Abstract
Context:
Enhancing quality of life (QoL) is a goal of palliative care. Existential QoL is an important aspect of this.
Objectives:
This study sought to advance our understanding of existential QoL at the end of life through examining levels of Preparation and Completion, subscales of the QUAL-E, and their associated factors.
Methods:
We used data from a multi-site study of 331 older cancer patients receiving palliative care. We examined levels of Preparation and Completion and their association with demographic, religious, and medical factors, and with the Patient Dignity Inventory (PDI).
Results:
Preparation and Completion scores were moderately high. In adjusted models, being 10 years older was associated with an increase of 0.77 in Preparation (p=.002). Non-white patients had higher Preparation (1.03, p=.01) and Completion (1.56, p=.02). Single patients reported Completion score 1.75 point lower than those married (p=.01). One-point increase in intrinsic religiousness was associated with a 0.86-point increase in Completion (p=.03). One-point increase in terminal illness awareness (TIA) was associated with 0.75-point decrease in Preparation (p=.001). A 10-point increase in symptom burden was associated with a decrease of 0.55 in Preparation (p<.001) and a decrease of 1.0 in Completion (p<.001). The total PDI score and all of its subscales were negatively correlated with Preparation (r from −.26 to −.52, all p<.001) and Completion (r from −.18 to −.31, all p<.001).
Conclusion:
While most patients reported moderate to high levels of existential QoL, a subgroup reported low existential QoL. TIA and symptom burden may be associated with lower existential QoL.
Keywords: Existential QoL, palliative care, cancer, dignity, symptom burden
Introduction
Enhancing quality of life (QoL) is a goal of palliative care (1). QoL includes physical, psychosocial, and existential dimensions (2). Existential QoL has been conceptualized by Gagnon and colleagues as addressing “questions about human existence and all that is connected to one’s reason for being” (3). Existential distress, one manifestation of existential QoL, was defined by Lo and colleagues as “distress that arises when the meaning and value of life is unclear” (4). This concept has been addressed at length in the literature. However, there is limited research on the broader topic of existential QoL, the focus of the present study. We operationalized existential QoL using the Preparation and Completion subscales of the Quality of Life at the End of Life (QUAL-E) measure (5). These subscales include topics such as concerns about the future, meaning, and value, which address Lo’s definition of existential distress and Gagnon’s definition of existential QoL (3, 4). We used these subscales to examine levels of existential QoL and its associated factors including demographic, medical, and religious factors, dignity, and terminal illness awareness (TIA).
Three prior studies of patients with cancer or other serious illness, which used the Preparation and Completion subscales of the QUAL-E, found high levels of existential QoL (6, 7, 8). However, Chochinov and colleagues found that existential distress was highly endorsed on the Patient Dignity Inventory (PDI) in a sample of 253 patients receiving palliative care (9). For example, participants reported “a problem,” “a big problem” or “an overwhelming problem” for several concerns including not being able to carry out important roles (37.5%), feeling no longer who they were (36.4%), and feeling a burden to others (24.1%) (9, 10). Bovero and colleagues reported that dignity-related existential distress was an issue for 18.8% of a sample of cancer patients near the end of life (11). Furthermore, existential issues including preparation and completion were rated as important among seriously ill patients, recently bereaved family members, physicians, and other care providers in a cross-sectional national survey of 1,462 people (2). Finally, existential distress is one of the most common factors associated with the desire for a hastened death (12). Although levels of existential QoL are poorly understood, low existential QoL is a significant problem for those experiencing it.
Existing literature has limited and sometimes contradictory findings about demographic and medical factors associated with existential QoL at the end of life. Two studies demonstrated no association between age and existential distress at the end of life (5, 13), but Bovero and colleagues found that younger patients had higher levels of dignity-related existential distress (11). Steinhauser and colleagues reported that women scored higher on the Preparation subscale of the QUAL-E (5), whereas Jones and colleagues reported that women had lower existential QoL scores on the McGill Quality of Life Questionnaire (MQOL) (13). Only one study focused on the associations of symptom burden and performance status with existential distress, finding no associations (4).
Prior research has examined spiritual well-being and existential QoL, but not other dimensions of religion. Several studies reported positive associations between spiritual well-being and existential QoL (4, 6, 15, 16). Research on the association between patient dignity and existential QoL is limited, but Chochinov and colleagues found existential distress did not correlate significantly with sense of dignity, as measured by the dignity item of the Structured Interview Assessment of Symptoms and Concerns in Palliative Care (10). No literature, to our knowledge, addresses TIA or time since diagnosis.
Several measures of QoL at the end of life exist, including the MQOL (17) and the Quality of Dying and Death Questionnaire (QODD) (18). In a systematic review of 29 instruments evaluating at least one domain of quality of life, most for patients receiving palliative care, the QUAL-E received one of the highest ratings (19, 20). Most literature using the QUAL-E has used the single-item global QoL question (20), the total score (21), or the Symptom Impact subscale (16), but not the Preparation and Completion subscales. Prior studies using the QUAL-E had samples with a single diagnosis (8, 16, 21, 22), or small sample size (≤30) (22, 23). Ten of the fourteen studies using the QUAL-E were conducted at one site, and all in one geographical region (5, 20, 24, 25).
Levels of existential QoL at the end of life are poorly understood, and the existing literature has limited findings on associated factors. Furthermore, few studies of existential QoL have reported results using subscales from the QUAL-E. The aim of the present study was to advance our understanding of existential QoL at the end of life by examining levels of Preparation and Completion, which are subscales of the QUAL-E, and their associated factors.
