Effects of Brief Nurse Advance Care Planning Intervention with Visual Materials on Goal-of-Care Preference of Japanese Elderly Patients with Chronic Disease: A Pilot Randomized-Controlled Trial

Abstract

Purpose:

Advance care planning is an important component of quality palliative care. In Asian countries, few randomized clinical trials have been reported. This pilot randomized-controlled trial examined the effects of brief nurse intervention with visual materials on the goal-of-care preference, cardiopulmonary resuscitation (CPR) preference, and designation of a health care proxy.

Methods:

This randomized clinical trial was performed from January to February 2018 on elderly Japanese patients with chronic disease. The patients were randomly assigned to a control group (brief nurse intervention using verbal descriptions) or intervention group (using visual materials). The primary endpoint was goal-of-care preference, and secondary outcomes included the following: (1) CPR preference, (2) presence of a designated health care proxy, (3) knowledge of CPR, and (4) readiness for advance care planning. Outcome measures were obtained at baseline and just after completion of the intervention.

Results:

A total of 220 patients were enrolled (117 in the intervention group and 103 in the control group). All patients completed post-intervention measurement. There was no significant difference between the groups in any of the outcome measures, while <5% of the participants wanted life-prolonging care as the goal of care at the baseline. Before/after comparisons indicated that, in both groups, the number of participants who designated a health care proxy significantly increased (29% to 65% vs. 22% to 52%, respectively; p < 0.001 each); and the knowledge and readiness scores significantly increased. Moreover, there was a significant increase in the number of patients who did not want CPR (55% to 67% with a terminal condition, p = 0.003; 67% to 80% with a bedridden condition, p < 0.001) in the intervention group.

Conclusions:

Brief nurse intervention increased documentation of a patient-designated health care proxy and improved the knowledge of CPR and patient readiness. Visual materials might help patients to imagine the actual situation regarding CPR.

Introduction

Advance care planning is an important component of quality palliative care.^1,2 In Western countries, multiple randomized-controlled trials demonstrated that a variety of advance care planning interventions are beneficial for patients.^3–6 The value of advance care planning has been recognized in Asian countries, and Taiwan and Korea have legalized the advance directive and health care proxy; but there have been few randomized clinical trials with a scientifically sound methodology.^7–10 Empirical research to identify the types of interventions and changes in outcomes they lead to is urgently needed for implementation.^7–10

This pilot randomized-controlled trial aimed to obtain information to plan a future clinical trial to investigate the effects of advance care planning intervention. As we consider the use of video as one attractive option to facilitate advance care planning,^11–17 we examined the effects of brief nurse intervention with versus without visual materials on multiple key outcomes related to advance care planning interventions, including the goal-of-care preference, CPR preference, and health care proxy, in elderly Japanese patients with chronic disease. This study was designed to compare the outcomes between groups, but we also aimed to examine whether this type of nurse intervention is feasible and whether the intervention improved outcomes based on before/after comparisons within the group. Our main hypothesis tested in this study was whether this brief nurse intervention with visual materials improved goal-of-care preference, CPR preference, and health care proxy in elderly Japanese patients with chronic disease, compared with the intervention without visual materials.

Subjects and Methods

This randomized clinical trial was performed involving elderly Japanese patients with chronic disease from January to February in 2018. The patients were randomly assigned to a control group (brief nurse intervention using verbal descriptions) or intervention group (using visual materials). Outcomes were obtained at the baseline and just after completion of the intervention using a web-based questionnaire. The Institutional Review Board of the Seirei Mikatahara General Hospital approved the ethical and scientific validity of this study (2016-1637). Written consent was obtained by all participants.

Participants

Inclusion criteria for the patients were as follows: (1) age of 65 years or older, (2) receiving regular medical treatment by general practice or hospital physicians for at least one chronic illness, including cardiovascular disease, malignant disease, pulmonary disease, or renal disease, and (3) capable of understanding and speaking Japanese. Treatment setting was thus primary care outpatient setting.

The participants were recruited from a commercial database (Rakuten research). This database consists of >1,000,000 Japanese who voluntarily registered, and participants were invited via the Internet by the Rakuten Cooperation if they met the inclusion criteria of the study. Each patient received a small incentive. Although we were aware of a potential bias related to the sampling, that is, participants were probably well educated and not so seriously ill, we decided to use this sample due to the pilot nature of this study.

