Palliative Family Conference Reduces the Risk of Death in Intensive Care Units and Cardiopulmonary Resuscitation at End of Life

Abstract

Background:

Palliative family conference (PFC) was included in the reimbursement of National Health Insurance to promote palliative care in Taiwan in 2012.

Objectives:

This study aimed to evaluate the impact of PFC on death in intensive care unit (ICU) and receiving cardiopulmonary resuscitation (CPR) within three days before death.

Design:

This is a cross-sectional study.

Subjects:

All patients who died in a public hospital and were admitted to ICU within 30 days before death, from 2013 to 2018, were included.

Measurements:

The medical records were analyzed to identify information on causes of death, receiving PFC, receiving palliative care consultation, death in ICU, and receiving CPR within three days before death. Multivariate logistic regression was used to assess the independent effects of receiving PFC on the risk of death in ICU and receiving CPR within three days before death.

Results:

For patients who died and those who did not die in ICU, the proportion of receiving PFC was 45.8% (1818/3973) and 55.0% (808/1468), respectively. For patients who received and those who did not receive CPR within three days before death, the proportion of receiving PFC was 23.9% (140/585) and 51.2% (2486/4856), respectively. PFC was associated with a reduced risk of death in ICU (adjusted odds ratio [AOR]: 0.842; 95% confidence interval [CI]: 0.717–0.988) and a reduced risk of receiving CPR within three days before death (AOR: 0.361; 95% CI: 0.286–0.456).

Conclusion:

PFC reduces the risk of receiving nonbeneficial aggressive intervention and may improve the quality of end-of-life care.

Introduction

In caring for people with advanced life-limiting illnesses, special attention should be paid to their families, with patients and their families comprising care units.¹ An important step is interdisciplinary integration and the establishment of good communication with families so as to create a bond of trust and thereby avoid conflicting information.² Effective communication can help patients clarify and resolve their problems and encourage their participation in the decision-making process.³ In caring for terminally ill persons, their families undoubtedly face socioemotional difficulties and physical exhaustion. Under certain circumstances, misunderstandings and conflicts tend to contribute to the difficulties of intrafamily relationships and interfere in the evolution of the prognosis of the patient.¹ Thus, a family conference, a moment of planned interlocution between patients, families, and health care teams, need to be conducted as a therapeutic instrument.⁴

The Taiwanese government has been actively developing palliative care since 1996.⁵ Taiwan's National Health Insurance (NHI), a universal public health insurance covering 99.5% of the population by 2010,⁶ began to cover home palliative care in 1996. Taiwan passed the “The Hospice Palliative Medical Act” (Natural Death Act) in 2000. In the same year, Taiwan NHI started covering inpatient palliative care programs, and in 2009, it expanded the coverage of palliative care for patients with advanced cancer or amyotrophic lateral sclerosis to patients with other advanced diseases.⁷ In addition to reimbursement for palliative care, the NHI began to cover palliative family conference (PFC) in 2012. Two PFC sessions can be paid for by the NHI when they fulfill the following requirements: (1) participation by the health care team and the patients and/or their family members lasting for at least one hour, (2) a detailed record describing the communication between the care team and the patient and/or family members, (3) documentation of the family conference as part of the medical records, and (4) signatures by all participants.^8–10

Life-support procedures received in the intensive care unit (ICU) may adversely affect the quality of life in the patient's final week. Preventing ICU admission in the last month of life is regarded as a parameter for high-quality end-of-life care.¹¹ However, relative to other settings, ICUs could potentially reduce suffering at this stage because ICUs are equipped with trained physicians who can access restricted drugs, medical staff of different professionals, and palliative care specialists.^12,13 A substantial proportion of cancer decedents had an ICU admission in the last 30 days of life.¹⁴

In addition to ICU admission within the last month of life, death in the ICU for patients with advanced illness warrants attention. One study reported high rates of nurse-perceived suffering and loss of dignity in patients during their final week in the ICUs of two academic care centers.¹³ Another research reported dissatisfaction with the amount and quality of care for dying patients in the ICU of a tertiary referral hospital.¹⁵ One study on oncology patients reported that patients who died in ICUs had lower scores of physical comfort than patients who died in non-ICU wards. In addition, death in ICUs was associated with an increased risk of post-traumatic stress disorder among caregivers.¹⁶ The prevention of death in ICUs for patients with end-stage life-limiting diseases is an important quality indicator of end-of-life care.¹⁷