Methods
Participants
This was a secondary analysis utilizing baseline data from 331 participants enrolled in a randomized clinical trial of Dignity Therapy (26). Participants were 55 years or older with a cancer diagnosis receiving outpatient palliative care at one of six hospitals across the US. Participants were required to be English literate and have a Palliative Performance Scale [PPS] score greater than 50 (27). Exclusion criteria included being legally blind or cognitively impaired (determined by a Mini Mental Status Exam [MMSE] score <24 (28)) or history of psychosis determined by record review. Patients enrolled in another similar intervention study were also excluded. The study was approved by the IRBs of all participating sites.
Measures
Existential QoL:
Existential QoL was measured using the Preparation and Completion subscales of the QUAL-E, a valid and reliable measure of QoL at the end of life (5). The items in these subscales encompass Gagnon’s definition of existential QoL, while addressing Lo’s definition of existential distress, an important manifestation of existential QoL (3, 4)
The 4-item Preparation subscale assesses an individual’s sense of integrity and concerns about being a burden to others. A sample item is “I have regrets about the way I have lived my life.” The 7-item Completion subscale assesses an individual’s sense of meaning and peace, and unfinished interpersonal business. A sample item is “I have been able to share important things with my family.” The items in both subscales are rated on a 5-point scale from: 1) ‘Not at all true for me’ to 5) ‘Completely true for me’. The items in the Preparation subscale are reverse scored. In the present study, the Cronbach’s alphas for the Preparation and Completion subscales were 0.66 and 0.81, respectively, and the correlation between the subscales was 0.23 (p<.001).
Demographic Variables:
Participants’ age, gender, race, marital status, education level, and income were obtained from interviews or medical record review. Due to small numbers of non-white participants, race was treated dichotomously. In the regression analysis education level and income were treated as ordinal level variables.
Medical Information:
Cancer type and time since diagnosis were obtained from the medical record. PPS was completed during screening (27). Pain, tiredness, nausea, depression, anxiety, drowsiness, appetite, well-being, and shortness of breath were assessed with the self-report 9-item Edmonton Symptom Assessment Scale (ESAS) (29, 30). Symptoms were rated from 0 (‘no symptoms’) to 10 (‘worst possible symptoms’). Item scores were added to create a total symptom burden score. In the present study, the Cronbach’s alpha for the total score was 0.81.
Terminal Illness Awareness:
TIA was measured with the question “How would you describe your current health status?” Response options included 1) ‘Relatively healthy,’ 2) ‘Seriously but not terminally ill,’ and 3) ‘Seriously and terminally ill.’ The item is a frequently used approach to structured assessment of prognostic awareness (31). In our regression analysis the item was treated as an ordinal measure with higher scores indicating greater awareness of the serious and terminal nature of one’s illness.
Religion:
Religious involvement was measured with three standard items (32): 1) religious preference (the category “no religion” included “spiritual but not religious”), 2) if the patient carried their religious beliefs over into all other dealings in life, also referred to as intrinsic religiousness, coded from ‘strongly disagree’ to ‘strongly agree’, and 3) frequency of private prayer in places other than at a house of worship, coded from ‘never’ to ‘more than once a day’. Items 2 and 3 are included in the popular DUREL measure of religion; evidence of their content and construct validity have been reported (33, 34). In our regression analysis these items were treated as ordinal measures with higher scores indicating greater intrinsic religiousness or frequency of private prayer.
Dignity-Related Distress:
Chochinov and colleagues stated that dignity-related distress “encapsulates a broad range of issues—physical, psychosocial, spiritual, and existential—that may influence a dying patient’s sense of dignity.” This was assessed with the PDI, a validated self-report 25-item measure of dignity-related distress (10). We also used the five PDI subscales. Questions were rated from 1 (‘not a problem’) to 5 (‘an overwhelming problem’). Item scores were summed; higher total scores represented more dignity-related problems. In the present sample, the Cronbach’s alpha for the total PDI was 0.93. Similarities between the items in the Preparation and Completion subscales of the QUAL-E and items in the PDI, especially in the existential and psychosocial subscales, suggest they measure similar constructs.
Statistical Analysis
The analysis began with descriptive statistics including mean, standard deviation, frequency, and percentage, for the sample characteristics and other study measures. Next, we examined the bivariate associations between study variables and Preparation and Completion. Finally, we used a series of linear regression models to examine the association of groups of factors with Preparation and Completion, specifically demographic factors, religious factors, TIA and medical factors. The item overlap between the PDI and the QUAL-E subscales suggested that the PDI measured a similar construct, rather than a meaningful explanatory variable, so it was not included in these models. We examined the correlations between the QUAL-E subscales and PDI (total and five subscales) to provide insight on their relationships.
Given the relatively large sample size of our study, our association analysis is robust against violation of normality assumption (35). We used variance inflation factor (VIF) to assess multicollinearity of our regression models and found the level of multicollinearity to be weak (VIF<2) (36). Component plus residual plots were used to verify that the linearity assumption holds well in our models (37). Score test was used to verify that there was no significant heteroskedasticity (38). Our dataset has a relatively small amount (<5%) of missing data, with percentage of missing entries for individual variables ranging between 0% and 21%. Multiple imputation was used for missing data processing, where 200 multiply imputed datasets were generated under fully conditional specification, on which inference was performed and then aggregated (39). The impact of missing data as measured by the fraction of missing information ranges from 1% to 25% (40).
