Supportive Care in Cancer

, 14:379

Is there a correlation between spirituality and anxiety and depression in patients with advanced cancer?


    • Department of Palliative Medicine, Elgar HouseSouthmead Hospital
  • Andrew N. Davies
    • Department of Palliative MedicineRoyal Marsden Hospital
Original Article

DOI: 10.1007/s00520-005-0892-6

Cite this article as:
McCoubrie, R.C. & Davies, A.N. Support Care Cancer (2006) 14: 379. doi:10.1007/s00520-005-0892-6


Aims and objectives

To establish whether there is a correlation between spirituality and anxiety and depression in patients with advanced cancer.

Patients and methods

Patients with a diagnosis of cancer at St. Peter’s day hospice in Bristol were asked to complete three questionnaires to assess anxiety, depression and spirituality. Informed consent was obtained. Anxiety and depression are indicated by the Hospital Anxiety and Depression Scale score, and spirituality is indicated by scores on the Spiritual Well-Being Scale (SWBS) and the Royal Free Interview for Spiritual and Religious Beliefs. As will be explained, religion and spirituality are generally recognised as having different meanings—religion entailing a relationship with a higher being, while spirituality can be thought of in terms of meaning and purpose in life.


Eighty-five complete data sets were obtained. A significant negative correlation was found between both anxiety and depression scores and overall spiritual well-being scores (p<0.0001). When the SWBS subscale scores were analysed individually, a significant negative correlation was found between the existential well-being scores and the anxiety and depression scores (p<0.001). However, no correlation was found between the religious well-being scores and anxiety or depression.


This study found a significant negative correlation between spirituality (in particular, the existential aspect) and anxiety and depression in patients with advanced cancer. Religious well-being and strength of belief had no impact on psychological well-being in this study.




It has long been postulated in the literature that spirituality and religion influence both physical and mental health [1, 7, 12, 28]. Increasingly, the importance of this relationship is being recognised [28, 30, 41]. The words spirituality and religion are often used synonymously but actually have different meanings [1, 9, 10, 12, 18, 26, 27, 34, 38, 40, 42].


The dictionary definition of religion is “beliefs and opinions concerning the existence, nature, and worship of a deity” or “a particular institutionalised or personal system of beliefs and practices relating to the divine” [11]. Religiosity pertains to participation in organised beliefs, rituals and practices of traditional religion [26, 34]. Thus, most religions are centred on a belief in a higher being and a desire to please that being [38]. Although different, spirituality and religion are not necessarily mutually exclusive [34]. They may coexist, or spirituality may exist in the absence of religious beliefs or practices [4, 8, 34, 37, 40, 41, 43].


Spirituality is defined in terms of life having a purpose, the search for meaning and the attempt to interpret personal illness and death in a way that makes sense [4, 15, 16, 38, 4042]. Despite the advocated holistic approach to patients with cancer, spiritual issues are frequently overlooked in day-to-day practice [1]. Spiritual activity may be prompted by personal experience and is usually heightened as individuals face deteriorating health, and ultimately, their own mortality [9, 30, 43]. Spiritual issues raised as a patient faces the end of life often include “Does life have a meaning?” and “Why me?” [38]. Some believe that this ‘making sense’ is crucial to making the most of living with dying, and that spirituality enhances one’s health through a deeper understanding of life’s meaning or purpose [25, 38].

Spiritual “well-being” has been suggested to comprise two dimensions—existential well-being and religious well-being [32]. Existential well-being refers to the horizontal or “this worldly” aspect of spirituality. It involves perception of life’s purpose and satisfaction. In contrast, religious well-being refers to the vertical or “other worldly” dimension of spirituality, i.e. the relationship between the person and a higher being [31, 35].

