Abstract
Context
Anxiety in terminal cancer is linked to diminished quality of life, yet overall it is poorly understood with regard to prevalence and relationship to other aspects of psychological distress.
Objectives
This study examines anxiety in terminally ill cancer patients, including the prevalence of anxiety symptoms, the relationship between anxiety and depression, differences in anxiety between participants receiving inpatient palliative care and those receiving outpatient care, and characteristics that distinguish highly anxious from less anxious patients.
Methods
Participants were 194 patients with terminal cancer. Approximately half (n = 103) were receiving inpatient care in a palliative care facility and half (n = 91) were receiving outpatient care in a tertiary care cancer center. The Hospital Anxiety and Depression Scale was used to assess anxiety and depression, and was administered along with measures of hopelessness, desire for hastened death, and social support.
Results
Moderately elevated anxiety symptoms were found in 18.6% of participants (n = 36) and 12.4% (n = 24) had clinically significant anxiety symptoms. Level of anxiety did not differ between the two treatment settings. However, participants receiving palliative care reported significantly higher levels of depression and desire for hastened death. A multivariate prediction model indicated that belief in an afterlife, social support, and anxiolytic and antidepressant use were unique, significant predictors of anxiety.
Conclusion
Severity of anxiety symptoms did not differ between the study sites, suggesting that anxiety may differ from depression and desire for hastened death in the course that it takes over the duration of terminal cancer.
Keywords: Anxiety, cancer, palliative care
Introduction
Anxiety is one of the most common reasons for psychiatric consultation in terminally ill cancer patients and has been linked to lower levels of quality of life, increased levels of insomnia, decreased trust in physicians and poor treatment compliance (1–4). Patients with advanced or terminal cancer often experience anxiety surrounding the treatment process, disease progression, uncontrolled pain, dying, and uncertainty as to what happens after death (5–7). Anxiety also can negatively impact recovery from medical procedures, and may even impact survival time in cancer patients (8). However, despite its significant impact on patient quality of life, the causes and consequences of anxiety have rarely been the focus of systematic research in patients with advanced cancer (9–11).
Only a handful of studies have focused specifically on anxiety in advanced or terminally ill cancer patients. Estimates of prevalence of anxiety disorders range from 2% to 14% of patients with advanced cancer (4, 12–14). Generalized anxiety disorder (4.8%–5.8%) and panic disorder (5.5%), the most common diagnoses, were found at rates that exceeded those found in the general adult population (3.1% and 2.7% respectively [11, 12]). Significant anxiety symptoms are estimated in 25% to 48% of advanced cancer patients (3, 14, 15). Elevated rates of anxiety disorders and symptoms are also more common in females, unmarried patients and those with poor physical functioning (4, 16).
Studies examining the relationship between disease severity and anxiety in terminal cancer have generated conflicting findings as to whether patients receiving only palliative care experience more (17) or fewer (16) symptoms of anxiety than those receiving potentially curative treatment. A review of the literature of psychiatric disorders in patients with advanced cancer (14) concluded that anxiety disorders were increasingly common as patients progressed from having advanced disease to terminal disease. Still other studies have observed no relationship between disease severity and anxiety (18). These inconsistent findings regarding the relationship between anxiety and disease severity warrant further investigation.
One of the challenges associated with identifying anxiety in advanced cancer patients is that anxiety is often entangled with other types of psychological distress such as depression, worry and negative affect (11, 19, 20). The frequency of co-morbid anxiety and depressive disorders in advanced cancer patients has resulted in the identification of a mixed anxiety/depression phenotype of depression in cancer patients (21). These findings exemplify the complex relationship between anxiety and depression in patients with a terminal illness and underscore the difficulties in studying anxiety at the end of life.
The potential implications of unidentified and under-treated anxiety symptoms necessitate more systematic research. By better understanding of the nature and prevalence of anxiety in terminally ill cancer patients, physicians and mental health professionals may be better able to identify patients at risk for experiencing anxiety symptoms, refer those with clinically significant symptoms for appropriate treatments, and perhaps avoid the diminished quality of life and reduced survival time related to anxiety. The present study attempts to address this need by investigating the prevalence of anxiety in a sample of terminally ill cancer patients. A primary study aim was to identify predictors of anxiety, with a focus on treatment-related variables and psychosocial variables that have been previously linked to psychosocial well-being at the end of life (22, 23). In keeping with previous findings in the literature, we expected that social support, religiosity and belief in an afterlife would be associated with lower levels of anxiety, and demographic factors such as female gender and being unmarried would be associated with higher levels of anxiety symptoms. Additionally, in an attempt to clarify the conflicting results in the literature concerning the relationship between anxiety and proximity to death, anxiety levels were compared between patients receiving inpatient palliative care who were closer to death and those receiving outpatient, life-prolonging care.
Methods
Terminally ill cancer patients were recruited from two New York City area hospitals. All eligible patients had a diagnosis of stage IV cancer and a life expectancy of less than one year. Roughly half of the participants were recruited following admission to a 200-bed palliative care hospital for terminally ill cancer patients; these patients were not receiving any life-prolonging care. The second half of the sample was recruited from a tertiary care cancer center, where they were receiving aggressive, life-prolonging or experimental interventions despite advanced disease. Participants recruited from the palliative care hospital typically had a life expectancy of less than six weeks while those recruited from the cancer center generally had a longer life expectancy (3 to 12 months). All participants were recruited as part of a larger study evaluating a measure of hopelessness in terminally ill cancer patients (24). Institutional Review Boards at Calvary Hospital, Memorial Sloan-Kettering Cancer Center, and Fordham University approved the study.
