Expectations in the Communication About Death and Dying From a Caregiver’s Perspective: The End-Of-Life Conversation – Expectation Scale

Abstract

Communication about the end-of-life (EOL) is often avoided, despite its association with positive outcomes for those affected and their caregivers. Negative expectations seem to be a mechanism through which avoidance is maintained. This study examined existing expectations in talking about death to a significant other and validated a caregiver’s version of the End-of-Life Conversation – Expectation Scale (EOLC-E-c) with a German community sample of N = 250. Participants most frequently expected EOL conversations to cause sadness in themselves and the other person. Confirmatory factor analysis supported the original three-factor structure: Expected own emotional burden (α = .92), expected emotional burden of the other person (α = .94), and communication self-efficacy (=.90). Internal consistency of the scale was excellent, and its validity was supported. Using the scale could be valuable for clinicians and researchers in understanding individuals’ expectations concerning EOL conversations and in developing targeted interventions focusing on expectation violation.

Keywords

expectations end-of-life communication death dying caregivers

Introduction

Despite the inevitability of death, communication about it is often avoided in Western societies. Conversations surrounding death and dying are frequently perceived as taboo, difficult, and distressing (Scheinfeld & Lake, 2019; Wildfeuer et al., 2015). Although the majority of the general population in Germany appears to consider death, dying, and grief to be important topics, only about half engage in discussions addressing these themes (Strupp et al., 2021).

In many Western societies, the processes of dying and death have been increasingly shifted into health care settings, resulting in a significant decline in family involvement (Sallnow et al., 2022). Nevertheless, communication between patients and caregivers remains a crucial aspect of end-of-life (EOL) experiences (Fenton et al., 2024). Such communication has been associated with increased alignment between individuals’ preferences and the care received (Detering et al., 2010), reductions in overtreatment (e.g., non-indicated ventilation and resuscitation) and earlier hospice enrollment. These outcomes are, in turn, linked to improved quality of end-of-life care and death (Wright et al., 2008; Yamaguchi et al., 2017). Additionally, less intensive medical treatment is not only associated with higher quality of life but also with reduced health care costs (Patel et al., 2018; Zhang et al., 2009).

Caregivers often bear responsibility for end-of-life medical treatment decisions, which can lead to emotional distress and persistent doubts about their choices (Wendler & Rid, 2011). Confidence in understanding the patient’s treatment wishes appears to mitigate these negative effects (Wendler & Rid, 2011). Furthermore, postponing or avoiding EOL conversations has been linked to adverse psychological outcomes for family members and significant others including increased stress (Detering et al., 2010), feelings of guilt (Jonasson et al., 2011), depressive symptoms (Bachner et al., 2021; Yamaguchi et al., 2017), anxiety (Detering et al., 2010), reduced personal growth (Generous & Keeley, 2022), bereavement-related distress (Haaksman et al., 2024), and complicated grief (Yamaguchi et al., 2017).

One potential explanation for the gap between the benefits of EOL conversations and their infrequent occurrence lies in individuals’ unfavorable expectations. Expectations are defined as probabilistic beliefs regarding the likelihood of future events, experiences, or outcomes (Panitz et al., 2021). Research has demonstrated that modifying expectations can lead to improved health-related outcomes (e.g., Rief et al., 2017; Shedden-Mora et al., 2020). In the context of EOL conversations, expectations such as the desire to protect oneself or others from emotional distress, and the belief that “positive thinking” precludes discussing death, have been identified as contributing to avoidance (Nagelschmidt et al., 2021). The modification of expectations seems to be a possible mechanism through which EOL conversations can be facilitated. Initial research suggests that expectations regarding EOL conversations can be modified through targeted interventions (Bendel et al., 2025). However, in order to design effective interventions, it is first necessary to identify the specific unfavorable expectations individuals hold. The present study utilizes and validates a newly developed questionnaire to provide a comprehensive overview of common and less frequent negative expectations surrounding conversations about death, with the aim of informing future interventions.

Several instruments assessing aspects of communication about death have been developed and validated including the communication apprehension about death scale (Carmack & DeGroot, 2016) and the death attitude profile-revised scale (Wong et al., 1994). However, these instruments do not explicitly address expectations. The End-of-Life Conversation-Expectation Scale (EOLC-E) was therefore developed, based on literature research and a subsequently evaluated by four experts from palliative care, clinical psychology, and developmental psychology, as the first tool specifically designed to assess expectations related to communication about death and dying (Bendel et al., 2022). Yet, like most existing questionnaires, the EOLC-E focuses solely on individuals’ expectations concerning their own death. Given that conversations inherently involve at least two participants, it is essential to also consider the expectations of those close to the dying person.

This study had two primary objectives. First, it aimed to explore prevailing expectations surrounding end-of-life conversations from the perspective of a significant other (i.e., any person important in the individual’s life) within a German online community sample. Second, it sought to validate an adapted version of the EOLC-E scale, focusing specifically on conversations with a significant other about their death. The psychometric properties and proposed factor structure of the adapted scale were assessed using this sample. Findings from this study contribute to a deeper understanding of caregivers’ expectations in the context of end-of-life communication and provide a foundation for the development of interventions aimed at addressing and modifying identified unfavorable expectations.