Spirituality and religion with respect to health

Spiritual well-being in patients with cancer has been found to positively correlate with subjective well-being [37], lower pain levels [44], a faster recovery time from intercurrent illness [44], fighting spirit [7], hope and positive mood states [12, 31], high self-esteem, social competence, purpose in life [35] and overall quality of life [7]. In other studies involving patients with terminal cancer, low levels of spirituality have been found to correlate with negative mood states, such as tension, anxious preoccupation, depression, anger, cognitive avoidance [7, 12, 24], as well as hopelessness and suicidal ideation [6, 30] and the desire for a hastened death [3]. Furthermore, spiritual suffering is often apparent in patients with physical symptoms that do not respond to conventional measures [38].

Religious commitment has been associated with both increased physical and mental health in review articles [14, 17, 18, 28, 44]. Physical health parameters reported to be influenced by religiosity in these reviews of the literature include greater longevity, less heart disease and hypertension, less tuberculosis, emphysema, and cirrhosis and a reduced incidence of cancer [14, 17, 18, 28, 44]. Religiosity has been reported to influence mental health with respect to lower levels of depression and anxiety [14, 17], increased self-esteem, better coping with illness [18, 28], lower suicide rates [17], less substance abuse [28], less divorce [28] and higher life satisfaction [17].

Anxiety and depression are the most commonly reported emotional symptoms in patients with cancer and frequently coexist [5, 22]. Thus, the reported prevalence of depression in patients with cancer ranges from 4.5 to 58% [20, 21, 29]. Many of the somatic manifestations of anxiety and depression are commonly found in terminally ill patients. This can lead to confusion in differentiating whether these symptoms are due to physical illness, psychological illness or the normal psychological responses seen towards the end of life [20, 21, 29]. Untreated anxiety and depression can lead to difficulty with symptom control, impaired ability to make treatment decisions, poor compliance with treatment, poor social interaction and impaired quality of life [21, 22]. Understanding and treating anxiety and depression in patients with advanced disease could therefore lead to improvements in their quality of life [21, 22]. The objective of the study was to determine whether there is any correlation between spirituality and anxiety and/or depression in patients with cancer. The hypothesis was that patients with cancer with high levels of spiritual well-being are less likely to be anxious and/or depressed.


This was a cross-sectional, observational study. All patients with a diagnosis of metastatic or incurable cancer attending St. Peter’s day hospice in Bristol who were able to give written informed consent were eligible for inclusion in the study.

The researcher (RM) personally approached all eligible day hospice patients once on their usual day of attendance. Once patients had given written consent, basic demographic details were collected. Patients were then asked to complete three questionnaires whilst at day hospice: (1) the Hospital Anxiety and Depression Scale (HADS) [45], (2) the Spiritual Well-Being Scale (SWBS) [35] and (3) the Royal Free Interview (RFI) for Spiritual and Religious Beliefs [27]. The researcher (RM) offered to read out the questions to the patient if so desired.

The HADS is the most widely used tool for screening for anxiety and/or depression in patients with advanced cancer [19, 21, 33]. It assesses how the patient has been feeling over the last week. The HADS contains two subscales (anxiety and depression), with each scale containing seven questions, leading to a subscale score of 0–21 and to an overall score of 0–42 (high scores correlate with morbidity). Cut-off points for diagnosing anxiety and depression have been derived for the general population [45], although there is some disagreement about the exact cut-off points for patients with advanced cancer [20, 33, 36]. To minimise criticism about the use of specific cut-off points, this study purposefully correlated the actual scores with other relevant measures (see below). The HADS is well validated, with high sensitivity and specificity, good internal consistency and high test–retest reliability and is very acceptable to patients [19, 23].

The SWBS is probably the best instrument available to assess spiritual well-being for both research and clinical purposes [2, 15]. There are 20 items, 10 measuring one’s sense of meaning and purpose [Existential Well-Being (EWB)] and 10 measuring one’s relationship to a higher being, “god” [Religious Well-Being (RWB)]. The questions are worded in the present tense, for example, “I feel that life is a positive experience”. Both subscale scores range from 10 to 60, resulting in an overall Spiritual Well-Being (SWB) score of 20–120, with high scores indicating high levels of spiritual well-being. It has been employed with various populations [2, 12, 15, 34, 35] and has been shown to be well validated for patients with cancer, with good internal consistency (0.78–0.94) and high test–retest reliability (0.82–0.99) [2, 12, 15, 34, 35].

