Abstract
Introduction
Over the past decades, advances in medical technology have led to greater survival in patients with chronic illnesses. One consequence of this is that health care providers find themselves facing higher proportions of geriatric patients with uncurable diseases than in the past. In addition to chronic diseases, these patients may also suffer from mental and psychosocial issues, and some illnesses or medical procedures may cause them to suffer. This makes the provision of holistic palliative care extremely important. Helping end-of-life (EOL) patients to achieve a “good death” is the aim of palliative and EOL care.
Although the concept of a “good death” is a crucial part in palliative and EOL care, there is no clear universal definition for this term. Instead, what is understood to constitute a good death differs based on culture and beliefs. Various working groups have proposed different definitions of a good death. For example, the Institute of Medicine defined a good death as “one that is free from avoidable death and suffering for patients, families and caregivers in general accordance with the patients’ and families’ wishes” (Field & Cassel, 1997, p. 24). In Thailand, the residents of which are primarily Buddhist, a good death consists of three elements: “Death with peacefulness, Natural death, and Death without worrisome” (Rodpal et al., 2007).
Physicians are key in the management of EOL patients’ treatment, their perceptions regarding what constitutes a good death are important in improving both patient care and the palliative care system as a whole (Heyland et al., 2006; Miyashita et al., 2007; Steinhauser et al., 2000; Yeun, 2005). However, physicians have their own personal moral, cultural, and emotional responses to death; they might find it difficult or frustrating to manage dying patients and their real life as their patient’s perspective might be different from theirs. In addition, components of care that are understood as enabling a good death can vary by culture and social background, meaning that physicians’ perspectives regarding a good death may differ by region, personal experience, and beliefs. Evaluation of these perspectives in various regions may help to improve the local palliative care systems.
Several studies have compared the opinions of physicians and patients, and have found differences in the two groups’ perceptions (Meier et al., 2016). For example, a systematic review of 36 studies examining peoples’ perspectives with regard to a good death (30 of which were from Western countries) found differences in patients’ and health care providers’ opinions in this regard (e.g., patients placed more importance on quality of life before death than did physicians; Meier et al., 2016). There have been several studies conducted in Asia to explore the components of a good death, but most were conducted in East Asian countries such as Japan, Korea, and Taiwan (Mori et al., 2018; Morita et al., 2015). Data with regard to what patients and physicians in Southeast Asia perceive as a good death are still lacking.
In Thailand, there have been some studies conducted comparing physicians’ and older patients’ opinions regarding the components of a good death. One study conducted in 2011 at a super-tertiary university hospital in central Thailand found significant differences between the wishes of physicians and patients in this area. The patients, for example, rated knowing the full truth about their illnesses as being more important than did the physicians. In addition, a significantly higher proportion of patients than physicians did not wish to receive treatment to prolong their lives (Kornbongkotmas et al., 2011).
However, different parts of Thailand could reflex different ideas about good death; for example, the prior study in central Thailand, the participants place importance on environmental and spiritual comfort (Srinonprasert et al., 2014), but another study from northern Thailand gave prominence to the association between ethics and place of death. In this condition, the good death was a prior to home death, whereas dying in the hospital was unethical (Stonington, 2012).
In recent years, the concepts of palliative and EOL care have become widely known, even to the general public, the perspective about this issue of both physicians and patients might be changed. Physician trainees are the future health care professionals who will continue to develop their expertise and become proficient in their fields. By improving palliative care systems and palliative care education for training physicians, it might be possible to increase their understanding as to their patients’ wishes at the end of their lives, which may benefit in identifying targets for intervention to increase the quality of the EOL care. Therefore, the objectives of this study were to examine and compare the perceptions of older patients and physician trainees with regard to what they feel constitutes a “good death” for older adults.
Method
Study Design and Participants
This questionnaire-based cross-sectional study was conducted at two tertiary university hospitals in central (Siriraj Hospital at Mahidol University in Bangkok) and northeast (Srinagarind Hospital at Khon Kaen University in Khon Kaen) Thailand from September 2017 to February 2018. This study included older patients (aged 60 years or above) who attended the geriatric clinic or general internal medicine outpatient clinic. The exclusion criteria were patients with dementia or severe depressive symptoms that compromised the patient’s decision-making capacity, life expectancy of less than 6 months, and inability to communicate in Thai, which were assessed by a team of researchers. All physician trainees included residents and fellows in training potentially involved in caring for EOL patients in the departments of internal medicine, surgery, psychiatry, obstetrics and gynecology, ophthalmology, otolaryngology, anesthesiology, orthopedic surgery, radiology, dermatology, or emergency medicine who were willing to participate in the study.