Methods

Participants and Procedure

Participants were recruited through in-person interactions, university mailing lists, and social networking sites including Instagram and WhatsApp. The study’s inclusion criteria consisted of being at least 18 years old and having sufficient command of German language to engage with the study material. After providing written informed consent, participants filled out an online questionnaire. A lottery with vouchers (2 vouchers à 50 Euro, 4 vouchers à 25 Euro) and feedback on the study results were offered as incentives for participation. The study was approved by the Ethical Review Committee of the Department of Psychology at Philipps University of Marburg (2021-32k-Amendment) and the study was carried out in accordance with the tenets stated in the Declaration of Helsinki. The study protocol was preregistered in the German Clinical Trials Register (DRKS00034105).

Measures

End-Of-Life Conversation - Expectation Scale – Caregiver Version (EOLC-E-c)

The End-of-Life Conversations – Expectations Scale: Caregiver version (EOLC-E-c) was developed by analogously adapting items from the original German-language scale designed by Bendel et al. (2022) to focus on the death of a significant other instead of one’s own death. This semantic adaptation was conducted by the first author, a native German speaker, and verified by a second native German-speaking researcher to ensure conceptual equivalence and linguistic accuracy. The items reported here are English translations of the German version used in the study. These translations were initially drafted by the research team and subsequently verified by a native English speaker. Instructions prompted participants to imagine having a conversation with a chosen significant other about the significant other’s death and to keep this person in mind while responding to the items. A sample item was “If I talk to a significant other about her/his death or her/his dying, it will make her/him sad”. Participants were asked to indicate their agreement on a 6-point Likert scale from 0 (strongly disagree) to 5 (strongly agree), higher scores indicating increased negative and unfavorable expectations.

Readiness for End-Of-Life Conversations

Participant’s readiness to engage in conversations about death and dying was assessed with the Readiness for End-of-Life Conversations Scale (REOLC; Berlin et al., 2021). A sample item was “I know what advantages talking about the end of my life holds”. Each item was assessed on a scale from 1 (absolutely disagree) to 6 (absolutely agree). Internal consistency was good in this study (a = .80).

Communication About the Death of the Other

Communication about death and illness in close relationships was assessed with the German version of the 5-item Caregiver Communication with Patients about Illness and Death (Bachner et al., 2008; Betker et al., 2024). For the current study, items were converted into the present tense and the broader term ‘significant other’ (nahestehende Person) instead of ‘patient’ (Patient) was used. A sample item was “I avoid talking with a significant other about his/her feelings and fears”. All items were answered on a 5-point Likert type scale from 1 (strongly disagree) to 5 (strongly agree), higher sum scores indicating increased avoidance of communication about death and illness. In the current study, internal consistency was good (a = .81).

Communication Apprehension About Death

Communication apprehension about death was measured using a German version of the 12-item Communication Apprehension About Death Scale (CADS; Carmack & DeGroot, 2016), which had previously been translated by Bendel et al. (2022) for the purpose of validating the initial German version of the EOLC-E Scale. The CADS consists of two subscales “communication apprehension” and “communication avoidance”. A sample item was “I avoid talking about death altogether”. Each item was assessed on a scale from 1 (strongly disagree) to 5 (strongly agree). Cronbach’s alpha in this sample was .92, demonstrating excellent internal consistency of the scale.

Fear of Death, Death Avoidance and Neutral Acceptance

Fear of death, death avoidance and neutral acceptance were assessed using the corresponding subscales in the German version of the Death Attitudes Profile – Revised (Jansen et al., 2019). A sample item was “Death is neither good nor bad”. Participants were asked to indicate their agreement on a 7-point Likert-type scale from 1 (strongly disagree) to 7 (strongly agree). In the current study, Cronbach’s alpha was .91 for “fear of death”, .90 for “death avoidance”, .62 for “death acceptance”.

Trait Anxiety and Trait Depression

Anxiety and depression were measured with the Patient Health Questionnaire-4 (Löwe et al., 2010), consisting of two items on anxiety (GAD-2) and two items on depression (PHQ-2). A sample item was “Over the last 2 weeks, how often have you been bothered by the following problems? Little interest or pleasure in doing things”. Participants were asked to indicate their agreement on a scale from 0 (not at all) to 3 (nearly every day). In the current sample, Cronbach’s alpha was 0.83 in this sample (Cronbach’s α = .74 for the PHQ-2, and Cronbach’s α = .76 for the GAD-2).

Sociodemographic Data, Experiences with Death, and EOL Communication

Sociodemographic data (including sex, age, nationality and education level), personal experiences with one’s own mortality and the death of a significant other, as well as prior engagement in EOL communication, were assessed with self-constructed dichotomous (yes/no) items. To explore reasons for the lack or avoidance of EOL conversations, participants were presented with a list of seven potential contributing factors. For each factor they were asked to indicate whether the factor was a reason (yes), was not a reason (no), or if they had already had these conversations often (I have already spoken about it often).

