deactivation prior to implantation appeared to be a particularly contentious issue, with some clinicians expressing concerns about the distal nature of this ‘advance decision’ that may or may not have to be implemented; however, others were concerned about the potential for information and/or discussions about deactivation carrying a risk of communicating mixed messages to patients.

However, we identified evidence of decisional regret associated with a lack of patient awareness about the long-term consequences of ICD therapy related to deactivation; for example, one patient reported that if he was aware of the issues around deactivation, then he would not have chosen to have an ICD. Clinicians also described patients/relatives experiencing difficulties in the deactivation process owing to a failure to engage with them about future possibilities that inhibited them from developing appropriate planning for end-of-life care in the context of ICD deactivation.

Potential trigger points for deactivation discussions embedded within a formalised protocol were suggested by clinicians to ensure that opportunities for appropriate and timely discussions about deactivation could occur with patients and families, with an emphasis placed on the need for regular monitoring and review over time in terms of the clinical effectiveness and appropriateness of continued ICD therapy, a process described by one palliative care clinician as ‘an entry and exit strategy’ when discussing the option of ICD therapy. Specific trigger points that could result in a deactivation discussion included diagnosis of an additional life-limiting condition, deterioration in heart failure, frequent hospitalisations and placement on a palliative care register, with the production of a DNAR order being a final trigger point. However, clinicians suggested that there might be difficulties with, and challenges to, introducing a formalised protocol with regard to identifying trigger points for historical cases in which patients had been living with an ICD for a number of years and are unaware of deactivation issues. Nevertheless, opportunities to identify historical cases could be facilitated via ‘trigger points’ such as appointments for battery replacement, or, where patients are being cared for in the community, appointments with their GPs (emphasising a potential role for primary care clinicians in deactivation discussions).

A lack of consensus on who is best placed to initiate discussions about deactivations at these trigger points was apparent from interviews with a range of clinical specialists involved in the ICD patient pathway. Assertions that cardiologists are best placed were based on their expertise with the device itself, but were countered by arguments that other clinicians, such as primary care clinicians and heart failure nurses, are more closely involved in an individual patient’s continuing care (and thus are more familiar with their preferences and values). There were also calls for greater involvement from palliative care clinicians, but a key barrier was a lack of perceived interprofessional collaboration in the care of cardiac patients in some of the research sites. Palliative care clinicians themselves expressed frustration at cardiologists who they felt often surrendered conversations about deactivation to them.

Consistent with implantation issues, patients and relatives valued having an expert and trusted clinician, with whom they had developed a positive therapeutic alliance, to initiate a discussion about deactivation. A role for heart failure nurses or physiologists in initiating deactivation conversations was highlighted by patients, as they had the most established relationships and continued contact with them. However, nurses are clinically trained, whereas physiologist training is focused on the technical elements of ICD function; therefore, currently nurses may be better placed and more likely to have had appropriate training for engaging patients/relatives in deactivation discussions than physiologists. For physiologists to take responsibility for initiating discussions about deactivation, this would involve taking on additional duties that, at present, are not recognised as part of the physiologist role.

The crucial role of patients’ preferences and values in making decisions about deactivation was acknowledged, including the benefits of decisions about deactivation being made in advance, which would provide patients with a sense of control over their situations. Interviews also emphasised that decision-making does not occur outside consultations in a social vacuum; rather, the patient’s family is often closely involved in deactivation discussions and decisions, frequently acting as a patient’s proxy towards the end of life if no advance discussion had taken place.

