Decision-Making of Patients With Implantable Cardioverter-Defibrillators at End of Life: Family Members’ Experiences

Abstract

Introduction:

Many patients with advanced heart failure (HF) experience the life-extending benefits of implantable cardioverter-defibrillators (ICD), but at the end stage of HF, patients may experience shocks with increasing frequency and change the plan for end-of-life (EOL) care including the deactivation of the ICD. This report describes family members’ experiences of patients with ICD making decisions at EOL. Understanding the decision-making of patients with ICD at EOL can promote informed decision-making and improve the quality of EOL care.

Methods:

This pilot study used a mixed methods approach to test the effects of a nurse-guided discussion in decision-making about ICD deactivation (turning off the defibrillation function) at the EOL. Interviews were conducted, audiotaped, and transcribed in 2012 to 2013 with 6 family members of patients with advanced HF and ICDs. Three researchers coded the data and identified themes in 2014.

Results:

Three main themes described family members’ experiences related to patients having HF with ICDs making health-care decision at EOL: decision-making preferences, patients’ perception on ICD deactivation, and communication methods.

Discussion:

Health-care providers need to have knowledge of patients’ decision-making preferences. Preferences for decision-making include the allowing of appropriate people to involve and encourages direct conversation with family members even when advance directives is completed. Information of ICD function and the option of deactivation need to be clearly delivered to patients and family members. Education and guidelines will facilitate the communication of the preferences of EOL care.

Introduction

Over 5 million people in the United States have heart failure (HF), and the American Heart Association¹ projects that over 8 million people will be diagnosed with HF by 2030. Despite improvements in HF diagnosis and management, mortality rates remain high, with a 5-year mortality rate of 50%.² Patients with HF are expected to gain life-extending benefits and quality-of-life improvement from an implantable cardioverter-defibrillators (ICDs). Currently, ICD is being implemented to 10 000 people every month.³ Implantable cardioverter-defibrillator is implanted for primary or secondary prevention. Primary is prophylaxis for patients without prior experience of sudden cardiac death (SCD). Secondary prevention is for those who survived SCD.^4,5

When HF progresses to end-stage disease, the frequency of arrhythmias and number of shocks from ICD may increase. The result would be that these patients with advance HF consider not only the decision of end-of-life (EOL) treatment but also the maintaining of defibrillator function in the “life-saving” ICD devices.^6–8 A recent Swedish study found that 31% of patients with ICDs received a shock in the last 24 hours of life.⁹ Many patients with advanced HF and ICD die in a hospital or hospice facility and have “do-not-resuscitate” orders in place at the time of the shock.^7,8 Consensus statements on ICD management stress the importance of thorough discussions with the patient or surrogate decision maker about ICD deactivation (turning off the defibrillation function).^5,10 However, the literature suggests that physicians do not consistently have timely discussions about ICD deactivation with patients and families.^11,12 An ICD shock can be painful and stressful for both patients and caregivers^13,14 and negatively impact the patient’s quality of life.¹⁰ One of the barriers to discussing the ICD deactivation was providers not feeling comfortable to suggest it^15,16 and patient and family hesitation to accept the suggestion of ICD deactivation.¹⁷ In addition, patients with advanced HF may lose mental capacity for decision-making and family members are not well informed by doctors to make decisions regarding ICD deactivation. As a result of these barriers, many patients and surrogates are not aware that deactivation is a possibility,¹⁷ and few studies have described family members’ experiences on decisions concerning patients with ICDs near the EOL.¹⁸ Understanding the patient and family experiences of decision-making about the ICDs at EOL can promote informed decision-making and improve quality EOL care.