Measurements

The primary endpoint was documentation of the goal-of-care preference, and secondary outcomes included the following: (1) CPR preference, (2) documentation of designated health care proxy, (3) knowledge of CPR, and (4) readiness for advance care planning. We determined these outcomes according to previous studies with some modifications suitable for Japanese health care systems based on a pilot test.^12–18

Goal-of-care preference included three choices: (1) life-prolonging care (aiming to prolong life at any cost and include all potentially indicated medical care such as CPR, intubation, mechanical ventilation, and care in an intensive care unit); (2) limited medical care (aiming to prolong life if pain and suffering are acceptable, but treatment would be withdrawn under patient-designated specific situations such as unacceptable pain and suffering, uncommunicable state, or totally dependent in daily activities); and (3) comfort care (aiming to maximize comfort and alleviate suffering).^12–18 The statement that “hospitalization, intravenous fluids, and antibiotics are not included in comfort care” in the original question^12–18 was deleted, because such medical intervention is frequently performed as a minimum standard of care even in homecare services and inpatient hospices in Japan. “Basic medical care” in the original question was thus changed to “limited medical care” as described above.

CPR preference was rated as “do not want,” “want,” or “unsure.” In these preferences, the participants were asked to rate their preference based on the situation of (1) current health care status, (2) terminally ill condition (incurable disease, estimated prognosis of one year), and (3) bedridden (incurable disease, estimated prognosis of one year, and totally dependent in daily activities).

The presence of a designated health care proxy was rated as present or absent. Knowledge of CPR was measured using five questions from Volandes's study.¹² In the original measure, six questions were used, but the pilot test suggested one question (question 4: comfort care is a type of medical care that can only be provided for cancer patients living in a hospice) was difficult to answer for our population, because the term hospice means inpatient hospice in Japan; therefore, we used the remaining five questions. The knowledge score was calculated as the total of correct answers to five questions, and thus ranged from 1 to 5, with a higher score indicating greater knowledge. Cronbach's alpha was 0.81.

Readiness for advance care planning was measured using a scale consisting of four questions (are you ready to think by yourself/talk with your family about the goal of care, CPR, designated proxy, and preferred place of end-of-life care) using a 4-point Likert-type scale from 1 (not ready) to 4 (completely ready). The readiness score was calculated as the mean of 4 items, and thus ranged from 1 to 4, with higher scores indicating greater willingness. Cronbach's alpha was 0.94.

Procedures and intervention

All interviews were conducted by trained six nurses. All nurses had five or more clinical experiences in oncology, and one of them was a certified oncology nurse. Each nurse had received four-hour interactive education, including actual interaction with a participant before the study from one of the authors.

As the participants were recruited from the Internet, the intervention was provided in a private room of a public conference facility in a public conference at the day that the participants were available. After informed consent was obtained, participants underwent a baseline assessment. Patient characteristics investigated were: sex, age, education, employment status, living status (alone or not), personal experience of hospital admission, personal experience of caring for a dying person, and presence of written advance directives. Randomization was performed based on a list, stratified by sex, generated by a computer randomization program. No blinding was performed.

The participants received a 30-minute (both groups) face-to-face intervention from the nurses. The conversation was performed one-to-one (i.e., no family members being present). None of the encounters was connected to the actual patients' hospitals or home physicians (i.e., patients' reports were not fed back to their physicians). In a brief introduction section, the interviewer explained why advance care planning was valuable (i.e., possibility of unexpected change in physical condition and that patients usually lose their chance of communicating their preference; advance care planning may increase the chance of receiving their preferred medical care). Thereafter, the concepts of the three types of goal of care (described as above) and details of a designated health care proxy and life-sustaining treatment such as CPR (what was carried out and efficacy) were explained. The description is included as Appendix A1. After all interactions were completed, the research nurses asked the participants again to clarify their goal-of-care preference, CPR preference, and whether they designated a health care proxy. During all processes, questions and interaction for clarification, as well as assistance in using a web interface, were permissible.

