- Email: [email protected]
Distinguishing Between Spiritual Distress, General Distress, Spiritual Well-Being, and Spiritual Pain Among Cancer Patients During Oncology Treatment
Accepted Manuscript Distinguishing between spiritual distress, general distress, spiritual well-being, and spiritual pain among cancer patients during oncology treatment Michael Schultz, MCC, Tehilah Meged-Book, MD, Tanya Mashiach, MA, Gil Bar-Sela, MD PII:
S0885-3924(17)30187-2
DOI:
10.1016/j.jpainsymman.2017.03.018
Reference:
JPS 9421
To appear in:
Journal of Pain and Symptom Management
Received Date: 6 October 2016 Revised Date:
21 February 2017
Accepted Date: 22 March 2017
Please cite this article as: Schultz M, Meged-Book T, Mashiach T, Bar-Sela G, Distinguishing between spiritual distress, general distress, spiritual well-being, and spiritual pain among cancer patients during oncology treatment, Journal of Pain and Symptom Management (2017), doi: 10.1016/ j.jpainsymman.2017.03.018. 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.
Spiritual distress compared to other measures: 1
ACCEPTED MANUSCRIPT Original Article 16-00636R1
Distinguishing between spiritual distress, general distress, spiritual well-being, and
RI PT
spiritual pain among cancer patients during oncology treatment
Michael Schultz MCC1, Tehilah Meged-Book MD2,
Division of Oncology, 3Statistical Department, Rambam Health Care Campus 2
Rappaport Faculty of Medicine, Technion-Israel Institute of Technology
M AN U
1
SC
Tanya Mashiach MA3, Gil Bar-Sela MD1,2
Haifa, Israel
TE D
Running title: Spiritual distress compared to other measures
Address for correspondence: Prof. Gil Bar-Sela
EP
Director, Integrated Oncology and Palliative Care Unit
AC C
Rambam Health Care Campus
POB 9602, Haifa 31096, Israel Tel: 972-4-7773128; Fax: 972-4-7773168 Email: [email protected] Accepted for publication: March 22, 2017. Number of tables: 4; figures: 1; references: 31 Word count: 3752
Spiritual distress compared to other measures: 2
ACCEPTED MANUSCRIPT
Abstract Background: Spiritual distress is present in ~25% of oncology patients. We examined the extent to which this measure is identical to a variety of other measures, such as spiritual well-
RI PT
being, spiritual injury, spiritual pain, and general distress. Methods: Structured interview of oncology outpatients over 12 months, approached nonselectively. The presence or absence of spiritual distress was compared against spiritual pain
SC
and two spiritual well-being tools: FACIT-Sp-12 and the Spiritual Injury Scale (SIS). We also examined whether a general distress Visual Analogue Scale sufficed to identify spiritual
M AN U
distress. Other questions concerned demographic and clinical data.
Results: Of 416 patients approached, 202 completed the interview, of whom 23% reported spiritual distress. All measures showed significant correlation (Receiver Operating Characteristic, area under the curve 0.79, SIS; 0.68, distress thermometer; 0.67, Facit-Sp-12),
TE D
yet none were identical with spiritual distress (sensitivity/specificity 64%/79%, SIS; 72%/76%, spiritual pain; 41%/76%, distress thermometer; 57%/72%, Facit-Sp-12). Of the FACIT-Sp-12 subscales, only Peace correlated with spiritual distress. A significant predictor
EP
of spiritual distress was patients’ self-evaluation of grave clinical condition (odds ratio 3.3, 95% confidence intervals 1.1-9.5). Multivariable analysis of individual measure items
AC C
suggests an alternative three-parameter model for spiritual distress: not feeling peaceful, feeling unable to accept that this is happening, and perceived severity of one's illness. Conclusions: The distress thermometer is not sufficient to identify spiritual distress. The Peace subscale of Facit-Sp-12 is a better match than the measure as a whole. The SIS is the best match for spiritual distress, although an imperfect one.
Spiritual distress compared to other measures: 3
ACCEPTED MANUSCRIPT Key words: spiritual distress; spiritual well-being; spiritual pain; oncology; distress thermometer; spiritual screening Introduction Spiritual distress and spiritual well-being can have a significant impact on key health
and wish to die, and improves quality of life (QoL)(2-6).
