Defining a Good Death in the Intensive Care Unit: A Systematic Review

Search strategy

The search was conducted in PubMed, ScienceDirect and Ebscohost. We did formulate a Population, Intervention or exposure, Comparison and Outcome (PICO) strategy to guide the search process in formulating the research question [Supplementary File 1]. The engaged keywords were settled on medical subject heading terms and the BOOLEAN operator. The search syntax was formed as follows: definition AND good death AND ICU.

Selection criteria

The inclusion criteria encompassed English-language quantitative or qualitative studies published from inception until August 30, 2024, that reported definitions of a good death in the ICU and were available in full text. Exclusion criteria included studies that focused exclusively on euthanasia.

Selection of studies

All relevant studies were imported into EndNote 20, and duplicates were removed. One reviewer (IH) conducted the screening process twice and was then independently checked by another reviewer (NU and BM). Any discrepancies were resolved through discussion for agreement (WW). Screening involved three steps [Figure 1]: Excluding irrelevant titles and abstracts, searching for full-text articles and reading full-text articles to identify those meeting inclusion criteria. Studies reporting definitions of good death in the ICU were included.

Close

Figure 1:

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses - 2020 flows diagram of studies included in the review.

Export to PPT

Risk of bias assessment

The risk of bias in the studies included in our systematic review was assessed by utilising the Joanna Briggs Institute Critical Appraisal Checklist for quantitative and qualitative studies^[8] in assessing the potential bias in study design, conduct and analysis. Three reviewers (IH, NU and BM) independently evaluated the study quality, resolving discrepancies through discussion. Any unresolved disagreements were arbitrated by a fourth reviewer (WW). To determine the score, the number of ‘yes’ items was divided by the total number of items and multiplied by 100%. Studies scoring over 70% were deemed high quality, those between 50% and 70% were medium quality, and those below 50% were considered low quality. Studies of low quality were excluded from this systematic review.^[9] All studies in this systematic review were classified as high quality, as they scored 85% and 90%. None of the studies have scores below 50%; thus, all studies are included in this systematic review. All studies clearly defined the definition of a good death in the ICU. We also accounted for the potential risk of sampling bias in the studies included in our systematic review by geographical region.

Data extraction

An Excel spreadsheet was utilised for data extraction. The sheet contained study details such as author names, publication year, design, location, participants, age, sample size and definition of good death. The data extraction was carried out by IH and cross-checked by NU and BM, with any disparities resolved through discussion (WW).

Data synthesis

After the data extraction, the information was synthesised by a reviewer (IH) to identify the definition of a good death in the ICU, which was then independently validated by another reviewer (NU and BM). In cases of disagreement, discrepancies were resolved through discussion between the reviewing authors. We consulted the third reviewer (WW) if consensus could not be reached.

Due to the heterogeneity of participants across the included studies, a meta-analysis of the data was not conducted. This kind of study was not feasible with our data. In addition, because qualitative and quantitative investigations were mixed, weighting was not done. However, we were able to aggregate frequencies across research to get the mean percentages for various areas of what is thought to be part of a good death, because all studies reported stakeholder frequency of responses that supported particular themes of a good death.

We generated common themes of a good death using the grounded theory-based coding consensus, co-occurrence and comparison method. Our thematic analysis was conducted inductively from the data, meaning that themes were identified directly from the content of the studies, without being guided by a pre-existing framework. This allowed us to stay close to the data and ensure that the themes reflected the experiences and perspectives of the participants in the studies we reviewed. However, we also recognise that existing theories and concepts related to EOLC in ICU settings may have informed the initial stages of analysis.

The process of identifying themes followed the standard procedure for thematic analysis: We first familiarised ourselves with the data by reading and re-reading the studies; initial codes were generated to capture key features of the data related to a good death in ICU settings; as patterns emerged, we grouped related codes into broader themes. These themes were refined through constant comparison and discussion amongst the research team to ensure that they accurately represented the data and captured the diversity of perspectives on a good death; after initial coding and theme identification, we revisited the studies to ensure that the final themes were robust and reflective of the data as a whole. We engaged in several rounds of refinement, checking that the themes were sufficiently inclusive of the different experiences and perspectives presented in the studies. In cases where discrepancies or ambiguities arose, the research team discussed and resolved them to reach consensus on the final themes.

We reported the rate of endorsement of each of the 5 codes within each of the sources (e.g., patients, family members and HCPs) and computed the percentages.