Statistical Analysis

All data was analyzed using IBM SPSS Statistics 28 with a statistically significant threshold of p ≤ .05 on two-tailed analyses, except for the confirmatory factor analysis, which was conducted using the lavaan package (Rosseel, 2012) for Jamovi (The Jamovi Project, 2025). Confirmatory factor analysis was performed to assess factorial validity and was based on the three-factor model presented by Bendel et al. (2022). Mardia’s standardized coefficient for skewness was 63.4, χ² (1540) = 2644, p < .001, and for kurtosis it was 507.0, z = 17.9, p < .001 indicating non-normality of the data. Consequently, the robust maximum likelihood method was used. To determine the model fit, the following standard fit requirements were considered: χ2-statistics, Comparative Fit Index (CFI), Standardized Root Mean Square Residual (SRMR), Root Mean Square Error of Approximation (RMSEA), Tucker-Lewis -Index (TLI). Cut-off values of CFI and TLI ≥0.90 were indicative of good fit and SRMR and RMSEA ≤0.08 were indicative of an acceptable fit (Brown, 2015; Schermelleh-Engel et al., 2003). Cronbach’s alpha was used to measure internal consistency of the total scale and its subscales. Pearson correlations were calculated between the EOLC-E-c scale and related measurements to examine validity of the scale.

Results

Sample Characteristics

313 individuals signed up for the study of which 63 participants were excluded due to implausibly fast completion speed (less than 7 min) or incompleteness of the questionnaire. Hence, the final sample included 250 participants (M_age = 37.40, SD = 16.39; 185 women, 59 men; 6 diverse). Participants were predominately German (93.6%), in a partnership or married (62.8%), had no children (66.8%), and had an A-levels degree or a university degree (88.8%). 33.6% of the sample indicated no religious denomination and 45.6% identified themselves as not religious. Most participants indicated having experienced the loss of a significant other (82.0%), and having engaged in a conversation about their own death (70.8%) or the death of a significant other (80.4%). Sample characteristics are displayed in Table 1.

Table 1.

Sample Characteristics

Variable	n	%
Gender
Female	185	74.0
Male	59	23.6
Diverse	6	2.4
Nationality
German	234	93.6
Other	16	6.4
Marital status
Single	79	31.6
Married/partnered	157	62.8
Divorced	10	4.0
Widowed	4	1.6
Children^a	83	33.2
Highest educational level
High school	79	31.6
University degree	143	57.2
Other	28	11.2
Employment
Employed	134	53.6
Student	93	37.2
Other	23	9.2
Religious^a	96	38.4
Chronic diseases^a	67	26.8
Death of a loved one^a	205	82.0
Faced own death^a	56	22.4
Faced death of a loved own^a	192	76.8
Conversation about own death^a	177	70.8
Conversation about loved one’s death^a	201	80.4

Note. N = 250. Participants were on average 37.4 years old (SD = 16.39).

^aReflects the number and percentage of participants answering “yes” to this question.

Expectations and Item Analysis

The total mean score of the EOLC-E-c items was M = 2.56 (SD = 0.88) indicating that participants’ agreement fell between somewhat disagree and somewhat agree. Participants most strongly agreed with the item “… it will make me sad” (M = 3.66, SD = 1.27; item 1) and “… it will make him/her sad” (M = 3.04, SD = 1.16; item 8). Participants least strongly agreed with the item “… it will be difficulty for me to ask her/him about her/his wishes (M = 1.51, SD = 1.46; item 20) and “… it will be difficult for me to show my vulnerability” (M = 1.78, SD = 1.60; item 18). Overall, participants agree stronger with items focusing on own emotional burden (M = 2.86, SD = 1.13; items 1-7) and emotional burden of the other person (M = 2.66, SD = 1.09; items 8-14) when engaging in EOL conversations and less with the expected communication self-efficacy (M = 2.11, SD = 1.20; items 15-20). Item characteristics are displayed in Table 2.

Table 2.

Item Characteristics and Results From a Factor Analysis of the EOLC-E-c

EOLC-E-c item	Item characteristics		Standardized factor loading
EOLC-E-c item	M	SD	1	2	3
If I talk to a significant other about her/his death or her/his dying, …
Factor 1: Expected own emotional burden
1. … it will make me sad	3.66	1.27	.769
2. … it will make me feel depressed	2.86	1.35	.792
3. … it will burden me emotionally	3.23	1.32	.855
4. … it will be difficult for me to stand	2.02	1.36	.796
5. … I will be afraid of her/his death	2.84	1.45	.781
6. … I will be afraid of her/his dying	2.96	1.40	.749
7. … it will make me feel insecure	2.44	1.43	.731
Factor 2: Expected emotional burden of the other person
8. … it will make her/him sad	3.04	1.16		.770
9. … it will make her/him feel depressed	2.71	1.21		.830
10. … it will burden her/him emotionally	2.99	1.28		.887
11. … it will be difficult for her/him to stand that	2.40	1.34		.861
12. … she/he will be afraid of her/his death	2.48	1.30		.846
13. … she/he will be afraid of her/his dying	2.52	1.30		.844
14. … it will make her/him feel insecure	2.47	1.29		.761
Factor 3: Expected communication self-efficacy
15. … I won’t know what exactly to talk about	2.11	1.40			.814
16. … it will be difficult for me to find the right words	2.60	1.39			.876
17. … it will be difficult for me to initiate the conversation	2.80	1.52			.813
18. … it will be difficult for me to show my vulnerability	1.78	1.60			.601
19. … it will be difficult for me to express my feelings	1.84	1.53			.622
20. … it will be difficult for me to ask her/him about her/his wishes	1.51	1.46			.652

Note. Standardized factor loadings after allowing residual covariances between items 5-6, 12-13, 18-19.