Results
Demographic characteristics are in Table 1. The sample was mostly white, married or partnered, and had at least some college education. Approximately three-fourths of the sample was Christian. Our sample had high functional status due to the inclusion criteria. The mean PPS score was 75.9 (SD=10.3) out of 100 (normal activity and work with no evidence of disease). The symptom burden was low, with a mean ESAS score of 26.1 (SD=15.7) out of 90 (worst possible symptoms). Approximately one-third of the sample considered themselves relatively healthy, another third reported being seriously but not terminally ill, and the last third seriously and terminally ill.
Table 1.
Sample Characteristics (N=331)
| Variable | Missing | Category | Value |
|---|---|---|---|
| Age | 0 | Mean (SD) | 66.4 (7.4) |
| Gender | 1 | Female | 186 (56%) |
| Male | 144 (44%) | ||
| Race | 1 | White | 237 (72%) |
| Other or Unknown1 | 93 (28%) | ||
| Marital status | 32 | Married/Partnered | 173 (58%) |
| Single | 126 (42%) | ||
| Education | 32 | High school or less | 73 (24%) |
| Some college/vocational | 90 (30%) | ||
| Bachelor’s degree | 65 (22%) | ||
| Graduate degree | 71 (24%) | ||
| Income | 54 | Less than $20,000 | 66 (24%) |
| $20,000 to $49,999 | 69 (25%) | ||
| $50,000 or more | 142 (51%) | ||
| Religious affiliation | 28 | Christian | 233 (77%) |
| Other religion2 | 37 (12%) | ||
| No religion (includes spiritual not religious) | 33 (11%) | ||
| Intrinsic religiousness | 30 | Strongly disagree | 23 (8%) |
| Disagree | 35 (12%) | ||
| Agree | 115 (38%) | ||
| Strongly agree | 128 (43%) | ||
| Private prayer | 29 | Less than once a month | 38 (13%) |
| Weekly-monthly | 41 (14%) | ||
| Daily | 83 (27%) | ||
| More than once a day | 140 (46%) | ||
| Cancer type | 24 | Lung | 45 (15%) |
| Breast | 37 (12%) | ||
| Myeloma | 21 (7%) | ||
| Prostate | 19 (6%) | ||
| Ovary | 18 (6%) | ||
| Pancreas | 18 (6%) | ||
| Colon or Rectum | 16 (5%) | ||
| Head or Neck | 14 (5%) | ||
| Leukemia | 9 (3%) | ||
| Kidney | 8 (3%) | ||
| Lymphoma | 7 (2%) | ||
| Uterus | 7 (2%) | ||
| Bladder | 6 (2%) | ||
| Liver | 5 (2%) | ||
| Sarcoma | 4 (1%) | ||
| Cervix | 2 (1%) | ||
| Stomach | 2 (1%) | ||
| Bone | 1 (0%) | ||
| Brain | 1 (0%) | ||
| Other | 40 (13%) | ||
| Multiple Types (see detail) | 27 (9%) | ||
| Time since diagnosis (years) | 70 | Mean (SD) | 4.0 (4.6) |
| Terminal Illness Awareness (TIA) | 2 | Relatively healthy | 120 (36%) |
| Seriously but not terminally ill | 112 (34%) | ||
| Seriously and terminally ill | 97 (29%) | ||
| Palliative Performance Scale (PPS) (0–100) | 0 | Mean (SD) | 75.9 (10.3) |
| ESAS total (0–90) | 9 | Mean (SD) | 26.1 (15.7) |
| Patient Dignity Inventory (PDI) total (25–125) | 24 | Mean (SD) | 42.4 (13.7) |
| Symptom distress (PDI) (6–30) | 7 | Mean (SD) | 12.2 (4.5) |
| Existential distress (PDI) (6–30) | 9 | Mean (SD) | 10.3 (4.2) |
| Dependency (PDI) (3–15) | 6 | Mean (SD) | 4.6 (1.8) |
| Peace of mind (PDI) (3 –15) | 3 | Mean (SD) | 4.8 (1.9) |
| Social support (PDI) (3–15) | 3 | Mean (SD) | 3.6 (1.2) |
| QUAL–E Preparation (4–20) | 4 | Mean (SD) | 15.0 (3.4) |
| QUAL-E Completion (8–35) | 6 | Mean (SD) | 26.7 (5.3) |
Other Race includes Black or African American 73 (22%), Asian 4 (1%), American Indian or Alaska Native 1 (<1%), Native Hawaiian or other Pacific Islander 1 (<1%), Multi-racial 3 (1%), and Other or Unknown 11 (3%).
Other religion includes Jewish 12 (4%), Buddhist 4 (1%), and Other 21 (7%)
ESAS = Edmonton Symptom Assessment Scale; QUAL-E = Quality of Life at End of Life
The mean total score of the Preparation subscale was 15.0 (SD=3.4; range 4–20). The mean total score of the Completion subscale was 26.7 (SD=5.3; range 8–35). These scores indicate moderate to high levels of existential QoL. The proportion of participants who were experiencing some distress on items in the Preparation subscale (responses of ‘quite a bit’ or ‘completely’) ranged from 6% (“I have regrets about the way I have lived my life”) to 27% (“At times I worry that I will be a burden to my family”). In the Completion subscale, the proportion of participants experiencing some distress (responses of ‘not at all’ or ‘a little bit’) ranged from 9% (“Despite my illness, I have a sense of meaning in life”) to 20% (“I have been able to help others through time together, gifts, or wisdom”) (Figures 1a,b).
