Family Caregivers' Emotional Preparedness for Death is Distinct from Their Cognitive Prognostic Awareness for Cancer Patients

Abstract

Background:

Cognitive prognostic awareness (PA) and emotional preparedness for a loved one's death are distinct but related phenomena. However, the distinction between these two concepts has not been studied in family caregivers.

Objective:

To examine whether these two concepts are distinct by comparing their evolution and predictors over cancer patients' last year.

Methods:

Agreement between emotional preparedness for death and cognitive PA was longitudinally evaluated for 309 family caregivers by percentages and kappa coefficients. Predictors of the two outcomes were evaluated by multivariate logistic regression models with the generalized estimating equation.

Results:

Agreement between family caregivers' emotional preparedness for death and cognitive PA decreased slightly (54.73%–43.64%) from 181–365 to 1–30 days before the patient's death, with kappa values (95% confidence interval) from −0.060 (−0.123 to 0.003) to 0.050 (−0.074 to 0.174), indicating poor agreement. Participants were more likely to report adequate emotional preparedness for death if they had financial sufficiency, more contact/communication with the patient, lower caregiving burden, and stronger perceived social support. Family caregivers were more likely to have accurate PA if they were 56–65 years old, the patient's adult child, and had more contact/communication with the patient and greater subjective caregiving burden.

Conclusions/Implications:

Family caregivers' emotional preparedness for death and cognitive PA were distinct, as supported by their poor agreement, lack of reciprocal associations, and two different sets of predictors. Health care professionals should facilitate family caregivers' accurate PA and cultivate their emotional preparedness for death by enhancing patient-family contact/communication and easing their caregiving burden to improve quality of end-of-life care.

Introduction

Preparing family caregivers of terminally ill patients for their loved ones' imminent death is a vital component of high-quality end-of-life (EOL) care^1,2 because it improves caregivers' bereavement outcomes.^3–8 Preparing for a loved one's death involves coping with medical, practical, psychosocial, and spiritual uncertainty permeating the patient's full EOL experience and the tension of managing EOL-caregiving demands, anticipatory grief about losing a longstanding relationship with the patient, and planning for life without the patient.^2,9–13 Thus, preparedness for death involves transitioning into cognitively acknowledging and emotionally accepting a loved one's inevitable death.^13,14

Cognitive acknowledgment and emotional acceptance of a loved one's death are distinct but related phenomena.^2,14 Accurately understanding a loved one's prognosis is a prerequisite for family caregivers' preparedness for death^15–17 and reduces their psychological distress before¹⁸ and after¹⁹ the loss. However, the distinction between these two concepts has not been studied in family caregivers. Furthermore, identifying factors facilitating/hindering adequate emotional preparedness for death and accurate prognostic awareness (PA) for family caregivers while they provide EOL caregiving per se is critical to guide development of effective interventions to improve EOL-care quality before and after the patient's death.

However, only a handful of studies^{4–6,8,16,17,19} examined factors facilitating or hindering family caregivers' emotional preparedness for death or PA. All these studies, except one,⁸ retrospectively assessed bereaved family members' outcomes only once after patient death. This approach leads not only to neglecting the dynamic nature of emotional preparedness for death and PA⁹ but also to recall bias, in which grief reactions color predictors of emotional preparedness for death or PA.^2,3,6,17 Furthermore, studies on predictors of emotional preparedness for death and PA should consider factors other than commonly examined demographic factors, for example, factors related to the larger caregiving context as suggested by the stress-appraisal-coping model²⁰ and supported by a recent systematic review²¹ and empirical evidence from Asian countries^22,23 where this issue has never been explored.

Therefore, the purposes of this study were to longitudinally examine whether Taiwanese family caregivers' emotional preparedness for death and cognitive PA are distinct by investigating their agreement over patients' last year and factors facilitating/hindering adequate emotional preparedness for death and accurate PA based on the stress-appraisal-coping model.²⁰

Methods

Study design and sample

This was a longitudinal, observational study on associations of preparedness for death with caregiving outcomes for family caregivers of terminally ill cancer patients. A consecutive sample of family caregivers was recruited (August 2015 through December 2018) from a medical center in northwest Taiwan and included in the analysis if their loved ones died by July 2019. Loved ones (patients) were recognized by their oncologist as terminally ill (i.e., their cancer was a late-stage disease that continually progressed and was unresponsive to repeated chemotherapy or targeted therapies). Caregivers were recognized by patients as the family member most involved in their daily care. Participants' data were collected by experienced, trained oncology nurses in face-to-face interviews at baseline and approximately every 2–4 weeks thereafter until they withdrew or their loved ones died. The Research Ethics Committee of the study site approved the research protocol (103-7015B). All participants provided written informed consent.