Instead, a narrative synthesis was conducted, following the reporting guidelines of Synthesis Without Meta-analysis [Supplementary File 2].^[10] This narrative synthesis aimed to identify and describe patterns within the data, focusing on similarities, differences and associations amongst study outcomes.^[11]

The synthesis specifically focused on identifying definitions of a good death in the ICU. To provide a more comprehensive synthesis of the evidence across studies, the identified definitions were categorised. This classification process was informed by previous studies on a good death. These categories included (1) definitions of a good death from the perspectives of patients, (2) definitions of a good death from the perspectives of family members and (3) definitions of a good death from the perspectives of HCPs.

Identification and selection of studies

A total of 2,774 records were identified in the database. After removing 629 duplicate records, we screened the remaining 2,145 records based on their titles and abstracts. Of these, 1,781 reports were excluded due to irrelevant titles and abstracts. We then assessed the full text of 364 studies, eliminating 85 studies. As a result, 279 studies were initially considered. However, upon further evaluation, 244 studies were excluded. Ultimately, 35 studies were included, as summarised in [Figure 1].

Characteristics of included studies

Thirty-five high-quality studies were included. The participants’ mean age ranged from 30.64 to 48.37 years old. The sample size ranged from 30 to 604. The characteristics of included studies are summarised in Figures 2 and 3. They were conducted in various regions, including Indonesia (1 study), Turkey (3 studies), Thailand (4 studies), Iran (6 studies), Korea (5 studies), UK (7 studies) and US (8 studies) [Figures 2 and 3].

Close

Figure 2:

Participant characteristics based on number of studies.

Export to PPT

Close

Figure 3:

Study location.

Export to PPT

Definitions of a good death in ICU

Definitions were categorised into themes and subthemes, with the frequency of each theme determined from the perspectives of patients, family members and HCPs. The term defines the characteristics that differentiate the phenomenon from a similar one. The relevant resources are used to identify the definition of a good death amongst ICU patients. ‘Being free from sufferings’ was the most supported theme in the literature, followed by the ‘withdrawing and withholding of life-sustaining technologies in ICU settings’, ‘privacy’, ‘family involvement’ and ‘receiving spiritual and cultural support’. Therefore, the conceptual definition of good death in ICU arrived at in this study was the condition of dying and death, including being free from sufferings, withdrawing and withholding of life-sustaining technologies in ICU settings, privacy, family involvement and having spiritual and cultural support.

Being free from sufferings

Being free from suffering is a vital part of a good death in the ICU. This attribute is free from physical and emotional suffering. The majority of suffering amongst ICU patients is pain.^[6,12-17]

Withdrawing and withholding of life-sustaining technologies in ICU settings

Life-sustaining technologies, such as mechanical ventilation, dialysis and other interventions, may affect the ability of patients to have a good death. These technologies, while life-saving in many circumstances, can sometimes prolong the dying process and may contribute to discomfort, emotional distress or a perceived loss of autonomy for both patients and their families. This theme includes the earlier cessation of treatment, unnecessary prolongation of life and less visible technology.^[18-22]

Privacy

Privacy is a comfortable environment. Patients die in a private room, accompanied by family.^[23-27]

Family Involvement

The involvement of the family is crucial in ensuring that the patient dies in a state of good death. This theme includes having an opportunity to say goodbye, participating in important care decisions, patients have chances to say their wishes to families, and patients are not alone while dying. Family members often experience profound grief, stress and a sense of helplessness, which can influence how they interpret the quality of the patient’s death. This includes aspects such as the opportunity for closure, the sense of peace and dignity in the patient’s final moments and the ability to participate in important care decisions.^[28-35]

Receiving spiritual and cultural support

Having spiritual and cultural support means patients having spiritual and cultural support from their families or healthcare providers to maintain their behaviours and traditions before death.^[36-45] The themes and sub-themes of the definition of a good death in the ICU in [Table 1]. However, discrepancies among the respondent groups were noted in the core themes: family members emphasised being free from suffering, while patients reported family involvement more frequently. Healthcare providers highlighted the importance of spiritual and cultural support. Differences in frequencies of themes amongst the stakeholder groups were greatest for withdrawing and withholding of life-sustaining technologies in ICU settings, which was rated more frequently in HCP’s perspective articles (80%) than in patient and family members’ perspective articles (44.44% and 38%, respectively) [Table 2]. The themes of good death in ICU as summarised in [Figure 4].

