Good death and end-of-life care preferences: perspectives from older Singaporeans Authors: Hyo Jung Lee, Inhye Jung, Shiwei Chen, Miriam Sang-Ah Park, Sabrina Ching Yuen Luk Published: 2026 ABSTRACT BackgroundReflecting on the rapidly ageing population with a growing need for end-of-life (EOL) care and ethnic diversity in Singapore, it is important to understand culturally specific views on the meaning of good death and end-of-life (EOL) care plans and preferences.MethodsThis study explores older Singaporeans’ perspectives on the meaning of good death and their end-of-life (EOL) care plans and preferences, by conducting 18 in-depth interviews with older Singaporeans. The data was analysed using an iterative thematic analysis involving independent coding followed by iterative discussion and refinement of themes.ResultsWe found four attributes of good death, including being free of pain or suffering, not being a burden on family members, accepting mortality, and fulfilling parental and spousal responsibility. As regards EOL care plans and preferences, we found that most of the older adults interviewed never had an open and detailed EOL conversation with their family members. Four barriers to EOL care planning or preference discussions were identified: the degree of readiness, fear of making them unhappy, leaving it up to family, and interpersonal and social influence.ConclusionsTo encourage older adults to make EOL care plan while they are healthy and competent, we suggest that the government promote professionally appropriate and family-engaged EOL care services tailored towards individuals, and open and ongoing conversations between older adults and their family members.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12904-026-02162-5. Introduction The demand for end-of-life (EOL) care is projected to increase as a result of the growth of ageing populations and the rising burden of chronic illnesses and disability [1]. Studies show that conversations about treatment preferences, including those involving family members, can improve the quality of life (QoL) of patients at EOL [2]. Nevertheless, initiating or having an EOL care conversation can be a challenging task for healthcare professionals [3]. Likewise, many older adults do not know how to initiate an EOL conversation while they still can, or feel nervous about having such conversations with loved ones or healthcare professionals [4]. Hence, experiencing a good death remains more of a hope than the standard of care for most older adults [5]. Currently, ‘there is little agreement as to what specifically constitutes a good death or successful dying’ (p. 262) [6]. More importantly, the difference among cultures seems to be even less discussed. Given the cultural diversity of its multi-ethnic population, Singapore faces a growing demand for EOL care as its population is ageing rapidly. Therefore, it is timely and appropriate to examine the meaning of good death and EOL care plans and preferences from the perspective of older Singaporeans. Findings from this study contribute to a better understanding of older Singaporeans’ perspectives on the meaning of a good death and may inform future research, policy, and practice related to Advance Care Planning (ACP), EOL decision-making, and family conversations about care preferences. Such efforts may ultimately support improvements in older adults’ quality of life and quality of death. Literature review Good death in older adults In practice, older adults’ willingness to talk about death and dying can lead to a greater likelihood of experiencing a good death through forward planning and informed EOL choices (p. 177) [7]. Studies indicate that older adults’ willingness to discuss death is affected by various factors, including personal experiences with loss, health conditions, and the presence of social networks [2, 8]. Besides, culture can shape people’s perceptions and attitudes towards death, influencing the degree of openness and feelings around both death and discussions of it. In traditional Chinese culture, for example, the subject of death is often considered taboo. Initiating a conversation about death is disrespectful and engaging in such conversation would bring bad luck to oneself and family [9]. In such a culture, the discussion of death with other people is discouraged. A ‘good death’ is believed to help alleviate distress for individuals at the EOL and those who cared for them [10]. Research indicates that the attributes of a good death are multifaceted, including a sense of control over the dying process, a comfortable and appropriate feeling, leaving a legacy and affirming the value of the dying person, being with family care, while reducing the burden on family members and optimising relationships with friends and family members, and a trusted relationship with care provider (pp. 281–282) [11]. Research also emphasizes a peaceful moment of death surrounded by family members and having their personal wishes respected with regard to EOL care and funeral matters to be perceived as a good death [12]. These findings suggest that a good death is complex in nature, encompassing physical, psychological, social, and existential