Measures

Outcome variables

Emotional preparedness for death

Emotional preparedness for death was measured by the five-item preparation for end-of-life subscale of the QUAL-E scale,^24,25 whose development was based on qualitative research on perceptions of a good death.¹ The preparation for end-of-life subscale assesses the extent of concerns about impending issues related to the loved one's forthcoming death, for example, the patient's care becoming a financial strain or burdening the family, one's own preparation for and ability to cope with the patient's death/dying, fear of the patient's death/dying, and regrets about the patient's life.

Each item, measured on a 5-point Likert scale, is reverse scored from 5 (poor preparation) to 25 (excellent preparation). To derive a binary measure of death preparation, the total score is dichotomized into “adequate” (≥20, the median score) and “inadequate” (<20) emotional preparedness for death.²⁶

Prognostic awareness

PA was measured by asking family caregivers at each assessment whether they knew their loved ones' prognosis, and if so, whether the disease (1) was curable; (2) might recur in the future, but their loved ones' life was not currently in danger; and (3) could not be cured, or their loved ones would probably die soon.²⁷ Family caregivers were recognized as having accurate PA only if they chose option 3; inaccurate PA reflected not knowing their loved ones' prognosis or choosing option 1 or 2.

This measure of cognitive PA was developed based on a literature review and Taiwanese physicians' cultural practice of prognostic disclosure. Death is a taboo topic in Asian culture²⁸; thus, Taiwanese physicians tend to disclose that a disease can be cured or not, which is perceived by Taiwanese patients and their family caregivers as less threatening, more acceptable, and more understandable than stating that the disease is at a terminal stage. Incurability in Taiwanese clinical practice indicates death, rather than survival, as the endpoint of the disease process. The validity of this measure is supported by its reflection of PA conceptualizations and measures used in a 34-study review of PA.²⁹

Independent variables

Factors facilitating/hindering family caregivers' adequate emotional preparedness for death and accurate PA were guided by the stress-appraisal-coping model,²⁰ including (1) contextual factors, (2) caregiving demands, (3) appraisal of the caregiving situation, and (4) available resources. Contextual factors are primarily time-invariant and were measured at enrollment. To reflect our longitudinal data, that is, each participant had subject-specific responses at different data-collection times, we explored time-varying covariates from the other three categories.

Contextual factors

Contextual factors were caregivers' demographics and level of contact/communication with the patient. Demographics included age (21–45, 46–55, 56–65, >65 years old), gender, marital status (married/unmarried), educational attainment (</≥ high school), financial sufficiency (making ends meet/financial strain), comorbidity (yes/no), and relationship with the patient (spouse/adult child/other). We measured contact/communication level with the patient by the following three items³⁰: (1) frequency of contact with the patient, (2) frequency of talking openly with the patient about his/her thoughts and feelings about his/her disease experiences or symptoms, and (3) knowledge of the patient's thoughts and feelings. Each item was assessed on a 5-point Likert scale, with higher scores indicating more contact with or better knowledge of the patient.

Objective caregiving demands

Objective caregiving demands were measured by the amount of assistance provided in personal care, homemaking, transportation, and health care on a 4-point scale (1 = none at all to 4 = a lot).³¹ Scores for objective caregiving demand range from 4 to 16; higher scores indicate greater objective caregiving load.

Appraisal of caregiving situation

This variable was assessed by caregivers' perceived caregiving burden. Caregiver subjective caregiving burden was measured by the Caregiver Reaction Assessment (CRA).³² Total score range = 24–120. Higher scores represent a stronger negative caregiving impact on schedule, health, finances, caregiver support, and self-esteem.

Available resources

Available resources were indicated by perceived social support and measured by the 19-item Medical Outcomes Study Social Support Survey (MOS-SSS).³³ The MOS-SSS assesses emotional, informational, tangible, and affectionate support, as well as positive social interaction.³³ For each subscale, total sums are computed, and the raw subscale scores are transformed to a 0–100 scale. Higher scores indicate better perceived social support.