The revised version of the RFI consists of 18 questions [27]. It does not focus on what the patient believes but rather on strength of spiritual belief. The numerical data available from this questionnaire include a reported strength of belief score between 0 and 10 and a calculated overall strength of belief score between 0 and 60. It has been shown to have acceptably high criterion validity, predictive validity, internal consistency (0.74–0.89) and test–retest reliability (0.94) [27]. There is no literature to date on its usefulness compared with other assessment tools or on the ease of completion for patients. The RFI was chosen as an additional tool as it provides different information to the SWBS.

The aim was to obtain at least 82 complete data sets to have 90% power to detect a correlation of 0.35 or above at the 5% level of significance. The data collected were entered into a database (SPSS), and non-parametric correlation coefficients were used to explore the associations within the data.

The study was undertaken as part of the researcher’s (RM’s) M.S. in Palliative Medicine at the University of Bristol. The study was approved by the relevant Local Research Ethics Committee.


One hundred and twenty-eight patients attended day hospice during the 3-month study period. Of these, 10 patients died and 14 became too unwell to attend the day hospice before being approached for the study. Six patients had non-cancer diagnoses, two were confused, three refused to participate and the investigator did not have the opportunity to approach six patients. Eighty-seven patients participated in the study, and 85 complete data sets were obtained. The incomplete data sets were from patients who were unable to complete questions on the religious well-being subscale of the SWBS, as they could not identify with the word “god”.

Demographic data

The mean age of the participating patients was 68 years (range 29–93 years), and there were 54 women and 31 men. There were 43 married (or living with partner), 30 widowed, 7 separated and 5 single patients. Eighty-two subjects were self-classified as white British, one as black British, one as black Caribbean and one as Chinese.

The cancer diagnoses were as follows: gastrointestinal (n=16), breast (n=13), lung (n=11), urological (n=10), CNS (n=10), gynaecological (n=9), haematological (n=8) and other diagnoses (n=8). Time since primary diagnosis was grouped into 0–6 months (n=8), 6–12 months (n=8), 1–2 years (n=19), 2–5 years (n=33), 5–10 years (n=10) and 10+ years (n=9).

The Eastern Cooperative Oncology Group (ECOG) performance statuses were ECOG 0 (fully active; n=2), ECOG 1 (restricted in physically strenuous activity, ambulatory; n=11), ECOG 2 (ambulatory, unable to carry out work activities but up and about >50% waking hours; n=55) and ECOG 3 (confined to bed or chair >50% waking hours; n=17).

On simple questioning, 17 patients considered themselves spiritual, 19 religious, 30 both spiritual and religious and 19 neither spiritual nor religious.

HADS data

Using the HADS, the mean score on the anxiety subscale was 7 (range 0–18), the mean score on the depression subscale was 6 (range 0–14) and the mean overall HADS score was 13 (range 2–32). On average, women scored more highly than men in all categories of the HADS. This difference was only significant for anxiety score (p<0.05 on t test for equality of means). Interestingly, there was a statistically significant negative association using Spearman’s rank correlation between the “time since diagnosis” category and the anxiety score (p<0.04) and the overall HADS score (p<0.02), i.e. the more distant the diagnosis, the higher the relevant score. There were no other statistically significant associations between any of the other demographic variables and the HADS scores in this population.

Spiritual well-being

Using the SWBS, the mean scores were 39 (range 10–60) for religious well-being, 40 (range 33–60) for existential well-being and 79 (range 41–120) for overall spiritual well-being. There were no statistically significant associations between any of the demographic variables and the SWBS scores in this population.

