Assessing the relationship between the distress levels in patients with irreversible terminal delirium and the good quality of death from the perspective of bereaved family

Abstract

Background

Research on achieving a good death for terminally delirious patients is scarce, with limited knowledge about the level of good death and influencing factors. This study investigates the level of good death among delirium patients, factors influencing it, and the correlation between distress, end-of-life care needs, and achieving a good death by surveying bereaved family members of deceased patients in Chinese hospitals.

Methods

This cross-sectional study from January 2022 to January 2024 was conducted among bereaved family members of patients using an online questionnaire. The questionnaires consisted of (1) participants’ demographic and disease-related questionnaires; (2) the Good Death Inventory (GDI) China version; (3) Terminal Delirium-Related Distress Scale (TTDS) China version; (4) the Care Evaluation Scale - short form (CES) China version. All data were analyzed using descriptive statistics, and the associated factors influencing good death were analyzed by multiple linear regression analyses.

Result

A total of 263 subjects were enrolled. More males (63.5%) participated than females (36.5%), the mean age was 75.35 ± 13.90 years. Good quality of death was significantly and negatively related to the distress in patients with irreversible terminal delirium (r = -0.458, P<0.01).The multiple linear regression model indicates that TDDS score, CES score, types of diseases, smoking history, nutritional deficiency are important factor affecting the good quality of death.

Conclusions

The good quality of death from the perspective of bereaved family, a negative correlation was found between the distress in patients with irreversible terminal delirium and good death. Medical staff should be more aware of good quality of death in patients, future research should expand sample sizes to include more demographic data, and explore the concept of a good death across different cultural contexts.

Background

Death and the quality of dying have always been issues of extreme concern for patients, families, and healthcare professionals [1]. Good quality of death, refers to “good death”, is one of the primary goals of palliative care [2]. What constitutes a good death or successful dying remains undefined. Previous scholars have included several aspects in defining a good death: it is an expected, accepted, and prepared death, free from pain and symptoms, in peaceful acceptance of one’s life and circumstances, with support from individualized and holistic care [3,4,5]. Achieving a good death is a significant objective and result in assessing end-of-life care for patients [1].

Most patients experience delirium in the dying phase, also known as terminal delirium, which often occurs during the final days of life and is unlikely to be reversed [6, 7]. The most recent evidence from systematic reviews and meta-analysis studies reports risk factors for terminal delirium to include older age, male sex, organ failure, fluid and electrolyte imbalances, suboptimal pain management, hypoxia, breathlessness, poor well-being with declining performance status [8, 9]. Terminal delirium is a distressing condition commonly experienced by hospitalized patients receiving end-of-life care, with the associated pain of death typically exacerbating physiological and psychological distress for patients and their family members, burdening family caregivers, and presenting a significant challenge for healthcare providers [10, 11]. Delirium complicates communication among patients, family members, and healthcare professionals, affecting treatment decision-making and hindering healthcare providers’ assessment of patients’ physical symptoms [12, 13]. Additionally, terminal delirium may lead to palliative care challenges, including palliative sedation and the delicate balance between managing precipitating factors and treatment burdens [14, 15].

Currently, a standard assessment tool for irreversible terminal delirium is currently missing, with no consensus on its management and treatment. Challenges in participating in assessments are faced by most patients experiencing terminal delirium due to rapid cognitive alterations, disruptions in consciousness, and distressing physiological symptoms [10, 16]. Therefore, in this study, we chose to survey the primary caregivers of dying patients to inquire about the patients’ experiences of distress, treatment, and care during terminal delirium, as they are the individuals who maintain the most direct and enduring intimate contact with the patients. In order to provide the most appropriate treatment and care for terminal delirium, how family members, healthcare providers, and terminally ill patients experiencing delirium make decisions is crucial for a good quality of death.

The primary goal in managing patients with terminal delirium is to assist both the patients and their families in coping with death. By gaining insights into end-of-life care experiences from the perspectives of both patients and families, healthcare institutions can identify and address the unmet needs of terminally ill patients and their families, thereby improving the quality of care [17]. Due to the confusion experienced by terminal delirium patients and their imminent death, their families may provide a more accurate assessment of overall satisfaction with end-of-life care services.

There is a lack of research on achieving good death for terminal delirium patients. Conclusions drawn from historical research indicate that, in terminal delirium, a considerable number of families experienced high levels of emotional distress and do not believe that their relatives have achieved a good death [18]. We have limited understanding of the level of the good death in terminally delirious patients, as well as the factors influencing it. The association between the distress levels in patients with irreversible terminal delirium and the good quality of death remains unsupported by evidence. Meanwhile, given the scarcity of end-of-life care centers in China, comprehensive hospitals or wards have emerged as the primary institutions for managing end-of-life care. This study investigates the level of good death among delirium patients, factors influencing it, and the correlation between distress, end-of-life care needs, and achieving a good death by surveying bereaved family members of deceased patients in Chinese hospitals.