RI PT
measures. Among advanced cancer patients, spiritual well-being reduces despair, depression,
Spiritual distress has a nursing diagnosis (NANDA International) of "Impaired ability to
SC
experience and integrate meaning and purpose in life through the individual’s connectedness with self, others, art, music, literature, nature, or a power greater than oneself" (7). This
M AN U
definition corresponds well with the consensus definition of spirituality: "Spirituality is a dynamic and intrinsic aspect of humanity through which persons seek ultimate meaning, purpose, and transcendence, and experience relationship to self, family, others, community, society, nature, and the significant or sacred. Spirituality is expressed through beliefs, values,
TE D
traditions, and practices" (8).
Spiritual well-being, on the other hand, is not formally defined. Although studies have found a strong inverse correlation between a diagnosis of spiritual distress and scores on varying
EP
measures of spiritual well-being (9-12), since lack of spiritual well-being is not a diagnosis we cannot say that spiritual distress and spiritual well-being are precise inverses of each other.
AC C
Various tools have been developed to measure spiritual well-being, examining differing elements of spirituality. Ideally, measures of spiritual well-being should examine a range of elements relating to intrapersonal, interpersonal, and transpersonal spirituality (13). It is worth examining each measure to see the extent to which it functions as a precise inverse correlate of spiritual distress. Facit-Sp-12, one of the most commonly used measures of spiritual well-being, has three subscales: Peace, Meaning, and Faith, focusing largely on the intrapersonal and somewhat on
Spiritual distress compared to other measures: 4
ACCEPTED MANUSCRIPT the transpersonal (14), and has shown validity in cross-cultural settings (15,16), including the Middle East (17). The Spiritual Injury Scale (SIS) has not been widely utilized but benefits from directly examining potential expressions of spiritual distress rather than indicating spiritual distress
RI PT
through low well-being scores. These stand-alone items relate to guilt, anger or resentment, sadness or mourning, lack of meaning, despair, unfairness, being troubled by lack of faith, and death fixation. The SIS correlates with depression and PTSD (18).
SC
Another measure which may or may not be identical to spiritual distress is spiritual pain.
Defined as "a pain deep in your soul (being) that is not physical," spiritual pain is common
M AN U
among advanced cancer patients and correlates with depression, anxiety, and lower QoL (1921). Mako's original study found three major components to spiritual pain, broken down into intrapersonal (despair and isolation), interpersonal (isolation), and transpersonal (despair and anxiety)(19).
TE D
All three measures share significant content elements with the diagnosis of spiritual distress, thus providing the conceptual basis that prompted our in-depth comparison. The key elements of the diagnosis of spiritual distress are a loss of meaning and a breakdown in relationships to
EP
oneself, others, and the world/the sacred. Despair and isolation dovetail with a loss of meaning and a breakdown of relationships. Faith ties in with the sacred. Peace speaks to one's
AC C
relationship with self. Unfairness and resentment reflect a break in one's relationship with the world or the sacred.
Finally, it is worth examining the extent to which spiritual distress is the same as general distress, both to understand the nature of spiritual distress and as an aid in screening. Various methods of screening for spiritual distress have been proposed (22). Several recent studies have found near-exact correspondence between the diagnosis of spiritual distress made by a trained clinician and the patient's own self-diagnosis (10-12), suggesting that
Spiritual distress compared to other measures: 5
ACCEPTED MANUSCRIPT screening for spiritual distress can be accomplished simply by asking the patient if he/she is experiencing spiritual distress and providing its definition. This became our "gold standard" for diagnosing spiritual distress. This study aimed to develop the precision of our terminology in studies of spiritual distress,
RI PT
by examining the extent to which a diagnosis of spiritual distress is identical with four
measures: Facit-Sp-12, SIS, spiritual pain, and the general distress thermometer. These four measures describe the state of a patient's spiritual well-being from substantially different
SC
directions, representing a wide range of approaches to identifying or describing spiritual
M AN U
distress.
