Death-related distress in adult primary brain tumor patients

Abstract

Background

A diagnosis of cancer may increase mortality salience and provoke death-related distress. Primary brain tumor (PBT) patients may be at particular risk for such distress given the certainty of tumor progression, lack of curative treatments, and poor survival rates. This study is the first to examine the prevalence of death-related distress and its correlates in PBT patients.

Methods

Adult PBT patients (N = 105) enrolled in this cross-sectional study and completed the Death Distress Scale (subscales: Death Depression, Death Anxiety, Death Obsession), Generalized Anxiety Disorder–7, and Patient Health Questionnaire–9. Prevalence and predictors of death-related distress, and the relationships of demographic variables to clusters of distress, were explored.

Results

The majority of PBT patients endorsed clinically significant death-related distress in at least one domain. Death anxiety was endorsed by 81%, death depression by 12.5%, and death obsession by 10.5%. Generalized anxiety was the only factor associated with global death-related distress. Cluster analysis yielded 4 profiles: global distress, emotional distress, resilience, and existential distress. Participants in the resilience cluster were significantly further out from diagnosis than those in the existential distress cluster. There were no differences in cluster membership based on age, sex, or tumor grade.

Conclusions

PBT patients appear to have a high prevalence of death-related distress, particularly death anxiety. Further, 4 distinct profiles of distress were identified, supporting the need for tailored approaches to addressing death-related distress. A shift in clusters of distress based on time since diagnosis also suggest the need for future longitudinal assessment.

For centuries, philosophers and researchers have contemplated the unique human experience of thanatophobia, or the fear of dying.¹ Although advancements in medical treatments have lengthened life expectancy, an increasing number of individuals are now faced with the challenge of “double awareness”—simultaneously confronting death while engaging and finding meaning in life.² This duality may be difficult to sustain for some individuals who feel dominated by or preoccupied with distress related to dying and death.³ Although death-related distress of at least moderate severity has been found to be common in nonclinical populations (reported by the majority of respondents in some studies), there is variation based on age, sex, socioeconomic status, and education.^4,5 However, these findings may not actually be comparable to rates in patients with advanced disease because death anxiety in healthy individuals appears to be a different phenomenon.⁶ Advanced cancer is a heterogeneous group of diseases that can increase mortality salience and provoke existential distress.⁷

Patients with primary brain tumors (PBTs) may be at heightened risk of distress due to the certainty of tumor progression, lack of curative treatments, and poor prognosis of their disease (the 5-year survival rate for glioblastoma multiforme ranges from 5% to 19%, anaplastic astrocytoma from 14% to 54%, and oligodendroglioma from 46% to 70%⁸). Sustaining “double awareness” and managing death-related distress may, therefore, be more challenging for PBT patients than those with other cancers. Based on data obtained from qualitative interviews, Adelbratt and Strang³ found that PBT patients and close kin both experienced emotional distress, contemplated existential issues during the treatment process, and reported living with constant fear related to their imminent dying process.

Death-related distress is a multidimensional construct made up of psychological and emotional factors that affect one’s daily functioning. Previously established dimensions of death-related distress include death anxiety, death depression, and death obsession.⁹ Death anxiety, which refers to the fear of death and dying, has been endorsed by approximately one-third of individuals with advanced cancer,^7,10 although rates vary across cancer types. Overall, individuals with elevated death anxiety report more general symptoms of anxiety and depression.^10–13

Death depression refers to feelings of sadness, despair, loss, and loneliness that occur in response to thoughts about death.¹⁴ The current literature indicates that the prevalence of death depression across cancer types may not vary significantly from that of the general population^15,16 and that it is associated with other typical depressive symptoms such as changes in sleep and appetite.¹⁷ Death depression has also been found to positively correlate with both state (transient) and trait (stable, personality related) anxiety in patients with breast, lung, or colon cancers.¹⁶

Death obsession is defined as repetitive thoughts, ruminations, persistent ideas, or intrusive thoughts centered around death.¹⁸ A study by Soo and Sherman¹⁹ that examined ruminative thought processes about cancer generally, found that it was most positively associated with depression, anxiety, and stress. These authors also reported a negative relationship between illness rumination and posttraumatic growth. A similar relationship between cancer rumination and emotional distress was found among women with ovarian cancer²⁰ and mixed-diagnosis cancer survivors.²¹ However, to date, there have been no reported studies measuring death obsession specifically in oncology patients.