Control group

In the control group, all information was given verbally using the standardized verbal descriptive (Appendix A1). We decided to adopt this type of intervention as a control group, because the intervention was not connected to actual clinical activities and setting a “no intervention” group as usual care was impossible.

Intervention group

In the intervention group, visual information was added to the verbal description using a PowerPoint presentation. The goal-of-care slide pictured a patient in an intensive care unit on a ventilator as a life-prolonging care image, and a comfortable patient with family being cared for at home as the comfort care image. The CPR slide showed the clinical situation of a patient receiving CPR and intubation.

Statistical analyses

The sample size needed for this study was 100 per group to detect a difference of 55% versus 30% in a terminal condition in the primary endpoint goal-of-care preference (comfort) with alpha = 0.05 and power = 80%.¹⁴

For comparisons, categorical endpoints were collapsed into two categories (goal-of-care preference: comfort vs. life-prolonging/limited medical care; CPR preference: do not want vs. want/unsure; presence of designated health care proxy: presence vs. absence). We categorized patients who reported being unsure regarding their CPR preference into those who reported “want,” because they would receive CPR as a default option.

First, demographic data were compared between the intervention and control groups using Student's t test or the chi-square test where appropriate. Second, for primary endpoints (i.e., goal-of-care preference), group differences were compared using post-intervention data by the chi-square test. As there was imbalance in age between groups, we conducted post hoc Cochran–Mantel–Haenszel tests that adjust for age and obtained essentially the same results (data not shown for secondary endpoints). Third, for secondary endpoints, changes from baseline to post-intervention were analyzed using the McNemar test for categorical variables and nonpaired Student's t test for continuous variables. In addition, before/after comparisons were performed for all endpoints using the McNemar test for categorical variables and paired Student's t test for continuous variables.

Due to the exploratory nature of this study, p-values of 0.05 were regarded as significant; no adjustment for multiple comparisons was performed. All analyses were performed using the SAS software, version 9.4 (SAS Institute, Cary, NC).

Results

In total, 220 patients were enrolled (117 in the intervention group and 103 in the control group). All patients completed post-intervention measurement. The participants' background was generally balanced (Table 1): ∼80% were male with high-level education culminating in college/university. The average age was 67.1 in the control group and 69.5 in the intervention group, with statistical significance.

Table 1.

Participants' Background

	Intervention arm (n = 117)		Control arm (n = 103)		p
	Mean	SD	Mean	SD	p
Age	69.5	3.6	67.1	6.0	<0.001
	n	%	n	%
Sex					0.32
Male	94	80.3	88	85.4
Female	23	19.7	15	14.6
Education					0.11
High school/elementary school	23	19.7	12	11.7
University/college	94	80.3	91	88.3
Employment status					0.80
Employed	88	75.2	79	76.7
Living status					0.97
Alone	15	12.8	13	12.6
Experience of admission	94	80.3	81	78.6	0.76
Experience of caring for a dying	47	40.2	36	36.0	0.43
Primary disease					0.094
Cardiovascular disease	92	78.6	93	90
Cancer	20	17.1	7	6.8
Pulmonary disease	3	2.6	1	1
Renal disease	2	1.7	2	1.9
Written advance directives	7	6.0	10	9.7	0.30

SD, standard deviation.

There was no significant group difference in the post-intervention goal-of-care preference between the groups (Table 2). Adjustment for age obtained the same results: p = 0.330 (current condition), p = 0.884 (terminal condition), and p = 0.703 (bedridden condition). At the baseline, <5% of the participants wanted life-prolonging care in terminal and bedridden situations.

Table 2.

Changes in Goal-of-Care Preference

	Intervention arm (n = 117)					Control arm (n = 103)					p^**
	Preintervention		Post-intervention		p^*	Preintervention		Post-intervention		p^*
	n	%	n	%	p^*	n	%	n	%	p^*
Current condition					<0.001					0.003	0.71
Life-prolonging	6	5.1	14	12.0		3	2.9	7	6.8
Limited medical care	42	35.8	60	51.3		35	34.3	53	51.4
Comfort care	61	52.1	41	35.0		58	56.3	43	41.7
Terminal condition					<0.074					0.18	0.86
Life-prolonging	4	3.4	5	4.3		2	1.9	0	0
Limited medical care	19	16.2	25	21.4		18	17.4	27	26.2
Comfort care	94	80.3	85	72.6		81	78.6	75	72.8
Bedridden condition					<0.82					0.83	0.99
Life-prolonging	1	0.9	3	2.6		0	0	0	0
Limited medical care	23	19.7	18	15.3		19	18.4	19	18.4
Comfort care	90	76.9	91	77.8		82	79.6	83	80.6

Before/after.