Methods Protocol
The study protocol was approved by our institutional Ethics Committee. Research staff
TE D
underwent a training session in the protocol and the scripted interview procedure. Enrollment ran between April 2014 and April 2015 in our Oncology Day Care Clinic. The benefits of this study population include diversity of oncologic condition and of demographics. Eligibility
EP
criteria included currently receiving oncologic treatment, ability to complete a 30 minute interview, and Hebrew comprehension.
AC C
Sample
Minimum sample size necessary was estimated using events per potential predictors calculation. Patients were approached non-selectively during times of research staff availability, for 95 periods of 1-2 hours each. All participants provided informed consent. With an average of 25 patients present and receiving treatment at any given time, and estimating that each patient was present on average during three periods of study administration, approximately 790 patients were present during periods of study
Spiritual distress compared to other measures: 6
ACCEPTED MANUSCRIPT administration, of whom 412 (52%) were approached. Of these, 202 participated; 115 were excluded for reasons of language comprehension; 37 were too ill to complete the interview; 17 were excluded for other reasons, including physical or cognitive impairment; 41 chose not to participate without specifying why.
RI PT
Measures Distress Thermometer
Participants responded to a general distress visual analogue scale from 0-10 ("distress
SC
thermometer"), in common use and with widespread validity (23-24), before being asked any
clinical indicator of distress. Facit-Sp-12
M AN U
questions relating specifically to spirituality or spiritual distress. Scores of 8 or higher are a
The Functional Assessment of Chronic Illness Therapy – Spiritual Well-Being 12 Item Scale (Facit-Sp) has been validated cross-culturally(16) and has a validated Hebrew translation. The
TE D
12 items are divided into three subscales: Peace, Meaning, and Faith (25). The 5-option Likert-type scale yields potential scores of 0-48. There is no established clinical cut-off score for poor spiritual wellbeing.
EP
SIS
The SIS, developed by Chaplain Gary Berg (18), contains 8 stand-alone indicators of spiritual
AC C
injury as described above. Of the four response options, responses of "often" or "very often" are considered by the tool's developer to be clinical indicators of spiritual injury. Positive responses to at least one item indicate a diagnosis of spiritual injury. Both the Facit-Sp and the SIS were used with the permission of their owners. New Israeli Items Three new items were hypothesized to be culturally sensitive, strong expressions of spiritual distress in Israel: being unable to accept what is happening, feeling like one is on his/her own
Spiritual distress compared to other measures: 7
ACCEPTED MANUSCRIPT in coping, feeling cursed. These items have not been validated and were included to expand our analysis beyond the American-developed tools included here. Spiritual Pain Patients were asked whether or not they were experiencing spiritual pain, defined as "a pain
RI PT
deep in your soul (being) that is not physical." We limited ourselves to a yes/no screening question, rather than the full measure (19), for purposes of comparison to our primary, dichotomous spiritual distress item.
SC
Spiritual Distress
As discussed above, it is possible to rely on a patient’s self-diagnosis regarding the presence
M AN U
or absence of spiritual distress(10-12), since self-diagnosis accords almost perfectly with professional diagnosis. Following the protocol of these studies, patients were asked whether or not they were experiencing spiritual distress and the definition of spiritual distress was read to them.
TE D
Medical Information
Medical information was both self-reported and abstracted from the medical file. Selfreported items included time since diagnosis, severity of illness, and purpose of current
EP
oncologic treatment. Abstracted medical data included diagnosis, staging, type of treatment, and purpose of treatment.
AC C
Statistical Analysis
We used receiver operator characteristics (ROC) analysis to assess the predictive ability of the various measures and to determine the optimal cut-off for quantitative parameters of various measures. The area under the ROC curve was used as a measure of multivariable model discrimination. Two-tailed p values of 0.05 or less were considered as statistically significant.
Spiritual distress compared to other measures: 8
ACCEPTED MANUSCRIPT We used bivariate logistic regression to calculate the odds ratios (OR) with 95% confidence intervals (95% CI), and p values for determining the associations of the various measures with spiritual distress. All variables from among patient characteristics and items within the larger measures which
RI PT
were found to be significantly associated with spiritual distress were selected as candidates for multivariable analysis. Multivariable forward stepwise logistic regression analysis was then performed.