Statistical analysis

Agreement between family caregivers' emotional preparedness for death and cognitive PA was evaluated by cross-tabulating these two outcomes over the patient's last year by both a percentage and a kappa coefficient for chance-corrected agreement.³⁴ Agreement (kappa value) was determined as poor (≤0.20), fair (0.21–0.40), moderate (0.41–0.60), substantial (0.61–0.80), or almost perfect (0.81–1.00).³⁵ To explore longitudinal changes in agreement between family caregivers' emotional preparedness for death and cognitive PA as well as their predictors over the patient's dying process, time proximity to patient death, that is, the period between death and data collection, was categorized as 1–30, 31–90, 91–180, and 181–365 days, the conventional intervals used in estimating terminally ill cancer patients' survival.

Predictors for the two outcome variables were identified by multivariate logistic regression models with the generalized estimating equation (GEE)³⁶ for time-varying and time-invariant independent variables and time proximity to patient death. The GEE uses robust standard error estimates to account for within-subject correlations³⁶ of outcome and time-varying independent variables during follow-ups and to accommodate uneven numbers of follow-ups, inconsistent intervals between subsequent data collections, and missing data for outcome variables. GEE uses all outcome variables available in each time period to construct the model, eliminating the need to delete observations in analyses or to impute missing data.³⁶ The regression estimate for each independent variable in the logistic regression models was exponentiated to transform into adjusted odds ratio (AOR) with 95% confidence interval (CI).

Results

Participant characteristics

Among the 384 patients with available family caregivers who agreed to participate, 371 family caregivers were enrolled (96.6% participation). At the end of follow-ups, 14 caregivers had withdrawn and 48 caregivers' loved ones were still alive. Participants whose loved ones died during the study (n = 309) comprised the final sample. The majority of participants were female (79.9%), >55 years old (54.1%), married (89.0%), educated at <high-school level (54.8%), had no chronic disease (71.5%), the patient's spouse (77.0%), living with the patient (90.2%), and the patient's primary decision maker (81.8%) (Table 1). After enrollment, caregivers participated on average 127.90 days (standard deviation [SD] = 163.86; median = 69.0). The following analyses are based on 1293 assessments with a mean interval between the last interview and patient death of 23.71 days (SD = 73.09, median = 13.0).

Table 1.

Family Caregiver Characteristics at Enrollment (n = 309)

Characteristic	n^a	%	Characteristic	n^a	%
Gender			Marital status
Male	62	20.1	Single	23	7.4
Female	247	79.9	Married/cohabiting	275	89.0
Age (years)			Separated/widowed	11	3.6
21–45	55	17.8	Financial sufficiency
46–55	87	28.2	Making ends meet	271	87.7
56–65	104	33.7	Financial strain	38	12.3
>65	63	20.4	Chronic disease
Educational level			Yes	88	28.5
≥High school	138	45.2	No	221	71.5
<High school	167	54.8	Adequate emotional preparedness for death
Relationship			Yes	158	51.1
Spouse	238	77.0	No	151	48.9
Child	36	11.7	Accurate PA
Other	35	11.3	Yes	230	74.4
Living with the patient			No	79	25.6
Yes	277	90.2	Primary decision maker
No	30	9.8	Yes	252	81.8
			No	56	18.2
Parameter	Mean	SD	Parameter	Mean	SD
Objective caregiving load	11.53	2.61	Subjective caregiving burden	66.05	8.77
Confidence in caregiving	7.46	1.31	Family intimacy	12.88	1.68
Social support	60.64	11.06

The total number of cases for each characteristic may not equal the sample size due to missing data.

PA, prognostic awareness; SD, standard deviation.

Agreement between family caregivers' emotional preparedness for death and cognitive PA over terminally ill cancer patients last year

Agreement between caregivers' emotional preparedness for death and cognitive PA decreased slightly from 54.73% to 43.64% from 181–365 to 1–30 days before the patient's death (Table 2). Kappa values for agreement between the two outcome variables ranged from −0.060 (−0.123 to 0.003) to 0.050 (−0.074 to 0.174), indicating poor agreement. Controlling for time-varying and time-invariant covariates, agreement between emotional preparedness for death and cognitive PA decreased significantly in the last month in reference to 181–365 days before death (AOR [95% CI] range = 0.67 [0.47–0.95], p = 0.026). Disagreement was most frequent when family caregivers had accurate PA but were not emotionally prepared for their loved ones' death, with a substantial increase from 27.16% to 49.71% from 181–365 to 1–30 days before the patient's death.

