Background

The global incidence and mortality rate of end-stage kidney disease (ESKD) are rising, particularly among older adults [1]. In Singapore, ESKD-related deaths have increased by 44.3% between 2011 and 2020 [2], a trend expected to continue due to an ageing population, high diabetes prevalence [3], and low kidney transplant rates [4].

While dialysis is an effective first-line treatment for ESKD, evidence suggests that it can pose significant challenges and stress for patients as their quality of life and prognosis deteriorate and symptom burden increases after years of undergoing dialysis [5]. Consequently, international guidelines recommend consideration of dialysis withdrawal, often referred to as discontinuation of dialysis, and the transition to end-of-life (kidney supportive care or palliative care) care [6, 7]. This care approach prioritizes symptom management and focuses on patient-oriented quality of life [8,9,10]. However, the decision to withdraw from long-term dialysis is complex and can be overwhelming due to the ethical and emotional nature of the decision [11, 12].

While there is a growing recognition of dialysis withdrawal for patients with poor quality of life and poor prognosis, considerable disparities exist in dialysis withdrawal rates across different regions globally [10, 13]. These variations can be attributed in part to divergent attitudes towards end-of-life care, which are influenced by cultural and religious differences among societies [14]. Most existing studies and guidelines have predominantly emanated from and focused on Western countries [11], potentially rendering their recommendations unsuitable to populations with diverse cultural backgrounds [14]. For example, the medical decision-making processes in Western societies are centered around individual autonomy while in Asian cultures, medical decisions are often highly influenced or led by families [14, 15].

Furthermore, existing studies, with the notable exception of Russ et al.’s study [12], have primarily focused on perspectives of individual stakeholders in isolation (i.e. patients [16,17,18], caregivers [19, 20], or healthcare providers (HCPs) [21,22,23]), or limited dyadic perspectives (i.e. patient-HCPs [24,25,26], or patient-family [27]) while neglecting the comprehensive triadic viewpoints. In addition, previous research has primarily focused on advance care planning, reasons for stopping dialysis, or end-of-life experiences regarding dialysis patients. However, there is a notable gap in the literature investigating the decision-making process and informational needs for making decisions regarding dialysis withdrawal from the perspectives of all stakeholders involved in decision-making.

In Singapore, although legislation addresses the termination of non-beneficial treatments [28] and advances have been made in the quality of palliative care planning and provision [29], cultural factors such as the taboo surrounding death [30, 31], strong familial influence [32, 33] and paternalistic medical decision-making [34] continue to hinder patient autonomy and informed decision-making regarding dialysis withdrawal and transition to end-of-life care. This qualitative paper aims to explore the triadic perspectives of patients on dialysis, their family caregivers, and renal HCPs in order to identify their information needs and evaluate the support required for decision-making regarding dialysis withdrawal and the adoption of end-of-life care.

Methods

The qualitative study followed the Consolidated Criteria for Reporting Qualitative Studies (COREQ) [35] guidelines for reporting (Supplement 1).

Setting and participants

Participants were recruited from the National Kidney Foundation (NKF), a not-for-profit organization operating dialysis centers across Singapore, and two major public hospitals: Singapore General Hospital and National University Hospital. The study included patients on hemodialysis (HD), family caregivers of HD patients, caregivers of deceased HD patients, and renal HCPs.

Eligible patients were identified from the medical records of the HD patients (referred to as patients henceforth) at eight dialysis centres. The inclusion criteria included being 21 years or older (the age of majority in Singapore), currently undergoing dialysis, being cognitively intact (assessed by an abbreviated mental test, by a physician or from medical records), agreeable to audio recording, and meeting one of the following: a) prognosis of fewer than 12 months as determined by a treating physician at the dialysis clinic using the surprise question: Would I be surprised if this patient died in the next 12 months?, b) tolerating dialysis poorly (physically or mentally) as identified by a treating physician at the dialysis clinic, or c) expressed a desire to discontinue dialysis.

Informal caregivers of eligible patients were eligible for the study if they were aged 21 years or older and provided care, ensured the provision of care, or made decisions regarding patient care without expecting financial compensation. Spouses and adult offspring caregivers were sampled to obtain diverse perspectives. Caregivers of deceased patients within 6 months of death were also included to enrich the data with their experiences regarding dialysis withdrawal and end-of-life care.

Eligible patients and caregivers were contacted in person or over the telephone by two medical social workers. Out of 44 participants approached (34 patients and 10 caregivers), 26 (24 patients and 2 caregivers) declined participation due to lack of interest, sensitivity to the topic, or lack of time.