SC
Statistical analyses were performed with SPSS (Statistics Products Solutions Services) 21.0
M AN U
software for Windows.
Results
Sample Characteristics and Spiritual Distress
Characteristics of the 202 study participants are summarized in Table 1. Most of the
TE D
participants were Jewish (82%). Participants were 60% female, 60% over age 60, and 60% had metastatic disease (stage IV). Forty-seven (23%) reported spiritual distress. Relationship Between Spiritual Distress and Four Measures
EP
One key means of seeing the relationship between each measure and spiritual distress is the area under the ROC curve, where an area of 0.5 indicates randomality and 1.0 indicates
AC C
identity. As seen in Table 2, Facit-Sp as a whole is significantly related to spiritual distress. However, when examining the Facit-Sp subscales independently, the Peace subscale shows high predictive ability without being identical to spiritual distress (p<0.001) and the Meaning subscale is statistically significant (p=0.008), whereas the Faith subscale shows no significant predictive ability of spiritual distress (p=0.42).
Spiritual distress compared to other measures: 9
ACCEPTED MANUSCRIPT To enable ROC analysis, we treated the SIS as a continuous variable, from 0-8 items showing the presence of spiritual injury. As seen in Figure 1, SIS was the measure most closely related to spiritual distress (p<0.001). The strength of the relationship between the general distress thermometer and spiritual
RI PT
distress is similar to that of the whole Facit-Sp. The distress thermometer identifies some elements of distress common to the spiritual distress measure, yet they are clearly not
identical. SIS, discussed below, was excluded from ROC analysis because it is a dichotomous
Spiritual Distress and the General Distress Thermometer
SC
measure. Figure 1 shows the comparison between the various measures.
M AN U
Fifty-six of 202 (28%) patients were experiencing general distress, as measured by the distress thermometer (score of 8+), 34% of whom were experiencing spiritual distress. This was a statistically significant predictor of spiritual distress (OR 2.2, 95% CI 1.1-4.5, p<0.03). However, although the absence of distress as measured by the thermometer is a fairly good
TE D
predictor of there not being spiritual distress (specificity=76%), the distress thermometer failed to identify spiritual distress in a majority of individuals. As a screening tool, it would have identified only 19/46 instances of spiritual distress (sensitivity=41%).
EP
Spiritual Distress and its Correlation with Patient Medical Characteristics In a bivariate analysis comparing spiritual distress and patient medical characteristics, only
AC C
two items significantly correlated with the presence of spiritual distress. These were: patients' perception of their illness' severity as being "quite serious" (OR 3.3, 95% CI 1.1-9.5, p<0.03) and patients' perception of their illness' severity as being worse than it actually was (patient report of "quite serious" together with localized-only documented disease (OR 6.3, 95% CI 1.1-35.7, p<0.04). No other medical or demographic item showed significant correlation.
Spiritual distress compared to other measures: 10
ACCEPTED MANUSCRIPT
Spiritual Distress and Facit-Sp-12 Because Facit-Sp-12 does not have clinical cutoff scores, we used ROC analysis to determine what they would be for our population. Using these scores, we found that 35% of patients had
RI PT
low scores for spiritual well-being, of whom 38.6% reported spiritual distress. However, the tool is still far from identical with spiritual distress, with sensitivity=57%, meaning that low scores on spiritual well-being using the Facit-Sp failed to identify nearly half of those
SC
experiencing spiritual distress. In the bivariate analysis, Facit-Sp as a whole and the Peace subscale each strongly correlated with spiritual distress, while the Faith subscale continued
M AN U
not to be significant. The meaning subscale, however, which in the ROC analysis had been significantly but more weakly associated with spiritual distress, showed no significant correlation in the bivariate analysis. Table 3 summarizes the bivariate significance, sensitivity, and specificity of the various measures examined.
TE D
Spiritual Distress and Spiritual Injury Scale (SIS)
Because each individual item of the SIS is a potential indicator of spiritual distress, we first examined them independently in the bivariate analysis and found that 6/8 were significantly
EP
correlated with spiritual distress (items regarding resentment, sadness/grief, lack of meaning, despair, unfairness, and frequent thoughts about death).