Confirmatory Factor Analysis

The three-factor model showed a non-acceptable fit (χ² (167) = 921, p < .001, CFI = .798, TLI = .770, SRMR = .110, RMSEA = .145). Given the unsatisfactory fit, the modification indices matrix was inspected. Items 18-19, 12-13, 5-6 showed the highest covariances. These item pairs share nearly identical phrasing and reflect very similar underlying context, which likely explains the overlap in their residual variance. Allowing residual covariances between these items resulted in an improved fit (χ² (164) = 470, p < .001, CFI = .917, TLI = .904, SRMR = .058, RMSEA = .094). The fit indices indicated that the respecified three-factor model provided an overall acceptable fit to the data, although the RMSEA remained above the conventional cutoff (Table 3).

Table 3.

Fit Indices of the Confirmatory Factor Analysis of the EOLC-E-c

	χ² (df)	CFI	TLI	SRMR	RMSEA	90%-CI of RMSEA
Model 1	921 (167)	.798	.770	.110	.145	.136-.154
Model 2	470 (164)	.917	.904	.058	.094	.084-.104

Note. Model 1 = initial model. Model 2 = allowing residual covariances between items 5-6, 12-13, 18-19. TLI = Tucker-Lewis Index; CFI = comparative fit index; SRMR = standardized root mean square residual; RMSEA = root-mean-square error of approximation; CI = confidence interval.

Internal Consistency

Excellent internal consistency was demonstrated for the EOLC-E-c scale (α = .92) and its subscales: expected own emotional burden (α = .92), expected emotional burden of the other person (α = .94), and communication self-efficacy (α = .90).

Validity

Convergent validity was supported by a medium to large correlation with scales assessing communication apprehension about death (r = .56, p < .01), avoidance of caregiver communication with patients about illness and death (r = .64, p < .01), readiness for end-of-life conversations (r = −.40, p < .01) fear of death (r = .48, p < .01) and death avoidance (r = .52, p < .01). However, a lower-than-expected correlation was found with the subscale of the DAP-GR assessing neutral acceptance (r = −.19, p < .01). In support of discriminant validity, small correlations with the PHQ-2 (r = .20, p < .01) and the GAD-2 measure (r = .15, p < .05) were found. Further correlations with the three subscales of the EOLC-E-c scale are shown in Table 4.

Table 4.

Correlations Between EOLC-E-c, its Subscales, and Related Measures

	CADS	CCID	REOLC	Fear of death	Death avoidance	Neutral acceptance	PHQ-2	GAD-2
EOLC-E-c, total score	.56**	.64**	−.40**	.48**	.52**	−.19**	.20**	.15*
Expected own emotional burden	.53**	.53**	−.39**	.47**	.47**	−.21**	.23**	.19**
Expected other person’s emotional burden	.23**	.28**	−.14*	.27**	.23**	−.09	.06	.01
Communication self-efficacy	.55**	.68**	−.40**	.37**	.50**	−.14*	.17**	.15*

Note. N = 250. CADS = Communication Apprehension About Death Scale; CCID = Caregiver Communication about Illness and Death scale; REOLC = Readiness for End-of-Life Conversations scale.

Correlations depict Pearson correlations.

*p < .05 (two-tailed). **p < .01 (two-tailed).

Additional Analyses

A multiple linear regression analysis was conducted to examine whether expectations concerning end-of-life conversations varied as a function of demographic characteristics, death-related experiences and death attitudes (see Table 5). The overall model was significant F (13, 236) = 12.38, p < .001, explaining 40.5% of the variance in expectations (R² = .405, adjusted R² = .373). Demographic variables including age, gender, educational level, religious belief, and caregiving experience were not significant predictors of negative expectations. Among the death-attitude variables, death avoidance (B = .196, p < .001) and death anxiety (B = .155, p < .001) significantly predicted increased negative expectations, whereas death acceptance did not. Prior experience with conversations about death also emerged as a significant predictor. Participants who previously engaged in conversations about their own death reported lower negative expectations (B = −.420, p < .001), and the same pattern was observed for conversations about the death of a close other (B = −.272, p = .032). In contrast, experience of loss, and confrontation with death, whether one’s own or that of a significant other, were unrelated to expectations (all ps > .41). Selected reasons for the lack or reduction of end-of-life conversations show that these discussions are not taking place due to the societal, cultural, familial, and social environments participants are part of, rather than because of negative portrayals in films, articles, or negative experiences shared by others (see Table 6).

Table 5.