Cardiopulmonary resuscitation (CPR) can be lifesaving for patients whose cardiac arrests resulted from reversible causes. However, the outcomes are poor for many patients. Survival-to-hospital discharge rates are less than 20% for in-hospital arrests¹⁸ and less than 10% for out-of-hospital cardiac arrests.¹⁹ For patients with advanced life-limiting diseases, the sudden halt of their heartbeat may be a part of the natural dying process. Distinguishing patients for whom CPR may be advantageous or may just prolong suffering is necessary.²⁰ CPR administration within 30 days before death is considered a poor quality of death.²¹ Performing an aggressive and unsuccessful resuscitation procedure toward the end of a person's natural life could result in the loss of dignity and extend their suffering. In comparison with receiving CPR within 30 days before death, receiving CPR within three days before death might imply an even worse quality of death for those with advanced life-limiting diseases.

In ICUs, PFCs have become increasingly important in end-of-life care planning because many patients in such cases lose their decision-making capacity.²² Through structured communication among patients, their families, and health care teams, PFCs may provide an opportunity to improve the quality of care provided to dying patients and their families. In the current study, we aim to evaluate the impact of PFCs on death in ICUs and CPR administration within three days before death among patients who died in hospitals and were admitted to ICUs within 30 days before death.

Materials and Methods

Study design and data source

This work is a cross-sectional study that uses the medical records of Taipei City Hospital (TCH). Electronic medical records were analyzed after ethics approval of the Institutional Review Board was obtained.

In addition to the organization of integrated multidisciplinary palliative care teams to provide palliative care for patients with terminal diseases since January 2015,²³ TCH has conducted a series of training projects to promote palliative care and PFCs for all health care providers. The projects include the evaluation of patients' palliative care needs, education on how to provide palliative care for patients with terminal illnesses, and the development of communication skills for PFCs. All health care providers are encouraged to attend these palliative training projects before they conduct PFCs. TCH held palliative training projects 13 times from 2015 to 2016. Of the 3299 health care providers at TCH during this period, 1912 individuals (58.0%) completed the palliative training projects.⁸

Ethics approval and consent to participate

This study was approved by the TCH Research Ethics Committee, and the approval number is TCHIRB-10807019-E. This study is a retrospective review of medical records, and consent to participate was waived by the TCH Research Ethics Committee.

Study participants

We included patients who died of end-stage cancers and other illnesses (including motor neuron diseases, brain diseases, dementia, heart failure, lung diseases, liver failure, and renal failure) in TCH and were admitted to the ICU within 30 days before death from 2013 to 2018.

Measurements

The medical records were analyzed to identify information on each patient's sex, age at death, year of death, causes of death, ICU admission within 30 days before death, receipt of PFC, receipt of palliative care consultation, death in ICU, and receipt of CPR within three days before death. Information on receiving PFC and receiving palliative care consultation was identified with relevant documentation on the medical records. CPR administration was defined as receiving cardioversion and/or cardiac massage according to relevant documentation on the medical records.

Palliative family conference

The family members and patients, if they had clear consciousness and were able to communicate appropriately, were notified about the clinical conditions that called for a discussion of the options for end-of life care in a family meeting setting. A PFC was held after obtaining consent from the patients or their families. The health care teams in the ICU held a family conference to discuss options for end-of-life care. In such a conference, they discussed current clinical conditions and prognoses, plans and goals of treatment, feasibility of receiving palliative care, do-not-resuscitate instructions, withholding and withdrawal of life-sustaining treatment, and advance directives. The scheme of PFCs is illustrated in Figure 1.

FIG. 1.

Scheme of palliative family conference.

Data analysis

The proportions of patients who received PFCs are calculated according to various patient characteristics. Multivariate logistic regression with adjustment for sex, age at death, year of death, cause of death, and receipt of palliative care consultation (for the outcome of death in the ICU, and receiving CPR within three days before death) was used to assess the independent effects of receiving PFCs on the odds of receiving palliative care consultation, risk of death in the ICU, and risk of receiving CPR within three days before death. Data were analyzed using SAS 9.4 software. A p-value <0.05 was considered statistically significant.