dimensions. Importantly, the perception of a good death is affected by one’s own cultural background and personal life experiences [13], making it a highly personalized and subjective concept. Recent studies conducted in Asian countries highlight the importance of having a non-suffering death, maintaining dignity and respect, active involvement in the dying process, and not burdening family members [14, 15]. In Singapore, Chan and Yau [16] found that death being perceived as a societal taboo led to few Chinese older adults initiating discussion about this topic. The study further revealed that ‘lesser financial assets and a strong belief in filial piety’ (p. 225) [16] especially hindered them from engaging in such discussions and led them to perceiving death preparation as unnecessary. Collectively, these studies highlight the diverse cultural, social, and personal factors that shape older adults’ perceptions of a good death across different Asian contexts, which underscores the need for culturally sensitive approaches to EOL care discussions. Older adults’ EOL care planning and preferences Not all older adults who have a proactive attitude towards death have made concrete EOL plans [17]. The gap between attitude and behaviour underscores the complexity of EOL planning. There are several ways to plan for one’s EOL, including ACP, Lasting Power of Attorney (LPA), and Advance Medical Directive (AMD). Some older adults take proactive steps to discuss EOL care preferences with their family members, which give them a sense of relief and a sense of security [8]. However, limited awareness and knowledge stops some from proactive planning [18]. Therefore, it is important to examine older adults’ current status and involvement in the EOL care process, as well as the underlying reasons behind their choices and behaviours, in order to gain a better understanding of their death preparation. To date, there are only two studies examining older adults’ views on EOL care in Singapore. A study by Malhotra et al. [19] examined the views of 63 middle-aged and older Singaporeans on good EOL care, identified eight components of good EOL care (e.g. having physical comfort, avoiding burden on family) and called for better pain management at the EOL, better training of doctors and caregivers, and the availability of affordable EOL options. Another study by Foo et al. [20] found that there were significant discrepancies between terminally ill older patients’ and their named surrogates’ EOL decisions, and it called for the use of AMD to safeguard patients’ autonomy in making EOL medical decisions. Given the rapidly growing number of older adults in Singapore, suffering caused by ill health in the aged population would consequently lead to a growing demand for EOL care in the country [21]. Since a review of literature shows that there is limited knowledge regarding older adults’ perceptions of good death and their views on EOL care plans or preferences in Singapore, this study aims to fill the research gap by answering the following questions: What is the meaning of good death from older adults’ perspective? What are older adults’ EOL care plans or preferences? Methods This study examines the views of older Singaporeans on the meaning of a good death and their EOL care planning or preferences. Data were collected through semi-structured, in-depth interviews with older Singaporeans. Ethical approval for this study was obtained from the Institutional Review Board of Nanyang Technological University, Singapore (Ref. No.: IRB-2019-09-027) on October 25, 2019. Data collection Participants were recruited through convenience sampling and snowball sampling using the following inclusion criteria: Singaporeans aged 60 years and above [22]; having no severe physical and mental illness that would prevent a one-hour conversation; and having at least one adult child living in Singapore. Participants were recruited through community flyers and snowball referrals. Interviews were conducted in private settings, either face-to-face or online, without the presence of others. No prior relationship existed between interviewers and participants. Participants received SGD 40 as a token of appreciation. A total of 18 older Singaporeans, after reading the study information sheet, gave informed consent to take part in this study. Confidentiality, anonymity, and data protection were assured for all participants. Three interviewers were involved in data collection. The principal investigator (HL, female) and one interviewer (male) had doctoral training and prior experience in qualitative research. The second interviewer (female) was a local researcher with professional training and experience in journalism and prior qualitative research experience at the master’s level. Data collection and preliminary analysis were conducted concurrently, and recruitment concluded when no substantially new themes emerged from successive interviews. A total of 12 face-to-face interviews and six online interviews were conducted between January 2020 and September 2021. Interviews were conducted in private settings, either in participants’ homes or via online platforms. Privacy was ensured by confirming that no other individuals