Materials
The 13-item, situation-based questionnaire used in this study was also used in a previous study (Limpawattana et al., 2019; Srinonprasert et al., 2014) to determine participants’ perspectives regarding EOL care. The questions were selected from studies conducted in various countries around the world that were related to patients’ EOL preferences. We recorded participants’ baseline characteristics including age, sex, education, income, marital status, underlying diseases, and health status. Participants’ previous experiences with EOL situations were also recorded, as were their opinions about various aspects of the EOL period including physical and psychological needs, autonomy issues, and closure of life affairs.
Procedure
Physicians and older patients were asked to complete a 13-item questionnaire using a five-category Likert-type scale. The patients were asked to imagine that they had an untreatable disease and life expectancy of less than 3 months, and then rate how much they would agree with the 13 statements listed on the questionnaire using a scale of 1 to 5 (5 = strongly agree, 4 = agree, 3 = neutral, 2 = disagree, 1 = strongly disagree). The physicians were asked to rate the same statements assuming they were treating a patient with an untreatable disease and life expectancy of less than 3 months. The physicians also filled out an additional questionnaire consisting of the same 13 items, which they were asked to complete based on their preferences regarding their own EOL care.
Statistical Analysis
For the demographic data, categorical variables were presented as numbers and percentages, and continuous variables were presented as mean and standard deviation. If the distribution of these data was not normal, medians and interquartile ranges were used instead. Univariate analysis was used to assess the proportional differences in categorical variables between the two groups of participants. McNemar’s test was used to compare between the physicians’ preferences regarding their own death and what they would do for their patients. The value of p less than .05 was considered to indicate a statistically significant difference.
The five possible responses to the 13 statements were combined into agree (for strongly agree and agree) and disagree (for neutral, disagree, and strongly disagree). The results were expressed as percentages for each group. The missing data were missing completely at random (MCAR), making them safe to remove, and were analyzed as missing.
All analyses were performed using STATA version 10.0 (StataCorp, College Station, TX, USA).
Results
A total of 608 patients and 523 physicians were enrolled in this study. The mean age was 71 years in the patient group and 28 years in the physician group. The baseline characteristics of the patients and physicians are shown in Tables 1 and 2, respectively. In the patient group, about 54% were female and most were Buddhist. Seventy percent of patients had previously been admitted into the hospital, and 77% had prior experience of seeing someone die. However, less than 60% had prior experience caring for an EOL patient.
Baseline Characteristics of Older Patients.
Note. IQR = interquartile range; n (%) = number and percentages of responses to each item; DM = diabetes mellitus; HT = hypertension; IHD = ischemic heart disease; COPD = chronic obstructive pulmonary disease.
n = 660; bn = 607; cn = 605.
Baseline Characteristics of Physicians.
Note. IQR = interquartile range; n (%) = number and percentages of responses to each question; EOL = end of life; CPR = cardiopulmonary resuscitation.
n = 523; bn = 518; cn = 522.
Most of the physicians included in the study were Buddhist (96.9%), and approximately a quarter of them were working in the internal medicine department (26%). Only about 40% of them had a history of serious illness, but more than 90% had experience seeing patients die, caring for EOL patients, or counseling patients or their families regarding EOL care. However, only 28% reported having adequate knowledge regarding palliative and EOL care. About 80% of the physicians thought it was appropriate to provide information to terminally ill patients and allow them to make the decisions in advance, but only 77% had ever asked terminally ill patients whether or not they wanted to have cardiopulmonary resuscitation (CPR) performed, and only 75% consistently obeyed patients’ wishes when they did not want CPR. Ninety-one percent of physicians asked the relatives of EOL patients whether or not to perform CPR on the patients. Other types of experience regarding EOL care are shown in Table 2.
Comparison of the Wishes of Older Adults for Their EOL Period and Physicians’ Expectations Regarding the Wishes of Their Patients
The responses of all participants to the questionnaire are shown in Table 3. The two groups provided similar responses to many of the items on the questionnaire. More than 90% of participants agreed about the patient receiving the full truth about their illness, naming a surrogate decision maker, providing relief from uncomfortable symptoms, treating the patient as a whole person, and having the patient’s loved ones around when needed. More than 80% of participants placed similar levels of importance on the patient not to be burden to their family and not to prolong the patient’s life. However, a higher proportion of patients than physicians expressed agreement with wishing for their family to know the full truth about their illnesses (93.94%, 77.39%, p < .001), wishing to be mentally aware toward the last hour of their life (87.58%, 67.62%, p < .001), and wishing to pass away at home (72.98%, 67.18%, p = .006). Conversely, more physicians placed importance on patients being able to complete unfinished business, preparing for death and saying goodbye to family and friends (95.21%, 88.22%, p < .001), and having religious rituals conducted at the end of their lives (81.23%, 72.44%, p = .005).
Comparison of the Wishes of Elder Adults and Physicians’ Expectations Regarding the Wishes of Their Older Patients in Their EOL Period Using Univariate Analysis.