Unfortunately, research conducted thus far may underestimate the emotional and death-related distress experienced by PBT patients, in part, because the research in neuro-oncology has been sparse. A recent systematic review revealed that PBT patients are disproportionately excluded from studies of such death-related distress,²² representing only 0.18% of all participants in these existentially themed investigations. Most of the studies included in this review intentionally eliminated PBT patients from participation or had exclusion criteria related to cognitive impairment or specific treatment parameters that may exclude their participation. As a result, there is little evidence-based guidance for screening, monitoring, and treating death-related distress in neuro-oncology.

The conceptual framework for this study was developed following the extensive systematic literature review suggesting limited inclusion of PBT patients in psycho-oncology existential literature²² as well as the authors’ clinical experience of working with PBT patients and assessing their quality-of-life needs. This study aimed to: (1) assess the prevalence of death-related distress in PBT patients; (2) explore the correlates and predictors of death-related distress, including demographic, medical, and psychological variables; and (3) using cluster analysis, determine whether distinct groups of distress could be identified in this population and whether these groups differ based on age, sex, tumor grade, and time since diagnosis. We hypothesized that death-related distress in PBT patients would be correlated with general distress; however, given the exploratory nature of this study, no specific hypotheses regarding prevalence or cluster patterns were proposed.

Methods

Design and Participants

Cross-sectional data were obtained at an urban National Cancer Institute–designated cancer center over the course of 2 consecutive summers (2018 and 2019). Patients were recruited during routine neuro-oncology clinic visits. Ethical approval was granted from the institutional review board and informed consent was obtained from all study patients. Inclusion criteria were: (1) confirmed primary brain tumor diagnosis via histopathology, (2) age older than 18 years, (3) a minimum of 2 weeks postsurgical resection or biopsy (if applicable), and (4) English speaking. The flowchart in Figure 1 illustrates the inclusion and attrition process. Demographic and medical variables were extracted from medical records following informed consent. Of the 198 PBT patients (grades I-IV) seen in clinic during the data collection time period, 132 were approached for participation, and 105 participants (M_age =50.4 years, SD = 15.1; range, 20-86 years) were included in the final statistical analysis.

Fig. 1.

Flowchart of Participant Recruitment.

Measures

Death Distress Scale

⁹The Death Distress Scale (DDS) is a 24-item self-report screening inventory of death-related distress in adults, derived from 3 previously validated questionnaires: the Death Anxiety Scale,²³ Death Depression Scale,²⁴ and Death Obsession Scale.¹⁸ The DDS retains this 3-factor structure, with each domain containing 8 statements that can be rated on a 5-point scale ranging from 1 (no) to 5 (very much). The score for each factor is based on the sum of all items within each domain yielding a death anxiety, death depression, and death obsession score. The following guidelines have been proposed to interpret the factor scores: 8 to 18 mild, 19 to 29 moderate, and 30 to 40 severe.²⁵ A total score can also be calculated to assess global existential distress (DDS-total), with a higher score indicative of greater distress. To our knowledge, there have been no large-scale validation studies for use of this measure in a cancer population.