were present and that conversations could not be overheard. A total of 15 interviews were conducted in English while the remaining three were conducted in Chinese. Each interview had an average duration of 45 min. All interviews were audio-recorded and professionally transcribed (or translated) into English by an external transcription service. For the three interviews conducted in Chinese, the interviewer reviewed the English translations for accuracy, and the coders cross-checked key transcripts to clarify nuances. Data analysis To systematically organize and analyse the qualitative data obtained from the participants, we employed thematic analysis [23]. The specific procedure [24] was as follows: three coders (HL, IJ, SC) independently read the transcripts and performed open coding, allowing them to interpret, analyse, and identify thematic patterns within the semantics of participants’ expressions. NVivo was used during the initial stage of coding to support data organization and facilitate the identification of emerging codes across interviews, while subsequent analysis was conducted manually through iterative comparison and collaborative refinement of themes. After reviewing the first few transcripts, the coders discussed their initial codes and collaboratively developed a coding scheme. Analysis continued until a consensus was reached. Meaningful passages were systematically organized to allow comparison across interviews and refinement of codes. Themes were iteratively mapped and documented throughout the analytic process. Subsequently, an independent coder (MP) reviewed the transcripts using the agreed-upon themes to confirm that they accurately captured the participants’ perspectives. Through this process, the researchers ultimately defined the characteristics of the participants’ views and mapped Singaporean older adults’ perceived concepts of a good death and their EOL-related factors. The research team had prior training and experience in qualitative research, and regular discussions were conducted to reflect on potential biases and to ensure that interpretations remained grounded in the data. Results Table 1 displays the demographic characteristics of all participants. The majority were married (72.2%) and of Chinese ethnicity (72.2%). About 44 per cent of the participants were aged between 60 and 64 years old, with the remaining aged between 65 and 69 years old (28%) or aged 70 years and above (28%). Half the participants were male. In terms of education, about 28% completed primary or secondary education, 22% had O-level or Polytechnic qualifications, and half the participants (50%) attended college. About 33 per cent of participants identified as Christian, with the remaining identified as Buddhist (17%), Hindu (11%), Muslim/Islam (11%), Catholic (11%), or as free thinker or having no religious affinity (17%). Eleven participants (61%) reported having experienced the recent loss of a family member or close friend. To ensure anonymity, we assigned the six-figure identifier to each participant. For example, T01_M_6064 means that interviewee number 1 is a male aged between 60 and 64 years. Table 1Participants’ demographic characteristics (N = 18)MenWomenTotalAge: 60–64 years old628 65–69 years old235 70 + years old145Marital status: Married8513 Divorced123 Widowed022Race/Ethnicity: Chinese5813 Indian202 Malay101 Other112Education: Primary or Secondary134 O-level or Polytechnic235 College639Religious affiliation: Hindu202 Muslim/Islam202 Catholic022 Christian246 Buddhist123 Free thinker/None213Recent loss of a family member or close friend4711Total9918 Table 1Participants’ demographic characteristics (N = 18)MenWomenTotalAge: 60–64 years old628 65–69 years old235 70 + years old145Marital status: Married8513 Divorced123 Widowed022Race/Ethnicity: Chinese5813 Indian202 Malay101 Other112Education: Primary or Secondary134 O-level or Polytechnic235 College639Religious affiliation: Hindu202 Muslim/Islam202 Catholic022 Christian246 Buddhist123 Free thinker/None213Recent loss of a family member or close friend4711Total9918 Participants’ demographic characteristics (N = 18) Table 2 outlines two major themes with relevant subthemes emerged from interviews. The first theme was the Meaning of Good Death, consisting of four key attributes of dying well. The second theme was the barriers (or facilitators) to EOL care planning or preference discussions, based on four distinct subgroups which varied in terms of their thoughts on EOL care planning, EOL conversations with family and their EOL plans. Table 2Themes and sub-themes emerged from interviewsThemesSub-themesThe meaning of good deathFree of pain or sufferingNot being a burden on family membersAccepting mortalityThe fulfilment of parental and spousal responsibilityThe barriers (or facilitators) to End-of-life (EOL) care planning or preference discussionsThe degree of readiness - it is too earlyThe fear of making them unhappyLeaving it up to familyInterpersonal and social influence Table 2Themes and sub-themes emerged from interviewsThemesSub-themesThe meaning of good deathFree of pain or sufferingNot being a burden on family membersAccepting mortalityThe fulfilment of parental and