Intergroup.

Regarding CPR preference, the group difference did not reach statistical significance (+12% vs. 0%, p = 0.063 in a terminal condition; +14% vs. 3.9% in a bedridden condition, p = 0.057; Table 3). In the intervention group, there was a significant increase in the number of patients who did not want CPR (55% to 67% in a terminal condition, p = 0.003; 67% to 80% in a bedridden condition, p < 0.001).

Table 3.

Changes in CPR Preference, Designated Health Care Proxy, Knowledge Score, and Willingness Score for ACP

	Intervention arm (n = 117)						Control arm (n = 103)						p^**
	Preintervention		Post-intervention			p^*	Preintervention		Post-intervention			p^*
	n	%	n	%	⊿	p^*	n	%	n	%	⊿	p^*
Do not want CPR
Current	18	15.4	23	19.7	4.3	0.059	12	11.7	20	19.4	7.7	0.033	0.51
Terminal condition	64	54.7	78	66.7	12	0.003	61	59.2	61	59.2	0	1.0	0.063
Bedridden	78	66.7	94	80.3	13.6	<0.001	78	75.7	82	79.6	3.9	0.29	0.057
Proxy designated	34	29.1	76	65	35.9	<0.001	23	22.3	54	52.4	30.1	<0.001	0.71
	Mean	SD	Mean	SD	⊿	p ^*	Mean	SD	Mean	SD	⊿	p ^*	p
Knowledge score (0–5)	3.7	1.1	4.6	0.7	0.9	<0.001	3.4	1.2	4.5	0.7	1.1	<0.001	0.19
Willingness score for ACP (1–4)
By themselves	3.2	0.52	3.5	0.55	0.26	<0.001	3.2	0.57	3.4	0.59	0.17	<0.001	0.18
With families	3.3	0.61	3.5	0.56	0.18	<0.001	3.4	0.62	3.5	0.39	0.14	0.001	0.51

Before/after.

Intergroup.

ACP, advance care planning; CPR, cardiopulmonary resuscitation.

A statistically significant difference was not observed in any of the remaining outcomes, that is, the number of participants who designated a health proxy, knowledge of CPR, and readiness for advance care planning (Table 3). Before/after comparisons identified a significant increase in these outcomes in both intervention and control groups. The number of participants who designated a health proxy increased from 29% to 65% in the intervention group and from 22% to 52% in the control group, respectively (p < 0.001 each).

Discussion

This was a pilot randomized-controlled trial that provides important insight for future studies.

Although there was no significant difference between the groups in any of outcome measures examined, brief nurse intervention increased documentation of a patient-designated health care proxy, and improved the knowledge of CPR and patient readiness from the baseline. Moreover, <5% of the Japanese elderly patients wanted life-prolonging care at the baseline, and there is a tendency through which the patients who received the intervention with visual materials were less likely to want CPR.

First, the percentage of our participants who wanted life-prolonging care seems to be smaller than that in Western clinical trials; <5% in this study versus 41% (patients with advanced heart failure), 33% (patients cared for in nursing homes), 26% (glioblastoma patients), and 13% (dementia vignette, preintervention).^12,14,15,17 This figure is close to the finding from a national survey in 2017 conducted by the Ministry of Health, Labour, and Welfare, Japan, which indicated that the majority of Japanese do not want life-prolonging measures in advanced stages of disease (i.e., want CPR: 11% for advanced cancer, 12% for heart failure, and 9.5% for dementia).¹⁰ These findings strongly suggest that the preference of the elderly Japanese public is to not receive life-prolonging care. However, multiple empirical studies reported marked conflicts in decision making about withholding or the withdrawal of life-prolonging measures in Asian countries, including Korea and Taiwan.^19–21 This may be because in Asian cultures, patients often hesitate to express their preferences directly and they are sometimes overridden by families when the patient becomes unable to communicate.^19–21 Therefore, rather than educating patients to clarify their preferences, future studies should focus on finding better methods to help patients share their preferences with their families and help families respect patient preferences.