Methods

Study Design and Aims

This article describes the qualitative portion of a mixed methods pilot study (Trial of Ascertaining Individual preferences for Loved Ones’ Role in End-of-life Decisions [the “TAILORED“ ICD study], N = 50). This pilot study tested the effects of an intervention of brief nurse-guided discussion regarding patient preferences for involving family in decision-making about ICD deactivation at the EOL. Interviews are conducted to further develop the “TAILORED ICD” study. This article reported the qualitative data of the study that described family members’ experiences of patients with ICD making decisions about their EOL care. A semistructured interview was used. An interview guide was developed to facilitate the sharing of experiences.

Sample and Setting

The TAILORED ICD study enrolled a convenience dyads sample of patients with advanced HF (New York Heart Association class III or IV with ICDs) and their family members. Patients were recruited from both inpatient and outpatient settings of a tertiary medical center in the mid-Atlantic region of the United States. In this institution, information about the option for ICD deactivation and/or utility is usually discussed by the cardiologist, at the discretion of the interventional cardiologist placing the device, for patients who are undergoing placement of a primary prevention ICD and have competing medical issues that may result in death such as end-stage renal disease and malignancy. The discussion usually occurs about modes of death and that sudden death is one of the modes. For those without competing medical issues, substantive discussions about ICD deactivation rarely occur. Six family members were interviewed. At the time of interview, 4 patients had passed away within 4 to 6 weeks of the interviews. The study was approved by the Institutional Review Board of Johns Hopkins University (NA_00010044) where the interviews took place. Written consents were obtained prior to research activities.

Data Collection

Interviews were conducted within 6 months of being enrolled in the TAILORED ICD study in 2012 to 2013. An interview guide with probes was developed for consistency. Interview questions focused on the perceptions of care and issues surrounding the decision-making at EOL. Examples of questions are: Who made the EOL care decisions? How were those decisions being made? Factors that influenced those decisions? Did the topic of deactivation of ICD come up? When? and how?. All interviews were conducted by the same person, a nurse researcher with an experience in conducting research on health-care decision-making at the EOL. Each interview lasted approximately 60 to 90 minutes.

Data Analysis

All interviews were digitally recorded, transcribed, and validated. After interview transcription and verification, transcripts were entered into a computer program (Nvivo 9; QRS International Ply Ltd, Victoria, Australia). Line-by-line analysis of the transcripts was guided by thematic approach.¹⁹ ML, AK, and MN independently coded each transcript. Descriptions related to the experience of EOL decisions were coded, and similar codes were combined into themes. Identified themes were discussed until a consensus was reached. Commonalities and differences among participants were reviewed. Trustworthiness of the study was assessed using the guidelines developed by Lincoln and Guba.²⁰ Truth-value: All participants were offered the opportunity to review and verify the outcome description of their discussion for clarification and credibility of the findings. Applicability: The experiences were described in detail to be useful to others in a similar situation. Consistency: The scripts and prompts were used for each interview to maintain consistency. Detail observation during the discussion was documented. Neutrality: A team approach was used to code and extract themes. Discrepancy was discussed with the team to resolve discrepancies.

Results

Five participants were white and 1 was African American. The mean age was 55 years, ranging from 18 to 84 years old. Half of the participants were women who had completed a high school-level education. Three of the women were spouses and 3 were children of the patients (Table 1). Four of the 6 patients had died at the time of the interview.

Table 1.

Family and Patient Demographic.

Patient							Family
Age	Sex	Race	Education	Ejection fraction	ICD fired	ICD placed	Age	Sex	Race	Education	Relationship
73	M	W	College	10%	1	90 mo	43	M	W	HS	Child
79	M	W	Grad. Scho	20%	none	144 mo	70	F	W	HS	Spouse
79	F	W	Grad. Scho	20%-25%	none	68 mo	84	M	W	Grad. Scho	Spouse
79	M	W	College	15%-20%	2	96 mo	50	F	W	College	Child
65	M	W	College	15%	2	120 mo	63	F	W	College	Spouse
43	F	AA	Grad. Scho	65%	2	132 mo	18	M	AA	Some HS	Child

Abbreviations: ICD, implantable cardioverter-defibrillator; M, male; W, white; mo, months that ICD has been placed; F, female; Grad. Scho, graduate school; HS, high school; AA, African American.