Note. p-value was significant at p < .05.
Comparison Between Physician’s Preferences for Their Own EOL Care and Those Regarding the EOL Care of Their Patients
This study also compared the physicians’ preference with regard to their own EOL care and what they would do for their EOL patients. The results are shown in Table 4. A higher proportion of physicians rated the following items higher for themselves than for their patients: family members knowing the full truth about their illnesses (87.3%, 77.3%, p < .001); being involved in decisions about treatment (98.6%, 96.3%, p = .005); naming a surrogate decision maker (92.5%, 87.7%, p = .002); not being a burden to family members (96.5%, 90.2%, p < .001); completing unfinished business, being prepared to die, and saying goodbye to family and friends (97.1%, 95.2%, p = .03); not receiving treatments to prolong life (92.7%, 84.9%, p = .001); and being mentally aware toward the last hour of life (75.5%, 67.6%, p < .001).
Comparison of Physicians’ Wishes for Their Own EOL Treatment and What They Expected Older Patients’ Wishes to Be Using McNemar’s Test Analysis.
Note. p-value was significant at p < .05.
Discussion
This study aimed to explore what constitutes a good death among physicians and patients, and the results showed that autonomy was the most crucial factor in EOL care. Almost all participants from both groups agreed that patients should receive the full truth about their illnesses, and the participants ranked this statement as being most important to a good death. In addition, most participants agreed that patients should be involved in the decision making about their treatment. These results were similar to those of another study conducted in Japan in 2007, which reported that patients knowing what to expect with regard to their condition and how long they will live were necessary components of good death (Miyashita et al., 2007). Although most physicians placed importance on patients’ autonomy, only half of them had informed a majority of their patients about their diagnoses and prognoses. Instead, more than 70% chose to inform the patients’ relatives (as shown in Table 2). The reason for this apparent contradiction was not explored in the current study. However, it may be partially due to physicians not having the skills to adequately communicate with EOL patients. The results of a 2008 study support this idea, reporting that physicians were not confident in their ability to discuss prognosis or CPR with patients (Sulmasy et al., 2008).
The opinions of participants in both groups were similar with regard to the alleviation of suffering and treating the patient as a whole person. More than 95% agreed with this component of EOL care, a result that is consistent with those of several previous studies from Thailand (Kongsuwan et al., 2012; Rodpal et al., 2007), China (Haishan et al., 2015), and various Western countries (Meier et al., 2016). Thus, these two general concepts are crucial and should be considered in the care of every EOL patient.
Opinions of participants did not differ significantly between the two groups with regard to not being a burden and not receiving the intensive treatment when the chance of survival is low. This finding is similar to those of previous studies from various countries around the world (Meier et al., 2016; Miyashita et al., 2007; Yun et al., 2018). Furthermore, this may be due to physicians having a good understanding of palliative concepts, as feeling like one is a burden to others is one of the factors associated with a bad death (Johnson et al., 2007; McPherson et al., 2007). According to a prior systematic review and meta-analysis, implementation of a palliative care program instead of the usual aggressive treatment to prolong life was an important factor in improving patients’ quality of life and lowering health care utilization (Kavalieratos et al., 2016).
Interestingly, there was a 20% difference in the responses of patients and physicians to the item concerning mental awareness toward the last hour of life. A large number of older patients prioritized this issue, but physicians did not. There may be a religious explanation for this. Most Thai Buddhists believe that being conscious, aware, and thinking of positive experiences is an important component of a good death, and that this may help the soul reach heaven (Rodpal et al., 2007). However, this issue was less important to physicians, even those who were Buddhist. It may be that the physicians were less likely to emphasize religious belief or that they had observed patients experiencing suffering, such as pain and deliriousness, toward the end of their lives, which led them to place more importance on experiencing a comfortable death. Moreover, different cultural beliefs might affect the results regarding these issues. One study that explored the patients’ and physicians’ perspective of good death in Korea found that being “mentally aware” was the least important component among those assessed (Yun et al., 2018).
Physicians in this study had a tendency to prefer more aggressive treatment for their patients than they would for themselves. In addition, despite the fact that most of the physicians in this study felt it was best to act in accordance with the patients’ wishes, only 75% indicated that they would refrain from administering CPR if the patient did not want it. These results are similar to those of a previous study from the United States (Periyakoil et al., 2014). That study concluded that there were many causes that led to these kinds of differences including physicians’ practice style, the culture within the health care system, and having new innovative technology for further treatments and interventions. These factors may have led physicians to provide intensive management until they were sure that the patient had died. Moreover, physicians tended to overestimate the survival time or prognoses of their EOL patients, leading to the unnecessary use of aggressive treatment. The fact that most of the physicians had seen terminally ill patients and realized that many medical procedures had caused them suffering and distress might be why they preferred palliative and symptomatic care for their own EOL period. However, there is evidence that higher care intensity in older EOL patients could incur significantly greater costs without reducing the mortality rate (Chaudhuri et al., 2017).