Generalized Anxiety Disorder

²⁶The Generalized Anxiety Disorder–7 Items (GAD-7) is a 7-item self-report inventory of anxiety symptoms in adults. The frequency of each symptom over the last 2 weeks is rated on a 4-point scale: 0 (not at all) to 3 (nearly every day). The sum of all item responses yields a total score of anxiety symptoms. The following guidelines have been proposed to interpret scores: 0 to 4 minimal, 5 to 9 mild, 10 to 14 moderate, and 15 to 21 severe anxiety. The American Society of Clinical Oncology recommends the GAD-7 for the assessment of anxiety symptoms among those with cancer.²⁷

Patient Health Questionnaire

²⁸The Patient Health Questionnaire–9 Items (PHQ-9) is a 9-item self-report inventory of major depressive symptoms in adults derived from the validated full Patient Health Questionnaire. The frequency of each symptom over the last 2 weeks is rated on a 4-point scale: 0 (not at all) to 3 (nearly every day). The sum of all items yields a total score of depressive symptoms. The following guidelines have been proposed to interpret scores: 0 to 4 minimal, 5 to 9 mild, 10 to 14 moderate, 15 to 19 moderately severe depression, and greater than 20 severe depression. The PHQ-9 was developed specifically for use in a health-care setting and demonstrates clinical utility in screening for depression among cancer patients.²⁹ As such, the PHQ-9 is one of the measures recommended by the American Society of Clinical Oncology for the assessment of depressive symptoms among those with cancer.²⁷

Statistical Analysis

Data analyses were conducted using SPSS version 26. Descriptive analyses of PBT patients’ demographic, medical, and emotional distress variables were performed. Frequencies were run to determine preliminary prevalence rates of clinically elevated death anxiety, death depression, and death obsession using the DDS published cutoffs.²⁵ Independent t tests and chi-square tests were used to compare demographic and medical profiles of those endorsing clinically significant levels of death-related distress (defined as an elevation on any 1 of the 3 DDS factors). Correlations were then calculated to determine the relationship between death-related distress (DDS-total) and general emotional distress (GAD-7 and PHQ-9). Next, to identify factors related to death-related distress, we used a forward stepwise multiple regression analysis that included demographic and medical variables as well as measures of depressive and anxiety symptoms. This selection was based on literature review,^10,22,30,31 diagnosis-related plausibility, and clinical experience. Finally, to identify the presence of distinct patterns of distress, a multivariate clustering analysis was conducted, using the 3 subscales of the DDS (death anxiety, death depression, and death obsession), the PHQ-9 (depression), and GAD-7 (generalized anxiety). Scores from theses rating scales were first standardized by converting to z scores and then analyzed using a k-means cluster analysis. An agglomerative, hierarchical cluster analysis was performed to determine the number of clusters, with squared Euclidean distances used in the proximity’s matrix and between-groups average linkage as the clustering method. SDs from the cluster center were interpreted as follows: ±0 to 0.49 equals near average, ±0.5 to 0.99 equals slightly above/below average, ±1 to 1.49 equals moderately above/below average, ±1.5+ equals well above/below average. Cluster group comparisons were completed using chi-square tests and post hoc analysis of variance (ANOVA) based on age, sex, time since diagnosis, and tumor grade. The critical value for all analyses was P ≤ .05.

Results

Sample Demographics

The sample was predominantly white (80%) and female (54%). Most patients were diagnosed with a glioblastoma multiforme (31%), followed by oligodendroglioma (20%), astrocytoma (20%), and meningioma (13%). Time since diagnosis ranged from less than 1 month to more than 22 years (M = 53 months). Of the sample, 72% underwent initial resection of tumor, 79% underwent radiation treatment, and 74% had completed at least one cycle of chemotherapy at the time of evaluation; these groups are not exclusive. Table 1 provides full demographic information of the sample.

Table 1.

Demographics of Primary Brain Tumor Sample (N = 105)