AC C
Taken as a whole, the SIS identifies spiritual distress if participants responded "often" or "very often" to one or more items, following the scoring system of the scale's creator. Of 202 patients, 110 (54.5%) showed spiritual injury, 39 (35.5%) of whom also expressed spiritual distress. We tested to see whether requiring two or more items showing injury would be a better measure of spiritual distress than just a single item and, indeed it was, yielding a very robust correlation with higher odds ratio (9.5) and higher combined sensitivity (64%) +
Spiritual distress compared to other measures: 11
ACCEPTED MANUSCRIPT specificity (79%). The SIS as a whole, using the preferable scoring system, shows both greater sensitivity and better specificity than Facit-Sp (Table 3). Spiritual Pain Spiritual pain, not surprisingly, shows very strong bivariate correlation with spiritual distress,
RI PT
but it is noteworthy that the two are not identical. Seventy-one (35%) patients reported
spiritual pain. Of those, 34 (47.9%) also reported spiritual distress. The balance of sensitivity (74%) and specificity (76%) in comparing spiritual pain with spiritual distress was greater
SC
than that of any of the other measures examined. Its OR, while higher than Facit-Sp or the distress thermometer, was still somewhat lower than that of SIS.
M AN U
New Israeli Items
All three of these items ("not able to accept that this is happening to me", p<0.001, OR 6.2, 95% CI 2.7-14.1; "on my own in dealing with this, "p=0.004, OR 4.2, 95% CI 1.6-10.9; "cursed" p<0.001, OR 6.1, 95% CI 2.4-15.4) were significant in the bivariate analysis.
TE D
Because these three items do not form a coherent tool, they are not included in Tables 2 and 3 where we present results from whole measures, but they were included in the multivariable analysis below. The item showing the best balance of sensitivity and specificity was being
EP
unable to accept what is happening (sensitivity 43%, specificity 86%). Multivariable Analysis
AC C
The analysis included patient characteristics (only one of which, perceived severity of illness, was significant in the bivariate analysis) and those items that were significant in the bivariate analysis from the SIS, the proposed Israeli items, and individual items from the Facit-Sp. We used the items individually in this analysis rather than as subscales to identify key elements that could form the basis of future study or of a new tool to be validated. Three items remained significant in this analysis: not feeling peaceful, not being able to accept that this is happening, and perceiving one's illness as being more serious (Table 4). This model is
Spiritual distress compared to other measures: 12
ACCEPTED MANUSCRIPT comparable to the best measures examined in terms of correlation with spiritual distress (ROC area under the curve 0.76, p<0.001, 95% CI 0.68-0.84).
Discussion
RI PT
Although it is likely that many of these results hold true universally, we acknowledge that there is significant cross-cultural variation in examining spirituality(16), which may limit the applicability of these findings to other cultures.
SC
One key question is whether or not spiritual distress and spiritual well-being are properly inverses of each other. This is complicated by the fact that spiritual distress is a diagnosis,
M AN U
while spiritual well-being is measured on a scale without set clinical cut-off points, and a variety of measures are used for spiritual well-being. Chaves compared three measures of spiritual well-being with diagnosed spiritual distress, and found that one did not match at all while the other two, although highly associated and showing very good sensitivity and
TE D
specificity, were not quite identical to spiritual distress (11). However, another study, using yet another measure for spiritual well-being, found a near-identical match between diagnosed
measure (12).
EP
spiritual distress and spiritual well-being, once the proper cutoff was established for that
We included Facit-Sp-12 as one measure of spiritual well-being because of its widespread use
AC C
in studies examining the medical impact of low spiritual well-being (2,3,26). Although the Facit-Sp was significantly associated with spiritual distress, it was much further from being identical to it than the measures mentioned above in other studies. Even after optimizing the cutoff score to ensure maximal similarity (following the standard approach of identifying the highest value for sensitivity plus specificity (27)), 43% of those in spiritual distress did not have low spiritual well-being as measured by Facit-Sp (i.e., 1–sensitivity). It would seem that there are other elements to a patient’s spiritual distress that this measure did not ask about.