Effects of Demographics, Death-Related Experiences and Death Attitudes on EOLC-E-c

	Estimate	SE	95% Cl		p
	Estimate	SE	LL	UL	p
Age	−.001	.003	−.006	.005	.845
Gender	.153	.083	−.011	.317	.067
Educational level	.009	.062	−.112	.130	.882
Religious belief	.068	.044	−.018	.154	.121
Experience caregiving	−.066	.103	−.270	.137	.521
Death avoidance	.196	.046	.105	.287	<.001**
Death anxiety	.155	.042	.073	.237	<.001**
Death acceptance	.020	.065	−.108	.149	.754
Conversation death, other	−.272	.126	−.521	−.024	.032*
Conversation death, own	−.420	.114	−.644	−.196	<.001**
Confrontation death, own	.090	.109	−.126	.306	.412
Confrontation death, other	−.002	.115	−.228	.224	.985
Experience of loss	.077	.123	−.165	.320	.530

Note. *p < .05 (two-tailed). **p < .01 (two-tailed).

Table 6.

Reasons for the Lack of EOL Conversations

	No	Yes
… it is rarely talked about in my family	26 (25.5%)	76 (74.5%)
… it is rarely talked about among my friends	19 (18.6%)	83 (81.4%)
… it is rarely talked about in my cultural community	33 (32.4%)	69 (67.6%)
… it is rarely talked about in the society I live in	24 (23.5%)	78 (76.5%)
… other people have told me about their negative experiences with it	92 (90.2%)	10 (9.8%)
… I have read negative things about it	96 (94.1%)	6 (5.9%)
… I have seen negative portrayals of it in films or on television	93 (91.2%)	9 (8.8%)

Note. N = 102. Subgroup only includes participants who did not indicate they have already spoken about EOL often.

Discussion

This study explored expectations surrounding communication about death and dying from the perspective of caregivers and validated the corresponding scale (EOLC-E-c) in a German online community sample. Most participants reported prior engagement in conversations about death. Overall, participants’ agreement lay between somewhat disagree and somewhat agree, as reflected by the average score of the EOLC-E-c scale. Agreement was highest for items referring to elicited sadness—both one’s own and that of the conversational partner—whereas disagreement was strongest for items referring to difficulties in inquiring about the conversational partner’s wishes and in displaying vulnerability.

After respecification, confirmatory factor analysis yielded support for the 20-item three-factor structure comprising: (1) expected own emotional burden, (2) expected emotional burden of the other person, and (3) communication self-efficacy. These factors underscore the shared anticipation of emotional strain for both the self and the conversation partner during EOL discussions.

Supporting its validity, the EOLC-E-c scale was positively associated with measures assessing fear of death, death avoidance, communication apprehension about death, and avoidance of conversations about death and illness. It was negatively associated with readiness for end-of-life conversations. Small correlations with depression (PHQ-2) and anxiety (GAD-2) measures support the scale’s discriminant validity, suggesting that the EOLC-E-c captures distinct constructs. Contrary to most previous findings (Bachner et al., 2021; Seifart et al., 2020; Strupp et al., 2021), the majority of participants in the current sample, reported prior engagement in EOL conversations – either about their own death or that of a significant other. This discrepancy may reflect a selection bias, as the study’s objective was transparently communicated, possibly deterring individuals who habitually avoid such conversations. A similar trend was observed in a recent Dutch study (Haaksman et al., 2024).

Interestingly, the findings suggest that mere confrontation with death (e.g., through witnessing or working around dying individuals) is not sufficient to explain interindividual differences in EOL communication expectations. Rather, prior conversations about death are associated with reduced negative expectations, suggesting that such experiences may be linked to differences in held beliefs. In contrast, confrontation without communication appears not to be associated with similar differences in expectations. These findings emphasize the communicative nature of expectation change. However, given the cross-sectional design, causal conclusions cannot be drawn. It remains unclear whether reduced negative expectations lead to more frequent EOL conversations—or whether engaging in such conversations reduces negative expectations. Longitudinal research is needed to determine the directionality of this relationship.

Participants in this study were predominantly German, highly educated and identified as either non-religious or Christian. Previous research has shown that one’s cultural background, religious beliefs, and educational level may also explain differences in the extent of EOL conversations (De Graaff et al., 2012; Piracha et al., 2024). To better understand predictors of EOL conversations and associated expectations, future studies should aim for more diverse samples in terms of culture, religion, and socioeconomic status. This is especially relevant given that participants cited sociocultural environments as barriers to engaging in EOL conversations. Understanding these barriers is essential for developing effective interventions and public education efforts. Internationally, several initiatives have emerged to promote open dialogue about death, such as festivals addressing death and grief (Endlich.menschlich, n.d.), death cafes (Miles & Corr, 2017), public health campaigns such as the “Dying to Know Day” in Australia (Proveda, n.d), and websites promoting conversations about death such as the website “Death Over Dinner” (Death Over Dinner, n.d). The latter may be particularly useful for individuals with low communication self-efficacy, as it offers conversational prompts and guidance on initiating discussions. Death Cafés, by providing informal spaces to talk about death, may help individuals confront and adjust their expectations about emotional burden and communication difficulty. Likewise, information campaigns could normalize death-related discussions and reduce avoidance by highlighting their benefits.