HCPs were recruited from three participating institutions based on the following criteria: being a nephrologist, palliative care physician, renal nurse, renal counsellor, medical social worker, or clinical psychologist currently providing care to dialysis patients. HCPs were nominated by department heads and invitation letters were sent via email. Among 18 HCPs approached, two did not respond while the rest agreed to participate.

Data collection

A semi-structured interview guide was developed based on the Ottawa Decision Support Framework (ODSF) decisional needs manual [36]. The ODSF guides researchers to assess and address decisional support needed for difficult decisions. The interview guides (Supplement 2) tailored for each stakeholder group, were reviewed by the study steering committee comprising HCPs (2 nephrologists, 2 medical social workers, 1 renal nurse, 1 renal counsellor), and 2 patient- and 1 caregiver-representatives.

In-depth interviews were conducted between February and October 2022. All interviewers were trained in conducting qualitative interviews and they had no previous relationship with the study participants. Patient and caregiver interviews were conducted in-person in English by the first author or in Chinese by two team members fluent in the language, in a quiet private room at their preferred place (dialysis centre or their homes). Patients and caregivers were interviewed separately to enable conversations to be open and candid. HCP interviews were conducted remotely over a video conferencing platform in English by the first and corresponding authors. The interviews lasted 20–60 min and were audio-taped, transcribed verbatim and (Chinese interviews) translated, and repeat interviews were not conducted. Field notes were summarized following the interview. Recruitment continued in conjunction with analysis, and data collection ended when no new information or ideas were generated.

Data analysis

Thematic analysis was conducted using the framework proposed by Braun and Clarke in nVivo11 [37]. Our deductive analysis and coding were based on the ODSF decisional needs manual [38]. The initial set of codes was derived a priori based on the ODSF operational and conceptual definitions. Two team members independently reviewed the transcripts and assigned sections of text to the pre-defined codes, and texts were mapped to the codes. Code categories were discussed during team meetings, and any discrepancies were resolved by a third team member. Themes and sub-themes were developed deductively from the ODSF framework. Periodic meetings were held until consensus on salient themes and sub-themes was reached among team members. Exemplar quotes were extracted to illustrate these themes. Data collection continued alongside data analysis until no new themes emerged. Participant review of transcripts was not included, as we had a priori codes that could potentially alter the interpretation of data.

Ethical considerations

Written informed consent was obtained from all participants. The study was approved by the National University of Singapore, Institutional Review Board (Ref no NUS-IRB-2021–749).

Results

A total of 34 individuals (10 patients, 8 caregivers, and 16 HCPs) participated in the study ranging from 31–80 years with 59% females. The participant characteristics are detailed in Table 1.

Table 1 Participant characteristics (N = 34)
Full size table

Overall, four main themes were identified. Table 2 presents the themes, subthemes, and minimally-edited verbatim extracts. Figure 1 presents a visual representation of these themes at the stakeholder level.

Table 2 Themes and subthemes with illustrative quotes on information needs and support for decision-making for dialysis withdrawal
Full size table
Fig. 1
figure 1

Decision-making needs and support for dialysis withdrawal and end-of-life care

Full size image

Theme 1: Poor knowledge and fatalistic perceptions

Perceptions of having no choice but to continue dialysis

Many patients and caregivers viewed dialysis as a lifeline, making it difficult to consider dialysis withdrawal under any circumstances. Patients often held fatalistic beliefs that they had “no choice” but to continue with dialysis “until they died”. Only a minority of patients mentioned that they would consider dialysis withdrawal due to treatment fatigue. HCPs noted that patients and caregivers often did not completely comprehend the possibility that continuing dialysis may be medically futile or untenable in the future.

Limited awareness regarding dialysis withdrawal and palliative care

Most patients and caregivers were unaware that dialysis withdrawal was a patient’s medical right and a viable option. They were unsure of available palliative care options following dialysis withdrawal. For instance, one caregiver believed that palliative care was exclusively for cancer. Consequently, some patients and caregivers feared being “left on their own”, or believed that they would suffer more if they stopped dialysis. As such, some patients associated dialysis withdrawal with “childish” thinking or “unreasonable” non-compliance. HCPs reported that most patients lacked “knowledge, awareness, and empowerment” to make decisions regarding dialysis withdrawal.

Theme 2: Inadequate support and resources for decision-making

Inadequate discussions regarding dialysis withdrawal and end-of-life care

Patients and caregivers often expressed a lack of conversations with their physicians about disease progression and end-of-life care. HCPs were perceived as “very busy” and “have no time” for in-depth discussions. Notably, almost all patients and caregivers reported that their physicians had not broached the possibility of dialysis withdrawal. Most patients reported not having advance care planning discussions, with some indicating a preference for ‘nature to take its course’.