Results

Table 1 shows the proportions of patients who received PFCs according to various patient characteristics. For patients who died in the years 2013–2014 and those who died in the years 2015–2018, the proportions of those who received PFCs were 3.8% (65/1730) and 69.0% (2561/3711), respectively. For patients who died of cancer and those who died of noncancer diseases, the proportions of those who received PFCs were 51.8% (270/521) and 47.9% (2356/4920), respectively. For patients who received and those who did not receive palliative care consultation, the proportions of those who received PFCs were 78.1% (421/539) and 45.0% (2205/4902), respectively. For patients who died and those who did not die in the ICU, the proportion of those who received PFCs were 45.8% (1818/3973) and 55.0% (808/1468), respectively. Among the patients who received and those who did not receive CPR within three days before death, the proportions of those who received PFCs were 23.9% (140/585) and 51.2% (2486/4856), respectively.

Table 1.

The Proportion of Receiving Palliative Family Conference According to Various Patient Characteristics (N = 5441)

	PFC				p ^a
	Yes		No
	n	%	n	%
Sex					0.000
Female	1134	51.2	1083	48.8
Male	1492	46.3	1732	53.7
Age (year)					0.000
<65	389	42.7	521	57.3
65–74	374	49.9	376	50.1
75–84	730	46.2	851	53.8
>84	1133	51.5	1067	48.5
Year					0.000
2013–2014	65	3.8	1665	96.2
2015–2018	2561	69.0	1150	31.0
Cause of death					0.048
Cancer	270	51.8	251	48.2
Noncancer	2356	47.9	2564	52.1
Palliative care consultation					0.000
Yes	421	78.1	118	21.9
No	2205	45.0	2697	55.0
Death in ICU					0.000
Yes	1818	45.8	2155	54.2
No	808	55.0	660	45.0
CPR within three days of death					0.000
Yes	140	23.9	445	76.1
No	2486	51.2	2370	48.8

Based on chi-square test.

CPR, cardiopulmonary resuscitation; ICU, intensive care unit; PFC, palliative family conference.

The odds of receiving palliative care consultation are presented in Table 2. Compared with patients who died in the years 2013–2014, those who died in the years 2015–2018 had higher odds of receiving palliative care consultation (adjusted odds ratio [AOR]: 1.696; 95% confidence interval [CI]: 1.211–2.376). Compared with patients who died of cancer, those who died of noncancer diseases had lower odds of receiving palliative care consultation (AOR: 0.274; 95% CI: 0.216–0.348). PFC was associated with increased odds of receiving palliative care consultation (AOR: 3.326; 95% CI: 2.549–4.340).

Table 2.

The Risk of Receiving Palliative Care Consultation According to Various Characteristics (N = 5441)

	Patient number	Palliative care consultation		Risk
	Patient number	n	%	OR	95% CI	p	AOR^a	95%CI	p
Sex
Female	2217	243	11.0	1.000			1.000
Male	3224	296	9.2	0.821	0.687–0.982	0.031	0.850	0.704–1.025	0.090
Age (year)
<65	910	89	9.8	1.000			1.000
65–74	750	74	9.9	1.010	0.730–1.397	0.953	0.835	0.595–1.173	0.299
75–84	1581	157	9.9	1.017	0.773–1.337	0.904	0.988	0.741–1.316	0.933
>84	2200	219	10.0	1.020	0.787–1.322	0.882	0.973	0.740–1.278	0.842
Year
2013–2014	1730	64	3.7	1.000			1.000
2015–2018	3711	475	12.8	3.821	2.924–4.993	0.000	1.696	1.211–2.376	0.002
Cause of death
Cancer	521	125	24.0	1.000			1.000
Noncancer	4920	414	8.4	0.291	0.232–0.364	0.000	0.274	0.216–0.348	0.000
Palliative family conference
No	2815	118	4.2	1.000			1.000
Yes	2626	421	16.0	4.364	3.531–5.393	0.000	3.326	2.549–4.340	0.000

With adjustment of sex, age at death, year of death, cause of death.