Second, the patients in the intervention group were more likely to not want CPR than those in the control group, although the difference did not reach statistical significance. Before/after comparisons revealed that the number of patients who did not want CPR significantly increased in the intervention group. This is consistent with previous studies from North America in which a video significantly affected patient goal-of-care and CPR preferences.^12–17 These findings indicate that visual materials improve patient understanding about the medical interventions and significantly influence patient preference. The weak effects of the visual materials in this study may have been due to the premature state of the materials, and visual tools developed by more rigorous processes are promising.

Third, the finding that >50% of the participants successfully designated a health care proxy even by simple intervention adopted in the control group is very promising. According to a national study in 2017, 22% of the Japanese general public have designated a health care proxy,¹⁰ and the figure in this study was similar (i.e., 22% vs. 29%, respectively). Our findings suggest that brief nurse verbal intervention is successful in determining a health care proxy in more than half of the elderly population. Simple interventions using verbal descriptions by nurses may be effective to aid patients in formally designating a health care proxy.

Of note, a relatively large population of the participants wanted “limited medical care” as their goal-of-care preference. This category was made for this study, aiming to describe medical care that “prolongs life if pain or suffering is acceptable; but treatment would be withdrawn under patient-designated specific situations, such as unacceptable pain/suffering, uncommunicable state, or fully dependent in daily activities.” In general, a considerable number of Japanese regard hospitalization, intravenous fluids, and antibiotics as the minimum standard of care even during receiving home or hospice care services, and health care costs are fully covered by national insurance.^22,23 Limited medical care, that is, life-prolonging care promoting comfort with a chance of withdrawal, may be a valuable option for some countries where intravenous fluids are regarded as the minimum standard of care and should be examined as an endpoint in future studies.

This study was preliminary and has considerable limitations. The patient sample was from an online commercial database and may be biased, for example, highly educated males. We assume, however, that this bias would not change the conclusion, as the key endpoint figures were similar to those of the national survey.¹⁰ The intervention materials were not made through a fully rigorous development process. Measurement outcomes, especially goal-of-care preference, were used after modification without a formal validation process. However, as the interpretation of another category of life-prolonging care in the original study^12–17 was essentially the same, we believe that this limitation does not change the overall conclusion. Outcome measurements suitable for the Japanese health care system are needed in future studies.

In conclusion, brief nurse intervention increased documentation of a patient-designated health care proxy and improved knowledge of CPR and patient readiness. The Japanese elderly who wanted life-prolonging care as a goal of care in a terminal condition was small, and visual materials might help patients to image the actual situation regarding CPR. The use of brief nurse intervention to clarify a health care proxy, finding a method to help patients share their preferences with their families and help families respect patient preferences, and developing visual materials to assist clarifying patient preference about CPR are promising.

Footnotes

Acknowledgments

The authors express special gratitude to Naoko Mikoshiba, RN, PhD, for the development and administration of this study. They also thank Yuko Yamamoto, Chiwako Atsukata, Shiho Wagatsuma, Miyuki Wakiya, Asako Futami, Reiko Takebayashi, and Sachiko Kushiya for their assistance with performing interviews.

Funding Information

This study was funded by JSPS KAKENHI Grant No. JP16H05214, and the national project for improving End-of-Life Care 2017.

Author Disclosure Statement

No competing financial interests exist.

References

Rietjens

JAC

, Sudore

, Connolly

, et al.: Definition and recommendations for advance care planning: An international consensus supported by the European Association for Palliative Care. Lancet Oncol, 2017; 18:e543–e551.

Jimenez

, Tan

, Virk

, et al.: Overview of systematic reviews of advance care planning: Summary of evidence and global lessons. J Pain Symptom Manage, 2018; 56:436–459.e25.

Bernacki

, Paladino

, Neville

, et al.: Effect of the serious illness care program in outpatient oncology: A cluster randomized clinical trial. JAMA Intern Med, 2019; 179:751–759.