HCPs reported several barriers to initiating these discussions, including a cultural taboo regarding death. To illustrate, a physician noted, “Singapore is a very Asian society. We do not talk much about terminal care and withdrawal of dialysis.” Furthermore, HCPs noted that discussions about end-of-life care preferences “is not something that they [patients] would volunteer” due to a "certain fear of death". They also anticipated resistance from patients and caregivers, as the concept of death was viewed as “not something they [patients and families] will accept”, rendering it as a rationale for why they “do not bring it up”. In addition, HCPs reported limited time for engagement with patients and a lack of skills in end-of-life communication.

Unclear decision-making roles: Patient autonomy vs physician dependency

Patients exhibited diverse preferences for decision-making roles regarding dialysis withdrawal. Some patients leaned towards shared decision-making, valuing discussions with their families and physicians. A subset of patients emphasized autonomy, viewing the decision as deeply personal and asserting that no one else should be “allowed to make a decision on behalf of myself”. Conversely, other patients preferred to delegate decision-making to their families, noting that “their [families] decision is much more important than my decision”. Alternatively, they entrusted their physicians with the responsibility, putting faith in their medical expertise and believing that physicians would act in their best interests. However, physicians instead preferred to provide objective medical information. They acknowledged the pivotal role of “patient’s values”, and emphasized that “patient’s decision is (the) most important” and “ultimately all medical decisions [should be] joint decision making”.

Caregiver dominance in decision-making

HCPs noted the significant influence caregivers have on patients, a dynamic that occasionally undermined patient autonomy. Some caregivers pressured patients to continue dialysis, noting that “we had to advise her [patient]; life had to go on”, even if this conflicts with patients' own desires. While caregivers have good intentions, HCPs noted that they often lacked understanding of patients’ fatigue and suffering on dialysis. This was articulated by an HCP’s observation that “a lot of times, even though they [families] decide for the good of the patient, that decision may not be what the patient actually wanted”.

Furthermore, some caregivers intentionally avoided discussing medical matters with patients to prevent them from feeling “hopeless and depressed” about their condition. Consequently, some patients felt more comfortable discussing dialysis withdrawal with HCPs rather than their caregivers. To navigate these communication barriers, HCPs sometimes assume the role of mediators between patients and caregivers. In particular, they felt compelled to advise caregivers that “we can only do our best to relay messages, but we hope that you can also listen to what your loved one has to say”.

Emotional support for decision-making

To support decision-making regarding dialysis withdrawal, HCPs emphasized the need to reassure patients and caregivers that, medical teams would “not abandon if they choose to stop dialysis” and continue to support patients in “passing on peacefully”. These include providing emotional assistance throughout patients’ EOL care, and preparing families for their loved ones’ imminent death, including bereavement support to cope with grief. A caregiver voiced that receiving such reassurance early on helped her loved one become “more receptive” towards dialysis withdrawal.

Instrumental, health, and social services for supporting end-of-life care

Patients desired information and assurances regarding symptom management and “about the support for [their] care” after dialysis withdrawal. HCPs stressed the importance of providing information about community resources, financial assistance schemes, and comprehensive care plans that meet patients’ and caregivers’ needs. HCPs also emphasized preparing caregivers to help them handle the practical and emotional demands of EOL care at home. A caregiver of a deceased patient recounted that having her loved one die at home would have been a “very terrifying” experience. She was ultimately grateful that her loved one passed away peacefully in a hospice.

Theme 3: complexity of decision-making, unclear timing, and unpreparedness

Dialysis withdrawal is a difficult decision

Another key theme was the ethicality and difficulty of deciding on dialysis withdrawal. Patients expressed that dialysis prolonged their life and the thought of dialysis withdrawal disturbed them due to the imminence of death. Caregivers often viewed dialysis withdrawal as “giving up” and one caregiver even likened it to "suicide". Many HCPs also acknowledged that dialysis withdrawal can be perceived as "choosing to die", "suicide", or "euthanasia", and added that discussing dialysis withdrawal is challenging due to these perceptions.