Table 2.

Agreement between Emotional Preparedness for Death and Cognitive Prognostic Awareness (n = 309)

	Time before death, days
	≤30 (n = 346)		31–90 (n = 417)		91–180 (n = 287)		181–365 (n = 243)
	Accurate PA
	No	Yes	No	Yes	No	Yes	No	Yes
Adequate emotional preparedness for death
No	15	172	32	158	27	84	38	66
Yes	23	136	47	180	43	133	44	95
Observed agreement (%)	43.64		50.84		55.75		54.73
Kappa (95% CI)	−0.060 (−0.123 to 0.003)		−0.041 (−0.119 to 0.038)		−0.001 (−0.111 to 0.108)		0.050 (−0.074 to 0.174)
Likelihood of agreement^*
AOR (95% CI)	0.67 (0.47 to 0.95)		0.86 (0.62 to 1.20)		1.10 (0.79 to 1.54)		Reference
Z score	−2.23		−0.88		0.57
p	0.026		0.378		0.572

Time-invariant (gender, age, marital status, educational attainment, financial sufficiency, comorbidity, relationship with the patient, and family intimacy) and -varying (objective caregiving demands, subjective caregiving burden, and social support) covariates were controlled in the generalized estimating equation model. Gray areas indicate primary disagreement. n = number of assessments in each period; n does not always equal the number of participants because they might have had multiple assessments or their loved ones might not have survived long enough for the caregiver to have assessed before death.

AOR, adjusted odds ratio; CI, confidence interval.

Predictors of adequate emotional preparedness for death for family caregivers of terminally ill cancer patients

Among the contextual factors, only financial status and contact/communication with the patient were associated with family caregivers' adequate emotional preparedness for death. Caregivers were more likely to report adequate emotional preparedness for death if they had financial sufficiency (AOR [95% CI]: 4.291 [1.995–9.230]) and more contact with and better knowledge of the patient (AOR [95% CI]: 1.193 [1.035–1.374]) (Table 3). Adequate emotional preparedness for death was not associated with accurate PA.

Table 3.

Factors Associated with Adequate Emotional Preparedness for Death

	AOR	95% CI	Z	p
Time proximity to death, days
≤30	1.096	0.711–1.691	0.42	0.678
31–90	1.149	0.780–1.691	0.70	0.483
91–180	1.360	0.912–2.029	1.51	0.132
181–365	Ref
Accurate PA
Yes	0.716	0.457–1.123	−1.45	0.146
No	Ref
Gender
Female	0.652	0.397–1.071	−1.69	0.091
Male	Ref
Age (years)
21–45	0.564	0.264–1.205	−1.48	0.139
46–55	0.906	0.499–1.646	−0.32	0.747
56–65	1.185	0.676–2.079	0.59	0.553
>65	Ref
Marital status
Unmarried	1.573	0.673–3.676	1.05	0.295
Married	Ref
Relationship with the patient
Spouse	0.594	0.224–1.571	−1.05	0.294
Child	1.748	0.624–4.895	1.06	0.288
Other	Ref
Educational attainment
≥High school	0.944	0.606–1.471	−0.25	0.799
<High school	Ref
Financial sufficiency
Sufficient	4.291	1.995–9.230	3.73	<0.001
Financial strain	Ref
Caregiver-patient contact/communication	1.193	1.035–1.374	2.43	0.015
Objective caregiving demand	0.906	0.861–0.954	−3.77	<0.001
Subjective caregiving burden	0.964	0.939–0.990	−2.67	0.008
Social support	1.038	1.020–1.057	4.12	<0.001

The other three categories of factors based on the stress-appraisal-coping model were all associated with family caregivers' adequate emotional preparedness for death (Table 3). Family caregivers who carried heavier objective caregiving demand (AOR [95% CI]: 0.906 [0.861–0.954]) and perceived greater caregiving burden (AOR [95% CI]: 0.964 [0.939–0.990]) were significantly less likely to report adequate emotional preparedness for death. In contrast, stronger perceived social support increased family caregivers' likelihood of adequate emotional preparedness for death (AOR [95% CI]: 1.038 [1.020–1.057]).