Hospital records were reviewed to determine whether patients were English-speaking, had sufficient cognitive functioning to be formally assessed or approached (described below), and to verify the presence of a terminal cancer diagnosis and life expectancy of less than one year. Eligible patients were free of significant cognitive impairment, as evidenced by a score greater than 20 on the Mini-Mental State Exam (MMSE) (25), and were able to provide informed consent. Interviews were conducted at the patient’s bedside, following clinic appointments, or while participants received outpatient chemotherapy. The study was discontinued if a participant became too ill to continue.
All participants who provided informed consent were interviewed to elicit relevant demographic information including age, sex, race, ethnicity, marital status, education and religion. Medical information, including primary cancer diagnosis, stage and date of diagnosis, use of mental health services and psychiatric medications, was obtained from the patient’s hospital record. Participants were then administered a series of self-report measures. Anxiety was measured using the anxiety subscale of the Hospital Anxiety and Depression Scale (HADS) (i.e., the HADS-A), a 14-item self-report measure developed for use in a medically-ill population (26). The HADS includes seven anxiety items and seven depression items, but does not include the somatic symptoms that typically confound assessment of anxiety and depression in medically ill patients (10). The HADS has demonstrated high reliability and validity in a broad range of medically-ill populations (27, 28), with a cut-off of 8 providing the optimal discrimination between patients with and without clinically significant anxiety symptoms. However, Zigmond and Snaith (26), in their original validation study, proposed more conservative cut-off scores, recommending a cut-off of 8 to 10 for identifying the “possible” presence of clinically significant anxiety symptoms and scores of 11 and above indicating “definite” clinically significant anxiety symptoms. Participants also completed the Beck Hopelessness Scale (BHS) (29), the Schedule of Attitudes toward a Hastened Death (SAHD) (30), and the Duke-UNC Functional Social Support Questionnaire (FSSQ) (31), a measure of the functional elements of the social supports of patients in a primary care or medical setting. For participants who were too ill to complete these self-report measures themselves, the items were read aloud by the research assistant.
Statistical Analysis
Descriptive statistics were calculated for the overall sample on the demographic, clinical and psychosocial variables. Differences on these demographic, clinical and psychosocial variables between the inpatient and outpatient samples were assessed using t-test and Chi-square statistics. Correlation coefficients were used to examine the relationship between anxiety and other psychosocial measures. Finally, multiple regression analysis was used to assess the statistical prediction of anxiety, first entering demographic and clinical correlates and subsequently, variables of interest (e.g., treatment setting, social support).
Results
A total of 248 patients consented to participate in the study, six were deemed ineligible (i.e., were receiving inpatient, life-prolonging care), and 48 participants did not complete the HADS. Of these 48 participants, 12 (25.0%) expired before completing the study, 15 (31.25%) were deemed too ill to continue either because of increasing confusion or disease progression, 20 (41.66%) elected to discontinue the study, and one (2.08%) participant did not complete the HADS because of administrative error. Seven participants omitted a single item of the HADS; for these cases, HADS scores were pro-rated to account for the missing item.
The average participant age was 62.4 years (standard deviation [SD] 12.8; range 28 to 94). The majority of the sample was female (54.6%, n = 106) and Caucasian. Participants were highly educated, and almost half of the participants were married. The majority of participants considered themselves to be religious and reported some degree of belief in an afterlife. Complete demographic characteristics are presented in Table 1.
Table 1.
Demographic Characteristics and Comparison of Samples
| Variable | Total Sample | Inpatient Palliative Care | Outpatient Life-Prolonging Care | Df | χ2 or t |
|---|---|---|---|---|---|
| M (SD) | M (SD) | M (SD) | |||
| Age | 62.4 (12.8) | 65.6 (13.5) | 58.8 (11.0) | 192 | 3.76a |
| Education (years) | 14.6 (3.0) | 13.7 (3.1) | 15.8 (2.7) | 187 | 4.88a |
| Months to death | 4.1 (4.1) | 1.9 (1.5) | 8.1 (4.3) | 136 | 12.47b |
|
| |||||
| n (%) | n (%) | n (%) | Df | χ2 or t | |
|
| |||||
| Gender | |||||
| Male | 88 (45.5) | 39 (37.9) | 49 (53.8) | 1 | 4.97b |
| Female | 106 (54.5) | 64 (62.1) | 42 (46.2) | ||
| Race | |||||
| White | 154 (79.8) | 78 (75.7) | 76 (83.5) | 5 | 6.24 |
| Black | 29 (14.9) | 19 (18.4) | 10 (11.0) | ||
| Asian | 7 (3.6) | 4 (3.9) | 3 (3.3) | ||
| Other | 4 (2.0) | 2 (1.0) | 0 (0.0) | ||
| Ethnicity | |||||
| Hispanic | 21 (10.8) | 14 (13.6) | 7(7.7) | 2 | 2.81 |
| Not | 172 (88.7) | 89 (91.7) | 83(91.2) | ||
| Hispanic | |||||
| Marital Status | |||||
| Married | 94 (48.5) | 31 (30.1) | 63 (69.2) | 3 | 32.05a |
| Single | 36 (18.6) | 26 (25.2) | 10 (11.0) | ||
| Divorced | 39 (20.1) | 25 (24.3.5) | 14 (15.4) | ||
| Widowed | 25 (12.9) | 21 (20.4) | 4 (4.4) | ||
| Religion | |||||
| Catholic | 83 (42.8) | 45 (43.7) | 38 (41.8) | 5 | 13.29b |
| Protestant | 25 (12.9) | 9 (8.7) | 16 (17.6) | ||
| Jewish | 37 (19.1) | 14 (13.6) | 23 (25.3) | ||
| Baptist | 9 (4.6) | 6 (5.8) | 3 (3.3) | ||
| Other | 24 (12.3) | 18 (17.5) | 6 (6.6) | ||
| None | 16 (8.2) | 11 (10.7) | 5 (5.5) | ||
| Religious | |||||
| Yes | 104 (53.6) | 61 (59.2) | 43 (47.3) | 3 | 6.61 |
| Somewhat | 37 (19.1) | 23 (25.3) | 14 (13.6) | ||
| No | 51 (26.3) | 25 (27.4) | 26 (25.2) | ||
| Belief in an Afterlife | |||||
| Yes | 114 (58.8) | 61 (59.2) | 53 (58.2) | 3 | 1.92 |
| Somewhat | 42 (21.6) | 22 (21.4) | 20 (22.0) | ||
| No | 36 (18.6) | 18 (17.5) | 18 (19.8) | ||
P < 0.01.