Unclear decisional timing

Although HCPs agreed that decision-making regarding dialysis withdrawal should be a "step-wise procedure", involving multiple discussions, they seemed to disagree about the timing of initiating these discussions. While some HCPs thought these discussions should start earlier around dialysis initiation, others recommended that the "optimal time is when medical problems or recurring issues arise". HCPs narrated cases of patients having suicidal thoughts, such as "wanting to die", and "to stop dialysis and walk away from home". However, HCPs also emphasized that these thoughts may arise in the "fit of the moment" or as a way of "venting frustrations about the dialysis process". They viewed these thoughts as cues to further investigate whether dialysis withdrawal should be considered for the patient. HCPs expressed that palliative care teams should be included to "ease the patient into the discussions" and "explore psychosocial aspects" to ensure that dialysis withdrawal is a "well-considered decision".

Unreceptive decisional stage and denial

Patients generally showed resistance to discussing dialysis withdrawal. They reflected that they had not considered dialysis withdrawal and would only consider it as a last resort. Patients stated that "life was normal on dialysis", and there were "no other ways to change a life". Some HCPs also observed that patients often become comfortable with the dialysis routine over time and are reluctant to stop it even when it is no longer beneficial. HCPs highlighted that patients and caregivers are often not in a mindset to consider dialysis withdrawal, perceiving it as a far-off decision. One bereaved caregiver mentioned that her loved one was in denial and reluctant to consider dialysis withdrawal even after her physicians recommended it.

Theme 4: internal emotions of decisional conflict and regret

Decisional conflict

Caregivers of deceased patients expressed being "unsure" or "afraid" when deciding whether to withdraw patients from dialysis, questioning the rightness or wrongness of their decisions. Caregivers whose loved ones ultimately underwent dialysis withdrawal reported that while it was a difficult and "cruel" choice, stopping dialysis turned out to be a "practical" choice that "relieved suffering". Physicians described the "guilt" that caregivers expressed when faced with a dialysis withdrawal decision, as they felt responsible for "hastening the death of their loved one".

Regret for a delayed decision to withdraw dialysis

Caregivers of deceased patients expressed feeling "selfish" for not fully considering their loved one’s perspective and regretting not making the dialysis withdrawal decision earlier. This internal conflict was exemplified by a caregiver’s reflection: “Did we make a mistake, should we have stopped her dialysis earlier”. Such introspections arose from a realisation that an earlier withdrawal might have reduced their loved one’s suffering. All caregivers of the deceased shared a dual emotional response following their loved one’s passing, marked by a feeling of sorrow in the face of a loss and a sense of relief that their loved one’s suffering had finally come to an end.

Discussion

This study examined the triadic perspectives of patients on dialysis, their caregivers, and renal HCPs, with a specific focus on decision-making concerning dialysis withdrawal and the subsequent transition to end-of-life care in Singapore. The findings revealed that patients and caregivers had a limited understanding of dialysis withdrawal and palliative care. Notably, dialysis withdrawal was associated with feelings of abandonment, lack of support, and increased suffering. These findings are consistent with the previous studies showing that patients are often unaware of dialysis withdrawal [39] and available end-of-life care options [18], or perceive dialysis withdrawal as a form of medical abandonment [5, 40].

The complexity of decision-making regarding dialysis withdrawal was influenced by existential, ethical, and familial factors, hindering open discussions and informed decision-making. In parallel to extant findings [5, 11], patients and caregivers struggled to consider dialysis withdrawal due to the imminence of death following withdrawal. Instead of viewing it as a means to relieve suffering, dialysis withdrawal was often dismissed as “giving up” or even akin to a form of suicide. While some studies showed that patients may express interest in dialysis withdrawal due to declining quality of life [12, 18, 41], most patients in our study had become accustomed to their dialysis routines even though some expressed treatment fatigue and suicidal thoughts. Consequently, the idea of contemplating end-of-life care remained a distant consideration, denying the necessity to discuss it until continuing dialysis would become medically untenable.

This denial was reinforced by the lack of end-of-life conversations between patients and their clinicians and/or caregivers. Although some patients preferred a paternalistic approach to their care, HCPs hesitated to have these conversations due to cultural taboos regarding death, anticipated resistance from patients and caregivers, insufficient time, and lack of skills in end-of-life communication. This finding resonates with other local studies with renal HCPs showing low frequencies of discussions regarding advance care planning [42] and inadequate palliative care training [43]. HCPs also raised concerns about caregivers not being able to fully comprehend patients’ treatment burden, and sometimes over-riding patient preferences to prolong their lives. Furthermore, some caregivers avoided discussing medical issues with patients to prevent negative emotions. Consequently, these cultural and familial barriers, along with the lack of end-of-life discussions from HCPs, impeded informed decision-making.