Figure 1.
A. Distribution of Preparation Subscale Item Responses
B. Distribution of Completion Subscale Item Responses
The bivariate analyses of factors associated with Preparation and Completion are in Tables 2 and 3. Age was positively associated with higher levels of Preparation (p<.001). Race and marital status were both associated with Completion, with non-white patients and married or partnered patients reporting higher levels of Completion (p=.04 and p=.03, respectively). Higher reports of symptom burden were associated with lower levels of both Preparation and Completion (p<.001 for both). Higher levels of functional status were associated with higher levels of Preparation (p=.01), and greater awareness of having a terminal illness was associated with lower levels of Preparation (p<.001) and Completion (p=.01). There was a weak association between intrinsic religiousness and Completion (p=.048). There were no significant associations with other measures of religion or time since diagnosis.
Table 2.
Correlations between QUAL-E subscales and continuous/ordinal predictors1
| Variable | Preparation | Completion | ||
|---|---|---|---|---|
| Correlation | p | Correlation | p | |
| Age | 0.19 | <.001 | −0.03 | .62 |
| Education | −0.10 | .08 | −0.07 | .22 |
| Income | 0.04 | .48 | −0.04 | .46 |
| Intrinsic religiousness | 0.06 | .28 | 0.11 | .05 |
| Private prayer | 0.06 | .29 | 0.08 | .18 |
| Time since diagnosis | −0.06 | .32 | −0.07 | .26 |
| ESAS total | −0.33 | <.001 | −0.32 | <.001 |
| Palliative Performance Scale (PPS) | 0.14 | .01 | 0.07 | .23 |
| Terminal illness awareness (TIA) | −0.20 | <.001 | −0.15 | .01 |
Values are Spearman correlation coefficients
QUAL-E = Quality of Life at End of Life; ESAS = Edmonton Symptom Assessment Scale
Table 3.
Bi-variate association between QUAL-E subscales and categorical predictors
| Variable | Category | Preparation | Completion | ||
|---|---|---|---|---|---|
| Mean (SD) | p | Mean (SD) | p | ||
| Gender | Female | 14.9 (3.2) | .58 | 26.3 (5.6) | .18 |
| Male | 15.1 (3.7) | 27.1 (5.0) | |||
| Race | Other | 15.6 (3.5) | .05 | 27.6 (5.7) | .04 |
| White | 14.8 (3.3) | 26.3 (5.1) | |||
| Marital status | Married/Partnered | 15.1 (3.3) | .55 | 27.2 (5.0) | .03 |
| Single | 14.8 (3.5) | 25.9 (5.7) | |||
| Religious affiliation | Christian | 15.1 (3.4) | .64 | 26.7 (5.5) | .96 |
| Other religion | 14.8 (3.3) | 26.5 (5.1) | |||
| No religion | 14.6 (3.4) | 26.9 (4.1) | |||
QUAL-E = Quality of Life at End of Life
Regression modeling results appear in Tables 4a and 4b. As in the bivariate analysis, age was positively associated with higher levels of Preparation (p≤.002). Adjusting for other variables, race was associated with both Preparation (p≤.02) and Completion (p≤.02), with non-white patients reporting higher levels of both. Married or partnered patients reported higher levels of Completion (p=.01). There was no significant association with gender, education, or income for either subscale. Of the associations between the three measures of religion and the Preparation and Completion subscales, only one was significant. TIA was associated with lower levels of Preparation (p=.001). Symptom burden was associated with lower levels of both Preparation (p<.001) and Completion (p<.001). Time since diagnosis had no significant association with either subscale. The adjusted R2 of the models are reported in the tables. The models including the symptom burden as an independent variable had the best fit, although the proportions of variance predictable are modest for all models, reflecting the fact that there are other important factors contributing to Preparation and Completion that are not included in our analyses.
Table 4A.
Regression Models for QUAL-E Preparation1
| Variables (reference group) | Model 1 | Model 2 | Model 3 | Model 4 |
|---|---|---|---|---|
| Age | 0.090 (p<.001) | 0.092 (p<.001) | 0.092 (p<.001) | 0.077 (p=002) |
| Gender (male) | −0.262 (p=.50) | −0.327 (p=.41) | −0.210 (p=.58) | −0.365 (p=.33) |
| Other race (white) | 1.121 (p=.01) | 1.072 (p=.02) | 1.014 (p=.02) | 1.030 (p=.01) |
| Single (married) | −0.106 (p=.82) | −0.108 (p=.82) | −0.132 (p=.78) | −0.046 (p=.92) |
| Education | −0.344 (p=.07) | −0.334 (p=.10) | −0.290 (p=.13) | −0.246 (p=.19) |
| Income | 0.424 (p=.15) | 0.422 (p=.16) | 0.378 (p=.20) | 0.245 (p=.40) |
| Other religion (Christian) | −0.061 (p=.92) | |||
| No religion (Christian) | 0.013 (p=.99) | |||
| Intrinsic religiousness | 0.209 (p=.43) | |||
| Private prayer | −0.015 (p=.95) | |||
| Terminal Illness Awareness (TIA) | −0.750 (p=.001) | |||
| Palliative Performance Scale (PPS) | 0.026 (p=.15) | |||
| ESAS total | −0.055 (p<.001) | |||
| Time since diagnosis | −0.014 (p=.76) | |||
| Adjusted R2 | 0.06 | 0.05 | 0.09 | 0.13 |
Values are unstandardized regression coefficients (p value)
QUAL-E = Quality of Life at the End of Life; ESAS = Edmonton Symptom Assessment Scale
Table 4B.