Several potential limitations of the current study should be acknowledged. Confirmatory factor analysis indicated an acceptable fit, however the RMSEA remained above conventional cutoffs, suggesting that future research should replicate these findings in a large sample and may consider shortening the item set to reduce redundancy. Furthermore, despite efforts to recruit a culturally diverse sample, such as recruiting during a cultural exchange evening, the sample was relatively homogenous. The German language of the questionnaire may have presented a hindrance in this regard. Future studies should replicate the study in other languages. The study validated the scale in the general population and not in an active caregiver population. Thereby limiting the direct clinical validity and generalizability of the scale for active or clinical caregiver populations. Future studies should replicate the current study with a sample of caregivers of individuals with life-limiting diseases and healthcare workers who are frequently confronted with death, in order to broaden the scope of its application.

Practical Implications

This study offers important implications for both research and practice. Investigating the underlying reasons for avoiding EOL conversations can inform future interventions and educational campaigns. Using the EOLC-E-c in conjunction with its parallel version could serve as a screening tool in clinical settings to identify individuals with high levels of negative expectations. Tailored interventions—such as cognitive restructuring—could then be implemented to reduce these expectations and facilitate EOL communication.

Conclusion

This study supports the validity and reliability of the EOLC-E-c scale in a German online community sample. The scale can be employed in both research and clinical contexts to assess expectations regarding end-of-life conversations with significant others. By identifying and addressing negative expectations, interventions can be developed to facilitate such conversations, potentially improving end-of-life experiences for both patients and caregivers.

Footnotes

ORCID iDs

Sophie Kievernagel

Yannik Bendel

Christian Schulz-Quach

Pia von Blanckenburg

Ethical Approval

Ethical Review Committee of the Department of Psychology at Philipps University of Marburg (2021-32k-Amendment), German Clinical Trials Register (DRKS00034105).

Consent to Participate

Written consent was provided.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Deutsche Forschungsgemeinschaft (DFG; German Research Foundation), project number 290878970-GRK2271, Project 10.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

Upon request.*

Author Biographies

Sophie Kievernagel, M.Sc., is a PhD candidate and is pursuing training as a cognitive-behavioral psychotherapist. Her research focuses on expectations concerning end-of-life communication.

Yannik Bendel, PhD, is a registered clinical psychologist. He works in an outpatient clinic, where he combines clinical practice with research on psychotherapy.

Christian Schulz-Quach, MD, is an Associate Professor and Clinical Educator in the Department of Psychiatry at the University of Toronto and a staff psychiatrist at the University Health Network, Princess Margaret Cancer Centre, Department of Supportive Care, Division of Psychosocial Oncology.

Pia von Blanckenburg, PhD, is a registered clinical psychologist with experience in both psychotherapy and psycho-oncology. She has been conducting research in the field of psycho-oncology for over ten years.

References

Bachner

Y. G.

Davidov

Carmel

(2008). Caregivers’ communication with patients about illness and death: Initial validation of a scale. OMEGA - Journal of Death and Dying, 57(4), 381–397. https://doi.org/10.2190/OM.57.4.d

Bachner

Y. G.

Guldin

M.-B.

Nielsen

M. K.

(2021). Mortality communication and post-bereavement depression among Danish family caregivers of terminal cancer patients. Supportive Care in Cancer: Official Journal of the Multinational Association of Supportive Care in Cancer, 29(4), 1951–1958. https://doi.org/10.1007/s00520-020-05685-6

Bendel

Pinquart

Schulz-Quach

von Blanckenburg

(2022). Expectations in the communication about death and dying: Development and initial validation of the end-of-life conversations – Expectations scale. Omega, 90(2), 710–725. https://doi.org/10.1177/00302228221110726

Bendel

Pinquart

Schulz-Quach

von Blanckenburg

(2025). Changing expectations toward end-of-life communication: An experimental investigation. Patient Education and Counseling, 131, 108571. https://doi.org/10.1016/j.pec.2024.108571

Berlin

Leppin

Nagelschmidt

Seifart

Rief

von Blanckenburg

(2021). Development and validation of the readiness for end-of-life conversations (REOLC) scale. Frontiers in Psychology, 12, 662654. https://doi.org/10.3389/fpsyg.2021.662654

Betker

Senßfelder

Knorrenschild

J. R.

Volberg

Berthold

Seifart

von Blanckenburg

(2024). Difficulties of cancer patients’ relatives in end-of-life discussions: Validation of a questionnaire. Journal of Pain and Symptom Management, 67(5), 420–428. https://doi.org/10.1016/j.jpainsymman.2024.02.003

Brown

T. A.

(2015). Confirmatory factor analysis for applied research, second edition. Guilford Publications.

Carmack

H. J.

DeGroot

J. M.

(2016). Development and validation of the communication apprehension about death scale. OMEGA - Journal of Death and Dying, 74(2), 239–259. https://doi.org/10.1177/0030222815598440

Death Over Dinner . (n.d.). Death over dinner. https://deathoverdinner.org/

10.

De Graaff

F. M.

Mistiaen

Devillé

W. L.

Francke

A. L.

(2012). Perspectives on care and communication involving incurably ill Turkish and Moroccan patients, relatives and professionals: A systematic literature review. BMC Palliative Care, 11(1), 17. https://doi.org/10.1186/1472-684X-11-17

11.