Summary of the barriers and facilitators to shared decision-making in the context of implantable cardioverter defibrillator therapy

Arguably the most salient barriers to SDM were discordant clinician and patient/relative preferences and views on information needs with regard to ICD therapy. These were strong contributing factors to: (1) patients expressing variable levels of knowledge about the severity/gravity of their condition and associated risk of sudden cardiac death, including details about the surgical implantation procedure for ICDs, and, importantly, the clinical rationale for ICD therapy (i.e. as a risk modifier as opposed to a risk eliminator for sudden cardiac death and not as a treatment to improve their symptoms); (2) clinicians’ variable disclosure and methods of conveying information about the benefits, risks and consequences of ICD therapy; (3) patients lacking awareness of choice, in particular, the reasonable ‘preference-sensitive’ option of no active therapy (no ICD); (4) clinicians avoiding discussions about end-of-life issues, which, to some extent, was driven by beliefs about the ‘illogical juxtaposition’ of presenting information about implantation alongside deactivation (beliefs that patients would be harmed by discussing end-of-life issues alongside discussing a device to prevent cardiac death); and (5) a lack of consensus among clinicians about when in the pathway a conversation about deactivation should occur (and who should do this) with perceived suboptimal interprofessional collaboration in the care of cardiac patients in some of the research sites.

Facilitators to SDM in the context of ICD therapy were close involvement of a patient’s relatives in the decision-making process (in the context of deactivation, this facilitated acting as a patient’s proxy towards the end of life); the existence of positive therapeutic alliances with expert and trusted clinicians (and the role of heart failure nurses, GPs and, possibly, physiologists in the deactivation discussions); the provision of sufficient time for patients and their relatives to reflect on information provided to them; clinicians using strategies such as information checks across multiple consultations; raising deactivation as a possibility (which also served to address unrealistic expectations about ICD therapy); and potential trigger points for deactivation discussions being embedded within a formalised protocol for appropriate and timely discussions about deactivation.

Interpretation of findings in relation to the literature on shared decision-making

Several themes identified in the current study resonate with findings of previous research on SDM about ICD therapy and the wider theoretical research on SDM.

The involvement of multiple actors in the decision-making process in the context of ICD therapy is consistent with the concept of ‘distributed decision-making’, which describes decision-making as a continuing event taking place over multiple encounters with a range of different people. 49 This was driven to a large extent by the complexity of the ICD pathway, the multiple clinicians involved, the close involvement of relatives and the widening of this complexity when end-of-life issues emerge. As seen in other preference-sensitive decision-making contexts, such as decision-making about thrombolysis in acute stroke care,42 patients valued the views and support of expert and trusted clinicians (in this context, heart failure nurses with whom many patients in the current study had developed a close and continuing relationship) and close family members.

Sepucha and colleagues50–52 describe the concept of decision quality, whereby preference/value-concordant decisions are made by patients who are knowledgeable about their condition, the options available to them and their personal preferences and values associated with the options. Our findings suggest that information exchange in SDM encounters should go beyond the presentation of options, with information on the clinical rationale for ICD therapy, including the patient’s diagnosis, also playing an integral part in engaging patients in SDM to enable them to understand their condition and probable prognosis. This is particularly important given that simple ICD therapy does not have symptom or quality-of-life benefit and the appropriateness of presenting a ‘do nothing’ option (i.e. no ICD therapy) as a valid choice.

Our findings concur with previous research that reported that patients/relatives want to be actively involved in decision-making about implantation and deactivation, and the majority want to be informed about deactivation at the pre-implantation stage,53 which can have important benefits such as an improved sense of control for patients, while preparing both patients and families for later end-of-life discussions. Patients and relatives want to be offered a choice and sought balanced information about the available options [pros and cons, in particular, the potential impact on psychological well-being and quality of life in the short and long term (deactivation)], and involvement in decision-making. Patients/relatives who participated in the group workshop unanimously agreed that patients would benefit from a clear overview of the pathway and associated decision points (see Figure 2) to facilitate the process of SDM.

However, we identified a range of issues that may have inhibited good SDM and prevented patients from making quality decisions about ICD therapy.

Consistent with previous research,15,16,24,54 several patients in our study held erroneous beliefs about the benefit of ICD therapy and seemed unaware of the adverse effects of ICDs and the potential threats to quality of life, such as the risk of psychological problems, driving restrictions and a negative impact on body image.