Methods and material

This cross-sectional study was approved by the Medical Ethic Committee at Zhongnan Hospital of Wuhan University, Hubei, China (2023063 K). The informed consent was obtained from the bereaved family members.

Participants

The participants of this study were bereaved family members of patients from January 2022 to January 2024. The inclusion criteria were as follows: adult patients with delirium during the final two weeks of life at hospital (based on a progress note and nursing records and using Diagnostic and Statistical Manual of Mental Disorders, 5 (DSM-V) criteria by primary physicians); bereaved family members aged ≥ 18 years who were the main caregivers of the patients, with basic ability in Chinese reading and writing and were able to complete the questionnaire. The exclusion criteria were as follows: patients died in the intensive care unit (ICU); bereaved family members could not be contacted via telephone, or refused to participate.

Data collection

Participants were recruited from multiple departments (excluding ICU) at Zhongnan Hospital of Wuhan University through convenience sampling. Two nurses reached out to potential participants who identified based on inclusion/exclusion criteria via phone to provide information about the survey. After obtaining consent, the online questionnaire link was sent to their email or WeChat for further communication. The introductory page of the questionnaire outlined the fundamental principles and objectives of the survey, enabling participants to make an informed decision on whether to take part and complete the consent form.

Measurements

Participants’ demographic and disease-related questionnaires included age, gender, marital status, educational level, primary disease, systemic infection, nutritional deficiency, comorbidity, etc. Additionally, the following study measures were used.

Good quality of death was evaluated by using the Good Death Inventory (GDI) Chinese version [19, 20]. The scale consists of 18 items in 10 main and 8 secondary domains. Bereaved family members were asked to rate the patients’ quality of death with seven-point Likert scale from four dimensions: physical and mental comfort, decision making and relation to medical staff, family relationship, and psycho-existential issues. A higher value indicated better quality of death. The psychometric characteristics of the Chinese version of the GDI were confirmed to be applicable China by Juanjuan Zhao. The Cronbach’s alpha coefficient for the questionnaire was 0.896 overall and ranged from 0.561 to 0.950 for the subdomains [20].

The Terminal Delirium-Related Distress Scale (TDDS) was developed by Megumi Uchida (2021) [21] to assess distress levels, care and treatment in patients with irreversible terminal delirium by a bereaved family member. Our team translated the original scale and conducted cross-cultural adaptation. The reliability of the scale was assessed with Cronbach’s α (0.634), with Cronbach’s α for each dimension ranging from 0.650 to 0.766, and structural validity was evaluated by confirmatory factor analysis. The Chinese version of TDDS was only used in this study, has not been published elsewhere. The TDDS comprises four dimensions: patient pain experience (6 items), patient communication ability (6 items), support and explanation from medical staff (10 items), and natural death (2 items), totaling 24 items. Each item is rated on a 6-point Likert scale from “strongly agree” to “strongly disagree,” with scores ranging from 1 to 6. The total score ranges from 24 to 144, with higher scores indicating a more severe perceived distress due to delirium.

The quality of end-of-life care was evaluated by the Care Evaluation Scale - short form (CES) and was modified to a 10-item scale: physical care by physicians, physical care by nurses, psycho-existential care, physician’s explanations to the patient, physician’s explanations to the family, environment, cost, consideration of family health, availability, and coordination and consistency [22]. The scale was completed by bereaved family members with six-point Likert scale. Higher scores indicated good care. This scale’s Cronbach’s alpha coefficient was 0.967 among Chinese people [23].

Statistical analysis

Demographic and disease-related data were analyzed using descriptive statistical methods, including means, standard deviation (SD), and frequency counts (percentages). The association between good quality of death, terminal delirium-related distress, and quality of end-of-life care was investigated utilizing Pearson correlation analysis. Variables that were identified as significant in the univariate analysis (P < 0.2) were further analyzed using multiple linear regressions with stepwise variable selection to identify the significant factors influencing good quality of death. The significance level was set at P < 0.05.

Results

Patient enrollment

From January 2022 to March 2024, of 998 patients who died during this study period, 801 patients (80%) were diagnosed with delirium during the final two weeks of life. Of them, 82 patients presented with pre-existing symptomatic organic brain or psychiatric disorders, while 19 patients lacked adult family members able to participate in the study. We thus contacted 700 family members of deceased patients by phone. Among them, 137 family members declined to participate. We distributed 563 online questionnaires, of which 219 were refused to be filled out. Additionally, 81 questionnaires were excluded due to incomplete responses, with more than half of the data missing. Ultimately, 263 valid questionnaires were analyzed (response rate 47%).