Close

Figure 4:

Conceptual model of good death in ICU.

Export to PPT

Being free from sufferings

The first and most prominent theme that emerged from our data is the desire to be free from suffering, a fundamental aspect of a good death. The physical discomfort associated with terminal illness, often exacerbated in the ICU by invasive procedures and mechanical interventions, remains a major concern for both patients and their families. The ability to alleviate pain and distress, through effective palliative care and symptom management, is a cornerstone of achieving a good death in the ICU setting. This finding aligns with previous studies, which have identified the relief of physical suffering as essential for a dignified and peaceful death.^[6,12-17] In addition to physical suffering, emotional and psychological distress are a significant component of the ICU experience, particularly for patients who are nearing the end of life. Patients and families in our study reported the emotional toll of uncertainty, fear and the sense of losing control over the dying process. The constant medical interventions, coupled with the often unpredictable course of critical illness, create an environment that can exacerbate anxiety, sadness and existential distress. A good death, therefore, also involves the mitigation of emotional suffering. Ensuring that patients and families feel informed, supported and heard can significantly alleviate the emotional burden of the dying process. Open and empathetic communication between the healthcare team and the family is critical, as it helps to establish realistic expectations and reduce the fear of the unknown. In this context, the role of the healthcare provider goes beyond medical expertise; it also involves offering emotional reassurance, being present with the patient and family and acknowledging the emotional gravity of the situation.^[6,12-17]

Our reviews suggest that emotional suffering is not limited to the patient but extends to family members as well. Family members often experience anticipatory grief, uncertainty and distress about the patient’s suffering, as well as the fear of impending loss. Providing emotional support to families through counselling, clear explanations and allowing for family involvement in decision-making can contribute significantly to reducing their psychological distress and help them feel more equipped to navigate the final stages of the patient’s life.

Withdrawing and withholding of life-sustaining technologies in ICU settings

The percentage of occurrence of this theme is highest from the healthcare professionals’ perspective. Life-sustaining technologies in ICU settings, such as mechanical ventilation, dialysis and other interventions, may affect the ability of patients to have a good death. These technologies, while life-saving in many circumstances, can sometimes prolong the dying process and may contribute to discomfort, leading to potential issues such as pain, sedation or feelings of being detached from one’s body, emotional distress or a perceived loss of autonomy for both patients and their families. In addition, the presence of technology can diminish the opportunity for patients to die in a more natural or peaceful setting, potentially altering the emotional and spiritual aspects of death. For families, witnessing a loved one attached to life-sustaining machines can create distress, feelings of helplessness and a complex decision-making environment.^[18-22]

HCPs manage the tension between the benefits of life-sustaining technologies and the desire for a good death using effective communication, shared decision-making and palliative care to guide families through difficult choices regarding whether to continue, limit or withdraw treatments. The importance of aligning treatment decisions with the patient’s values and preferences, as well as providing adequate support to families in these emotionally charged situations, will also be emphasised. HCPs must strike a balance between extending life through technological interventions and ensuring that the patient’s quality of life, particularly their comfort and dignity, is maintained during the dying process. This includes recognising when technology may no longer be in the patient’s best interest and considering a shift towards palliative care measures that prioritise symptom management and emotional support.^[18-22]

This theme is closely tied to the medical decision-making process and reflects the importance of aligning care with the patient’s prognosis and goals. Many participants in our study highlighted the significance of having treatments discontinued when they were deemed futile or when the burdens of continued intervention outweighed the potential benefits. This theme underscores the tension between medical technology and the preservation of dignity, as the ICU is often perceived as a space where aggressive treatment may be overemphasised at the expense of quality of life. The decision to withhold or withdraw life-sustaining treatments is fraught with ethical and emotional challenges, yet it remains an essential aspect of ensuring that the death process respects the patient’s wishes and values.^[18-22]

Privacy

Privacy emerged as a critical factor in defining a good death, especially given the high-stress, often crowded, nature of ICU environments. Both patients and their families expressed the importance of maintaining dignity and personal space in the final moments of life. The ICU, with its constant monitoring, visits and medical procedures, can be an inhospitable environment for maintaining privacy, particularly in the moments of death. This finding emphasises the need for ICU teams to be mindful of the emotional and psychological impact of surveillance and constant medical presence. Ensuring privacy, even if only in the final hours, provides families with the space to grieve and patients with the sense of autonomy and dignity that they deserve at the end of life. Privacy, in this context, extends beyond physical space, encompassing the ability to have intimate, unimpeded moments with loved ones and to experience the final phase of life in a way that preserves the patient’s personhood.^[23-27]