spousal responsibilityThe barriers (or facilitators) to End-of-life (EOL) care planning or preference discussionsThe degree of readiness - it is too earlyThe fear of making them unhappyLeaving it up to familyInterpersonal and social influence Themes and sub-themes emerged from interviews The meaning of good death We identified four attributes of dying well: being free of pain or suffering; not being a burden on family members; accepting mortality; and fulfilling parental and spousal responsibilities. First, many participants thought that good death was having an easy and smooth dying process without pain or suffering. Good death meant ‘no struggle, no pain’ (T3_W_6064), dying in one’s sleep without any illness (T2_M_6064; T14_M_6064), or a quick and painless death (T12_M_6064). Second, all participants said that a good death meant not being a burden on their spouse, children and other family members. For example, having a quick death could free a person from becoming a burden to his/her loved ones (T12_M_6064). As reflected in the following excerpt, such a burden referred to both caregiving and financial burden. Number one, you should not be a burden on anyone when you go … You should not make anyone financially … hard. … Number two, the caregiving should not be very exhausting. (T15_M_6064) Number one, you should not be a burden on anyone when you go … You should not make anyone financially … hard. … Number two, the caregiving should not be very exhausting. (T15_M_6064) Third, good death meant accepting mortality. In our data, accepting mortality was multi-dimensional, encompassing cognitive acceptance of death as inevitable, a sense of psychological life completion (e.g., absence of regret, resolved conflicts), and, for some participants, spiritual or religious surrender. These dimensions were interconnected but varied in emphasis across participants, reflecting their life experiences. Death was often understood as a natural and unavoidable event or part of life, with a common sentiment being that once the time comes, it comes anyway – nothing can change it, or it does not matter in the broader context of the universe. As one participant said, death ‘is a process in everyone’s life, it will happen to everyone, and no one can escape it’ (T05_M_6569). People should ‘just let nature take its course’ (T7_M_7074). Accepting death (i.e. ‘When it is time to die, I will die’ (T18_W_7579)) could lead to people making most of the time they had and having a happier and more satisfying life. Hence, some participants said that good death meant having the opportunity to do things they wanted to do in their lifetime (T3_W_6064), resolving feelings of hate (T10_W_6569), having ‘no regrets or worries’ (T17_M_7074), no unresolved conflicts or issues (T3_W_6064; T14_M_6064), and ‘meeting or fulfilling wishes at the time of death or dying’ (T10_W_6569). Having mental preparation for death, accepting mortality, led to embracing death without fear and viewing death as another adventure or phase of life. For instance, one participant showed mental preparedness by saying ‘(when) time is up and you just know that any moment you can go, and you are not frightened’ (T3_W_6064). Another participant mentioned the idea of having ‘freedom,’ being liberated from her own body (T10_W_6569). This acceptance of death could be affected by one’s religiosity, which often includes a degree of religious beliefs and everyday practices. As reflected in the excerpt below, religious beliefs such as those in Muslim and Christian, played a key role in helping people accept his/her fate and believe that death is determined by God. This led to people having a sense of comfort and inner peace. good death also means being well prepared psychologically and spiritually for your death. As Muslims we believe that before you die, you should repent all your sins and do good to others… as Muslims, we really believe in pre-destination. So, your date of death [or] birth has all [been] determined by God. …you have to accept your fate. (T1_M_6064) good death also means being well prepared psychologically and spiritually for your death. As Muslims we believe that before you die, you should repent all your sins and do good to others… as Muslims, we really believe in pre-destination. So, your date of death [or] birth has all [been] determined by God. …you have to accept your fate. (T1_M_6064) When the Lord wants to get you, you just surrender and go. …I never thought that there’s a choice. (T9_W_6569) When the Lord wants to get you, you just surrender and go. …I never thought that there’s a choice. (T9_W_6569) Fourth, good death meant having fulfilled the parental and spousal responsibilities. Participants said that raising children to become independent and well-educated, pursue a good career path, and have their own family indicated their achievement as parents. Hence, they would not have to worry about their children. Some participants said that it was important to ensure that their spouse was financially secure and healthy and could carry on after their death. The idea that I die is quite fine… Because literally after my son went to NS [National Service], I’m like, [sigh] okay, I can die now. (T10_W_6569) The