Clinicians’ values (beliefs/attitudes) and their patients’ preferences for information and involvement in decision-making were often discordant with each other, in particular with respect to the timing of deactivation discussions. We found evidence that clinicians relied on their clinical judgement to ‘diagnose’ their patients’ preferences for information about ICDs, including the method of conveying risk information (e.g. Dr Oak described how he used elastic language in the first instance, followed by numerical information if perceived to be ‘helpful’, to convey risk information to patients) and their desired level of involvement in decision-making. However, research has consistently shown that clinicians are typically very poor at accurately ‘diagnosing’ their patients’ preferences. 55,56 In the SDM literature, this has become known as the ‘silent misdiagnosis’, as clinicians are often unaware of their patients’ preferences or make incorrect assumptions and are unaware of the consequences that this might have for patients. 57

This silent misdiagnosis was evident in our findings and has important implications for SDM in the context of ICD implantation and deactivation. Misdiagnosing patient/family member preferences can:

• Act as a salient barrier to engaging patients and relatives in SDM, which may result in patients receiving treatment and care that they do not want, or would not choose if they were fully informed.

• Foster the development of unrealistic expectations of ICD therapy, in particular when risk information (outcome probabilities) is presented using different methods that can have a differential impact on perceptions of risk and decision-making in both clinicians and patients. 58

• Lead to missed opportunities for optimising the costs of ICD treatment to the NHS by reducing any unwarranted variation in their use across the UK,59 as previous research has shown that patients often choose less conservative options when they are informed about the available options. In the context of our study, we found evidence that patients being considered for ICD implantation were often unaware of the option to ‘do nothing’.

We found evidence of another type of silent misdiagnosis in the context of ICD therapy, with regard to clinicians’ perceptions of a patient’s future disposition in terms of the impact an ICD may have on their psychological health and well-being. In the context of ICD therapy, this may provide an argument for the wider involvement of psychologists as further sources of ‘expertise’, which patients could use to more accurately identify their treatment preferences in order to inform their decision-making at the pre-implantation stage. In addition, for ICD patients who experience anxiety and depression following implantation, greater awareness and referral (self or clinician) to the NHS Improving Access to Psychological Therapies service may be a further preference-sensitive option that could be presented to patients.

We identified evidence of clinicians’ gatekeeping of information regarding ICD therapy. In addition to unmet information needs regarding implantation and deactivation, the option of not having an ICD was not always presented as a valid choice to patients. One possible reason for this is that clinicians may be less inclined to offer no active therapy (no ICD) as an option to patients who have experienced a cardiac event (SP). Furthermore, such patients themselves may be less inclined to consider having no intervention as their experiences of a cardiac event may instil in them a more acute perception of the potential benefits of ICDs. In PP, the balance of risks and benefits, the course of underlying disease and varying patient values may mean that not having an ICD is a more valid option.

A further key barrier to the provision of information on risks and benefits at the pre-implantation stage was the view held by some clinicians in our study that providing patients with information about ICDs prior to attendance at the implanting centre may raise patient expectations of the device, when later they may be deemed to be ineligible for an ICD by the implanting cardiologist.

Congruent with previous research,11,24 we identified a reticence among clinicians about raising the issue of deactivation with patients and relatives (especially pre implantation), including a lack of consensus on who should initiate discussions about deactivation and their timing across the pathway. In previous research,6,13,14,53 many patients and relatives reported not being informed about deactivation until a patient was nearing the end of life, when early discussion (pre implantation) would have been their preference. In contrast, several clinicians in our study held the view that discussing deactivation at the pre-implantation stage would cause unnecessary distress or send mixed messages about ICDs; however, this is also contrary to research that reported that the great majority of patients did not believe that this would lead to increased anxiety in patients or relatives. 53 Inattention to deactivation in NICE guidelines2 and a regional policy for the deactivation/reactivation of ICDs26 is likely to have been another factor influencing clinicians’ decisions to refrain from discussing deactivation of ICDs with patients and relatives.