Spiritual distress compared to other measures: 13
ACCEPTED MANUSCRIPT However, it is also not the case that all patients showing low spiritual well-being were found to be in spiritual distress. In fact, a majority of those with low spiritual well-being as measured by Facit-Sp were not in spiritual distress. Thus, both as an expression of spiritual
demonstrably far from being identical with spiritual distress.
RI PT
distress and as a predictor of spiritual distress, Facit-Sp is statistically significant yet is
Spiritual pain, which by its name sounds like it would be similar to spiritual distress, was indeed found to be more nearly identical with spiritual distress. In the bivariate analysis,
SC
spiritual pain was much more strongly associated with spiritual distress than either the FacitSp or the distress thermometer. We found similar levels of spiritual pain to those in other
M AN U
studies (20,21). However, here, too, the terms were far from identical. One-quarter of those in spiritual distress are not in spiritual pain (i.e., 1–sensitivity), and over half of those in spiritual pain did not report spiritual distress.
The Spiritual Injury Scale, given that it directly asks about items that would seem to be strong
TE D
indicators of distress (as indicated by the name "injury"), was expected to show high predictive ability. That is to say, a patient showing spiritual injury certainly ought to show spiritual distress. This is measured by specificity, which can be logically interpreted as the
EP
extent to which, if the secondary measure is present, then the primary measure will also be present. This hypothesis was fairly well supported by our findings, especially when using the
AC C
alternate scoring system suggested by our statistical analysis. In Berg's scoring system, as long as even one item shows injury, the patient is considered to be experiencing spiritual injury (18). Using this system, spiritual injury was found to be a very common expression of spiritual distress, but a relatively poor predictor of spiritual distress because of the cumulative effect of the "false positives" contained within each item. However, if we raise the bar for diagnosing spiritual injury to two items showing injury, this tool remains sufficiently good at
Spiritual distress compared to other measures: 14
ACCEPTED MANUSCRIPT capturing the expression of spiritual distress (64% sensitivity) while becoming a very good predictor of spiritual distress (79% specificity). Finally, we examined the similarity of spiritual distress and general distress. As a screening tool, the distress thermometer (DT) is fairly accurate (i.e., high scores on the DT correlate
RI PT
with spiritual distress most of the time, similar to results reported elsewhere (28)), but it would still fail to identify those in spiritual distress over half the time (i.e., only 41% of
people in spiritual distress scored high on the DT). It could be used in conjunction with other
SC
approaches but seems not to suffice by itself. Where it is already in routine use, a high score on the DT could be used as a strong indicator for referring to spiritual care.
M AN U
Given that neither spiritual well-being as measured by Facit-Sp, spiritual pain, spiritual injury, nor general distress were sufficiently identical to spiritual distress, we performed a multivariable analysis to see if we could suggest the beginnings of an alternate measure that would be a better stand-in for spiritual distress, which could be further studied and validated.
TE D
The model generated by the multivariable analysis included three items: peace ("I feel peaceful"), ability to accept that this is happening, and patient perceived severity of illness, each of which we will now discuss.
EP
Inner peace or serenity is increasingly proving significant in examining the impact of illness on patients' well-being. It is particularly noteworthy that, in the present study, of the three
AC C
subscales of the Facit-Sp – Peace, Meaning, and Faith – only Peace was significant in all tests. In several studies showing medical benefit of spiritual well-being for cancer patients and the elderly, only the Peace and Meaning subscales were significant, but not the Faith subscale (2,3,15). In a Middle Eastern cancer study of health related QoL, Peace was the strongest factor followed by Meaning, while Faith was not significant. A number of studies have begun focusing specifically on Peace, using variants of the single-item Steinhauser measure ("I feel at peace") to demonstrate significant associations with improved medical condition (1,6) and
Spiritual distress compared to other measures: 15
ACCEPTED MANUSCRIPT to measure the validity of assessment tools (29). Other studies show a correlation between peace and spiritual well-being (30). The present study buttresses this approach of using peacefulness as the single most medically important measure of spiritual well-being. The second item persisting in the multivariable model, an inability to accept that this is
RI PT
happening, was a new item hypothesized to be significant in Israel. In Middle Eastern,
Islamic-influenced societies, a commonly heard theme in the face of illness is the belief that whatever happens is God's will and that man's job is to accept God's will (31). Thus, in this
SC
religious cultural context, an inability to accept what is happening would seem to be an
expression of spiritual distress, of not finding comfort within commonly shared spiritual
M AN U
beliefs.