Family involvement

The emotional and psychological needs of family members during the dying process are crucial. The involvement of family members in care decisions can significantly impact the overall experience of death and, consequently, the definition of a good death. The concept of a good death includes not only the patient’s experience but also the relational and emotional aspects that influence how families perceive the good death. HCPs adopt more family-centred approaches in ICU settings to better support family members throughout the dying process. This includes fostering open, empathetic communication, involving families in care decisions and offering psychological and emotional support. By attending to both the patient’s and the family’s needs, healthcare providers can help facilitate a more holistic approach to EOLC that aligns with the goals of a good death for all involved.

The opportunity for patients to say goodbye to their loved ones is another essential aspect of a good death in the ICU, as highlighted by our study participants. The ICU, with its focus on life-saving interventions, often places patients in situations where they may not be able to interact meaningfully with their families before death. This disconnection can lead to a sense of unfinished business, both for patients and their families. Ensuring that patients have the chance to communicate and say farewell to loved ones is an emotional and relational aspect of a good death that contributes to closure and emotional healing. This theme resonates with the broader literature on EOLC, where the importance of family involvement and communication has been identified as a significant contributor to both the quality of death and the grieving process. Allowing family members to witness the final moments and participate in decision-making fosters a sense of shared experience and collective mourning, which can be essential for both emotional closure and family resilience.^[28-35]

Receiving spiritual and cultural support

Finally, the theme of spiritual and cultural support highlights the role of belief systems in shaping what constitutes a good death. For many patients and families, the presence of spiritual guidance or the accommodation of cultural practices provides comfort and meaning in the face of terminal illness. In the ICU, where medical care often focuses on the physical body, the need for spiritual care can be overlooked. Our findings suggest that access to chaplaincy services, prayer or cultural rituals can significantly enhance the experience of dying, offering a sense of peace, transcendence and continuity beyond the medicalised environment of the ICU. The importance of spiritual and cultural care has been increasingly recognised in palliative care, with studies emphasising that attending to these needs can improve patient satisfaction, reduce suffering and provide emotional solace to families.^[46]

The importance of spiritual and cultural support in EOLC, particularly in ICU settings, where patients may face complex medical decisions and heightened emotional stress. Spiritual and religious beliefs can significantly shape a patient’s and their family’s understanding of a good death, providing comfort, meaning and a sense of peace.^[36-45]

Examples of religious and spiritual practices such as Islam: For Muslim patients, spiritual practices such as prayer (Salat), the recitation of the Shahada (testimony of faith) and the presence of a religious leader or imam are essential at the end of life. Christianity: Many Christian patients and families may find comfort in the rituals of prayer, the presence of a chaplain or the administration of sacraments such as communion or anointing of the sick. In ICU settings, patients may desire to receive spiritual support from clergy or family members to affirm their faith and provide peace in their final moments. Hinduism: In Hindu tradition, the concept of death is often viewed as a transition to the next life. Rituals such as chanting mantras, having a priest or family member perform specific prayers, and the idea of a proper ritual for the body (such as bathing and anointing the body) are key aspects of a good death. In the ICU, family members may focus on ensuring that the patient’s body is treated with respect, often requesting the opportunity to perform or witness these rituals. Buddhism: Buddhist patients may emphasise the importance of mindfulness, meditation and the presence of a spiritual leader (monk or teacher) to guide them through the process of dying. Indigenous Spiritual Practices: For patients from Indigenous cultural backgrounds, the concept of death is often deeply connected to family, community and ancestral spirits. Spiritual rituals may include ceremonies that honour the deceased, such as smudging, prayer or offerings to spirits. In ICU settings, family members may request space for these rituals to be performed, and the presence of family and community can play a significant role in how the dying process is viewed.^[36-45]