Epstein

, Duberstein

, Fenton

, et al.: Effect of a patient-centered communication intervention on oncologist-patient communication, quality of life, and healthcare utilization in advanced cancer: The VOICE randomized clinical trial. JAMA Oncol, 2017; 3:92–100.

Clayton

, Butow

, Tattersall

, et al.: Randomized controlled trial of a prompt list to help advanced cancer patients and their caregivers to ask questions about prognosis and end-of-life care. J Clin Oncol, 2007; 25:715–723.

Johnson

, Butow

, Bell

, et al.: A randomised controlled trial of an advance care planning intervention for patients with incurable cancer. Br J Cancer, 2018; 119:1182–1190.

Chen

, Liang

, Ho

, et al.: Association between palliative care and life-sustaining treatments for patients with dementia: A nationwide 5-year cohort study. Palliat Med, 2018; 32:622–630.

Chung

, Wong

, Kiang

, et al.: Knowledge, attitudes, and preferences of advance decisions, end-of-life care, and place of care and death in Hong Kong. A population-based telephone survey of 1067 adults. J Am Med Dir Assoc, 2017; 18:367.e19–367.e27.

Aoki

, Miyashita

, Yamamoto

, et al.: Patient experience of primary care and advance care planning: A multicentre cross-sectional study in Japan. Fam Pract, 2017; 34:206–212.

10.

Ministry of Helath, Labor, and Welfare: A national survey about end-of-life in Japan. www.mhlw.go.jp/toukei/list/dl/saisyuiryo_a_h29.pdf

11.

Sudore

, Boscardin

, Feuz

, et al.: Effect of the PREPARE website vs an easy-to-read advance directive on advance care planning documentation and engagement among veterans: A randomized clinical trial. JAMA Intern Med, 2017; 177:1102–1109.

12.

El-Jawahri

, Podgurski

, Eichler

, et al.: Use of video to facilitate end-of-life discussions with patients with cancer: A randomized controlled trial. J Clin Oncol, 2010; 28:305–310.

13.

Volandes

, Paasche-Orlow

, Mitchell

, et al.: Randomized controlled trial of a video decision support tool for cardiopulmonary resuscitation decision making in advanced cancer. J Clin Oncol, 2013; 31:380–386.

14.

El-Jawahri

, Paasche-Orlow

, Matlock

, et al.: Randomized, controlled trial of an advance care planning video decision support tool for patients with advanced heart failure. Circulation, 2016; 134:52–60.

15.

Volandes

, Paasche-Orlow

, Barry

, et al.: Video decision support tool for advance care planning in dementia: Randomised controlled trial. BMJ, 2009; 338:b2159.

16.

El-Jawahri

, Mitchell

, Paasche-Orlow

, et al.: A randomized controlled trial of a CPR and intubation video decision support tool for hospitalized patients. J Gen Intern Med, 2015; 30:1071–1080.

17.

Volandes

, Brandeis

, Davis

, et al.: A randomized controlled trial of a goals-of-care video for elderly patients admitted to skilled nursing facilities. J Palliat Med, 2012; 15:805–811.

18.

Sudore

, Heyland

, Lum

, et al.: Outcomes that define successful advance care planning: A Delphi panel consensus. J Pain Symptom Manage, 2018; 55:245–255.e8.

19.

Yun

, Lee

, Park

, et al.: Use of a decision aid to help caregivers discuss terminal disease status with a family member with cancer: A randomized controlled trial. J Clin Oncol, 2011; 29:4811–4819.

20.

Morita

, Oyama

, Cheng

, et al.: Palliative care physicians' attitudes toward patient autonomy and a good death in East Asian countries. J Pain Symptom Manage, 2015; 50:190–199.e1.

21.

Chiu

, Hu

, Huang

, et al.: Prevailing ethical dilemmas in terminal care for patients with cancer in Taiwan. J Clin Oncol, 2009; 27:3964–3968.

22.

Morita

, Shima

, Adachi

; Japan Palliative Oncology Study Group: Attitudes of Japanese physicians toward terminal dehydration: A nationwide survey. J Clin Oncol, 2002; 20:4699–4704.

23.

Morita

, Tsunoda

, Inoue

, Chihara

: Perceptions and decision-making on rehydration of terminally ill cancer patients and family members. Am J Hosp Palliat Care, 1999; 16:509–516.