AOR, adjusted odds ratio; CI, confidence interval; OR, odds ratio.

The risk of death in the ICU is shown in Table 3. Compared with patients <65 years of age, those who were 65–74 years of age (AOR: 0.710; 95% CI: 0.556–0.905), 75–84 years of age (AOR: 0.694; 95% CI: 0.563–0.857), and >84 years of age (AOR: 0.450; 95% CI: 0.369–0.549) had lower risks of death in the ICU. Compared with patients who died of cancer, those who died of noncancer diseases had a higher risk of death in the ICU (AOR: 1.792; 95% CI: 1.468–2.188). PFC (AOR: 0.842; 95% CI: 0.717–0.988) and palliative care consultation (AOR: 0.244; 95% CI: 0.201–0.296) were associated with reduced risks of death in the ICU.

Table 3.

The Risk of Death in Intensive Care Units According to Various Characteristics (N = 5441)

	Patient number	Death in ICU		Risk
	Patient number	N	%	OR	95% CI	p	AOR^a	95% CI	p
Sex
Female	2217	1574	71.0	1.000			1.000
Male	3224	2399	74.4	1.188	1.02–1.341	0.005	1.084	0.955–1.231	0.213
Age (year)
<65	910	739	81.2	1.000			1.000
65–74	750	563	75.1	0.697	0.551–0.881	0.003	0.710	0.556–0.905	0.006
75–84	1581	1191	75.3	0.707	0.577–0.865	0.001	0.694	0.563–0.857	0.001
>84	2200	1480	67.3	0.476	0.394–0.574	0.000	0.450	0.369–0.549	0.000
Year
2013–2014	1730	1318	76.2	1.000			1.000
2015–2018	3711	2655	71.5	0.786	0.689–0.896	0.000	1.041	0.875–1.238	0.649
Cause of death
Cancer	521	309	59.3	1			1.000
Noncancer	4920	3664	74.5	2.000	1.661–2.410	0.000	1.792	1.468–2.188	0.000
Palliative family conference
No	2815	2155	76.6	1.000			1.000
Yes	2626	1818	69.2	0.689	0.611–0.777	0.000	0.842	0.717–0.988	0.036
Palliative care consultation
No	4902	3748	76.5	1.000			1.000
Yes	539	225	41.7	0.221	0.184–0.265	0.000	0.244	0.201–0.296	0.000

With adjustment of sex, age at death, year of death, cause of death, receiving palliative care consultation.

The results regarding the risk of receiving CPR within three days before death are listed in Table 4. Compared with female patients, male patients had a higher risk of receiving CPR within three days before death (AOR: 1.253; 95% CI: 1.037–1.513). Compared with patients <65 years of age, those who were 75–84 years of age (AOR: 0.634; 95% CI: 0.499–0.805) and >84 years of age (AOR: 0.349; 95% CI: 0.272–0.448) had lower risks of receiving CPR within three days before death. Compared with patients who died of cancer, those who died of noncancer diseases had a higher risk of receiving CPR within three days before death (AOR: 1.706; 95% CI: 1.178–2.469). PFC (AOR: 0.361; 95% CI: 0.286–0.456) and palliative care consultation (AOR: 0.128; 95% CI: 0.057–0.290) were associated with reduced risks of receiving CPR within three days before death.

Table 4.

The Risk of Receiving Cardiopulmonary Resuscitation within Three Days before Death According to Various Characteristics (N = 5441)