Regression Models for QUAL-E Completion1
| Variables (reference group) | Model 1 | Model 2 | Model 3 | Model 4 |
|---|---|---|---|---|
| Age | 0.025 (p=.54) | 0.032 (p=.43) | 0.026 (p=.52) | 0.004 (p=.91) |
| Gender (male) | −0.641 (p=.29) | −0.873 (p=.16) | −0.599 (p=.33) | −0.786 (p=.18) |
| Other race (white) | 1.766 (p=.01) | 1.697 (p=.02) | 1.679 (p=.01) | 1.555 (p=.02) |
| Single (married) | −1.861 (p=.01) | −1.895 (p=.01) | −1.883 (p=.01) | −1.748 (p=.01) |
| Education | −0.310 (p=.29) | −0.403 (p=.20) | −0.267 (p=.36) | −0.145 (p=.61) |
| Income | −0.319 (p=.51) | −0.349 (p=.47) | −0.357 (p=.46) | −0.582 (p=.21) |
| Other religion (Christian) | 0.431 (p=.66) | |||
| No religion (Christian) | 1.700 (p=13) | |||
| Intrinsic religiousness | 0.855 (p=.03) | |||
| Private prayer | −0.086 (p=.82) | |||
| Terminal illness awareness (TIA) | −0.609 (p=.09) | |||
| Palliative Performance Scale (PPS) | −0.005 (p=.86) | |||
| ESAS total | −0.102 (p<.001) | |||
| Time since diagnosis | −0.043 (p=.52) | |||
| Adjusted R2 | 0.03 | 0.04 | 0.04 | 0.11 |
Values are unstandardized regression coefficients (p value)
QUAL-E = Quality of Life at the End of Life; ESAS = Edmonton Symptom Assessment Scale
The correlations between the Preparation and Completion subscales and the PDI and its subscales are shown in Table 5. We found a strong association between Preparation and the total PDI, and a moderate association between Completion and the total PDI. Associations between the Preparation and Completion subscales and the PDI subscales ranged from weak to strong, but Preparation and the PDI subscales overall had stronger correlations than Completion and the PDI subscales.
Table 5.
1 Correlation between Patient Dignity Inventory (PDI; total and subscales) and Preparation and Completion1
| Variable | Preparation | Completion | ||
|---|---|---|---|---|
| Correlation | p | Correlation | p | |
| PDI total | −0.52 | <.001 | −0.31 | <.001 |
| Symptom distress | −0.47 | <.001 | −0.24 | <.001 |
| Existential distress | −0.51 | <.001 | −0.29 | <.001 |
| Dependency | −0.31 | <.001 | −0.18 | .001 |
| Peace of mind | −0.44 | <.001 | −0.27 | <.001 |
| Social support | −0.26 | <.001 | −0.29 | <.001 |
Values are Spearman correlation coefficients
Discussion
Using a large, multi-site sample of patients with advanced cancer receiving palliative care, we described levels of Preparation and Completion, which are subscales of the QUAL-E, and their associated factors. In our sample, existential QoL was moderate to high, although a proportion of participants experienced some distress. Preparation and Completion scores in this sample were nearly identical with two large studies of patients with cancer or serious illness (6, 7). In our sample, the mean total Preparation score was 15.0 (SD=3.4), while the other studies reported scores of 15.0 (SD=3.4) and 18.0 (SD=4.0). The mean total Completion score was 26.7 (SD=5.3), whereas the other studies reported scores of 26.0 (SD=5.1) and 26.1 (SD=5.5). Additionally, a study of 41 Destination Therapy congestive heart failure patients reported Preparation and Completion scores of 13.61 (SD=4.36) and 27.87 (SD=4.12) respectively (8).
These findings may indicate that existential QoL is generally high in patients with advanced illness near the end of life. Our findings may also reflect a selection bias towards better health and lower distress among research participants (cf. 41, 4). Our sample in particular included patients with better health, due to the inclusion criteria. In contrast, 18.8% of cancer patients near the end of life with a Karnofsky Performance Status of less than 50 reported dignity-related existential distress as a problem or major problem on the existential distress subscale of the Italian version of the PDI, indicating low existential QoL in that population (11).
Three demographic variables were associated with Preparation or Completion in the regression analysis: non-white people scored higher in both subscales, older people scored higher in Preparation, and married or partnered people scored higher in Completion. While these differences are statistically significant, they are small and may not be clinically significant. We did not find any significant association between gender or socioeconomic status and either subscale. This is in contrast with Steinhauser and colleagues, who reported no significant association between ethnicity or marital status and either subscale, but found that women reported significantly higher levels of Preparation (5), and with Jones and colleagues, who reported that women had lower existential QoL scores on the MQOL (13). Our findings on age aligned with Bovero and colleagues, who reported that younger patients experience more dignity-related existential distress (11). Our findings on marital status may reflect the relational nature of Completion subscale items (e.g. “There is someone in my life with whom I can share my deepest thoughts”). However, we would also expect an association with Preparation, given that it also includes relational items (e.g. “I worry that my family is not prepared to cope with the future”). This discrepancy deserves further study.