Interviews with bereaved caregivers further revealed the complex nature of the decision. These interviews illuminated a range of emotions, including feelings of guilt stemming from not knowing how to proceed, regret for delaying dialysis withdrawal, and eventual relief following a peaceful death.

Overall, the research findings suggest that discussions and decision-making surrounding dialysis withdrawal and end-of-life care are primarily shaped by familial dynamics (e.g., dominant family role in decision-making), and cultural factors (e.g., taboo nature of discussing end-of-life issues), rather than individual factors (e.g., socio-economic status, age). In addition, institutional factors, such as lack of adequate time and training in communication skills for handling, emerged as significant factors contributing to the study outcomes.

Our findings unveil a complex web of emotional barriers among all stakeholders, hindering meaningful discussions and adequate support for end-of-life care and dialysis withdrawal. Patients reported grappling with the fear of abandonment and death, caregivers described the daunting prospect of being unable to meet their loved ones’ end-of-life care needs, while clinicians expressed concerns about causing distress or hastening death. These fears fuel a collective avoidance of conversations surrounding end-of-life care and dialysis withdrawal, possibly exacerbated by patients resorting to denial as a coping mechanism. Recognizing the profound effect of emotional challenges, we suggest interventions such as emotional counseling and support groups facilitated by allied health professionals. These initiatives can be used to address and alleviate the fears and anxieties experienced by patients and caregivers, fostering a more informed and emotionally prepared decision-making process.

Our findings underscore the imperative for HCPs to adopt a strategic approach that centers on delivering clear, compassionate, and culturally-sensitive information about dialysis withdrawal, especially the potential benefits and drawbacks of continuing versus stopping dialysis. To counteract the prevalent misconception that dialysis withdrawal equates to medical abandonment, it is crucial for HCPs to offer clarification and inform about unwavering support that would be available throughout palliative and end-of-life care. Given the ethical and emotional complexities surrounding dialysis withdrawal and end-of-life decisions, it is essential for HCPs to receive training in undertaking serious illness conversations and guidelines that facilitate better identification of patients who could benefit from these discussions. Building skills in effective communication, cultural competence, and addressing existential concerns can equip providers to guide patients and caregivers through these challenging decisions. HCPs should also be provided with guidelines to assess when patients and their families are ready to have these conversations.

Future research can focus on identifying the specific types of training for HCPs that are most effective in facilitating meaningful and effective end-of-life conversations. In addition, identifying patient preparedness for conversations centred around advance care planning in the context of dialysis withdrawal and the transition to end-of-life care is another avenue for investigation. This understanding could potentially lead to more personalized and effective communication strategies tailored to each patient’s unique emotional state and circumstances. These tools and training materials can all be part of a decision aid that can be developed to facilitate congruous and shared decision-making regarding dialysis withdrawal.

The study findings should be interpreted within the context of the study limitations. Findings derived from qualitative research are by nature prone to a degree of potential subjectivity. Despite efforts to engage a wide sample of stakeholders, we did not include patients on peritoneal dialysis and patients dialyzed at hospitals, who may have different perspectives and decisional needs.

The study has notable strengths. First, there is limited empirical research on the perspectives of multiple stakeholders on this topic. By including patients on dialysis, caregivers, and renal HCPs, as well as purposively sampling bereaved caregivers, we obtained a comprehensive understanding of the decision-making needs for dialysis withdrawal and experiences of EOL care. Second, the qualitative approach allowed us to examine the lived experiences and complexities associated with dialysis withdrawal decision-making[44]. Third, using the ‘surprise question’ helped identify a selective population of patients with poor prognoses who may face a decision to withdraw from dialysis in the near future. Last, we ensured study rigor in the methodology and reporting adhering to the COREQ guidelines. Additionally, incorporating a sound framework to guide systematic data collection and analysis enabled a more precise identification of decisional needs.

Conclusions

Withdrawing from dialysis was viewed as a difficult decision due to the imminence of death and reluctance towards discussing end-of-life care. The lack of information and limited conversations regarding dialysis withdrawal and end-of-life care, coupled with disagreements regarding decision-making roles contributed to the complexity. While educational materials on dialysis withdrawal have been developed and utilized in Western contexts, there currently exists no decision aids specifically tailored for dialysis withdrawal. Our study highlights the importance of developing a culturally-sensitive educational and decision aid to prepare patients and their caregivers for dialysis withdrawal and transition to end-of-life care. These tools can also empower HCPs to initiate challenging yet crucial end-of-life conversations. The ultimate goal is to foster shared and informed decision-making, aligned with patient values, and sensitive to cultural contexts.