	Patient number	Receiving CPR within three days before death		Risk
	Patient number	N	%	OR	95% CI	p	AOR^a	95% CI	p
Age
Female	2217	192	8.7	1.000			1.000
Male	3224	393	12.2	1.464	1.221–1.756	0.000	1.253	1.037–1.513	0.019
Age (year)
<65	910	158	17.4	1.000			1.000
65–74	750	109	14.5	0.809	0.620–1.056	0.119	0.905	0.688–1.191	0.478
75–84	1581	178	11.3	0.604	0.479–0.761	0.000	0.634	0.499–0.805	0.000
>84	2200	140	6.4	0.323	0.254–0.412	0.000	0.349	0.272–0.448	0.000
Year
2013–2014	1730	272	15.7	1.000			1.000
2015–2018	3711	313	8.4	0.494	0.415–0.587	0.000	0.943	0.767–1.160	0.579
Cause of death
Cancer	521	34	6.5	1.000			1.000
Noncancer	4920	551	11.2	1.805	1.263–2.584	0.000	1.706	1.178–2.469	0.000
Palliative family conference
No	2815	445	15.8	1.000			1.000
Yes	2626	140	5.3	0.300	0.246–0.366	0.000	0.361	0.286–0.456	0.000
Palliative care consultation
No	4902	579	11.8	1.000			1.000
Yes	539	6	1.1	0.084	0.037–0.189	0.000	0.128	0.057–0.290	0.000

With adjustment of sex, age at death, year of death, cause of death, receiving palliative care consultation.

Discussion

Among the patients who died in the hospital and were admitted to the ICU within 30 days before death, those who died in the years 2015–2018 received more PFCs than the patients who died in the years 2013–2014 (69.0% vs. 3.8%). PFCs were associated with increased odds of receiving palliative care consultation (AOR: 3.326; 95% CI: 2.549–4.340), reduced risk of death in the ICU (AOR: 0.842; 95% CI: 0.717–0.988), and reduced risk of receiving CPR within three days before death (AOR: 0.361; 95% CI: 0.286–0.456).

TCH created integrated multidisciplinary palliative care teams to promote palliative care among patients with advanced life-limiting diseases. This hospital has been actively integrating palliative care into public health policies and practices since January 2015.²³ The present study found a substantial increase in the proportion of patients receiving PFCs after the implementation of multidisciplinary palliative care.

In our study, PFCs were associated with increased odds of receiving palliative care consultation. Being understanding about the needs, doubts, anxieties, and fears of patients and their families is important to establish a good orientation in care planning. Effective communication can help patients clarify and resolve their problems and encourage their active participation in the decision-making process and in the search for alternative solutions to problems.³ Through structured communication, PFCs can provide an opportunity to improve the quality of end-of-life care for dying patients and their families.

Previous studies reported that PFCs resulted in an improvement in the quality of care with the implementation of individualized plans of care with consideration on the values and preferences of patients and their families.²⁴ Adjustments in the care plan can be made according to the evolution of the clinical picture and patient acceptance.²⁵ Studies in a university tertiary care hospital reported that intensive communication among the care team, patients, and their families held within 72 hours of ICU admission remarkably reduced the median length of ICU stay and was associated with a reduced need for critical care by patients who died.^26,27 In our study, PFCs were associated with a reduced risk of death in the ICU.

Previous studies also reported an inconsistency in the preferences for end-of-life care between patients and their family caregivers.^28,29 One study in Taiwan reported that the overall consistency was 42.28% for preferences for end-of-life life-sustaining medical treatment, including CPR, intravenous therapy, nasogastric feeding tube, ICU admission, blood transfusion, tracheostomy, and hemodialysis, between hospitalized elderly patients and their primary family caregivers. With regard to the decision on CPR at the end-of-life stage, the percentage of patients who chose “Yes,” “No,” and “Not sure” were 3%, 88%, and 9%, respectively. The corresponding figures for their family care givers were 16%, 48%, and 36%.²⁸

Given that severely ill patients lose their decision-making capacity and that the wishes of family members and physicians may not be consistent with those of the patients,^28,29 discussion in advance about do-not-resuscitate orders should be promoted to respect patients' autonomy. For patients in ICUs, decisions on resuscitation are of paramount importance and should be clear to guide the resuscitation or do-not-resuscitate plan in case of cardiac arrest.³⁰ When no definitive directive is provided, that is, no do-not-resuscitate decision was documented, the standard procedure is to perform CPR regardless of whether this intervention is consistent with patient values.³¹ PFCs enhance effective communication, and the documentation of the decisions on resuscitation in the event of cardiac arrest could reduce the risk of receiving useless CPR toward the end of life.¹

One study on health care cost reduction by implementing PFCs with the decision to withdraw life-sustaining treatments reported that PFCs within seven days of ICU admission were associated with an increased likelihood of signing a do-not-resuscitate order and an increased likelihood of the decision to withdraw life-sustaining treatments.⁹ One study in Taiwan reported that PFCs reduced the likelihood of receiving life-sustaining treatments during the last three months of life.⁸ In our study, PFCs were associated with a reduced risk of receiving CPR within three days before death.