In models adjusted for demographic factors, we found an association between overall symptom burden and existential QoL. Lo and colleagues found no significant correlation between physical symptom burden, as measured by six items of the ESAS, and existential distress as measured by the Existential Distress Scale (4). We also found that TIA was associated with lower levels of Preparation. There has been no previous research on TIA and existential QoL.
In previous studies, investigators found a correlation between existential QoL and spiritual well-being. Lo and colleagues found a positive association between the FACIT-Sp and the Preparation and Completion subscales of the QUAL-E (6), and Grünke found a positive association between the FACIT-Sp and total German QUAL-EC-P score (14). There is item overlap between the FACIT-Sp Meaning/Peace subscale and the QUAL-E, which could explain these associations. However, Lo and colleagues analyzed the Faith subscale of the FACIT-Sp separately, which does not have item overlap, and found that the positive association remained with both subscales (6).
In contrast to these studies, we measured three key dimensions of religious involvement (religious affiliation, intrinsic religiousness, and frequency of private prayer), and found only one significant association between these measures and the Preparation and Completion subscales. Of note, our sample had high levels of religious involvement. Considering the items in the Preparation and Completion subscales, one might assume a positive association with multiple measures of religious involvement. The lack of a statistically significant association does not necessarily imply a lack of a clinically significant association, especially given the modest number of patients with no religion in our sample. The descriptive statistics, however, seemed to suggest the existential QoL difference between patients with different religious preference to be relatively minor. Nonetheless, further research with a larger sample is needed to provide a more definitive answer.
Chochinov and colleagues found no significant correlation between the existential distress factor of the PDI and sense of dignity (10). In contrast, we found associations between both QUAL-E subscales and the PDI total and all subscales. Given that the constructs and underlying items they measure are very similar, this is not surprising.
This study has several limitations. Our sample had a low symptom burden, likely due to the inclusion criteria. Therefore, it may not be representative of the population of older cancer patients receiving palliative care. Our cross-sectional study design precludes interpretation of causal relationships. While the total ESAS score is used as a measure of symptom burden (30), its use in this way has not been validated. Additionally, our measure of religious involvement was not comprehensive. We encourage further research on aspects of religion and spirituality associated with existential QoL.
Clinicians should be aware that while existential QoL is moderate to high in many patients, a subgroup of patients experience low levels of existential QoL. Thus, it is important to inquire about existential QoL and refer to a mental health or spiritual care professional as indicated. Although the causal direction is not established, TIA and symptom burden may decrease existential QoL, and clinicians should be attentive to changes associated with these variables.
In this large multisite study, we advanced the understanding of existential QoL in older cancer patients receiving palliative care, including important clinical and religious factors associated with Preparation and Completion. In our sample, in which all participants had high functional status and many had low symptom burden, levels of existential QoL were moderate to high. We found a negative association between symptom burden and both Preparation and Completion, and a negative association between TIA and Preparation, indicating that these variables may lower existential QoL. We found limited associations between religious involvement and existential QoL.
Further investigation of factors associated with existential QoL in samples with lower functional status is warranted, to deepen our understanding of these variables in a more representative sample. Furthermore, a longitudinal study would allow investigation of the trajectories of existential QoL over the course of a patient’s illness and establish predictors. Finally, the relationship between religious involvement and existential QoL deserves further research.
Key Message.
Among older cancer patients receiving palliative care, most reported moderately high levels of existential QoL. However, a subset of the population had low existential QoL. Clinicians should be attentive to existential QoL, particularly as terminal illness awareness and symptom burden evolve.
Acknowledgements
Funding:
The authors thank the patients and clinicians for giving of their time to advance science. This research was made possible by Grant Number 1R01CA200867 from the National Institutes of Health (NIH), National Cancer Institute (NCI) and National Institute of Nursing Research (NINR). The information in this article is solely the responsibility of the authors and does not necessarily represent the views of the NIH, NCI, or NINR. The final peer-reviewed manuscript is subject to the National Institutes of Health Public Access Policy.