The final item in the multivariable model was patients' perception of the severity of their illness. The data from the medical file, by contrast, did not significantly correlate with spiritual distress, in line with previous findings (12). This finding also supports the connection
TE D
between spiritual well-being and patients' subjective feelings regarding their physical wellbeing (2,17). Patients who see their situation as being worse than it actually is are, unsurprisingly, more likely to despair of the possibility of a continued life of purpose and
EP
connectedness, and vice versa.
AC C
Limitations
Given that cultural influence on spiritual distress' expression is central to this study, ideally we would have liked to study the major subgroups in Israel separately, including Arabs (21% of the national population) and Russian immigrants (15%). Unfortunately, the sample size was not large enough to enable us to draw significant conclusions regarding the expression of spiritual distress within these subgroups examined separately.
Spiritual distress compared to other measures: 16
ACCEPTED MANUSCRIPT
Conclusion Spiritual distress is a significant source of suffering for many patients with cancer, especially those perceiving their situation as grave. It is closely related to but distinct from general
RI PT
distress, spiritual pain, and two common measures of spiritual well-being. Feeling at peace is a key element of not being in spiritual distress and continues to show promise, in this as in
SC
other recent studies, as a simple research measure.
Disclosures and Acknowledgements
M AN U
Thank you to the UJA/Federation of New York for their ongoing support of the spiritual care service and research efforts at Rambam Health Care Campus. Thank you to Professor George
AC C
EP
TE D
Fitchett of Rush University for his assistance in thinking through the study design.
Spiritual distress compared to other measures: 17
ACCEPTED MANUSCRIPT
References 1. Park C, Aldwin C, Choun S, et al. Spiritual peace predicts 5-year mortality in congestive heart failure patients. Health Psychol 2016;35:203-210. doi: 10.1037/hea0000271.
RI PT
2. McClain CS, Rosenfeld B, Breitbart W. Effect of spiritual well-being on end-of-life despair in terminally-ill cancer patients. Lancet 2003;361:1603-1607.
3. Nelson CJ, Rosenfeld B, Breitbart W, Galietta M. Spirituality, religion, and depression in
SC
the terminally ill. Psychosomatics 2002;43:213-220.
4. Brady MJ, Peterman AH, Fitchett G, Mo M, Cella D. A case for including spirituality in
M AN U
quality of life measurement in oncology. Psychooncology 1999;8:417-428. 5. McCoubrie RC, Davies AN. Is there a correlation between spirituality and anxiety and depression in patients with advanced cancer? Support Care Cancer 2006;14:379-385. 6. Rabow MW, Knish SJ. Spiritual well-being among outpatients with cancer receiving
TE D
concurrent oncologic and palliative care. Support Care Cancer 2015;23:919-923. 7. Herdman T. (Ed.). North American Nursing Diagnosis Association International—Nursing Diagnoses: Definitions and classification 2009–2011. Oxford: Wiley-Blackwell, 2009.
EP
8. Puchalski C, Vitillo R, Hull S, Reller N. Improving the spiritual dimension of whole person care: reaching national and international consensus. J Palliat Med 2014;17:642-
AC C
656. doi: 10.1089/jpm.2014.9427. 9. Simão TP, Lopes Chaves Ede C, Campos de Carvalho E, et al. Cultural adaptation and analysis of the psychometric properties of the Brazilian version of the Spiritual Distress Scale. J Clin Nurs 2016;25:231-239. doi: 10.1111/jocn.13060. 10.
Caldeira S, Timmins F, de Carvalho EC, Vieira M. Clinical validation of the nursing diagnosis spiritual distress in cancer patients undergoing chemotherapy. Int J Nurs Knowl. 2015 Jun 11. doi: 10.1111/2047-3095.12105.
Spiritual distress compared to other measures: 18
ACCEPTED MANUSCRIPT 11.
Chaves Ede C, Carvalho EC, Beijo LA, Goyatá SLT, Pillon SC. Efficacy of different instruments for the identification of the nursing diagnosis spiritual distress. Rev Lat Am Enfermagem 2011;19:902-910.