Demographic			Frequency	Mean/Percentage
Age, y			105	50.36 (range, 20-86)
Sex
	Male		48	45.7
	Female		57	54.3
Race
	White		57	80.0
	African American		19	18.1
	Other		2	1.9
Tumor type
	Meningioma		14	13.3
		Grade I	10	71.4
		Grade II	3	21.4
		Grade III	1	7.1
	Astrocytoma		21	20.0
		Grade I	3	14.3
		Grade II	14	66.7
		Grade III	4	19.0
	Oligodendroglioma		21	20.0
		Grade II	12	57.1
		Grade III	9	42.9
	Glioblastoma multiforme		33	31.4
		Grade IV	33	31.4
	Other		16	15.2
Tumor grade
	Low	Grades I and II	52	49.5
	High	Grades III and IV	53	50.5
Tumor hemisphere
	Left		55	52.4
	Right		39	37.1
	Both		11	10.5
Time since Dx, mo			105	53.3 (range, 1-264)
		> 1 year	31	29.5
		1-3 years	32	30.5
		3-5 years	10	9.5
		5-10 years	17	16.2
		10+ years	15	14.3
Therapy
	Resection		76	72.4
		Biopsy	26	24.8
		Gross total	19	18.1
		Subtotal	45	42.9
		Unknown	15	14.2
	Radiation		83	79.0
	Chemotherapy		78	74.3
GAD-7	Minimal		59	56.2
	Mild		28	26.7
	Moderate to severe		18	17.1
PHQ-9	Minimal		47	44.8
	Mild		28	26.7
	Moderate to severe		30	28.6

Abbreviations: Dx, diagnosis; GAD-7, Generalized Anxiety Disorder questionnaire–7 Item; PHQ-9, Patient Health Questionnaire–9 Item.

Death-Related Distress

Participants’ global death-related distress scores ranged from 26 to 116 (M = 46.1, SD = 15.9) on the DDS. Of the 105 participants, 81.0% (n = 85) endorsed moderate to severe levels of distress on at least 1 of the 3 subscales, 17.1% (n = 18) on at least 2 of the 3 subscales, and 5.7% (n = 6) on all 3. Nineteen percent (n = 20) denied any type of death-related distress. There were no differences in death-related distress based on age, sex, or tumor grade. Those who endorsed moderate to severe levels of death-related distress were more recently diagnosed (M = 44.1 months since diagnosis, SD = 56.5) than those who reported minimal to mild levels (M = 92.3 months since diagnosis, SD = 86.8; t(103) = 3.06, P < .05).

Death anxiety

The mean score on the death anxiety scale (M = 22.4, SD = 5.5) was in the moderate range. The results presented in Figure 2 show that 19% of PBT patients endorsed mild death anxiety and 81% endorsed moderate to severe symptoms.

Fig. 2.

Prevalence of Moderate to Severe Existential Distress in a Primary Brain Tumor Sample. Distress endorsement followed suggested guidelines of greater than 18 raw score.²²

Death depression

The mean score on the death depression scale (M = 12.0, SD = 6.8) was in the mild range. The results presented in Figure 2 show that 87.6% of PBT patients endorsed mild death depression and 12.4% endorsed moderate to severe symptoms.

Death obsession

The mean score on the death obsession scale (M = 11.7, SD = 6.3) was in the mild range. The results presented in Figure 2 show that 89.5% of PBT patients endorsed mild death obsession and 10.5% endorsed moderate to severe symptoms.

Association of Emotional Distress With Death-Related Distress

Generalized anxiety

Overall, the GAD-7 sample mean was mild (M = 5.3, SD = 5.5); 56.2% (n = 59) endorsed minimal, 26.7% (n = 28) endorsed mild, and 17.1% (n = 18) endorsed moderate to severe symptoms (see Table 1). There was a positive association between global death-related distress (DDS-total) and generalized anxiety, (r[103] = 0.62, P < .001).

Depression

Overall, the PHQ-9 sample mean was mild (M = 6.71, SD = 5.9); 44.8% (n = 47) endorsed minimal, 26.7% (n = 28) endorsed mild, and 28.6% (n = 30) endorsed moderate to severe symptoms (see Table 1). There was a positive association between global death-related distress (DDS-total) and depression, (r[103] = 0.539, P < .001).