Footnotes
Disclosures
DW is Founder and Chairman of eNURSING llc, a company with no interests in the research reported in this manuscript.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.WHO Definition of Palliative Care. (2012, January 28). Retrieved October 10, 2020, from https://www.who.int/cancer/palliative/definition/en/
- 2.Steinhauser KE, Christakis NA, Clipp EC, et al. Factors considered important at the end of life by patients, family, physicians, and other care providers. JAMA. 2000. Nov 15;284(19):2476–82. doi: 10.1001/jama.284.19.2476. PMID: 11074777. [DOI] [PubMed] [Google Scholar]
- 3.Gagnon P, Fillion L, Robitaille MA, et al. A cognitive-existential intervention to improve existential and global quality of life in cancer patients: A pilot study. Palliat Support Care. 2015;13(4):981–990. doi: 10.1017/S147895151400073X [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lo C, Panday T, Zeppieri J, et al. Preliminary psychometrics of the Existential Distress Scale in patients with advanced cancer. Eur J Cancer Care (Engl). 2017. Nov;26(6). doi: 10.1111/ecc.12597. Epub 2016 Oct 24. PMID: 27778415. [DOI] [PubMed] [Google Scholar]
- 5.Steinhauser KE, Clipp EC, Bosworth HB, et al. Measuring quality of life at the end of life: validation of the QUAL-E. Palliat Support Care. 2004. Mar;2(1):3–14. doi: 10.1017/s1478951504040027. PMID: 16594230. [DOI] [PubMed] [Google Scholar]
- 6.Lo C, Burman D, Swami N, et al. Validation of the QUAL-EC for assessing quality of life in patients with advanced cancer. Eur J Cancer. 2011. Mar;47(4):554–60. doi: 10.1016/j.ejca.2010.10.027. Epub 2010 Dec 1. PMID: 21126869. [DOI] [PubMed] [Google Scholar]
- 7.Steinhauser KE, Alexander S, Olsen MK, et al. Addressing Patient Emotional and Existential Needs During Serious Illness: Results of the Outlook Randomized Controlled Trial. J Pain Symptom Manage. 2017. Dec;54(6):898–908. doi: 10.1016/j.jpainsymman.2017.06.003. Epub 2017 Aug 10. PMID: 28803082. [DOI] [PubMed] [Google Scholar]
- 8.Woodburn JL, Staley LL, Wordingham SE, et al. Destination Therapy: Standardizing the Role of Palliative Medicine and Delineating the DT-LVAD Journey. J Pain Symptom Manage. 2019. Feb;57(2):330–340.e4. doi: 10.1016/j.jpainsymman.2018.11.007. Epub 2018 Nov 15. PMID: 30447385. [DOI] [PubMed] [Google Scholar]
- 9.Chochinov HM, Hassard T, McClement S, et al. The landscape of distress in the terminally ill. J Pain Symptom Manage. 2009. Nov;38(5):641–9. doi: 10.1016/j.jpainsymman.2009.04.021. Epub 2009 Aug 26. PMID: 19713069. [DOI] [PubMed] [Google Scholar]
- 10.Chochinov HM, Hassard T, McClement S, et al. The patient dignity inventory: a novel way of measuring dignity-related distress in palliative care. J Pain Symptom Manage. 2008. Dec;36(6):559–71. doi: 10.1016/j.jpainsymman.2007.12.018. Epub 2008 Jun 24. PMID: 18579340. [DOI] [PubMed] [Google Scholar]
- 11.Bovero A, Sedghi NA, Opezzo M, et al. Dignity-related existential distress in end-of-life cancer patients: Prevalence, underlying factors, and associated coping strategies. Psycho-oncology. 2018. Nov;27(11):2631–2637. DOI: 10.1002/pon.4884. [DOI] [PubMed] [Google Scholar]
- 12.Hudson PL, Kristjanson LJ, Ashby M, et al. Desire for hastened death in patients with advanced disease and the evidence base of clinical guidelines: a systematic review. Palliat Med. 2006. Oct;20(7):693–701. doi: 10.1177/0269216306071799. PMID: 17060268. [DOI] [PubMed] [Google Scholar]
- 13.Jones JM, Cohen SR, Zimmermann C, Rodin G. Quality of life and symptom burden in cancer patients admitted to an acute palliative care unit. J Palliat Care. 2010. Summer;26(2):94–102. PMID: 20718394. [PubMed] [Google Scholar]
- 14.Grünke B, Philipp R, Vehling S, et al. Measuring the Psychosocial Dimensions of Quality of Life in Patients With Advanced Cancer: Psychometrics of the German Quality of Life at the End of Life-Cancer-Psychosocial Questionnaire. J Pain Symptom Manage. 2018. Mar;55(3):985–991.e1. doi: 10.1016/j.jpainsymman.2017.11.006. Epub 2017 Nov 16. PMID: 29155291. [DOI] [PubMed] [Google Scholar]
- 15.Rabow MW, Knish SJ. Spiritual well-being among outpatients with cancer receiving concurrent oncologic and palliative care. Support Care Cancer. 2015. Apr;23(4):919–23. doi: 10.1007/s00520-014-2428-4. Epub 2014 Sep 19. PMID: 25231797. [DOI] [PubMed] [Google Scholar]
- 16.Prince-Paul M Relationships among communicative acts, social well-being, and spiritual well-being on the quality of life at the end of life in patients with cancer enrolled in hospice. J Palliat Med. 2008. Jan-Feb;11(1):20–5. doi: 10.1089/jpm.2007.0119. PMID: 18370887. [DOI] [PubMed] [Google Scholar]
- 17.Cohen SR, Sawatzky R, Russell LB, et al. Measuring the quality of life of people at the end of life: The McGill Quality of Life Questionnaire-Revised. Palliat Med. 2017. Feb;31(2):120–129. doi: 10.1177/0269216316659603. Epub 2016 Jul 18. PMID: 27412257. [DOI] [PubMed] [Google Scholar]