Spiritual well-being vs HADS scores

The existential well-being subscale scores were strongly negatively correlated with the HADS anxiety subscale scores, the HADS depression subscale scores and overall HADS scores (Fig. 1; Table 1). However, there was no correlation between the religious well-being subscale scores and any of the HADS scores (Table 1). The overall spiritual well-being scores were also strongly negatively correlated with the HADS anxiety subscale score, the HADS depression subscale score and the overall HADS score (Table 1).
Fig. 1

Correlation between overall HADS scores and SWBS EWB scores

Table 1

Correlation between SWBS and HADS scores



Anxiety score

Depression score

Overall HADS score

RWB score

 Pearson correlation




 Significance (two-tailed)




EWB score

 Pearson correlation




 Significance (two-tailed)




Overall SWB score

 Pearson correlation




 Significance (two-tailed)




SWBS Spiritual Well-Being Scale, HADS Hospital Anxiety and Depression Scale, RWB Religious Well-Being, EWB Existential Well-Being, SWB Spiritual Well-Being

Strength of belief

From the RFI, the average score for reported strength of belief was 7 (range 0–10), and the average score for the calculated overall strength of belief was 37 (range 6–60). There was a highly significant correlation (p<0.0001) between the strength of belief scores from the RFI and both religious well-being and overall spiritual well-being scores from the SWBS (Table 2). This relationship did not exist with existential well-being.
Table 2

Correlation between SWBS and RFI strength of belief scores



Strength of belief (Qu 3, RFI)

Total strength of belief (sum of Qu 3, 7–11, RFI)

RWB score

 Pearson correlation



 Significance (two-tailed)



EWB score

 Pearson correlation



 Significance (two-tailed)



Overall SWB score

 Pearson correlation



 Significance (two-tailed)



SWBS Spiritual Well-Being Scale, RFI Royal Free Interview, Qu Question, RWB Religious Well-Being, EWB Existential Well-Being, SWB Spiritual Well-Being

Strength of belief vs HADS scores

Neither the reported strength of belief (from the RFI) nor the overall strength of belief scores (from the RFI) demonstrated any relationship with the HADS scores.


General considerations

Taking into consideration the well-recognised difficulties in recruiting palliative care patients into research, this study achieved a reasonable inclusion rate. Targeting the day hospice population was probably beneficial in that the patients are generally less unwell than the hospice inpatients, as they need to be independent and able to get in and out of a car to be able to attend. The achievement of 85 complete data sets from 87 participants was largely due to the meticulous checking of the questionnaires to ensure that all the answers were complete and asking the patients to address any omissions.

The range of cancer diagnoses showed a higher number of patients with haematological malignancies and a lower number of lung malignancies than might be expected in a general population. The time since diagnosis category has its limitations, but the ethics committee requested its inclusion. Common sense might suggest that a patient who is very recently diagnosed with cancer is likely to have high levels of anxiety or depression. As time goes by, distress may wane temporarily before peaking again at times of disease recurrence or as the patient nears the terminal phase.

Assessment of spirituality

It has been suggested that spirituality ought to be assessed routinely [8]. This may be done informally during the course of any consultation or more formally using validated assessment tools. In general, the patients found the RFI to be lengthy and difficult to complete. Many struggled with differentiating between spirituality and religion despite a paragraph within the questionnaire explaining the difference. Furthermore, it yielded little in the way of meaningful quantitative data for the purposes of the study. The SWBS was much less onerous for patients. Moreover, the two subscale scores were extremely useful in that the questions were worded specifically to differentiate between the existential and the religious. However, the two patients excluded from data analysis did find the religious questions impossible to answer because of their absolute inability to identify with a “god”. The religious well-being subscale of the SWBS has previously been criticised in that its wording is incomprehensible for people with no concept of who or what “god” is [15]. In this study, it is perhaps surprising that there were only 2 who felt unable to complete this subscale, considering that there were 20 patients who reported themselves to be neither religious nor spiritual. Thus, overall, the SWBS appears to be a more useful tool than the RFI for assessing spirituality in the palliative care population.