P < 0.05.
The most common cancer diagnoses included lung (18.0%; n = 35), followed by pancreas (14.9%; n = 29), colorectal (12.3%; n = 24), and breast (11.9%; n = 23). All other diagnoses occurred in less than 5% of the sample. Anxiolytics and antidepressants were the most frequently used psychotropic medications, with 23.7% (n = 46) taking anxiolytics and 24.7% (n = 48) taking antidepressants; 9.8 % (n = 19) were taking both anxiolytic and antidepressant medications. Sleep aids were prescribed to 19.6% (n = 38) of participants.
Of the 194 participants who were administered the HADS, 103 (53.1%) were recruited from the inpatient palliative care facility and 91 patients (46.9%) from the outpatient tertiary care facility (i.e., were receiving life-prolonging care). At the time of data analysis, dates of death were available for 88 (85.4%) participants from the inpatient palliative care facility and for 50 (54.9%) of the participants from the outpatient, life-prolonging care sample. As expected, the palliative care inpatients were significantly closer to death than those receiving outpatient, life prolonging care. They were also significantly older, more likely to be female, had fewer years of education, and were less likely to be married. The two samples also differed significantly by religious affiliation; however, no clear pattern emerged, and they did not differ by level of belief in an afterlife or by degree of religiosity. Table 1 displays the demographic and medical characteristics of the sample across the two study sites.
Prevalence and Correlates of Anxiety
The overall mean for the HADS-A subscale was 5.58 (SD 4.31; range 0–21). Nearly 70% (n = 134) of study participants reported anxiety levels in the average range, falling below the cutoff for identifying clinically significant anxiety (i.e., 8 or above). However, 18.6% of the sample (n = 36) obtained HADS-A scores between 8 and 10, suggesting the presence of moderate anxiety symptoms. An additional 12.4% of participants (n = 24) reported high levels of anxiety (11 or greater), indicating severe anxiety symptoms. Individual item analysis revealed that the most highly endorsed item on the HADS-A was the presence of worrying thoughts (mean = 1.03; see Table 2); however, given that only 30% of the sample reported elevated levels of anxiety symptoms, this mean was quite low. This pattern of endorsement held when participants were separated into those with elevated levels of anxiety symptoms (HADS-A ≥ 8) and those with lower levels (HADS-A < 8). The presence of worrying thoughts was the most strongly endorsed item, particularly by those with elevated anxiety (Table 2).
Table 2.
Item Statistics for the HADS-A Subscale
| Total Sample (n = 193) | HADS-A < 8 (n = 133) | HADS-A ≥ 8 (n = 60) | |
|---|---|---|---|
| Item | M (SD) | M (SD) | M (SD) |
| 1. I feel tense or wound up | 0.88 (0.84) | 0.53 (6.62) | 1.65 (0.73) |
| 3. I get a frightened feeling as if something awful is about to happen | 0.80 (0.93) | 0.40 (0.60) | 1.69 (0.93) |
| 5. Worrying thoughts go through my mind | 1.03 (1.01) | 0.59 (0.59) | 2.02 (1.24) |
| 7. I can sit at ease and feel relaxed | 0.87 (0.87) | 0.56 (0.70) | 1.55 (0.83) |
| 9. I get a sort of frightened feeling like “butterflies” in the stomach | 0.54 (0.76) | 0.29 (0.48) | 1.08 (0.94) |
| 11. I feel restless as if I have to be on the move | 0.96 (0.97) | 0.65 (0.81) | 1.67 (0.94) |
| 13. I get sudden feelings of panic | 0.51 (0.71) | 0.24 (0.44) | 1.10 (0.83) |
As expected, anxiety was significantly positively correlated with other measures of psychological distress including depression, hopelessness, and desire for hastened death, and negatively correlated with social support. In addition, women reported significantly higher levels of anxiety than men. Participants who reported no belief in an afterlife had the lowest level of anxiety, followed by those with a definite belief in an afterlife. Participants who reported an unsure or “somewhat” belief in an afterlife reported significantly higher levels of anxiety. When belief in an afterlife was collapsed into a two-level variable (no belief versus somewhat/definite belief), participants who reported any level of belief in an afterlife reported higher levels of anxiety than those who did not (complete data presented in Table 3). Anxiety was not associated with other demographic variables (i.e., age, race, ethnicity, education, marital status and religion). Antidepressant and anxiolytic use also was associated with higher levels of anxiety. There also were no significant associations between anxiety and any other medical variables, including time since diagnosis, proximity to death, or primary cancer site.