This study has several limitations. First, the data were based on a single hospital, and the results could not be generalized to hospitals with different characteristics. Further studies that integrate the situations in other hospitals could comprehensively demonstrate the effects of PFCs. Second, some socioeconomic characteristics and preferences for end-of-life care of patients and their families were not evaluated; thus, the results might be confounded. Third, although we adjusted cause of death to assess the independent effects of PFCs on outcomes, the clinical conditions of patients could vary. Considering the similarities in disease severity, we included patients who died in the hospital and were admitted to the ICU within 30 days before death in our study. Last, we have no information about the willingness, preparedness, and understanding of participation in PFCs among patients and their families. In addition, the experience and skills required for PFCs among health care teams were not evaluated. These drawbacks may compromise the success of PFCs. Further research should explore the facilitators and barriers of PFCs and identify the factors associated with successful PFCs.

Conclusions

Among the patients who died in TCH and were admitted to the ICU within 30 days before death, PFCs were associated with a reduced risk of death in the ICU and a reduced risk of receiving CPR within three days before death. Through effective communication among patients, their families, and health care teams, PFCs may reduce the risk of receiving nonbeneficial aggressive intervention and provide an opportunity to improve the quality of end-of-life care.

Footnotes

Authors' Contributions

Study concept and design: P.-Y.K., M.-C.K., and S.-J.H.; Acquisition of data: H.-Y.L., C.-M.H., and C.-Y.T.; Analysis and interpretation of data: P.-Y.K., H.-Y.L., C.-M.H., M.-C.K., and S.-J.H.; Drafting of the article: P.-Y.K., C.-C.C., M.-C.K., and S.-J.H.; and Critical revision of the article: P.-Y.K., M.-C.K., L.-C.W., and S.-J.H.

Funding Information

This study was supported by a grant from the Department of Health, Taipei City Government (Grant No.: 10801-62-009). The Department of Health, Taipei City Government, had no role in conducting the study and interpreting the results.

Author Disclosure Statement

No competing financial interests exist.

References

Silva

RSD

, Trindade

GSS

, Paixão

GPDN

, et al.: Family conference in palliative care: Concept analysis. Rev Bras Enferm, 2018; 71:206–213.

Galriça Neto I and Trindade

: Family meetings as a means of support for patients. Eur J Palliat Care, 2007; 14:105–108.

Araújo

MMT

and Silva

MJP

: Communication strategies used by health care professionals in providing palliative care to patients. Rev Esc Enferm USP, 2012; 46:626–632.

Fineberg

: Preparing professionals for family conferences in palliative care: Evaluation results of an interdisciplinary approach. J Palliat Med, 2005; 8:857–866.

Tang

, Chen

, Huang

, et al.: Hospice utilization in Taiwan by cancer patients who died between 2000 and 2004. J Pain Symptom Manag, 2007; 33:446–453.

Hsing

and Ioannidis

: Nationwide population science: Lessons from the Taiwan National Health Insurance Research Database. JAMA Intern Med, 2015; 175:1527–1529.

Kang

, Lin

, Hwang

, et al.: Impact of hospice care on end-of-life hospitalization of elderly patients with lung cancer in Taiwan. J Chin Med Assoc, 2012; 75:221–226.

Yen

, Lee

, Hu

, et al.: An interventional study for the early identification of patients with palliative care needs and the promotion of advance care planning and advance directives. J Pain Symptom Manag, 2020; 59:974–982.

Chang

, Jerng

, and Chen

: Reduction of healthcare costs by implementing palliative family conference with the decision to withdraw life-sustaining treatments. J Formos Med Assoc, 2020; 119:34–41.

10.