Sample characteristics

As shown in Table 1, among the 263 participants, more males (63.5%) participated than females (36.5%), the mean age was 75.35 ± 13.90 years, the proportion of patients ≥ 60 is 85.6%. Most of the patients were married (91.3%) with low education level (60.9%). 55% of patients were hospitalized for more than 10 days, the mean length of hospital stay was 12.92 ± 13.72 days. Types of diseases included cancer (32.3%), digestive disease (11%), circulation disease(10.6%), respiratory disease (38%), urinary diseases (3%), hematological diseases (4.9%). The majority had opioid analgesics (61.6%), systemic infection (92.4%), hypoxia (87.1%), nutritional deficiency (76%), comorbidity (81%). For the bereaved family members, the mean age was 48.93 ± 6.75 years. The majority (71.1%) were the children of patients. Table 2 is about the characteristics of the patients’ family members.

Correlations and descriptive statistics

Based on the findings from Table 1, significant differences (P < 0.05) were observed in the GDI scores among types of diseases, smoking history, drinking history, systemic infection, hypoxia, nutritional deficiency. Table 3 displays the correlation between the GDI score and the TDDS score (r = -0.458, P<0.01), as well as the relationship between the GDI score and the four dimensions of the TTDS scale. Table 4 reveals the GDI scores is associated with the CES score (r = 0.362, P<0.01), and is also related to the following three items: physician’s explanations to the family (r = 0.209, P<0.01), cost (r = 0.223, P<0.01), consideration of family health (r = 0.158, P<0.05).

Table 1 Good death differences among different sociodemographic and disease-related characteristics (n = 263)

Table 2 Demographic information of the family member of patients

Table 3 Summary of the associations between good death and irreversible terminal delirium (n = 263)

Table 4 The associations between good death and end-of-life care (n = 263)

Factors influencing good death

Variables demonstrating statistically significant results in univariate tests (based on a p-value < 0.2) are chosen as candidate variables for inclusion in the multivariate analysis. The multiple linear regression model indicates that TDDS score, CES score, types of diseases, smoking history, nutritional deficiency are important factor affecting the good quality of death, explaining 39.4% of the total variation in the GDI score (Table 5).

Table 5 The factors influencing good death according to multiple linear regression analysis (n = 263)

Discussion

Our study found that the GDI score for patients with irreversible terminal delirium was 76.68 ± 6.52, slightly lower than in the study by Juanjuan Zhao et al. [20]and the deceased patients in Japanese hospitals [24]. This finding prompted us to contemplate the differences in healthcare and end-of-life care services across regions. The study by Zhao was conducted in Guangdong [20], a region with relatively strong economic development in China. Conversely, our study was carried out in a central city with slightly inferior economic conditions compared to Guangdong. Economic status often influences the quality of healthcare services, with most of the central region lacking end-of-life care services. This indicates disparities in the allocation of healthcare resources across regions, necessitating further attention and improvement to ensure that patients receive more detailed and humane care in their final days.

Significant variations in GDI scores were noted across different disease categories in a univariate analysis. This research included patients with either cancer or respiratory diseases, with a notable increase in respiratory disease cases due to the COVID-19 pandemic in China in 2022. Notably, cancer patients had the highest GDI scores at 78.39 ± 6.83. This finding may reflect the longer treatment duration characteristic of cancer. Cancer patients in the advanced stages of the disease, who have been aware of their illness for a longer period, tend to have a stronger subjective experience of the tumor. They endure more suffering from the disease, and as the duration of their illness lengthens, they also have sufficient time to undergo a psychological adaptation period [25]. This process enhances their psychological resilience, leading to a shift in their attitude towards a peaceful departure when the realization of an incurable condition sets in. Our study revealed that the dimension of patient suffering in scale is perceived by family members as being present when the patient is delirious. This may be attributed to the fact that terminal delirium often coexists with various physical distress states, such as severe pain, respiratory distress, and multi-organ failure. The participants were bereaved family members, as the patients themselves were not involved in the assessment. This decision was based on several factors. Firstly, the patients in this study were in the terminal and irreversible delirium stage, which left them with compromised consciousness and extreme vulnerability, rendering their active participation unfeasible. Secondly, existing literature suggests that family members of patients may offer valuable insights into the end-of-life experiences of the patients and can better address their end-of-life needs [26]. Additionally, unique cultural practices in Asian countries such as Japan [24, 27], South Korea [28], and China [23]often involve withholding information about impending death from patients and excluding them from end-of-life decision-making processes. In China, the traditional ethical concept of “rebirth taboo death” profoundly shapes societal norms, placing greater emphasis on material pursuits during life while overlooking the genuine needs of individuals facing the end of life [29]. Consequently, the prevalence of the “rebirth taboo death” concept leads to a predominant focus on material care and life extension for terminally ill patients, often neglecting their psychological well-being [29,30,31].