Table 3.
Correlates of Anxiety (HADS-A)
| Variable | M (SD) | df | t or F |
|---|---|---|---|
| Gender | |||
| Female | 6.32 (4.45) | 192 | 2.67a |
| Male | 4.68 (3.99) | ||
| Race | |||
| White | 5.73 (4.39) | 3 | 0.660 |
| Black | 4.72 (3.29) | ||
| Asian | 6.43 (6.92) | ||
| Other | 4.25 (2.06) | ||
| Ethnicity | |||
| Hispanic | 6.90 (4.82) | 191 | 1.48 |
| Not Hispanic | 5.43 (4.24) | ||
| Marital Status | |||
| Married | 5.44 (4.14) | 3 | 0.436 |
| Single | 5.33 (4.81) | ||
| Divorced | 6.28 (4.62) | ||
| Widowed | 5.36 (3.80) | ||
| Religion | |||
| Catholic | 5.59 (3.90) | 5 | 0.291 |
| Protestant | 5.56 (5.71) | ||
| Jewish | 6.05 (4.36) | ||
| Baptist | 4.56 (3.57) | ||
| Other | 5.71 (4.33) | ||
| None | 4.81 (4.50) | ||
| Religious | |||
| Yes | 5.84 (4.16) | 3 | 1.70 |
| Somewhat | 6.30 (4.80) | ||
| No | 4.67 (4.18) | ||
| Belief in an afterlife | |||
| Yes | 5.46 (4.18) | 190 | 3.73b |
| Somewhat | 7.19 (4.87) | ||
| No | 4.25 (3.52) | ||
| Anxiolytics | |||
| Yes | 8.15 (4.60) | 189 | 4.90a |
| No | 4.75 (3.93) | ||
| Antidepressants | |||
| Yes | 7.19 (4.46) | 189 | 3.05b |
| No | 5.03 (4.18) | ||
|
| |||
| r | |||
|
| |||
| Age | −0.07 | ||
| Education (years) | −0.04 | ||
| Depression (HADS-D) | 0.54a | ||
| Hopelessness (BHS) | 0.54a | ||
| Desire for Hastened Death (SAHD) | 0.53a | ||
| Social Support (FSSQ) | −0.27a | ||
HADS = Hospital Anxiety and Depression Scale; SAHD = Schedule of Attitudes Toward Hastened Death; BHS =Beck Hopelessness Scale; FSSQ = Duke-UNC Functional Social Support Questionnaire.
P < 0.01.
P < 0.05.
Anxiety and Depression
In order to investigate the relationship between anxiety and depression, participants were identified as having elevated levels of anxiety symptoms only (HADS-A ≥ 8), elevated levels of depressive symptoms only (HADS-D ≥ 8), elevated levels of mixed anxiety and depressive symptoms (HADS-A ≥ 8 and HADS-D ≥ 8) or normal levels of anxiety and depression (HADS-A < 8 and HADS-D < 8). Mixed anxiety and depressive symptoms was the most prevalent subtype (n = 31, 16.0%) while 14.9% (n = 29) reported only elevated anxiety symptoms and 13.4% (n = 26) reported only elevated depressive symptoms.
These groups differed by several demographic, clinical and psychological variables. Analysis of variance results show a significant main effect for hopelessness, desire for hastened death, social support, and proximity to death (Table 4). Tukey HSD (Honestly Significant Difference) post hoc tests were conducted to determine which groups were significantly different. Participants with mixed anxiety and depression reported significantly more hopelessness and desire for hastened death. Elevated depression only was associated with lower levels of social support and closer proximity to death. Participants were more likely to be prescribed anxiolytics if they reported purely anxiety symptoms rather than mixed symptoms.
Table 4.
Correlates of Anxiety and Depression Subtype
| Low Anxiety & Depression (n = 107) | Elevated Anxiety Only (n = 28) | Elevated Depression Only (n = 24) | Elevated Anxiety & Depression (n = 29) | |||
|---|---|---|---|---|---|---|
| M (SD) | M (SD) | M (SD) | M (SD) | df | F | |
| Desire for hastened death (SAHD) | 1.53 (2.16) | 2.58 (2.66) | 5.00 (4.90) | 7.12 (4.73) | 3 | 27.69a |
| Hopelessness (BHS) | 3.16 (3.49) | 5.79 (3.71) | 6.76 (4.52) | 10.58 (5.02) | 3 | 30.24a |
| Social support (FSSQ) | 50.15 (7.66) | 47.07 (6.84) | 45.33(12.22) | 46.55 (10.08) | 3 | 3.97b |
| Proximity to Death (months) | 5.08 (4.72) | 4.44 (4.11) | 2.15 (2.03) | 2.76 (2.69) | 3 | 3.97a |
|
| ||||||
| n (%) | n (%) | n (%) | n (%) | df | χ2 | |
|
| ||||||
| Anxiolytics | ||||||
| No | 86 (59.3) | 15 (10.3) | 22(15.2) | 22 (15.2) | 3 | 12.83a |
| Yes | 19 (41.3) | 14 (30.4) | 4 (8.7) | 9 (19.6) | ||
| Treatment Setting | ||||||
| Inpatient | 50(48.5) | 9 (8.7) | 22 (21.4) | 22 (2) | 3 | 22.02a |
| Outpatient | 58 (63.7) | 20 (22.0) | 4 (4.4) | 9 (9.9) | ||
HADS = Hospital Anxiety and Depression Scale; SAHD = Schedule of Attitudes Toward Hastened Death; BHS =Beck Hopelessness Scale; FSSQ = Duke-UNC Functional Social Support Questionnaire.