Predictors of accurate PA for family caregivers of terminally ill cancer patients

Among the contextual factors, family caregivers' age, their relationship with the patient, and contact/communication with the patient were associated with family caregivers' accurate PA. Family caregivers 56–65 years old were significantly more likely to have accurate PA (AOR [95% CI]: 2.403 [1.130–5.113]) than those who were over 65 years old (Table 4). Adult-child caregivers had a greater likelihood of accurate PA (AOR [95% CI]: 4.147 [1.004–17.131]) than spouse or other family caregivers. Family caregivers having more contact with or better knowledge of the patient were more likely to have accurate PA (AOR [95% CI]: 1.223 [1.052–1.420]). Adequate emotional preparedness for death was not associated with accurate PA.

Table 4.

Factors Associated with Accurate Prognostic Awareness

	AOR	95% CI		Z	p
Time proximity to death, days
≤30	5.215	3.190	8.525	6.59	<0.001
31–90	2.262	1.568	3.263	4.37	<0.001
91–180	1.712	1.177	2.488	2.81	0.005
181–365	Ref
Adequate emotional preparedness for death
Yes	0.790	0.570	1.096	−1.41	0.158
No	Ref
Gender
Female	0.941	0.466	1.899	−0.17	0.864
Male	Ref
Age (years)
21–45	0.684	0.275	1.702	−0.82	0.414
46–55	1.652	0.738	3.697	1.22	0.222
56–65	2.403	1.130	5.113	2.28	0.023
>65	Ref
Marital status
Unmarried	0.941	0.313	2.829	−0.11	0.914
Married	Ref
Relationship with the patient
Spouse	0.571	0.216	1.509	−1.13	0.259
Child	4.147	1.004	17.131	1.97	0.049
Other	Ref
Educational attainment
≥High school	0.805	0.433	1.498	−0.68	0.494
<High school	Ref
Financial sufficiency
Yes	1.116	0.548	2.275	0.30	0.762
No	Ref
Caregiver-patient contact/communication	1.223	1.052	1.420	2.63	0.009
Objective caregiving demand	1.014	0.958	1.074	0.47	0.637
Subjective caregiving burden	1.034	1.004	1.065	2.21	0.027
Social support	0.992	0.973	1.011	−0.82	0.410

Among the other three categories of factors based on the stress-appraisal-coping model, only perceived caregiving burden was associated with family caregivers' accurate PA (Table 3). Family caregivers who perceived greater caregiving burden (AOR [95% CI]: 1.034 [1.004–1.065]) were significantly more likely to have accurate PA.

Discussion

Family caregivers' emotional preparedness for death and cognitive PA agreed poorly over cancer patients last year, with more significant deterioration observed in the last month than 181–365 days before the patient's death. These two related concepts were differentiated by two different sets of predictors based on the stress-appraisal-coping model²⁰ and lack of reciprocal associations.

Family caregivers' emotional preparedness for death and cognitive PA did not agree beyond chance (43.64%–55.75%) over cancer patients last year, and kappa values showed poor agreement between these two concepts. Disagreement primarily came from caregivers' lack of emotional preparedness for death even if they accurately understood their loved ones' poor prognosis. These results confirm that accurate PA (a cognitive preparedness for death) does not guarantee adequate emotional preparedness for death.³⁷ Furthermore, agreement between these two concepts decreased substantially in the patients last month (Table 2).

Family caregivers' increasingly accurate understanding of their loved ones' poor prognosis as his/her death approached (Table 4) may have been due to the patient's severe symptom distress and profound functional decline as the cancer progressed,¹⁹ coupled with oncologists' tendency to disclose prognosis late in patients' terminal-illness trajectory.³⁸ However, tasks to emotionally prepare for a loved one's death are broader and more highly psychologically laden^2,9–12 than becoming cognitively aware of the loved one's poor prognosis. Under current clinical practice³⁹ and without active interventions, family caregivers' accurate PA does not automatically translate into adequate emotional acceptance of their loved ones' forthcoming death.

Effective interventions are urgently needed to help family caregivers develop emotional preparedness for death. Developing emotional preparedness for death may enable family caregivers to adequately cope with concerns that their loved ones' dying process will impose a financial strain or physical and psychological burdens on the family, dispel their fears about the patient's death/dying and regrets about the patient's life, and empower them not only to survive the loss of the relationship with the patient but also to plan for a life without him/her.