Researches indicate that the incidence of delirium among patients receiving palliative care can be as high as 58% [32], with rates ranging from 42–88% [33]in terminally ill patients, and even reaching 93% [34]. The TTDS scale utilized in this study not only assesses patients’ cognitive function and psychiatric symptoms but also encompasses four additional dimensions: support for families and esteeming a patient highly, ability to communicate, adequate explanation of treatment and care for delirium and dying in a natural way, and generally at peace and secure.

Our study shows that TTDS score is negatively correlated with the GDI score, indicating that the higher distress levels in patients with irreversible terminal delirium, the lower their scores on good death. This is consistent with research showing that delirium is an independent risk factor influencing the prognosis of late-stage cancer patients and also serves as an independent predictor of mortality [13]. Delirium, when severe, is more likely to lead to serious symptoms such as short-term memory impairment, severe hallucinations, and perceptual disturbances. Managing delirium poses a significant challenge in end-of-life care, causing immense distress to both the dying individual and their family as it is often underestimated or overlooked [10]. Consequently, this results in patients receiving inappropriate treatment or going untreated.

We used CES - short form to assess Chinese bereaved family members, aiming to measure the structure and quality of end-of-life care. In this study, the mean CES total score of patients was lower than that of patients in Japanese [35] and Korea [36]. And there is a correlation between the total scores of CES and GDI. This finding is reasonable, as receiving better end-of-life care can enhance the quality of life for terminally ill patients [17]. And GDI score is also related to the three items: physician’s explanations to the family, cost, consideration of family health. This may be attributed to the perspective of family members, who consider involvement in patient treatment decisions crucial, reflecting the extensive engagement of Chinese patient families in end-of-life care [23]. Simultaneously, it indicates that patient families desire greater attention from healthcare providers regarding their physical and mental well-being, which is vital in end-of-life cancer care.

Not surprisingly, nutritional deficiency reported poor quality of life. In end-of-life patients, the exacerbation of nutritional depletion due to increased expenditure and decreased intake worsens malnutrition. This worsening of malnutrition further intensifies the nutritional depletion in patients, leading to a vicious cycle [37]. Considering the close relationship between nutritional deficiency and good quality of life, the importance of addressing nutritional needs in end-of-life care to improve patient outcomes is underscored.

In cases of terminal irreversible delirium, patients often struggle with impaired cognitive function and communication abilities, making it difficult for them to understand medical decisions and express their needs. This presents a significant challenge for healthcare providers assessing these patients, as delirium symptoms can disrupt their behavior and speech, complicating the evaluation process. Our primary aim in end-of-life care is to improve patients’ quality of life, achieving good death and ensure they receive dignity and comfort in their final moments. Poor management of delirium not only affects patients’ comfort and the quality of care but also increases the emotional burden on medical teams and family members. Therefore, effective management of delirium in terminal patients is crucial for providing optimal care and support during their end-of-life stage, ultimately enhancing their quality of life. To better achieve this objective, it is crucial to actively involve family caregivers in the terminal care team. Family members possess a deeper understanding of the patients’ needs and preferences, and their participation can offer more personalized care while fostering trust and continuity between the medical team and patients. By supporting the involvement of family caregivers, we can establish a more intimate and supportive care environment, providing terminal patients with comprehensive and effective care [38]. Healthcare professionals can designate specific times for communication to guarantee that relatives have the opportunity for direct conversations. They may extend invitations to family members to engage in discussions about care decisions, acknowledging their perspectives and offering details on resources for palliative care and emotional support. Furthermore, healthcare workers can facilitate support groups for end-of-life care, enabling family members to interact with others who have undergone comparable experiences, thereby facilitating the sharing of thoughts and emotions.

There are several limitations. This study only represents the perspectives of caregivers of bereaved family members and is subject to memory bias. We did not investigate the mean number of months of bereavement, this bias may stem from emotional influences or the passage of time affecting the accuracy of recollections. Furthermore, the survey response rate was only 43%, introducing potential non-response bias and limiting the generalizability of the sample results. And the scales we used also have limitations, and we suggest that future research consider developing or validating more general scales to accommodate participants experiencing different types of loss. Additionally, the contact with relatives of deceased patients was conducted solely through the internet, without employing random sampling methods, further restricting the representativeness of the study sample and potentially leading to biased results.

Conclusion

Our study evaluates the good quality of death from the perspective of bereaved family, a negative correlation was found between the distress in patients with irreversible terminal delirium and good death. Among them, the quality of end-of-life care was also a key factor associated with good death. Future research should expand sample sizes to include more demographic data, investigate the specific locations of patient deaths, and explore the concept of a good death across different cultural contexts. This approach will facilitate the development of more personalized end-of-life care plans and provide a comprehensive understanding of the meaning and practices associated with a good death.