P < 0.01.
P< 0.05.
Anxiety and Treatment Setting
As displayed in Table 5, anxiety symptom severity was virtually identical across the two treatment samples. Interestingly, depression, social support and desire for hastened death did differ significantly between the two study sites, with the palliative care sample reporting significantly higher levels of distress than the life-prolonging care sample on both measures.
Table 5.
Psychosocial Characteristics: Comparison of Samples
| Variable | Palliative Care | Life-Prolonging care | df | t |
|---|---|---|---|---|
| Anxiety (HADS-A) | 5.54 (4.21) | 5.62 (4.45) | 192 | 0.12 |
| Depression (HADS-D) | 6.83 (4.29) | 4.11 (3.80) | 192 | 4.65a |
| Desire for hastened death (SAHD) | 4.15 (4.10) | 1.80 (3.10) | 191 | 4.45a |
| Hopelessness (BHS) | 5.79 (4.84) | 4.59 (4.61) | 192 | 1.77 |
| Social support (FSSQ) | 49.74 (7.67) | 47.18 (9.74) | 186 | 2.01b |
HADS = Hospital Anxiety and Depression Scale; SAHD = Schedule of Attitudes Toward Hastened Death; BHS =Beck Hopelessness Scale; FSSQ = Duke-UNC Functional Social Support Questionnaire.
P < 0.01.
P < 0.05.
Although there were no differences on overall HADS-A scores between the two treatment settings, when the two treatment settings were compared based on the four categories of anxiety and depression mentioned above, there were significant differences between the groups. The participants receiving outpatient, life-prolonging care were more purely anxious whereas those receiving inpatient palliative care reported greater symptoms of mixed anxiety and depression (Table 4).
Multivariate Prediction of Anxiety
A hierarchical multiple regression analysis was used to develop a model for identifying variables associated with participants’ HADS-A scores. Because a primary goal of this study was to determine if anxiety level was related to treatment setting, all variables that differed significantly between the two treatment settings were included in the first step of the regression model (i.e., as covariates). This included age, gender, years of education and marital status (married/cohabitating versus single, separated, widowed or divorced). This model, including only the covariates, was not statistically significant, F(4, 173) = 1.97, P = 0.10, R2 = 0.044, but gender was significantly associated with HADS-A score (Table 6). Treatment setting (life-prolonging or palliative care) was entered on the second step of the model, but this too was not significantly associated with HADS-A scores (F(5,172) = 1.57, P = 0.17, R2 change = 0.0%).
Table 6.
Final Model Predicting HADS-A
| Step | Variable | B | SE B | B | df | t |
|---|---|---|---|---|---|---|
| 1. | Age | −0.02 | 0.03 | −0.06 | 1 | −0.72 |
| Gender | −0.69 | 0.64 | −0.08 | 1 | −1.08 | |
| Education (years) | 0.06 | 0.11 | 0.04 | 1 | 0.52 | |
| Marital Status | −0.28 | 0.75 | −0.03 | 1 | −0.37 | |
| 2. | Treatment Setting | 0.70 | 0.67 | −0.08 | 1 | −1.05 |
| 3. | Belief in an afterlife | 1.78 | 0.79 | 0.16 | 1 | 2.24b |
| Anti-depressant use | 1.92 | 0.74 | 0.19 | 1 | 2.59a | |
| Anxiolytic use | 2.52 | 0.73 | 0.19 | 1 | 3.46a | |
| Social support | −0.10 | 0.04 | −0.20 | 1 | −2.66a |
HADS-A = Hospital Anxiety & Depression Scale–Anxiety subscale; Social support = Duke-UNC Functional Social Support Questionnaire (FSSQ)
P < 0.01.
P< 0.05.
Finally, additional variables that had a bivariate association with anxiety (use of anxiolytics and antidepressants, social support and belief in an afterlife) were entered simultaneously into the model to identify the unique contribution of these variables beyond the effect of demographic and medical variables. The final model was statistically significant, F(9,168) = 5.67, P < 0.001, R2 change =18.9%, with use of anxiolytics, antidepressants, belief in an afterlife, and social support significantly associated with HADS-A score. Gender was no longer significant in the final model.
Discussion
The present study provides a comprehensive examination of anxiety in terminally ill patients with cancer. Participants obtained a wide range of scores on the HADS-A (0 to 21), with the majority of participants reporting low levels of anxiety and others experiencing severe and clinically significant anxiety. Although infrequently studied, these rates of clinically significant anxiety symptoms are consistent with a recent prevalence study examining anxiety disorders in terminally ill cancer patients (11).
Surprisingly, severity of anxiety symptoms did not differ by treatment setting used as an approximation of illness severity or proximity to death. Moreover, illness severity (sample) was not a significant predictor of anxiety level in any of the multiple regression models. Although a true examination of the relationship between illness severity and anxiety would require a longitudinal analysis, these null findings suggest that patients do not become more anxious as they approach death, and differs from previous cross-sectional studies that found an association between anxiety and disease severity (14,17). However, these prior studies examined differences between patients with different stages of cancer (i.e., early versus advanced illness stage) whereas the present study included only patients with stage IV cancer, studying the relationship between patients receiving outpatient care versus those receiving only palliative care in an inpatient setting. Perhaps any increase in anxiety occurs at an earlier stage of disease progression, such as between early and advanced disease (e.g., at the time of identifying progressive or metastatic disease), rather than at the time of stopping potentially curative treatment.