Spiritual vs psychological well-being

This study adds to the evidence that spiritual well-being is interlinked with psychological well-being. The strong negative correlation between existential/spiritual well-being scores and HADS scores is striking, particularly in the light of the lack of correlation between religious well-being scores and HADS scores. In other words, patients with high levels of existential/spiritual well-being were less likely to be anxious or depressed. Evidently, the ability of an individual to make sense of his circumstances and find meaning and purpose when faced with life-threatening illness has far more impact on psychological well-being than does religious faith. Moreover, strength of belief did not appear to have an influence on the patients’ psychological state. This is not in keeping with findings from other studies in which religion has historically been thought to be beneficial for psychological health as detailed earlier. Other authors have also disputed these findings [7, 8, 39]. Sloan et al. [39] claim that the published work lacks consistency and that the evidence for an association is both inconsistent and weak. One might assume that patients with strong religious beliefs have a greater acceptance of death and less death anxiety, but research to date does not support this [8]. Indeed, religion may increase anxiety through doctrines on eternal damnation [38]. Our study certainly found no link between religious well-being and anxiety and depression.

Although this work clearly demonstrates a negative correlation between anxiety and depression and spiritual well-being, it does not give any clues as to the direction of causality. Indeed, they are likely to have a reciprocal relationship in that psychological state probably exerts an influence on spirituality, while spirituality is also likely to have an impact on psychological well-being. This has implications for patient care. Recognising that existential (spiritual) needs and psychological state are interrelated is the initial step, leading to acknowledgment that these existential matters need addressing. If the direction of causality can be established in future research and spiritual well-being can be shown to lower anxiety and depression levels, there is a wealth of opportunity for addressing existential and spiritual matters as a form of treatment for these psychological illnesses. Moreover, it could be argued that addressing spiritual suffering could result in better psychological and physical well-being and, therefore, improved quality of life for patients with advanced cancer.

Spiritual care

It would be wrong to assume that an individual without a religious faith or practice would also have no spiritual need or understanding. In addition to the need for meaning and purpose in life, spiritual needs include the need for love and good relationships with self and others, the need for forgiveness, hope, joy, love, peace, dignity and trust [1, 9, 34, 43]. These enable the individual to transcend present circumstances and to connect with people, surroundings, and powers outside of self, irrespective of the presence or absence of religious belief. If these needs are not met, the prospect of illness and impending death, together with the resulting dependency and loss of social role, can trigger spiritual distress in patients and their relatives [1, 15].

It has been suggested that general spiritual care can be provided by any member of the multidisciplinary team and includes presence, listening and compassion [1, 13, 25, 42, 43]. As professionals, it is important to pick up on clues from patients, as informal spiritual assessment can take place anytime—for example, discussions around the search for meaning and making sense or fear or what the future holds may indicate spiritual distress [1]. Patients can then be helped to address this distress through presence, empathy and understanding from professionals [1, 9, 13] as well as through helping the patient to think about things that renew inner peace and comfort or through measures such as relaxation, meditation, music or, for some, through religion [1].

As physicians, should we be addressing spiritual and existential issues with patients who present with symptoms of anxiety and depression, rather than simply prescribing anxiolytics and antidepressants? Could a clinical intervention increase a patient’s level of spiritual awareness and well-being, and thereby decrease their psychosocial distress [15]?


This study suggests that in this population of patients with advanced cancer, there is a significant negative correlation between spiritual well-being, particularly the existential aspect, and anxiety and depression. In other words, patients with high levels of existential/spiritual well-being are less likely to be anxious or depressed. Religious well-being did not have a significant association with anxiety and depression in this population. By helping patients to address existential issues, we may improve their spiritual well-being and, consequently, their psychological well-being.


We would like to thank Christopher Foy and Rosemary Greenwood for statistical assistance and Prof. Michael King, Rev. Peter Speck, Rev. Mark Cobb, Prof. Mari Lloyd-Williams and Dr. Bill Noble for other additional assistance.

Copyright information

© Springer-Verlag 2005