Nevertheless, this withholding (intentionally or not) can have negative consequences, such as patients experiencing decisional regret and poor satisfaction with care received. We found evidence of decisional regret associated with a lack of patient awareness about the long-term consequences of ICD therapy related to deactivation. As well as a salient barrier to patient engagement in SDM, it is also felt by some authors that withholding information violates three principles that underpin international clinical research ethical guidelines, namely, respect for persons or autonomy, beneficence or a favourable balance of risks and potential benefits and justice or a fair distribution of the benefits and burdens of research. 60

How can the silent misdiagnosis and gatekeeping of information be addressed to overcome these key barriers to patient engagement in SDM? The literature can be dichotomised into ‘clinician pull’ and ‘patient push’ (or vice versa) interventions, including strategies targeting both sides of the consultation.

Patients’ preferences for involvement is influenced by demographics, experiences of illness and medical care, their beliefs on involvement and the interactions and relationships they experience with health-care professionals, which are likely to develop and change over time. 61,62 Good SDM requires two kinds of expertise, namely clinicians’ knowledge of efficacy and outcomes of treatment options and patients’ expertise on their values and preferences with regard to the available treatment options. 63 This assumes that patients possess both the willingness and capability to engage in SDM. Arguably, the latter is a prerequisite of the former, with active involvement in SDM requiring patients and clinicians to have acquired a complex range of knowledge, behaviours (skills) and confidence to enact those skills. This does not preclude the value of decision aids, but, as Flynn64 asks, ‘can a decision aid be a useful adjunct to the clinical process if patients lack awareness of the value of SDM; and perhaps crucially do not have the experience, knowledge, skills and/or confidence to engage meaningfully in discussions with clinicians about their healthcare?’.

Patient activation is a behavioural concept, which is important for active patient engagement and participation in SDM about their treatment and care, and is defined as ‘an individual’s knowledge, skill, and confidence for managing their health and health care’. 65 Low patient activation is an important barrier to SDM and there is evidence that this can be ameliorated with behavioural interventions,66 which appear to be more effective when they are tailored to support the individual’s level of activation (tailored coaching) and focus on skills (e.g. question formulation) and confidence building,67 which have been shown to be effective in increasing patient activation for up to 12 month post intervention. 68

Clinician push interventions can take many forms to increase the willingness and capability for SDM in order to actively support their patients to engage in SDM and, ideally, will be underpinned by an evidence-based implementation strategy to embed SDM in clinical systems and patients pathways. 69 A multicentre implementation project, MAking Good decisions In Collaboration (MAGIC) programme,70 underpinned by a theory-based model of SDM skills,69 was designed to explore how SDM can be embedded in both primary and secondary care settings

More than 270 clinical and non-clinical staff (including NHS management and general practice staff) participated in MAGIC. They were provided with the following active ingredients of the MAGIC programme:

• skills development and engagement, for example introductory and advanced skills development workshops for participating clinicians

• support with developing and implementing (or adapting existing) decision-support tools (brief patient decision aids)

• facilitation and peer support, for example action learning forums that involved regular meetings with MAGIC staff and clinical teams to share experiences and learning

• support to involve patients, which included activities such as convening patient forums and a patient activation intervention designed to increase awareness of SDM and as a prompt to initiate SDM discussions known as ‘Ask 3 Questions’ (What are my options? What are the possible benefits and risks? How can we make a decision together that is right for me?).

An independent evaluation of MAGIC71 found evidence that the programme successfully improved participants’ understanding and awareness of SDM and developed their skills and confidence to use SDM in practice; supported clinical teams to test and develop SDM tools and approaches; made changes to clinical practice; and provided opportunities to hear and understand the patient experience of SDM, which (via role-play exercises) facilitated connections between the theory of SDM and its use in routine practice. This challenged a key barrier to SDM: ‘we are doing this already’. The importance of training was heavily emphasised, with a key learning point being the aphorism ‘attitudes and skills trump tools’.

The lack of formal incentives for clinicians to engage in SDM has been discussed in the literature; for example, the existing incentive structure ‘Qualities and Outcomes Framework’72 can, in effect, disincentivise engagement in SDM. 71,73 However, there is a recent legal precedent from the UK Supreme Court in 2015, whereby clinicians are now required to take ‘reasonable care to ensure that the patient is aware of any material risks involved in any recommended treatment, and of any reasonable alternative or variant treatments’. 74 This precedent in UK law provides a strong incentive for clinicians to refrain from information gatekeeping and may lead to improved SDM by increasing their willingness (and wider infrastructure support) to engage in continuing professional development on SDM.