Detering

K. M.

Hancock

A. D.

Reade

M. C.

Silvester

(2010). The impact of advance care planning on end of life care in elderly patients: Randomised controlled trial. BMJ, 340, c1345. https://doi.org/10.1136/bmj.c1345

12.

Endlich.menschlich . (n.d.). Stadtfestival & Public Health und Palliative Care Kongress 2024. https://endlich-menschlich.ch/festival/

13.

Fenton

A. T.

Blashill

Revette

A. C.

Mann

Reblin

Enzinger

A. C.

Ornstein

K. A.

Jain

Manz

C. R.

Tulsky

J. A.

Wright

A. A.

(2024). You want to shield your kids’: Patients’ and their adult-children’s serious illness conversations. Journal of Pain and Symptom Management, 69(5), 435–443. https://doi.org/10.1016/j.jpainsymman.2024.12.014

14.

Generous

M. A.

Keeley

M. P.

(2022). Exploring the connection between end-of-life relational communication and personal growth after the death of a loved one. Omega, 84(3), 792–810. https://doi.org/10.1177/0030222820915312

15.

Haaksman

Ham

Brom

Baars

Van Basten

J.-P.

Van Den Borne

B. E. E. M.

Hendriks

M. P.

De Jong

W. K.

Van Laarhoven

H. W. M.

Van Lindert

A. S. R.

Mandigers

C. M. P. W.

Van Der Padt-Pruijsten

Smilde

T. J.

Van Zuylen

L. C.

Van Vliet

L. M.

Raijmakers

N. J. H.

(2024). Open communication between patients and relatives about illness & death in advanced cancer—Results of the eQuiPe study. Supportive Care in Cancer: Official Journal of the Multinational Association of Supportive Care in Cancer, 32(4), 214. https://doi.org/10.1007/s00520-024-08379-5

16.

Jansen

Schulz-Quach

Eisenbeck

Carreno

D. F.

Schmitz

Fountain

Franz

Schäfer

Wong

P. T. P.

Fetz

(2019). German version of the death attitudes profile-revised (DAP-GR) – Translation and validation of a multidimensional measurement of attitudes towards death. BMC Psychology, 7(1), 61. https://doi.org/10.1186/s40359-019-0336-6

17.

Jonasson

J. M.

Hauksdóttir

Nemes

Surkan

P. J.

Valdimarsdóttir

Onelöv

Steineck

(2011). Couples’ communication before the wife’s death to cancer and the widower’s feelings of guilt or regret after the loss – A population-based investigation. European Journal of Cancer, 47(10), 1564–1570. https://doi.org/10.1016/j.ejca.2011.01.010

18.

Löwe

Wahl

Rose

Spitzer

Glaesmer

Wingenfeld

Schneider

Brähler

(2010). A 4-item measure of depression and anxiety: Validation and standardization of the patient health questionnaire-4 (PHQ-4) in the general population. Journal of Affective Disorders, 122(1–2), 86–95. https://doi.org/10.1016/j.jad.2009.06.019

19.

Miles

Corr

C. A.

(2017). Death cafe: What is it and what we can learn from it. Omega, 75(2), 151–165. https://doi.org/10.1177/0030222815612602

20.

Nagelschmidt

Leppin

Seifart

Rief

von Blanckenburg

(2021). Systematic mixed-method review of barriers to end-of-life communication in the family context. BMJ Supportive & Palliative Care, 11(3), 253–263. https://doi.org/10.1136/bmjspcare-2020-002219

21.

Panitz

Endres

Buchholz

Khosrowtaj

Sperl

M. F. J.

Mueller

E. M.

Schubö

Schütz

A. C.

Teige-Mocigemba

Pinquart

(2021). A revised framework for the investigation of expectation update versus maintenance in the context of expectation violations: The ViolEx 2.0 model. Frontiers in Psychology, 12, 726432. https://doi.org/10.3389/fpsyg.2021.726432

22.

Patel

M. I.

Sundaram

Desai

Periyakoil

V. S.

Kahn

J. S.

Bhattacharya

Asch

S. M.

Milstein

Bundorf

M. K.

(2018). Effect of a lay health worker intervention on goals-of-care documentation and on health care use, costs, and satisfaction among patients with cancer: A randomized clinical trial. JAMA Oncology, 4(10), 1359–1366. https://doi.org/10.1001/jamaoncol.2018.2446

23.

Piracha

N. Z.

Nickel

L. B.

Quryshi

Salah

Padela

A. I.

(2024). Muslims and end-of-life healthcare in Non-Muslim majority nations: A systematic literature review. Journal of Pain and Symptom Management, 67(4), e299–e312. https://doi.org/10.1016/j.jpainsymman.2024.01.004

24.

Proveda . (n.d.). About dying to know day. Proveda. https://proveda.com.au/community-programs/dying-to-know-day/

25.

Rief

Shedden-Mora

M. C.

Laferton

J. A. C.

Auer

Petrie

K. J.

Salzmann

Schedlowski

Moosdorf

(2017). Preoperative optimization of patient expectations improves long-term outcome in heart surgery patients: Results of the randomized controlled PSY-HEART trial. BMC Medicine, 15(1), 4. https://doi.org/10.1186/s12916-016-0767-3

26.