The only commonality in the factors facilitating or hindering family caregivers' adequate emotional preparedness for death and accurate PA is the extent of contact with and knowledge of the patient. Adequate emotional preparedness for death and accurate PA were more likely to be reported by family caregivers having more contact/communication with the patient and better knowledge of the patient's feelings and thoughts. These findings are consistent with reports that husbands' preparedness for their wife's¹⁶ death was determined by the couple's communication, parents' preparedness for their child's death¹⁷ was predicted by their communication, and women were less likely to have a short period of being intellectually aware of their husband's impending death if the couple shared emotional concerns.¹⁹

Greater subjective caregiving burden was negatively and positively associated with family caregivers' development of adequate emotional preparedness for death and accurate PA, respectively. Caregivers who are overwhelmed by providing EOL caregiving (carrying a heavy objective caregiving demand and great subjective caregiving burden) may not adequately handle anticipatory grief and look forward to their future.^6,12 In contrast, family caregivers' greater subjective caregiving burden may facilitate the realization that their loved ones' physical condition is deteriorating and he/she will soon die.

Family caregivers who reported financial sufficiency^3,6 might enjoy better monetary resources to ease worries about their loved ones' illness imposing a financial strain that would burden their families. Similarly, caregivers' greater perceived social support⁴⁰ might facilitate adjustment to anticipatory grief to move on to a future life without their loved ones, increasing their likelihood of being emotionally adequately prepared for their loved ones' forthcoming death. Adult-child⁸ and 56-to 65-year-old family caregivers were more likely than those related to the patient in other ways (e.g., a parent or sibling) and >65 years old to have accurate PA, probably due to their stronger sense of mastery and personal resources⁴¹ and Asian physicians' tendency to disclose patients' poor prognosis to their adult children based on the doctrine of filial piety.⁴²

Study strengths and limitations

Our study contributes to knowledge about agreement between the related concepts of emotional preparedness for death and cognitive PA by prospectively, longitudinally investigating their evolution and predictors, while family caregivers cared for terminally ill cancer patients over their last year. However, the study had several limitations. We recruited family caregivers from one hospital in Taiwan, limiting generalization of our findings to national and international target populations. We could not establish a cause-effect relationship between our outcome variables and their predictors in this observational study and many unmeasured factors have not yet taken into considerations, for example, physician-family EOL-care discussions,^16,17,19 health care professional support,^16,17,19 and personal coping capacities or strategies.⁴³

Conclusions and Implications

Our findings indicate that emotional preparedness for death and cognitive PA for family caregivers of terminally ill cancer patients are two distinct concepts as evident by their poor agreement (especially in the patient's last month when their condition significantly deteriorates), not being reciprocally associated, and having different sets of predictors. Health care professionals should realize that family caregivers may increasingly, accurately understand their loved ones' prognosis as the patient's death approaches, but may not be emotionally prepared for their loved ones' death. Health care professionals should not only prepare family caregivers to provide EOL caregiving but also cultivate their emotional preparedness for death. Strategies to lighten family caregivers' subjective caregiving burden and enhance patient-caregiver communication and close interactions may improve caregiving outcomes and facilitate family caregivers' transition into cognitively acknowledging and emotionally accepting their loved ones' forthcoming death to probably enhance bereavement adjustment.

Footnotes

Authors' Contributions

Conception and design: S.T.T., W.C.C., W.C.C., C.H.H., J.S.C., and F.H.W.; provision of study material or patients: W.C.C., W.C.C., C.H.H., and J.S.C.; collection and/or assembly of data: S.T.T., W.C.C., W.C.C., C.H.H., J.S.C., and F.H.W.; data analysis and interpretation: S.T.T., W.C.C., W.C.C., C.H.H., J.S.C., and F.H.W.; article writing: S.T.T.; final approval of article: S.T.T., W.C.C., W.C.C., C.H.H., J.S.C., and F.H.W.

Acknowledgments

No funding source had any role in designing and conducting the study: collecting, managing, analyzing, and interpreting the data or preparing, reviewing, or approving the article.