Our data showed that the physicians’ attitudes toward EOL care had changed significantly from those in the 2011 study previously mentioned (Kornbongkotmas et al., 2011). First, there were increases in the proportions of physicians who wished patients to receive the truth about their illnesses and felt that patients should make decisions about their own treatment. This suggests that more participants in this study prioritized respecting patients’ autonomy compared with the prior study. In addition, the proportion of physicians who supported not administering life-prolonging treatment also increased by about 10% from that in the prior study, thus reducing the gap between patients’ and physicians’ perspectives. The reason for this may be palliative knowledge becoming more widespread in Thailand. Over the past decade, the medical curriculum has come to place greater emphasis on palliative and EOL care. Medical students and physicians-in-training are being trained in the ability to initiate discussion about advanced care plans with the patients and their relatives. Moreover, palliative service programs have been established at various levels of the health care system and at hospitals in all parts of Thailand. Bridging the gap between patients’ desires and the perspectives of health care providers may help reduce conflicting opinions regarding treatment and allow patients to achieve a good death. Moreover, this could potentially reduce the amount of nonessential treatment that patients receive and the suffering that those treatments may cause.
This study also found that more than 70% of patients wished to pass away at home. This proportion was higher than that found in a previous study conducted in central Thailand, which reported that only about half of patients wished to pass away in their homes, as well as a systematic review from 33 countries, which reported that most people preferred to die at home. In addition, another descriptive study reported that only about 40% of cancer patients preferred to die at home (Gomes et al., 2013; Waghorn et al., 2011). This discrepancy may be due to cultural differences, and the change over time in Thailand might reflect improvements in the EOL resources and services available in Thai communities. We also found that a lower proportion of physicians expected patients to want to pass away at home than patients who indicated their desire to do so. Thus, it is important that patients and physicians discuss this issue while the patients are still capable of making such decisions.
Awareness of patients’ wishes and advance directive plans are extremely important in the adequate provision of EOL care. Despite this, a previous study conducted in the United States found that 28% of people who had documented advance directives had not shared these with families or health care providers (Inoue et al., 2019), making it the responsibility of the health care providers to interpret their wishes.
Our study found that it was important for physicians to not only be aware/understand the concept of a good death, but also to possess the skills necessary to communicate effectively with their EOL patients. Therefore, the palliative care system in Thailand could be improved by providing Thai residents/fellows-in-training more experience engaging in discussion with EOL patients regarding their wishes.
Conclusion
Although the opinions of patients and physicians regarding EOL care differed in some areas, they exhibited greater similarity than in a previous study. The current scheme to promote knowledge regarding palliative and EOL care appears to have been successful. Better communication between patients and physicians may help to close any remaining gaps.
Limitations and Further Study
This study was limited to central and northeast Thailand, so the results may not represent the views of people in all areas of the country, which differ in terms of both culture and beliefs. Thus, further research conducted in other parts of the country is warranted. Moreover, the participants in the patient group were not EOL patients, and some participants had not had experience with EOL care. Thus, the results in the patient group might differ from the perspectives of terminally ill patients. In addition, the current study did not explore the reasons for the differences in physicians’ perspective of a good death and their real-life practice. Further studies should be conducted to evaluate this issue. Finally, the questionnaire in this study was developed in a Buddhist population, so its application in other cultural contexts might be limited, and its reliability and validity were not addressed.
Footnotes
Acknowledgements
The authors would like to thank Dr. Dylan Southard (Research Affairs, Faculty of Medicine, Khon Kaen University, Thailand) for editing the article. They would also like to thank the Sleep apnea research group, Khon Kaen University, Khon Kaen, Thailand for their support.
Authors’ Contributions
P.L. made a substantial contribution to the conception and design of this work, as well as to the acquisition, analysis, and interpretation of data; V.S. contributed to the conception of this study and approved the final version of the article; S.K. and K.Y. contributed to the acquisition of data; M.M. drafted the article and revised it critically for important intellectual content; contributed to the study design, analysis, and interpretation of data; and approved the version of the article to be published.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the Neuroscience Research and Development Group (Khon Kaen University, Thailand).
Research Ethics
Approval from Siriraj Hospital and the Khon Kaen University (KKU) Institutional Review Board were obtained (reference numbers Si 621/2017 and HE 601309, respectively). The KKU Ethics Committee determined that the project could be exempted based on KKU’s announcement number 1877/2559 involving survey procedures and interview procedures and the observation of public behavior, and the requirement for informed consent was waived. The Siriraj Institutional Review Board considered expedited this study and has approved the final protocol, according to which participants were informed of the study. This study was reported according to the STROBE guidelines.