Rosseel

(2012). Lavaan: An R package for structural equation modeling. Journal of Statistical Software, 48(2). https://doi.org/10.18637/jss.v048.i02

27.

Sallnow

Smith

Ahmedzai

S. H.

Bhadelia

Chamberlain

Cong

Doble

Dullie

Durie

Finkelstein

E. A.

Guglani

Hodson

Husebø

B. S.

Kellehear

Kitzinger

Knaul

F. M.

Murray

S. A.

Neuberger

O’Mahony

Tutu van Furth

Lancet Commission on the Value of Death . (2022). Report of the lancet commission on the value of death: Bringing death back into life. Lancet (London, England), 399(10327), 837–884. https://doi.org/10.1016/S0140-6736(21)02314-X

28.

Scheinfeld

Lake

P. K.

(2019). Why is it just So hard? Making sense of end-of-life communication between adult children and their terminally ill parental figures. Omega, 83(3), 611–629. https://doi.org/10.1177/0030222819859163

29.

Schermelleh-Engel

Moosbrugger

Müller

(2003). Evaluating the fit of structural equation models: Tests of significance and descriptive goodness-of-fit measures. 8(2).

30.

Seifart

Riera Knorrenschild

Hofmann

Nestoriuc

Rief

von Blanckenburg

(2020). Let Us talk about death: Gender effects in cancer patients’ preferences for end-of-life discussions. Supportive Care in Cancer: Official Journal of the Multinational Association of Supportive Care in Cancer, 28(10), 4667–4675. https://doi.org/10.1007/s00520-019-05275-1

31.

Shedden-Mora

M. C.

Pan

Heisig

S. R.

von Blanckenburg

Rief

Witzel

Albert

U.-S.

Nestoriuc

(2020). Optimizing expectations about endocrine treatment for breast cancer: Results of the randomized controlled PSY-BREAST trial. Clinical Psychology in Europe, 2(1), Article e2695. https://doi.org/10.32872/cpe.v2i1.2695

32.

Strupp

Köneke

Rietz

Voltz

(2021). Perceptions of and attitudes toward death, dying, grief, and the finitude of life—A representative survey among the general public in Germany. Omega, 84(1), 157–176. https://doi.org/10.1177/0030222819882220

33.

The Jamovi Project . (2025). Jamovi. https://www.jamovi.org

34.

Wendler

Rid

(2011). Systematic review: The effect on surrogates of making treatment decisions for others. Annals of Internal Medicine, 154(5), 336–346. https://doi.org/10.7326/0003-4819-154-5-201103010-00008

35.

Wildfeuer

Schnell

M. W.

Schulz

(2015). Talking about dying and death: On new discursive constructions of a formerly postulated taboo. Discourse & Society, 26(3), 366–390. https://doi.org/10.1177/0957926514564739

36.

Wong

P. T. P.

Reker

G. T.

Gesser

(1994). Death attitude profile—revised: A multidimensional measure of attitudes toward death. In Death anxiety handbook: Research, instrumentation, and application (pp. 121–148). Taylor & Francis.

37.

Wright

A. A.

Zhang

Ray

Mack

J. W.

Trice

Balboni

Mitchell

S. L.

Jackson

V. A.

Block

S. D.

Maciejewski

P. K.

Prigerson

H. G.

(2008). Associations between end-of-life discussions, patient mental health, medical care near death, and caregiver bereavement adjustment. JAMA, 300(14), 1665–1673. https://doi.org/10.1001/jama.300.14.1665

38.

Yamaguchi

Maeda

Hatano

Mori

Shima

Tsuneto

Kizawa

Morita

Yamaguchi

Aoyama

Miyashita

(2017). Effects of end-of-life discussions on the mental health of bereaved family members and quality of patient death and care. Journal of Pain and Symptom Management, 54(1), 17–26. https://doi.org/10.1016/j.jpainsymman.2017.03.008

39.

Zhang

Wright

A. A.

Huskamp

H. A.

Nilsson

M. E.

Maciejewski

M. L.

Earle

C. C.

Block

S. D.

Maciejewski

P. K.

Prigerson

H. G.

(2009). Health care costs in the last week of life: Associations with end-of-life conversations. Archives of Internal Medicine, 169(5), 480–488. https://doi.org/10.1001/archinternmed.2008.587

Expectations in the Communication About Death and Dying From a Caregiver’s Perspective: The End-Of-Life Conversation – Expectation Scale – Caregiver (EOLC-E-c)

Abstract

Keywords

Introduction

Methods

Participants and Procedure

Measures

End-Of-Life Conversation - Expectation Scale – Caregiver Version (EOLC-E-c)

Readiness for End-Of-Life Conversations

Communication About the Death of the Other

Communication Apprehension About Death

Fear of Death, Death Avoidance and Neutral Acceptance

Trait Anxiety and Trait Depression

Sociodemographic Data, Experiences with Death, and EOL Communication

Statistical Analysis

Results

Sample Characteristics

Expectations and Item Analysis

Confirmatory Factor Analysis

Internal Consistency

Validity

Additional Analyses

Discussion

Practical Implications

Conclusion

Footnotes

ORCID iDs

Ethical Approval

Consent to Participate

Funding

Declaration of Conflicting Interests

Data Availability Statement

Author Biographies

References