Suggested areas to consider in the future development of services

• The clinical rationale for offering ICD therapy would benefit from being clearly communicated to patients and their relatives as early as possible in the patient pathway by skilled clinicians supported, where available, by codesigned information and tools, including live demonstrations and the manipulation of devices to support informed consent and SDM.

• Patients desire more information on the surgical procedure to implant ICDs to support informed consent and SDM, which can facilitate the development of positive/protective preparatory coping strategies to minimise any impact on psychological well-being (decisional regret) and perceived body image.

• Increased access to, and a greater role for, psychological support from appropriately qualified mental health professionals throughout the ICD pathway.

• Signposting to support groups for patients with lived experience of heart failure/ICD therapy and their relatives, while acknowledging that these options are preference-sensitive and that patients would benefit from being provided with sufficient information to make an informed values-based decision.

• There is a lack of consensus/ownership among clinicians about who should take responsibility for discussing deactivation. To address this, the issue of deactivation could be introduced early in the pathway and formalised within a standardised protocol with specific trigger points to enable more detailed and timely discussions with patients and relatives about deactivation.

Recommendations for future research

We identified a number of important areas that are prerequisites for good SDM and for maximising the likelihood of patients making ‘high-quality’ values-based decisions. The following would be fruitful avenues for further research:

• Developing and evaluating the use of risk communication tools that convey balanced information on the benefits, risks/adverse effects and short- and long-term consequences of ICD implantation and deactivation (identified by our research as being important for patients), including training on SDM skills and risk communication techniques for clinicians involved in the treatment and care of ICD patients.

• Developing and evaluating training for clinicians (SDM, risk communication and delivering bad news to patients/engaging in difficult conversations, which should be viewed as a key skill as part of the SDM process in this context).

• Developing theory-based patient activation interventions for patients and their relatives early on in the pathway.

• Developing ICD-specific decision quality measures (implantation and deactivation).

• Developing appropriate interventions to prepare patients (in particular those in tertiary care) and family members for SDM with cardiologists throughout the ICD pathway provided by clinicians with frequent patient contact (e.g. heart failure nurses) who can act as centralised points of contact outside tertiary consultations.

• Updating existing decision support for ICD implantation and deactivation decisions informed by our findings (available options and associated risks, benefits and consequences) as part of national decision-support material to reduce any unwarranted regional/national variation in ICD therapy and other health-care domains. Such tools would benefit from being codesigned in partnership with patients and clinicians in both primary and secondary care. A decision-support tool for ICD deactivation would potentially benefit from being embedded within advance care planning processes to ensure sharing of this information with all clinical staff across care settings (e.g. the Electronic Palliative Care Co-ordination systems that are being rolled out nationally).

• Exploration of the role of primary care clinicians and information sharing between and within primary and secondary care clinicians involved in the care of ICD patients, including palliative care.

• Developing a central role for heart failure nurses, physiologists and primary care clinicians in terms of providing ongoing care/initiating deactivation issues.

• Fostering evidence-based strategies for interprofessional learning and collaboration between cardiology and palliative care teams in the context of ICD therapy.

• Carrying out multicentre service evaluation studies of enhanced ICD pathway (training, tools and role extensions described above).

All the decisions around ICD therapy in the current study were taken in particular points in time and are, therefore, shaped by that context. Future research might consider the use of longitudinal interviews to explore the trajectory of experiences.