National Health Insurance Administration: Taiwan Payment of advance care planning meeting. 2015. https://www.nhi.gov.tw/Content_List.aspx?n=BC4B6B42238D5D7A&topn=D39EB72BOBDFA15 (Last accessed November 28, 2021).

11.

Zhang

, Nilsson

, and Prigerson

: Factors important to patients' quality of life at the end of life. Arch Intern Med, 2012; 172:1133–1142.

12.

Angus

and Truog

: Toward better ICU use at the end of life. JAMA, 2016; 315:255–256.

13.

, Lief

, Berlin

, et al.: Beyond pain: Nurses' assessment of patient suffering, dignity, and dying in the intensive care unit. J Pain Symptom Manage, 2018; 55:1591–1598.e1

14.

Bekelman

, Halpern

, Blankart

, et al.: Comparison of site of death, health care utilization, and hospital expenditures for patients dying with cancer in 7 developed countries. JAMA, 2016; 315:272–283.

15.

Choi

, Park

, Kang

, et al.: The quality of dying and death for patients in intensive care units: A single center pilot study. Crit Care Med, 2019; 34:192–201.

16.

Wright

, Keating

, Balboni

, et al.: Place of death: Correlations with quality of life of patients with cancer and predictors of bereaved caregivers' mental health. J Clin Oncol, 2010; 28:4457–4464.

17.

, Ko

, Huang

, et al.: Trends of utilization of palliative care and aggressive end-of-life care for patients who died of cancers and those who died of noncancer diseases in hospitals. J Pain Symptom Manage, 2020; 60:1136–1143.

18.

Meaney

, Nadkarni

, Kern

, et al.: Rhythms and outcomes of adult in-hospital cardiac arrest. Crit Care Med, 2010; 38:101–108.

19.

Perkins

and Cooke

: Variability in cardiac arrest survival: The NHS Ambulance Service Quality Indicators. Emerg Med J, 2012; 29:3–5.

20.

British Medical

Association

, Resuscitation Council (UK), Royal College of

Nursing

: Decisions Relating to Cardiopulmonary Resuscitation. A Joint Statement from the British Medical Association, the Resus-citation Council (UK) and the Royal College of Nursing. J Med Ethics, 2001; 27:310–316.

21.

Earle

, Neville

, Landrum

, et al.: Evaluating claims-based indicators of the intensity of end-of-life cancer care. Int J Qual Health Care, 2005; 17:505–509.

22.

Rabow

, Hauser

, and Adams

: Supporting family caregivers at the end of life: “They don't know what they don't know.”. JAMA, 2004; 291:483–491.

23.

Huang

, Huang

, Woung

, et al.: The 2017 Taipei declaration for health-promoting palliative care. J Palliat Med, 2018; 21:581–582.

24.

Hudson

, Quinn

, O'Hanlon

, et al.: Family meetings in palliative care: Multidisciplinary clinical practice guidelines. BMC Palliat Care, 2008; 7:1–12.

25.

Powazki

, Walsh

, Hauser

, et al.: Communication in palliative medicine: A clinical review of family conferences. J Palliat Med, 2014; 17:1167–1177.

26.

Lilly

, De Meo

, Sonna

, et al.: An intensive communication intervention for the critically ill. Am J Med, 2000; 109:469–475.

27.

Lilly

, Sonna

, Haley

, et al.: Intensive communication: Four-year follow-up from a clinical practice study. Crit Care Med, 2003; 31:S394–S399.

28.

Chuang

, Shyu

, Weng

, et al.: Consistency in end-of-life care preferences between hospitalized elderly patients and their primary family caregivers. Patient Prefer Adherence, 2020; 14:2377–2387.

29.

Chan

HYL

, Ng

JSC

, Chan

, et al.: Effects of a nurse-led post-discharge advance care planning programme for community-dwelling patients nearing the end of life and their family members: A randomised controlled trial. Int J Nurs Stud, 2018; 87:26–33.

30.

Cook

, Rocker

, Marshall

, et al.: Levels of care in the intensive care unit: A research program. Am J Crit Care, 2006; 15:269–279.

31.

Teixeira

and Cardoso

PRC

: How to discuss about do-not-resuscitate in the intensive care unit?. Rev Bras Ter Intensiva, 2019; 31:386–392.