Three themes, discussed subsequently, emerged to describe family members’ experiences of patients with ICDs and making decisions at EOL: decision-making preferences, perception on ICD deactivation, and communication methods. Figure 1 summarizes the main themes.

Figure 1.

Family members’ experiences of patients with implantable cardioverter-defibrillators making decisions at end of life.

Family members in this study reported that patients prefer a shared decision-making style with family members or deferred the decision-making to a physician. The concept of ICD deactivation was relatively new or unheard of to patients and family members. Only 1 family member had some discussion of ICD deactivation with physician, and the patient wanted to keep the defibrillator functioning at EOL. Two different communication methods were described: direct conversation and written advance directives (AD). Family members having direct conversations with patients regarding their EOL wishes experience more confidence in patients’ decision compared to relying on a written AD only.

Decision-Making Preference

Decision-making preference refers to patients’ preferences of involving others in decision-making. Family members had varying experiences with this but primarily reported that patients prefer to have shared decision-making with family members. Some patients deferred decision-making to the physician.

Shared decision-making.

Shared decision-making is patients and family members making decisions as equal partners.

He [the patient] wanted us always all, you know, making the decisions together [patients and family members] … (, daughter, patient deceased)

Deferred decision-making.

Some family members reported that patients relied on physicians to make health-care decisions.

Well, she [the patient] had not really understood what the choice was.… And ah, he [the doctor] told her, yes, she should definitely have that [the ICD] put in.… And ah, he finally talked her into it. She wasn’t sure if she wanted it. (husband, patient deceased)

Perception of ICD Deactivation

The deactivation of ICD is turning off the defibrillation function of an ICD. Most of the family members were surprised by the idea of deactivation and reported that the possibility of deactivation of patients’ ICD had never been considered. Only 1 family member reported that the patient was aware of the option to deactivate. And with the knowledge of ICD deactivation, patients preferred to keep the ICD active at EOL.

Um, I don’t know. I’ve not thought about that. I would say that [the patient] would still be more comfortable with it [ICD] there [at the end of life] than with it gone.… (patient alive)

Um, we never had any conversations about it [ICD deactivation] … (son, patient deceased)

Because if [the patient] is having an abnormal rhythm, [the patient] needs help. So, it’ll [the ICD] help. It’ll—it’ll help [at the end of life] … (husband, patient deceased)

Yeah, that was actually a very hard decision [ICD deactivation] too. Because … it’s like pullin’ the plug … I think for families, that is a difficult thing … (daughter, patient deceased)

Communication Methods

Direct conversation.

Family members who had direct conversations with patients about EOL care experience were more confident in patients’ decisions. Two family members described a prearranged use of a simple sign or expression of the patients to indicate what they wanted about treatments even when patients cannot verbally communicate with family members so that family members would not misunderstand patients.

I had an agreement that, you know, I would let him know if there was just nothing else they could do and he would let me know if he was just tired of fighting … he gave us the thumbs up, thumbs up and a big smile and, you know, we just knew that he just wasn’t ready to stop fighting. (son, patient deceased)

Even when he couldn’t talk anymore … he was able to, you know, motion to use facial expressions to let us know what was goin’ on, what he wanted. (daughter, patient deceased)

Written AD.

An AD is a written document reflecting the patient’s preferences for EOL care. No family member reported that ICD deactivation was documented in the patient’s AD. This study found mixed results on family members’ experiences with the effectiveness of AD. One family member reported that the patients’ AD was very clear in expressing patient’s decisions and found it helpful. However, 2 family members expressed that they had not discussed the AD with the team and just assumed that medical team understands and respects patients decisions listed in AD.