idea that I die is quite fine… Because literally after my son went to NS [National Service], I’m like, [sigh] okay, I can die now. (T10_W_6569) Upon your death when your children or your wife have to continue paying for the house, that is a big burden, especially if your wife [is] not working and then you expect your son to contribute to pay for the house. Or if you owe something, your family have to pay [for you], that is a burden. For me, I don’t have that thing. …my house [has] already [been] paid off, … and my children are grown up. They all have good jobs, … all [are] in good health. So I don’t have to feel worried about them. (T2_M_6064) Upon your death when your children or your wife have to continue paying for the house, that is a big burden, especially if your wife [is] not working and then you expect your son to contribute to pay for the house. Or if you owe something, your family have to pay [for you], that is a burden. For me, I don’t have that thing. …my house [has] already [been] paid off, … and my children are grown up. They all have good jobs, … all [are] in good health. So I don’t have to feel worried about them. (T2_M_6064) The barriers (or facilitators) to EOL care planning or preference discussions We found that there was some level of dissonance between what participants described as a good death and their degree of action. We first classified participants into four distinct subgroups which varied in terms of their engagements with, or thoughts on, EOL care planning, EOL conversations with family, and their EOL plans. These four subgroups were: the ‘Never thought of it’ group; the ‘Have thought of but never talked about it’ group; the ‘Have talked about it, but not documented it’ group; and the ‘Thought of it, talked about it and documented it’ group. These subgroup classifications were informed by prior literature on levels of engagement in EOL planning [18] and derived from participants’ responses to whether they had considered, discussed, or documented their EOL care preferences. They illustrate not only whether older adults hold EOL care preferences, but also how such preferences are communicated, postponed, or documented in practice. First, participants in the ‘Never thought of it’ group referred to those who never thought of EOL care planning. Second, participants in the ‘Have thought of but never talked about it’ group referred to those who thought of EOL planning, but had yet to discuss it openly with family or friends. Third, participants in the ‘Have talked about it, but not documented it’ group referred to those who had EOL conversations with family and friends but had yet to document their EOL care plans. Fourth, participants in the ‘Thought of it, talked about it and documented it’ group referred to those who thoroughly thought of EOL care, had EOL conversations with the family or friends, and had documented their EOL care preferences. The different degree of actions were affected by various factors, including individual beliefs or readiness, siblings or friends who had EOL care plans, family members’ attitude towards EOL care plans, and one’s knowledge of EOL care options. Among these factors, we identified four barriers (or facilitators) to EOL care planning or preference discussions: the ‘degree of readiness - it is too early’; the ‘fear of making them unhappy’; ‘leaving it up to family’; and ‘interpersonal and social influence.’ First, the ‘degree of readiness - it is too early’ for EOL care planning or preference discussions was a key determinant of inaction. In our study, many participants, especially those in their 60s with no critical health conditions, reported that they did not have EOL care planning in mind. They thought that it was too early to talk about it, not the right time to think about it and there was no reason to think about it now (T9_W_6569; T11_M_6064; T13_M_6569). As a result, they were unaware of EOL options available such as the LPA (T18_W_7579) and thought that EOL care planning was unnecessary until a critical health issue arose. Some participants in the ‘Have thought of but never talked about it’ group had thought about managing and distributing their property, Central Provident Fund (CPF) savings, money and other assets after death (T1_M_6064; T2_M_6064; T5_M_6569; T12_M6064). However, they had not given much thought to EOL care or choosing proxy decision-makers because they thought that it was ‘too early to talk about death’ (T2_M_6064) and there was ‘no reason to talk about it now’ when they were currently healthy (T12_M_6064). Some participants did not have an EOL conversation with family because of not being mentally prepared for it (T4_W_7579). They also said that, although nothing was discussed between them and their family members (T1_M_6064; T2_M_6064; T04_W_7579; T12_M_6064), they would have EOL conversations with the family when the time came and they were ready to talk about it (T12_M_6064). Second, the ‘fear of making them unhappy’ serves as a barrier to EOL planning or preference discussions. Participants in the ‘Have thought of but never talked about it’ group had the perception that it could be “a very serious talk” (T3_W_6064) and the fear of bringing