Funding Information

This work was supported by the Ministry of Science and Technology (MOST 104-2314-B-182-027-MY3), National Health Research Institutes (NHRI-EX109-10704PI), and Chang Gung Memorial Hospital (BMRP888).

Author Disclosure Statement

The authors declare no financial or other conflict of interest. We confirm that the article has not been and will not be submitted, in part or entirety, elsewhere for publication and all authors meet criteria for authorship as stated in the uniform requirements for articles submitted to biomedical journals. We confirm that if accepted, the article will not be published elsewhere in the same form, in English or in any other language, including electronically, without the written consent of the copyright holder.

References

Steinhauser

, Christakis

, Clipp

, et al.: Factors considered important at the end of life by patients, family, physicians, and other care providers. JAMA, 2000; 284:2476–2482.

Hebert

, Prigerson

, Schulz

, Arnold

: Preparing caregivers for the death of a loved one: A theoretical framework and suggestions for future research. J Palliat Med, 2006; 9:1164–1171.

Hebert

, Dang

, Schulz

: Preparedness for the death of a loved one and mental health in bereaved caregivers of patients with dementia: Findings from the REACH study. J Palliat Med, 2006; 9:683–693.

Hauksdottir

, Steineck

, Furst

, Valdimarsdottir

: Long-term harm of low preparedness for a wife's death from cancer—A population-based study of widowers 4-5 years after the loss. Am J Epidemiol, 2010; 172:389–396.

Barry

, Kasl

, Prigerson

: Psychiatric disorders among bereaved persons: The role of perceived circumstances of death and preparedness for death. Am J Geriatr Psychiatry, 2002; 10:447–457.

Caserta

, Utz

, Lund

, et al.: Cancer caregivers' preparedness for loss and bereavement outcomes: Do preloss caregiver attributes matter?. Omega (Westport), 2019; 80:224–244.

Kim

, Carver

, Spiegel

, et al.: Role of family caregivers' self-perceived preparedness for the death of the cancer patient in long-term adjustment to bereavement. PsychoOncol, 2017; 26:484–492.

Schulz

, Boerner

, Klinger

, Rosen

: Preparedness for death and adjustment to bereavement among caregivers of recently placed nursing home residents. J Palliat Med, 2015; 18:127–133.

Hebert

, Schulz

, Copeland

, Arnold

: Preparing family caregivers for death and bereavement. Insights from caregivers of terminally ill patients. J Pain Symptom Manage, 2009; 37:3–12.

10.

Breen

, Aoun

, O'Connor

, et al.: Family caregivers' preparations for death: A qualitative analysis. J Pain Symptom Manage, 2018; 55:1473–1479.

11.

Olson

: Indefinite loss: The experiences of carers of a spouse with cancer. Eur J Cancer Care, 2014; 23:553–561.

12.

Hovland-Scafe

, Kramer

: Preparedness for death: How caregivers of elders with dementia define and perceive its value. Gerontologist, 2017; 57:1093–1102.

13.

McLeod-Sordjan

: Death preparedness: A concept analysis. J Adv Nurs, 2013; 70:1008–1019.

14.

Prigerson

, Maciejewski

: Grief and acceptance as opposite sides of the same coin: Setting a research agenda to study peaceful acceptance of loss. Br J Psychiatry, 2008; 193:435–437.

15.

Jennings

, Galupo

, Cartwright

: The role of postformal cognitive development in death acceptance. J Adult Devel, 2009; 16:166–172.

16.

Hauksdottir

, Valdimarsdottir

, Furst

, et al.: Health care-related predictors of husbands' preparedness for the death of a wife to cancer—A population-based follow-up. Ann Oncol, 2010; 21:354–361.

17.

Valdimarsdottir

, Kreicbergs

, Hauksdottir

, et al.: Parents' intellectual and emotional awareness of their child's impending death to cancer: A population-based long-term follow-up study. Lancet Oncol, 2007; 8:706–714.

18.

Burridge

, Barnett

, Clavarino

: The impact of perceived stage of cancer on carers' anxiety and depression during the patients' final year of life. PsychoOncol, 2009; 18:615–623.

19.

Valdimarsdottir

, Helgason

, Furst

, et al.: Awareness of husband's impending death from cancer and long-term anxiety in widowhood: A nationwide follow-up. Palliat Med, 2004; 18:432–443.

20.

Lazarus

, Folkman

: Stress, Appraisal and Coping. New York: Springer Co., 1984.