Predictors of Existential Distress

A forward stepwise multiple regression was conducted to evaluate whether sex, age, tumor grade, time since diagnosis, symptoms of anxiety or symptoms of depression predicted global death-related distress. At step 1 of the analysis, symptoms of anxiety (GAD-7) were entered into the regression equation and were significantly related to death-related distress F(1,98) = 47.53, P < .001; 32.7% of the variance of the death-related distress could be accounted for by generalized anxiety scores. Sex, age, tumor grade, time since diagnosis, or symptoms of depression were not significant and were thus not entered into the model (P > .05).

Cluster Analysis

A multivariate clustering analysis was conducted, using the 3 subscales of the DDS (death anxiety, death depression, and death obsession), the PHQ-9 (depression), and GAD-7 (generalized anxiety). Convergence was reached in 4 iterations. Univariate ANOVAs indicated that the clustered groups differed significantly on all classifying variables (P < .001). The final cluster centers are visually depicted in Figure 3. Sample size for each cluster ranged from 5 to 62. For the remainder of this article, the clusters will be referred to as global distress (n = 5; cluster 1; high levels on all 5 scales), emotional distress (n = 20; cluster 2; high levels of anxiety and depression with relatively normal levels of death-related distress), resilience (n = 18; cluster 3; low scores on all measures, especially death anxiety), and existential distress (n = 62; cluster 4; high levels on death anxiety and death depression with low levels of depression and anxiety).

Fig. 3.

Final Clusters for 4-Cluster Solution. GAD-7 indicates Generalized Anxiety Disorder scale–7 Item; PHQ-9, Patient Health Questionnaire–9 Item.

Correlates of Cluster Membership

One-way ANOVA and chi-square tests were used to determine whether significant differences existed among the clusters in regard to age, sex, tumor grade, and time since diagnosis. Time since diagnosis was significantly different based on cluster membership [F(3,101) = 4.29, P < .01] with the resilience group having greater amount of time since diagnosis (M = 100.56 months; SD = 87.69) than the existential distress group (M = 41.15 months; SD = 57.28). No differences across the 4 clusters were found in patients’ age (F(3,101) = 2.45, P = .068), sex (χ ² (3) = 1.13, P = .771), or tumor grade (χ ² (9) = 7.61, P = .574).

Discussion

The present quantitative study is the first to examine death-related distress in PBT patients, a population that historically has been excluded from the existential distress literature.²² The most common death-related distress factor endorsed was death anxiety. Eighty-one percent of PBT patients reported moderate to severe levels of death anxiety—8 times more often than either death obsession or death depression. This rate is much higher than other advanced cancers, in which 32% to 45% endorse elevated levels of death anxiety.^7,10,32

The present study also examined the relationship between death-related distress and more general emotional distress. Analyses revealed significant positive correlations between global death-related distress and general symptoms of anxiety and depression. This is consistent with research in other cancer populations, in whom increased anxiety and depression were associated with heightened existential distress.^{10–13,16,17,33} Clinically significant symptoms of depression and anxiety are twice as common in cancer patients as in the general population,³⁴ and these symptoms can have adverse effects on psychological well-being,³⁵ compliance during medical treatment,³⁶ relational distress in caregivers,³⁷ and health-care utilization.³⁸ The combination of anxiety and depression with death-related distress has the potential to be highly disturbing to PBT patients and adversely affect their overall functioning. Further, it should be noted that although generalized anxiety was the only significant factor associated with global death-related distress, it accounted for only one-third of the variance, providing evidence that death-related distress is a separate construct and discrete issue for PBT patients.

Lastly, this study examined whether distress in PBT patients could be clustered in a meaningful way based on symptoms of death-related distress, generalized anxiety, and depression. Four clusters were found: 1) global distress, 2) emotional distress, 3) resilience, and 4) existential distress. These cluster profiles suggest PBT patients respond differently when faced with the challenge of sustaining “double awareness.” Some PBT patients appear to cope well with no evidence of distress (resilience); others display typical psychological profiles of depression and generalized anxiety, without evidence of death-specific distress (emotional distress); and a small number of patients show high levels of distress on all measures across the board (global distress). However, the majority of PBT patients experienced moderate to high levels of death anxiety and death depression (existential distress), but not clinically elevated levels of generalized anxiety and depression. These clusters suggest that some patients with PBT are preoccupied and distressed by thoughts of death, whereas others experience more generalized anxiety and depression without specific existential crises. The high prevalence of the existential distress group exemplifies death-related distress as an independent concept and warrants future screening of death-related distress beyond general depression and anxiety.