discomfort to family members (T13_M_6569) or disclosing too much personal and confidential information to family members (T1_M_6064). Therefore, using humor or a casual tone helped ease discussions about death and dying. Some participants in the ‘Have talked about it, but not documented it’ group said that they joked about death and the EOL conversation with the family and friends were carried out in a non-serious manner (T3_W_6064; T15_M_6064). Besides, some participants in the sample group felt more comfortable having EOL conservations with friends rather than family members for fear of making their loved ones worry or burdening their loved ones with anticipated responsibilities. One participant worried that her young children had been traumatized by the EOL conversations she had with them (T10_W_6569). While participants in the ‘Thought of it, talked about it and documented it’ group made considerable efforts to plan for their EOL care, they still faced challenges in fully engaging their families in these conversations, often balancing the need for death preparation with concerns about causing discomfort or misunderstandings among their loved ones. One participant in this group also shared a bidirectional perspective on having EOL conversations with family members. EOL conversations may make family members feel uncomfortable or family members may not take EOL conversations seriously or fully understand her intention (T8_W_6064). I had a conversation with my husband. What are you gonna do when I’m dead? It didn’t go very far. I don’t think he was very comfortable. I remember talking about this with the kids and then my son rolled his eyes, there goes mama again. But I feel that it needs to be talked about. (T8_W_6064) I had a conversation with my husband. What are you gonna do when I’m dead? It didn’t go very far. I don’t think he was very comfortable. I remember talking about this with the kids and then my son rolled his eyes, there goes mama again. But I feel that it needs to be talked about. (T8_W_6064) Third, ‘leaving it up to family’ acts as a barrier to EOL planning or preference discussions in two distinctive ways: firstly, some participants who did not have EOL conversation with family members believed that the spouse and children understood what they wanted (T4_W_7579) and would make the best EOL decisions for them. I think they are big enough they can make decision for me… So I have good faith in my children. (T02_M_6064) I think they are big enough they can make decision for me… So I have good faith in my children. (T02_M_6064) In addition, despite having a detailed EOL care plan, participants still preferred letting their family members make EOL decisions for them (T8_W_6064; T14_M_6064). On the other hand, some in the ‘Have talked about it, but not documented it’ group avoided providing family members with detailed instructions about what they wanted at EOL because of the fear that such instructions could be a burden to the family (T15_M_6064). Similarly, one participant in the ‘Thought of it, talked about it and documented it’ group also stated that while the conversation could help her and her loved ones better prepare for her death, she worried that the provision of detailed instructions about EOL care might burden her loved ones (T8_W_6064). Fourth, ‘interpersonal and social influence’ on EOL planning or conversations is the last barrier (or facilitator) identified. Recognizing that their siblings, relatives and friends had started EOL care planning motivated participants to initiate EOL conversations with friends and family members. However, such conversations often focused more on financial matters (e.g. property distributions to loved ones) or funeral arrangements, and EOL care preferences had not yet been thoroughly discussed. But I did tell my husband, don’t waste much money on funeral thing. Dying is dying, I don’t want you to spend so much money on [it]… Simple one would do. But prayer is important (T3_W_6064). But I did tell my husband, don’t waste much money on funeral thing. Dying is dying, I don’t want you to spend so much money on [it]… Simple one would do. But prayer is important (T3_W_6064). The participants in the ‘Thought of it, talked about it and documented it’ group particularly mentioned that knowing that their siblings or peers were making EOL care plans, spousal encouragement, government promotion, and attending local events which explained and promoted EOL care or ACP options motivated them to make their EOL care plan and to have the conversations with others. Discussion In this study, we explored the meaning of good death from the perspective of older adults in Singapore. The findings reveal four attributes of good death: being free of pain or suffering, not being a burden on family members, accepting mortality, and fulfilling parental and spousal responsibility. Pain and unbearable suffering can lead to the loss of one’s ability to ‘do basic things that are considered to bestow meaning on one’s life’ (p. 335) [25] and to the loss of one’s identity and dignity [25]. The first attribute of good death, free of pain or suffering, indicates that effective pain control is an important measure to improve QoL