In addition, cultural context influences the ICU experience and the definition of a good death. Death rituals and practices vary significantly across cultures, and these differences can shape how a ‘good death’ is perceived in ICU settings. For instance, in some cultures, death is viewed as a transition to the afterlife that requires specific rituals or spiritual practices. For example, in many Asian cultures, the family may place great emphasis on preparing the deceased’s body and performing rites that ensure a peaceful transition. In contrast, some Western cultures may focus more on medical interventions aimed at sustaining life, sometimes at the cost of delaying the natural process of death. These cultural variations can influence not only the family’s expectations of a good death but also the care decisions made in the ICU.^[36-45] Family involvement in the decision-making process at the end of life is another aspect influenced by cultural context. In some cultures, especially in many Asian, African and Latin American communities, family decision-making is a collective process, with the extended family often playing a significant role. This contrasts with more individualistic cultural perspectives, such as in some Western societies, where the patient’s autonomy and individual wishes may be prioritised. In ICU settings, the family’s role in deciding whether to continue life-sustaining treatments or transition to palliative care can be influenced by cultural norms around family structure, authority and responsibility. This may impact the perceived quality of death for both the patient and the family, as certain cultural expectations around family involvement may either ease or complicate the decision-making process.^[36-45]

Furthermore, the role of spirituality and religion in defining a good death is deeply influenced by cultural context. For example, in many Christian cultures, spiritual practices such as prayer, the presence of a chaplain and rituals like communion or anointing the sick are integral to the experience of dying. In contrast, Hindu or Buddhist patients may seek guidance from spiritual leaders and engage in rituals that facilitate a peaceful transition to the afterlife, such as meditation or chanting mantras. Similarly, Islamic patients may focus on specific prayers, such as the recitation of the Shahada, or request the presence of an imam. Cultural differences in spiritual practices can influence both how families and patients experience the ICU environment and how they define a good death.^[36-45]

The varying cultural expectations and practices can significantly affect how patients and families experience death in ICU settings. In cultures where death rituals are highly significant, the inability to perform these rituals in the ICU – due to technological constraints or time limitations can create emotional distress and contribute to the feeling of an incomplete or unsatisfactory death. In addition, cultural expectations surrounding family involvement in decision-making can create tensions if healthcare providers are not sensitive to these dynamics. For example, in cultures where family members expect to be deeply involved in care decisions, excluding them from discussions or making decisions without their input can result in a sense of alienation and dissatisfaction with the dying process. To better support patients and families from diverse cultural backgrounds, HCPs must be trained in cultural competence and sensitivity. Understanding and respecting cultural differences in death rituals, family involvement and spiritual practices can help ensure that the dying process is handled in a way that aligns with the values and needs of the patient and their family. In ICU settings, where care is often highly medicalised, incorporating cultural and spiritual practices into care plans can help facilitate a more holistic, meaningful and peaceful death experience.^[36-45]

Strengths

To our knowledge, it is the first review to investigate the definitions of a good death in ICU. A comprehensive understanding of the definition of a good death in the ICU is crucial for nurses to improve EOLC for patients.

Limitations

The limitation of this study is the heterogeneity of the included studies, particularly in terms of study populations and methodologies. While this diversity provides a broad perspective, it also presents challenges in synthesising the findings and drawing generalised conclusions. We address this by acknowledging the limitations in the generalisability of the results and suggest that future studies explore more homogeneous populations for more specific insights.

Implications

This study highlights the importance of cultural competence in ICU settings, suggesting that healthcare providers need to be more attuned to the diverse religious, cultural and emotional needs of both patients and families. By incorporating culturally sensitive care into end-of-life planning, HCPs can facilitate a more positive and individualised experience of death. In addition, the findings suggest a need for policies that support the inclusion of spiritual and cultural practices in EOLC protocols in ICU settings. This would ensure that patients and families have the necessary support to navigate the emotional and spiritual aspects of dying, in addition to the medical aspects. Healthcare systems may benefit from the integration of chaplaincy services, cultural liaisons and other support systems in ICU care. Furthermore, future studies should explore more homogeneous populations for more specific insights.

Recommendations

We recommend that ICU care teams adopt a more family-centred care approach in end-of-life situations. This includes involving families in decision-making and ensuring that their emotional, psychological and spiritual needs are addressed alongside the patient’s medical needs. Furthermore, we recommend that healthcare providers receive training in cultural competence. This training should focus on understanding diverse cultural attitudes towards death and the rituals and practices that accompany the dying process. In addition, to increase the generalisability of findings, we recommend expanding research on ICU EOLC to diverse populations, particularly in low-income and non-Western countries. This would provide a broader understanding of how different cultural, social and economic factors influence the definition of a good death.