Although anxiety did not differ between the two samples, there were significant group differences in other measures of psychological distress. Participants in the palliative care sample reported significantly higher levels of depression, hopelessness and desire for hastened death. These results indicate that many aspects of psychological distress do increase as patients approach death, even if anxiety does not. This finding is particularly interesting because anxiety is significantly correlated with other measures of psychological distress, despite having a different association with disease progression. Hence, it appears that the emergence of clinically significant anxiety may occur early in the course of a life-threatening illness, whereas depression and desire for hastened death do not emerge until much later, after the severity of one’s illness has become evident. Differences between patients with pure anxiety and those with elevated anxiety and depression support this assertion. Patients with elevated anxiety and depression were more hopeless and reported greater desire for hastened death than those who were purely anxious. Additionally, patients receiving inpatient palliative care were more likely to have elevated anxiety and depression and those receiving outpatient, life-prolonging treatment were more likely to report elevated anxiety only. Depression may become more of a factor as patients acutely approach death. Differences in the pharmaceutical management of anxiety and depression also may be a factor in differences in patterns of elevated anxiety and depressive symptoms between the two sites. Participants receiving inpatient palliative care may be more aggressively treated for anxiety symptoms at the end of life, and anxiety symptoms may more quickly respond to pharmaceutical intervention than depressive symptoms.
Another aim of this study was to identify the demographic, clinical and psychosocial characteristics of highly anxious patients. Among the strongest predictors of anxiety were social support, psychotropic medication use, and belief in an afterlife. Not surprisingly, social support was negatively associated with anxiety in both univariate analyses and multiple regression models, a finding that is consistent with an extensive research literature (32, 33). The impact of psychotropic medication use also was expected, as anxiolytic and antidepressant medications are well-established for the treatment of anxiety symptoms. What was unexpected, however, was the negative association between belief in an afterlife and anxiety. Although we anticipated that belief in an afterlife might buffer the impact of advanced disease on anxiety, our data indicated precisely the opposite effect; patients with a belief in an afterlife were more anxious (on average) than patients who did not hold this belief. This relationship is inconsistent with the existing literature (23) and seems to be primarily driven by patients who report a tentative or “somewhat” belief in an afterlife. This indecision seems to be related to increased anxiety symptoms and may provide a point of intervention for pastoral care services.
There are several limitations that influence the interpretation of the study results. The rates of anxiety found in this study (as well as previous studies) may have been artificially lowered by the participants’ use of anxiolytic medications, which was comparable across the two study sites and by possible selection bias. Anxiolytic medications are frequently used and efficacious in reducing symptoms of anxiety (34). Nearly one-quarter of participants at each site were prescribed anxiolytic medications, which would likely lower the level of anxiety experienced by these patients. As patients experiencing heightened levels of anxiety are likely to be prescribed anxiolytic medications, they may report low levels of anxiety. Thus, measures of current distress may not adequately measure a patient’s underlying level of anxiety in patients who are “successfully treated”(35).
Because of the potential confounding influence of somatic symptoms, it is difficult to accurately diagnose anxiety and mood disorders in medically ill populations (9, 10, 36). Thus, we used the HADS, a reliable, valid self-report measure to identify anxiety symptoms, which has been used in many other studies of anxiety in patients with advanced cancer (3, 15, 16). However, unlike the Structured Clinical Interview for DSM Disorders, which is considered the gold standard for diagnosing psychiatric disorders (4), it is not possible to draw definite diagnostic conclusions using the HADS. We identified patients with elevated levels of anxiety symptoms who would likely meet diagnostic criteria for an anxiety or adjustment disorder. Therefore, we are not only identifying patients at risk for “pure” anxiety disorders, but adjustment disorders as well, which are the most common diagnosis in advanced cancer patients (14, 21, 37, 38). Furthermore, the present study was limited by the unavailability of data concerning pain and functional limitations. The presence of uncontrolled pain and symptom burden are significantly correlated with psychological disorders in advanced cancer patients (16, 39–41). However, recent findings may in fact indicate a weak relationship between physical symptom burden and anxiety (15).
Additional study limitations included the reliance on self-report data. Aside from chart review to determine current medications patients, provided all information. It is also important to note that because the patients receiving palliative care were bed-ridden and significantly sicker than those receiving life-prolonging care, the questionnaires were read aloud to those participants and may have influenced reporting of psychological distress. Studying psychological distress in terminally ill cancer patients presents many challenges including the declining physical health of the participants. Many patients were too ill or cognitively limited to participate in the study, and others did not complete the study because of disease progression and increasing confusion.
Despite the lack of association between disease severity and anxiety, and study limitations, the high rates of anxiety and other forms of psychological distress observed highlight the importance of developing adequate methods for identifying and treating distress. Further, research addressing the optimal ways to bolster patient resources (e.g., social support, uncertainty regarding afterlife beliefs) may help improve quality of life for this vulnerable population.
Acknowledgments
Support for this study was provided by the National Cancer Institute (Grant No. CA-101940, Barry Rosenfeld, Principal Investigator).
Many thanks to the following colleagues for their help in collecting data and managing the study: Jennifer Abbey, Joanna Fava, Megan Olden, Emily Sachs, Lia Amakawa and Maryann Santasiero. The authors also would like to thank the study participants, terminally ill men and women who gave of themselves to help us better understand psychological distress at the end of life.
Footnotes
Disclosures
The authors declare no conflicts of interest.