of older patients at EOL and ‘an essential element of a good death’ (p. 824) [26]. We suggest EOL care focus on effective assessment of pain (e.g. location, duration, frequency, and intensity of pain) and both acute and chronic pain control through pharmacological measures, nonpharmacological measures (e.g. acupuncture), and psychosocial measures (e.g. art and music therapy, mindfulness-based techniques) [27]. The second attribute that older adults do not want to be a burden on their family members indicates that they place a high value on family relationships and have a desire to avoid bringing inconvenience, additional stress, or suffering to their family members due to unsettled matters, such as finances or care [19]. The third attribute of good death, accepting mortality, indicates that older adults view death ‘as a reality that is neither feared nor welcomed’ (p. 128) [28]. Death acceptance could set older adults free from death anxiety and energize them to live a meaningful and abundant life [29]. Thus, ‘living well and dying well are interconnected in important ways because of the intimate relationships between the meanings of life and the meaning of death’ (p. 69) [29]. This attribute of good death also indicates that older adults accept death with faith rather than fear. For some participants, this appears to be connected to their religiosity, leading to the belief that the timing of death is determined by God [30] and ‘death was seen as a natural response to God’s calling’ (p. 15) [30]. From the perspective of older adults, a good death ‘does not only encompass the dying moment, but also the life a person led before this turning point’ (p. 46) [31]. This resonates with the last attribute of good death, having fulfilled the parental responsibilities, which indicates that older adults perceive a good death as having a sense of achievement after their duties as a parent towards a child’s upbringing, care, and welfare are fulfilled. Similarly, the finding of having fulfilled spousal responsibilities indicates that a good death is perceived by older adults as ensuring the welfare and well-being of their spouses after their death. Based on older adults’ perceptions on EOL care planning or preference, we identified four distinct subgroups: the ‘Never thought of it’ group; the ‘Have thought of but never talked about it’ group; the ‘Have talked about it, but not documented it’ group; and the ‘Thought of it, talked about it and documented it’ group. Participants reported four unique motivations and barriers to engaging in meaningful EOL discussions and planning by using different ways of revealing (or unrevealing) their preferences or wishes: the ‘degree of readiness - it is too early’; the ‘fear of making them unhappy’; the ‘leaving it up to family’; and ‘interpersonal and social influence.’ Understanding these diverse attitudes towards EOL care preparations helps us highlight the importance of targeted support strategies. We found that participants’ concern about burdening their family members through detailed EOL care instructions constitutes an important yet underexplored factor that may hinder engagement in EOL care planning. Many participants prioritized the well-being of family members over their own EOL care preferences. Such concerns may also reflect culturally shaped expectations surrounding family caregiving [32]. However, family members who ‘usually stood in as surrogate decision-makers’ (p. 1076) [33] for the incapable patient may feel pressured to make treatment decisions under time constraints or feel unprepared for making treatment decisions with limited knowledge on the treatment and its outcomes [33]. They may feel guilty or grievous for not knowing the patient’s care preferences [33]. They may also experience emotional distress when they disagree with family members or healthcare providers about the appropriate treatment the patient should receive [18]. Hence, an open and honest EOL conversations with the family could result in achieving goal-oriented and self-directed care and reduce the burden of surrogate decision-making for the family [34]. In fact, having an EOL conversation with the family should not be considered as ‘callous or rude’ (p. 450) [35], but rather ‘an act of love’ (p. 450) [35]. We suggest that the government and organizations promoting ACP frame EOL conversations with the family members as an act of love or care. This approach may help individuals feel more comfortable discussing their EOL care preferences with their spouses and children. In addition, the government may consider conceptualizing ACP as a ‘preventative health behavior’ (p. 247) [18], and normalizing and upstreaming the ACP discussion into usual health promotion activities to encourage the general public to openly talk about future healthcare decisions and wishes [36]. In this regard, providing training programs for healthcare professionals could help them develop the necessary skills for having EOL conversations with older adults. In addition to these policy-level efforts, our findings suggest a role for individuals and families in initiating EOL conversations, particularly given the relational and family-centered orientation reflected