The directive described the entire process from the beginning. Just, I guess, it gave me complete directions on what she would like to be done. (son, patient alive)

Um, I don’t know [what the patient decided] but, you know—he had everything written down that he wanted … (wife, patient deceased)

Well, in previous admissions, and I guess it was in the paperwork [AD] … I mean I didn’t know that much about what decisions were being made. But they [physicians] seemed to have everything under control … (, husband, patient deceased)

Discussion

Family members varied in the amount of involvement that they had over patients’ health-care decisions including the deactivation of ICD. This variability has also been found in another study.²¹ The challenges for patients who choose a shared decision-making approach in the decision-making process are that many physicians have little experience with this. One study of 1450 physicians caring for patients with advanced HF revealed that most physicians had limited conversation with patients’ family members on health-care decisions. An interprofessional team approach could be used to overcome the barriers and improve the communication.^22–24

The fact that most family members in this study were not aware of the option to deactivate the ICD or why it might be desired at the EOL is consistent with other studies²⁵ reported that 65% of patients maintained the ICD defibrillation function even when they had issued the do-not-resuscitation orders. Still other studies reported that patients with an ICD did not understand how the ICD functioned and had unrealistic expectations of ICD therapy (defibrillation function) at EOL.^21,26,27 Some patients in these studies believed that they needed the ICD therapy most at EOL when they were being shocked and would not deactivate ICD at the EOL even after being shocked multiple times. In fact, the ICD does not improve heart functioning, so shocking a failing heart may not be beneficial to patients. Current clinical guidelines recommend ongoing discussions with patients and their family members regarding treatment options, risks, and benefits of the ICD therapy.²⁸ The need for education and support among patients with ICDs was also emphasized in the scientific statement from the American Heart Association and other organizations.²⁹ Health-care providers should reinforce information about the function of the ICD and options for deactivation. An interprofessional approach should be used to initiate these conversations and provide appropriate education for the patient and family. Conversations of ICD deactivation preference should be encouraged and included in the EOL care discussion and AD documentation.³⁰

Communication is crucial in health-care decision-making. Family members valued open, honest communication patients. Family members who had direct conversation with patients about preferences for care at the EOL experience more confidence in patients’ health-care decisions. Therefore, health-care providers should encourage direct conversation even when patients have AD documentations prepared.^24,31,32

Patients with advance HF and ICDs not only have to make difficult decisions about their EOL care but they also need to consider the deactivation of their ICD device, adding complexity in the decision-making process at their EOL. Health-care providers need to have knowledge about patients’ decision-making preferences and include appropriate family members in the health-care decision-making process. Providers also need to have training to acquire the skills to communicate clearly of the function of ICD function, the option of deactivation, and risk and benefits of deactivation. Guidelines on palliative and EOL care for persons with ICD offer suggestions for wording for these conversations²⁹ and will help give providers confidence to have these discussions.³⁰ Finally, a written documentation of an AD may not replace the open, honest direct conversation of EOL decision-making discussion.³²

Limitations

The primary limitation of this pilot study was the small sample size, the reported themes should be interpreted carefully. However, the fact that themes recurred even in this small sample suggests that these themes should be considered and could be used in further studies. Another limitation is that most family members were white. Family members of other racial and ethnic groups may have different experiences with decisions at the EOL. A strength of these interviews is that they were undertaken near the EOL or soon after the patient’s death, minimizing recall bias.

Conclusion

Deactivation of ICD at EOL is a difficult decision. To support patients and family members to make informed decision, providers could utilize an interprofessional approach to improve prognostic communication, enhance understanding of ICD functions, and deactivation option. Advance directives could be an effective tool to support the decision-making at EOL. Providers should encourage patients to discuss their AD with their family members to promote the understanding of their EOL wishes. Providers play an active role in the completion of AD, and training and education in communication skills will prepare providers to be actively engaged in advance care planning discussion, and decision-making process with patients having advanced HF and their family members. This pilot study informs a larger study to explore patient and family decision-making regarding ICD deactivation at the EOL.