Strategy to translate findings into continuous professional development learning and training

Our approach to mobilisation of knowledge is multifaceted. First, a summary of the study findings will be produced for the Arrhythmia Alliance and Cardiomyopathy UK for distribution to patients/relatives. We shall also host a half-day dissemination event in 2016 in collaboration with the Arrhythmia Alliance and Cardiomyopathy UK, at which patients, family members and clinicians can discuss the study findings. Second, we shall present the findings at meetings of the Arrhythmia Alliance and other relevant patient organisations, as well as work with our advisory group and the Northern England Cardiovascular Network, to publicise the findings to clinicians and academics more broadly. Third, we shall present the findings to practitioners and academics in the field at national and international symposia. To engage with practitioners in the cardiology community, we shall present the findings at the Heart Rhythm Congress, and to connect with other academic audiences we plan to present findings at the Palliative Care Congress, the International Shared Decision Making Conference and the British Sociological Association Medical Sociology Conference. We shall also publish the study findings in a peer-reviewed, high-impact-factor, journal.

We shall work with the Northern England Cardiovascular Network and specialist palliative care teams to embed the patient voice into their operational policy for the deactivation of ICDs. We shall also cascade learning from this study to NICE to inform national guidelines on ICDs.

We shall share our interview findings with a team led by Dr Dawn Stacey at the University of Ottawa to inform the design and evaluation of a planned pilot RCT of an ICD decision-support tool.

Contributions of authors

Holly Standing (Research Assistant) contributed to data collection (observations, interviews and workshops), data interpretation and the drafting of the report.

Catherine Exley (Professor of Qualitative Health Research) developed the study protocol and contributed to the study design, data interpretation and the drafting of the report.

Darren Flynn (Senior Research Associate and Practitioner Health Psychologist) contributed to design and delivery of interactive group workshops, data interpretation and drafting of the report.

Julian Hughes (Honorary Professor of Philosophy of Ageing and Consultant in Old Age Psychiatry) developed the study protocol and contributed to the study design, data interpretation and the drafting of the report.

Kerry Joyce (Senior Research Associate) led PPI, developed the study protocol and contributed to the study design and data collection (observations and interviews).

Trudie Lobban MBE (Founder and Trustee of Arrhythmia Alliance: The Heart Rhythm Charity) developed the study protocol and contributed to the study design, data interpretation and the drafting of the report.

Stephen Lord (Consultant Cardiologist) developed the study protocol and contributed to the study design, data interpretation and the drafting of the report.

Daniel Matlock (Consultant in Palliative Medicine) developed the study protocol and contributed to study design, data interpretation and the drafting of the report.

Janet M McComb (Consultant Cardiologist) developed the study protocol and contributed to the drafting of the report.

Paul Paes (Consultant in Palliative Medicine) contributed to the study design, data interpretation and the drafting of the report.

Richard G Thomson (Professor of Epidemiology and Public Health) developed the study protocol and contributed to study design, data interpretation and the drafting of the report.

Outputs and dissemination

Data sharing statement

Data can be obtained from the corresponding author.

Disclaimers

This report presents independent research funded by the National Institute for Health Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, NETSCC, the HS&DR programme or the Department of Health. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, NETSCC, the HS&DR programme or the Department of Health.

Opt-out form

(PDF download)

Observation grid

Date: Clinician:

Observation coding frame

• Implantation.

Deactivation

• Reversibility of decision.

• Desire to keep it activated.

• Patient decision versus clinical.

• Clinical barriers:

  • triggers for discussion

  • identification of potential deactivation patients

  • difficulties ascertaining where patient is in discussions

  • negotiation between different clinician groups

  • clinician avoidance of end-of-life discussions

  • need to raise profile of ICDs

  • inappropriate terminology

  • need to routinise discussions.

• Timing:

  • advanced discussions

  • ‘right’ time for discussions.

• Actors involved:

  • cardiologists

  • palliative care.

• Ethical issues:

  • deactivation = suicide

  • active removal of treatment.

• Understanding of deactivation:

  • confusion of process

  • patient misunderstandings

  • clinician misunderstandings

  • significance of the heart.

Invitation letter

(PDF download)

Participant information sheet

(PDF download)

Consent form

(PDF download)

Interview schedules

Clinician cover letter for bereaved relative interviews

(PDF download)

Health-care professional workshop slides

(PDF download)

Patient/family member workshop slides and materials

(PDF download)