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 citable 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.Ballantyne JC, editor. Pain Clinical Updates. Seattle: International Association for the Study of Pain; 2009. The Cancer Patient with Anxiety and Chronic Pain. Pain; p. 6. [Google Scholar]
- 2.Stark D, Kiely M, Smith A, et al. Anxiety disorders in cancer patients: their nature, associations, and relation to quality of life. J Clin Oncol. 2002;15(20):3137–3148. doi: 10.1200/JCO.2002.08.549. [DOI] [PubMed] [Google Scholar]
- 3.Smith EM, Gomm SA, Dickens CM. Assessing the independent contribution to quality of life from anxiety and depression in patients with advanced cancer. Palliat Med. 2003;17:509–513. doi: 10.1191/0269216303pm781oa. [DOI] [PubMed] [Google Scholar]
- 4.Spencer R, Nilsson M, Wright A, et al. Anxiety disorders in advanced cancer. Cancer. 2010;116:1810–1819. doi: 10.1002/cncr.24954. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Roth A, Nelson CJ, Rosenfeld B, et al. Assessing anxiety in men with prostate cancer: further data on the reliability and validity of the Memorial Anxiety Scale for Prostate Cancer (MAX-PC) Psychosomatics. 2006;47:340–347. doi: 10.1176/appi.psy.47.4.340. [DOI] [PubMed] [Google Scholar]
- 6.American Cancer Society. [Accessed 2009.];Cancer, anxiety and fear. Available from http://www.cancer.org/docroot/MBC/content/MBC_4_1x_Intro_Anxiety.asp Updated August 2008.
- 7.Passik SD, Kirsh KL, Rosenfeld B, McDonald MV, Theobald DE. The changeable nature of patients’ fears regarding chemotherapy: implications for palliative care. J Pain Symptom Manage. 2001;21:113–120. doi: 10.1016/s0885-3924(00)00249-9. [DOI] [PubMed] [Google Scholar]
- 8.Groenvold M, Petersen MA, Idler E, et al. Psychological distress and fatigue predicted recurrence and survival in primary breast cancer patients. Breast Cancer Res Treat. 2007;105:209–219. doi: 10.1007/s10549-006-9447-x. [DOI] [PubMed] [Google Scholar]
- 9.Jacobsen PB, Breitbart W. Psychosocial aspects of palliative care. Cancer Control. 1996;3:214–222. doi: 10.1177/107327489600300303. [DOI] [PubMed] [Google Scholar]
- 10.Pessin H, Rosenfeld B, Breitbart W. Assessing psychological distress near the end of life. Am Behav Sci. 2002;46:357–372. [Google Scholar]
- 11.Wilson KG, Chochinov HM, Skirko MG, et al. Depression and anxiety disorders in palliative cancer care. J Pain Symptom Manage. 2007;33:118–129. doi: 10.1016/j.jpainsymman.2006.07.016. [DOI] [PubMed] [Google Scholar]
- 12.Kadan-Lottick NS, Vanderwerker LC, Block SD, Zhang B, Prigerson HG. Psychiatric disorders and mental health service use in patients with advanced cancer. Cancer. 2005;104:2872–2881. doi: 10.1002/cncr.21532. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kissane DW, Grabsch B, Love A, et al. Psychiatric disorder in women with early stage and advanced breast cancer: a comparative analysis. Aust N Z J Psychiatry. 2004;38:320–326. doi: 10.1080/j.1440-1614.2004.01358.x. [DOI] [PubMed] [Google Scholar]
- 14.Miovic M, Block S. Psychiatric disorders in advanced cancer. Cancer. 2007;15(110):1665–1676. doi: 10.1002/cncr.22980. [DOI] [PubMed] [Google Scholar]
- 15.Teunissen SCCM, de Graeff A, Voest EE, de Hass JCJM. Are anxiety and depressed mood related to physical symptom burden? A study in hospitalized advanced cancer patients. Palliat Med. 2007;21:341–346. doi: 10.1177/0269216307079067. [DOI] [PubMed] [Google Scholar]
- 16.Mystakidou K, Tsilika E, Parpa E, et al. Assessment of anxiety and depression in advanced cancer patients and their relationship with quality of life. Qual Life Res. 2005;14:1825–1833. doi: 10.1007/s11136-005-4324-3. [DOI] [PubMed] [Google Scholar]
- 17.Simon AE, Thompson MR, Flashman K, Wardle J. Disease stage and psychosocial outcomes in colorectal cancer. Colorectal Dis. 2009;11:19–25. doi: 10.1111/j.1463-1318.2008.01501.x. [DOI] [PubMed] [Google Scholar]
- 18.Lichtenthal WG, Nilsson M, Zhang B, et al. Do rates of mental disorders and existential distress among advanced stage cancer patients increase as death approaches? Psychooncology. 2009;18:50–61. doi: 10.1002/pon.1371. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Buck JM, Kelly WE, Silver NC. An investigation of the relationship between depression and worry: a research note. Individual Differences Research. 2008;6:120–122. [Google Scholar]
- 20.Le Fevre P, Devereux J, Smith S, Lawrie SM, Cornbleet M. Screening for psychiatric illness in the palliative care inpatient setting: a comparison between the Hospital Anxiety and Depression Scale and the General Health Questionnaire-12. Palliat Med. 1999;13:399–407. doi: 10.1191/026921699671260095. [DOI] [PubMed] [Google Scholar]