21.

Alam

, Hannon

, Zimmermann

. Palliative care for family caregivers. J Clin Oncol, 2020; 38:926–936.

22.

Wen

, Chen

, Chou

, et al.: Family caregivers' subjective caregiving burden, quality of life, and depressive symptoms are associated with terminally ill cancer patients' distinct patterns of conjoint symptom distress and functional impairment in their last 6 months of life. J Pain Symptom Manage, 2019; 57:64–71.

23.

Cheung

DSK

, Ho

KHM

, Cheung

, et al.: Anticipatory grief of spousal and adult children caregivers of people with dementia. BMC Palliat Care, 2018; 17:124.

24.

Steinhauser

, Bosworth

, Clipp

, et al.: Initial assessment of a new measure of quality of life at the end of life (QUAL-E). J Palliat Med, 2002; 5:829–842.

25.

Steinhauser

, Clipp

, Bosworth

, et al.: Measuring quality of life at the end of life: Validation of the QUAL-E. Palliat Support Care, 2004; 2:3–14.

26.

Wentlandt

, Burman

, Swami

, et al.: Preparation for the end of life in patients with advanced cancer and association with communication with professional caregivers. PscyhoOncol, 2012; 21:868–876.

27.

Wen

, Chou

, Chen

, et al.: Evolution and predictors of patient-caregiver concordance on states of life-sustaining treatment preferences over terminally ill cancer patients' last 6 months of life. J Palliat Med, 2019; 22:25–33.

28.

McDermott

, Selman

: Cultural factors influencing advance care planning in progressive, incurable disease: A systematic review with narrative synthesis. J Pain Symptom Manage, 2018; 56:613–636.

29.

Chen

, Kuo

, Tang

: Current status of accurate prognostic awareness in advanced/terminally ill cancer patients: Systematic review and meta-regression analysis. Palliat Med, 2017; 31:406–418.

30.

Lobchuk

. Degner LF: Symptom experience: Perceptual accuracy between advanced-stage cancer patients and family caregivers in the home care setting. J Clin Oncol, 2002; 20:3495–3507.

31.

Emanuel

, Fairclough

, Slutsman

, et al.: Assistance from family members, friends, paid care givers and volunteers in the care of terminally ill patients. N Engl J Med, 1999; 341:956–963.

32.

Given

, Given

, Stommel

, et al.: The Caregiver Reaction Assessment (CRA) for caregivers to persons with chronic physical and mental impairments. Res Nurs Health, 1992; 15:271–283.

33.

Sherbourne

, Stewart

: The MOS social support survey. Soc Sci Med, 1991; 32:705–714.

34.

Bland

, Altman

: Statistical methods for assessing agreement between two methods of clinical measurement. Lancet, 1986; 8476:307–310.

35.

Landis

, Koch

: The measurement of observer agreement for categorical data. Biometrics, 1977; 33:159–174.

36.

Diggle

, Liang

, Zeger

: Analysis of Longitudinal Data. New York: Oxford University Press, 2002.

37.

Ray

, Block

, Friedlander

, et al.: Peaceful awareness in patients with advanced cancer. J Palliat Med, 2006; 9:1359–1368.

38.

Keating

, Landrum

, Rogers

Jr , et al.: Physician factors associated with discussions about end-of-life care. Cancer, 2010; 116:998–1006.

39.

Chiarchiaro

, Buddadhumaruk

, Arnold

, White

: Quality of communication in the ICU and surrogate's understanding of prognosis. Crit Care Med, 2015; 43:542–548.

40.

Nijboer

, Tempelaar

, Triemstra

, et al.: The role of social and psychologic resources in caregiving of cancer patients. Cancer, 2001; 91:1029–1039.

41.

Kim

, Knight

: Caregiving subgroups differences in the associations between the resilience resources and life satisfaction. J Appl Gerontol, 2018; 37:1540–1563.

42.

Jeong

, Shin

, Park

: What we talk about when we talk about caregiving: The distribution of roles in cancer patient caregiving in a family-oriented culture. Cancer Res Treat, 2019; 51:141–149.

43.

Higgins

, Prigerson

: Caregiver evaluation of the quality of end-of-life care (CEQUEL) scale: The caregiver's perception of patient care near death. PLoS One, 2013; 8:e66066.