in participants’ responses. Healthcare providers may also support such conversations by creating culturally sensitive spaces. Moreover, in the Singaporean context, the emphasis on the role and importance of family in a person’s good death suggests a need to propose theoretical models of death and dying that take family dynamics and relationships into account. Existing models based on data from Western countries, which highlight individual autonomy, may need to further reflect cultural dimensions, such as parent-child relational features or expectations and shared responsibilities within the family members that are distinctive across different cultural settings. While the current Western cultural discussions emphasize the necessity of viewing individual autonomy through a relational lens, particularly in the EOL care setting [37, 38], our findings could suggest a basis for developing relational autonomy for older adults and their well-being in both Western and Eastern cultures. Notably, recent arguments support the idea that minimizing the burden on family members can be seen as an expression of one’s autonomy [38]. From this perspective, participants in this study can be regarded as exercising agency on their behalf. The agentic self both is impacted by the cultural norms and impacts upon the dynamic processes that establish the appropriate model of agency [39]. Therefore, a support system that enables them to fulfil their lives according to their own values and preferences must be established, highlighting the need for a well-developed EOL care system. Taken together, while several themes identified in this study resonate with prior research conducted in Asian contexts [14–16, 19, 40, 41]–particularly the emphasis on family-centred decision-making and relational considerations in EOL Care, where individuals may prioritize harmony and defer decision-making to family members [41, 42]–our findings also highlight two additional points. First, our findings point to an underlying ambivalence, in that some participants perceived that even providing detailed EOL instructions, often assumed to reduce family burden, might itself constitute an additional potential burden for their family members. Second, differentiating participants by their levels of engagement in EOL planning and conversations helps illustrate how such concerns or ambivalence may (or may not) lead to concrete planning behaviours. These findings point to the relational and process-oriented nature of EOL decision-making in later life. A limitation of this study is that the pace of data collection was disrupted several times during the COVID-19 pandemic due to the implementation of national safety measures, which made it difficult to recruit more participants from diverse demographics and backgrounds. For instance, we targeted community-dwelling older adults, yet, male participants tended to be relatively younger, only one man was 70 years and above – reflecting the current difference in the life expectancy between men and women to some extent. Still, the voices from men in old-old age group are needed. Although recurring thematic patterns were observed across interviews, a larger sample may have provided additional perspectives, particularly from minority ethnic groups. In addition, due to the design of the broader parent study, participants were required to have at least one adult child. As such, the perspectives of older adults with no children were not captured in this study. Future studies may explore their perspectives in light of changing family structures. Another limitation worth mentioning is that the meaning of death and dying appears to be heavily influenced by one’s religious beliefs, as indicated by some participants. But in this study, that perspective has not been further analysed because it goes beyond the study’s interest. Furthermore, some participants may have found the topic of the interview somewhat difficult and may not have disclosed their views as openly and honestly as we would have liked them to have. Future studies should employ methods that encourage participants to disclose their thoughts more openly. Conclusion To facilitate open and honest EOL conversations among older adults, families, and healthcare professionals, a better understanding of older adults’ perspectives on good death is warranted. Such understanding ensures that older individuals’ own needs and preferences are reflected in the EOL care planning process. This also provides the foundation for developing relevant programs and services in community. Hearing the voice of older Singaporeans, we learned that the major concerns around good death and EOL care include being a burden to family members and creating additional suffering or distress for them during the process of dying. This finding might be especially relatable in Singaporean context, expanding the idea of individual autonomy through a relational dynamic within family should be worth exploring in other cultural settings as well. This can eventually help improve the QoL of older adults at EOL and offer them an opportunity to have the ‘good death’ they wish for. Supplementary Information Supplementary Material 1. Supplementary Material 1.