- 18.Downey L, Curtis JR, Lafferty WE, Herting JR, Engelberg RA. The Quality of Dying and Death Questionnaire (QODD): empirical domains and theoretical perspectives. J Pain Symptom Manage. 2010;39(1):9–22. doi: 10.1016/j.jpainsymman.2009.05.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Albers G, Echteld MA, de Vet HC, et al. Evaluation of quality-of-life measures for use in palliative care: a systematic review. Palliat Med. 2010. Jan;24(1):17–37. doi: 10.1177/0269216309346593. Epub 2009 Oct 20. PMID: 19843620. [DOI] [PubMed] [Google Scholar]
- 20.Steinhauser KE, Bosworth HB, Clipp EC, et al. Initial assessment of a new instrument to measure quality of life at the end of life. J Palliat Med. 2002. Dec;5(6):829–41. doi: 10.1089/10966210260499014. PMID: 12685529. [DOI] [PubMed] [Google Scholar]
- 21.Zimmermann C, Swami N, Krzyzanowska M, et al. Early palliative care for patients with advanced cancer: a cluster-randomised controlled trial. The Lancet. 2014. May 17;383(9930):1721–30. doi: 10.1016/S0140-6736(13)62416-2 [DOI] [PubMed] [Google Scholar]
- 22.Adorno G, Wallace C. Preparation for the end of life and life completion during late-stage lung cancer: An exploratory analysis. Palliat Support Care. 2017. Oct;15(5):554–564. doi: 10.1017/S1478951516001012. Epub 2017 Jan 18. PMID: 28095941. [DOI] [PubMed] [Google Scholar]
- 23.Keall RM, Butow PN, Steinhauser KE, Clayton JM. Nurse-facilitated preparation and life completion interventions are acceptable and feasible in the Australian palliative care setting: results from a phase 2 trial. Cancer Nurs. 2013. May-Jun;36(3):E39–46. doi: 10.1097/NCC.0b013e3182664c7a. PMID: 23047796. [DOI] [PubMed] [Google Scholar]
- 24.Steinhauser KE, Alexander SC, Byock IR, et al. Do preparation and life completion discussions improve functioning and quality of life in seriously ill patients? Pilot randomized control trial. J Palliative Med. 2008. Nov;11(9): 1234–1240. [DOI] [PubMed] [Google Scholar]
- 25.Wilkinson A, Slatyer S, McCullough K, Williams A. Exploring the quality of life at the end of life (QUAL-E) instrument with Australian palliative care hospital patients: hurdles and directions. J Palliat Care. 2014. Spring;30(1):16–23. PMID: 24826439. [PubMed] [Google Scholar]
- 26.Kittelson S, Scarton L, Barker P, et al. Dignity therapy led by nurses or chaplains for elderly cancer palliative care outpatients: protocol for a randomized controlled trial. JMIR Res Protoc. 2019;8(4):e12213. doi: 10.2196/12213. PMID: 30994466; PMCID: PMC6492061. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Anderson F, Downing GM, Hill J, Casorso L, Lerch N. Palliative performance scale (PPS): a new tool. J Palliat Care. Spring 1996;12(1):5–11. [PubMed] [Google Scholar]
- 28.Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. Nov 1975;12(3):189–198. [DOI] [PubMed] [Google Scholar]
- 29.Watanabe SM, Nekolaichuk CL, Beaumont C. The Edmonton Symptom Assessment System, a proposed tool for distress screening in cancer patients: development and refinement. Psycho-oncology. 2012. Sep;21(9):977–85. doi: 10.1002/pon.1996. Epub 2011 Jun 13. PMID: 21671304. [DOI] [PubMed] [Google Scholar]
- 30.Davis MP, Lagman R, Parala A, Patel C, Sanford T, Fielding F, et al. Hope, Symptoms, and palliative care: do symptoms influence hope? Am J Hosp Palliat Care. 2017;34(3):223–232. [DOI] [PubMed] [Google Scholar]
- 31.Applebaum AJ, Kolva EA, Kulikowski JR, Jacobs JD, DeRosa A, Lichtenthal WG, Olden ME, Rosenfeld B, Breitbart W. Conceptualizing prognostic awareness in advanced cancer: a systematic review. J Health Psychol. 2014;19(9):1103–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Multidimensional measurement of religiousness/spirituality for use in health research. John E. Fetzer Institute National Institute on Aging Working Group; 1999. [Google Scholar]
- 33.Koenig HG, Büssing A. The Duke University Religion Index (DUREL): a five-item measure for use in epidemological studies. Religions. 2010; 1(1): 78–85. doi: 10.3390/rel1010078 [DOI] [Google Scholar]
- 34.Lace JW, Handal PJ. Confirming the tripartite structure of the Duke University Religion Index: a methodological approach. J Relig Health. 2018; 57(2): 704–716. doi: 10.1007/s10943-017-0556-0. [DOI] [PubMed] [Google Scholar]
- 35.Williams MN, Grajales CA, Kurkiewicz D. Assumptions of multiple regression: Correcting two misconceptions. Practical Assessment, Research, and Evaluation. 2013;18(1):11. [Google Scholar]
- 36.Belsley DA, Kuh E, Welsch RE. Regression diagnostics: Identifying influential data and sources of collinearity. John Wiley & Sons; 2005. Feb 25. [Google Scholar]
- 37.Fox J Applied regression analysis and generalized linear models. Sage Publications; 2015. Mar 18. [Google Scholar]
- 38.Cook RD, Weisberg S. Diagnostics for heteroscedasticity in regression. Biometrika. 1983;70(1):1–0. [Google Scholar]
- 39.Buuren SV, Groothuis-Oudshoorn K. Mice: multivariate imputation by chained equations in R. Journal of statistical software. 2010:1–68. [Google Scholar]
- 40.Van Buuren S Flexible imputation of missing data. CRC press; 2018. Jul 17. [Google Scholar]
- 41.Steinhauser KE, Clipp EC, Hays JC, Olsen M, et al. Identifying, recruiting, and retaining seriously-ill patients and their caregivers in longitudinal research. Palliat Med. 2006;20(8):745–54. doi: 10.1177/0269216306073112. PMID: 17148529. [DOI] [PubMed] [Google Scholar]