- 21.Brintzenhofe-Szoc KM, Levin TT, Li Y, Kissane DW, Zabora JR. Mixed anxiety/depression symptoms in a large cancer cohort: prevalence by cancer type. Psychosomatics. 2009;50:383–391. doi: 10.1176/appi.psy.50.4.383. [DOI] [PubMed] [Google Scholar]
- 22.Chen H, Jiang C, Liu Q, Liu Y, Li Y. The study of anxiety, depression and related factors in terminal cancer patients. Chin J Clin Psychol. 2002;10:108–110. [Google Scholar]
- 23.McClain-Jacobson C, Rosenfeld B, Kosinski A, et al. Belief in an afterlife, spiritual well-being and end-of-life despair in patients with advanced cancer. Gen Hosp Psychiatry. 2004;26:484–486. doi: 10.1016/j.genhosppsych.2004.08.002. [DOI] [PubMed] [Google Scholar]
- 24.Rosenfeld B, Pessin H, Lewis C, et al. Assessing hopelessness in terminally ill cancer patients: development of the Hopelessness Assessment in Illness Questionnaire (HAI) Psychological Assessment. doi: 10.1037/a0021767. in press. AU: ANY UPDATE? [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.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. 1975;12:189–198. doi: 10.1016/0022-3956(75)90026-6. [DOI] [PubMed] [Google Scholar]
- 26.Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand. 1983;67:361–370. doi: 10.1111/j.1600-0447.1983.tb09716.x. [DOI] [PubMed] [Google Scholar]
- 27.Bjelland I, Dahl AA, Haug TT, Neckelmann D. The validity of the Hospital Anxiety and Depression Scale. An updated literature review. J Psychosom Res. 2002;52:69–77. doi: 10.1016/s0022-3999(01)00296-3. [DOI] [PubMed] [Google Scholar]
- 28.Brennan C, Worrall-Davies A, McMillan D, Gilbody S, House A. The Hospital Anxiety and Depression Scale: a diagnostic meta-analysis of case-finding ability. J Psychosom Res. 2010;69:371–378. doi: 10.1016/j.jpsychores.2010.04.006. [DOI] [PubMed] [Google Scholar]
- 29.Beck AT, Weissman A, Lester D, Trexler L. The measurement of pessimism: the Hopelessness Scale. J Consult Clin Psychol. 1974;42:861–865. doi: 10.1037/h0037562. [DOI] [PubMed] [Google Scholar]
- 30.Rosenfeld B, Breitbart W, Stein K, et al. Measuring desire for death among patients with HIV/AIDS: the Schedule of Attitudes toward Hastened Death. Am J Psychiatry. 1999;156:94–100. doi: 10.1176/ajp.156.1.94. [DOI] [PubMed] [Google Scholar]
- 31.Broadhead WE, Gehlbach SH, de Gruy FV, Kaplan BH. The Duke-UNC Functional Social Support Questionnaire. Measurement of social support in family medicine patients. Med Care. 1988;26:709–723. doi: 10.1097/00005650-198807000-00006. [DOI] [PubMed] [Google Scholar]
- 32.Kelly B, Burnett P, Pelusi D, et al. Terminally ill cancer patients’ wish to hasten death. Palliat Med. 2002;16:339–345. doi: 10.1191/0269216302pm538oa. [DOI] [PubMed] [Google Scholar]
- 33.O’Mahony S, Goulet J, Kornblith A, et al. Desire for hastened death, cancer pain and depression: report of a longitudinal observational study. J Pain Symptom Manage. 2005;29:446–457. doi: 10.1016/j.jpainsymman.2004.08.010. [DOI] [PubMed] [Google Scholar]
- 34.Derogatis LR, Feldstein M, Morrow G, et al. A survey of psychotropic drug prescriptions in an oncology population. Cancer. 1979;44:1919–1929. doi: 10.1002/1097-0142(197911)44:5<1919::aid-cncr2820440555>3.0.co;2-t. [DOI] [PubMed] [Google Scholar]
- 35.Jacobsen PB, Bovbjerg DH, Redd WH. Anticipatory anxiety in women receiving chemotherapy for breast cancer. Health Psychol. 1993;12:469–475. doi: 10.1037//0278-6133.12.6.469. [DOI] [PubMed] [Google Scholar]
- 36.Periyakoil VS. Panic disorder at the end of life. J Palliat Med. 2007;10:483–485. doi: 10.1089/jpm.2006.9977. [DOI] [PubMed] [Google Scholar]
- 37.Grabsch B, Clarke DM, Love A, et al. Psychological morbidity and quality of life in women with advanced breast cancer: a cross-sectional survey. Palliat Support Care. 2006;4:47–56. doi: 10.1017/s1478951506060068. [DOI] [PubMed] [Google Scholar]
- 38.Akechi T, Okuyama H, Nishiwaki Y, Uchitomi Y. Major depression, adjustment disorder, and post-traumatic stress disorder in terminally ill cancer patients: associated and predictive factors. J Clin Oncol. 2004;22:2817–2823. doi: 10.1200/JCO.2004.08.149. [DOI] [PubMed] [Google Scholar]
- 39.Breitbart W. Suicide risk and pain in cancer and AIDS patients. In: Chapman CR, Foley KM, editors. Current and emerging issues in cancer Pain: Research and practice. New York: Raven Press; 1993. pp. 49–65. [Google Scholar]
- 40.Chochinov HM, Wilson KG, Enns M, et al. Desire for death in the terminally ill. Am J Psychiatry. 1995;152:1185–1191. doi: 10.1176/ajp.152.8.1185. [DOI] [PubMed] [Google Scholar]
- 41.Mystakidou K, Parpa E, Katsouda E, et al. Pain and desire for hastened death in terminally ill cancer patients. Cancer Nurs. 2005;28:318–324. doi: 10.1097/00002820-200507000-00014. [DOI] [PubMed] [Google Scholar]