One notable difference found between these profiles was time since diagnosis. Patients in the resilience cluster were significantly further from diagnosis compared to the existential distress cluster. This finding suggests that as time passes, some patients may learn to adapt and accept the diagnosis, becoming more resilient. These findings may be similar to seminal longitudinal research in bereaved individuals that found most individuals were resilient, and, over time, all trajectories reported decreased symptoms of grief.³⁹ Future research should investigate longitudinal trajectories of generalized and death-related distress in PBT patients to better understand adjustment over time. Of note, tumor grade or disease severity did not contribute to differences between the clusters, although this may be due to the heterogeneity of the sample and the sample size. As evident in other oncology studies, it is possible that the patterns observed in this cross-sectional sample of PBT patients are related to premorbid mental health and personality (eg, neuroticism, optimism),¹⁰ factors related to family and culture,⁴⁰ or the recency of treatment events (eg, diagnosis, disease progression).^13,16,17,41 This highlights the importance of examining death-related distress in PBT across the disease trajectory.

Despite the multiple patterns identified, 81% of patients endorsed at least one subscale of death-related distress and most patients (n = 62) fell into the existential distress cluster. Taken together, these findings suggest an unaddressed type of distress in PBT patients, and that further exploration into the best way to support patients with death-related distress is warranted. The findings also suggest that nonspecific interventions to relieve symptoms of anxiety and depression in the PBT population may not be sufficient to alleviate the profound death-related distress they may experience. It may be important for providers to screen for both death-related distress and general emotional distress to determine the need for referral to psychotherapy and the specificity of the intervention that is required.

Study Limitations

This study is not without limitations. First, the findings may have been limited by self-selection, because the level of distress may affect patients’ willingness to participate. During recruitment, participants were warned that the topic might be distressing and were told that they could withdraw at any point. It is possible that this caution contributed to the decision of some not to participate (n = 25), particularly those with the highest or lowest levels of death-related distress. Additionally, patients were not screened for aphasia and/or cognitive dysfunction as criteria to participate.

Another limitation of this study is its cross-sectional design. Previous research has shown that emotional distress following bereavement or trauma is linked to pre-event factors, as well as distinct trajectories postevent.³⁹ In that regard, we found in the present study that there was an association between the clusters and the time since diagnosis. Future research should integrate a longitudinal model with consideration of time since diagnosis, progression of disease, physical symptoms, treatment status, and proximity to death.

Lastly, there may be limitations related to the DDS measure,⁹ which is a screener for death-related distress, and may not be specific to the distress experienced by PBT patients or cancer patients in general. However, the findings of this study using the DDS did establish that death anxiety is more common among PBT patients than in other advanced-cancer populations. These results confirm the need to investigate predictors of death anxiety, their associated symptoms, impact on quality of life, and trajectory across disease progression in PBT patients. Interventions can then be tailored and evaluated to address this salient issue in neuro-oncology.

Conclusion

The threat of mortality is an ongoing concern for individuals with advanced cancer and may adversely affect their quality of life. Moderate to severe death anxiety was present in the majority of PBT patients and has been shown to contribute substantially to overall distress. Specialized psychosocial interventions may be of value to relieve death-related distress in PBT patients. There is clearly a need for addressing death-related distress and providing supportive care for neuro-oncology patients. These findings support further research on the trajectory of this distress, its relationship to more generalized distress, and the response to therapeutic interventions. Longitudinal research should consider the effects of prediagnosis and postdiagnosis factors. The present study supports the need for screening death-related distress and greater attention to psychosocial and palliative care in neuro-oncology clinic populations.

Funding

This work was supported by Virginia Commonwealth University, School of Medicine.

Conflict of interest statement

None declared.