12.
Caldeira S, Carvalho EC, Vieira M. Between spiritual wellbeing and spiritual distress:
RI PT
possible related factors in elderly patients with cancer. Rev Lat Am Enfermagem 2014;22:28-34. 13.
Benito E, Oliver A, Galiana L, et al. Development and validation of a new tool for the
SC
assessment and spiritual care of palliative care patients. J Pain Symptom Manage 2014;47:1008-1018.e1. doi: 10.1016/j.jpainsymman.2013.06.018.
Brady MJ, Peterman AH, Fitchett G, Mo M, Cella D. A case for including spirituality in
M AN U
14.
quality of life measurement in oncology. Psychooncology 1999;8:417-428. 15.
Agli O, Bailly N, Ferrand C. Validation of the Functional Assessment of Chronic Illness Therapy-Spiritual Well-being (FACIT-Sp12) on French Old People. J Relig Health
16.
TE D
2016 Mar 14. [Epub ahead of print]
Selman L, Harding R, Gysels M, Speck P, Higginson IJ. The measurement of spirituality in palliative care and the content of tools validated cross-culturally: a
EP
systematic review. J Pain Symptom Manage 2011;41:728-753. doi: 10.1016/j.jpainsymman.2010.06.023. Lazenby M, Khatib J, Al-Khair F, Neamat M. Psychometric properties of the Functional
AC C
17.
Assessment of Chronic Illness Therapy--Spiritual Well-being (FACIT-Sp) in an Arabicspeaking, predominantly Muslim population. Psychooncology 2013;22:220-227. doi: 10.1002/pon.2062. 18.
Berg G. The relationship between spiritual distress, PTSD and depression In Vietnam combat veterans. J Pastoral Care Counsel 2011;6:1-11.
Spiritual distress compared to other measures: 19
ACCEPTED MANUSCRIPT 19.
Mako C, Galek K, Poppito SR. Spiritual pain among patients with advanced cancer in palliative care. J Palliat Med 2006;9:1106-1113.
20.
Delgado-Guay MO, Hui D, Parsons HA, et al. Spirituality, religiosity, and spiritual pain in advanced cancer patients. J Pain Symptom Manage 2011;41:986-994. doi:
21.
RI PT
10.1016/j.jpainsymman.2010.09.017.
Delgado-Guay MO, Chisholm G, Williams J, et al. Frequency, intensity, and correlates of spiritual pain in advanced cancer patients assessed in a supportive/palliative care
22.
SC
clinic. Palliat Support Care 2016;14:341-348. doi: 10.1017/S147895151500108X.
Blanchard JH, Dunlap DA, Fitchett G. Screening for spiritual distress in the oncology
M AN U
inpatient: a quality improvement pilot project between nurses and chaplains. J Nurs Manag 2012;20:1076-1084. doi: 10.1111/jonm.12035. 23.
Vitek L, Rosenzweig MQ, Stollings S. Distress in patients with cancer: definition, assessment, and suggested interventions. Clin J Oncol Nurs 2007;11:413-418. Snowden A, White CA, Christie Z, et al. The clinical utility of the distress thermometer:
TE D
24.
a review. Br J Nurs 2011;20:220-227. 25.
Murphy PE, Canada AL, Fitchett G, et al. An examination of the 3-factor model and
EP
structural invariance across racial/ethnic groups for the FACIT-Sp: a report from the American Cancer Society's Study of Cancer Survivors-II (SCS-II). Psychooncology
26.
AC C
2010;19:264-272. doi: 10.1002/pon.1559. Wilson CS, Forchheimer M, Heinemann AW, Warren AM, McCullumsmith C. Assessment of the relationship of spiritual well-being to depression and quality of life for persons with spinal cord injury. Disabil Rehabil 2016;25:1-6. 27.
Hegel MT, Collins ED, Kearing S, et al. Sensitivity and specificity of the Distress Thermometer for depression in newly diagnosed breast cancer patients. Psychooncology 2008;17:556-560.
Spiritual distress compared to other measures: 20
ACCEPTED MANUSCRIPT 28.
Agarwal J, Powers K, Pappas L, et al. Correlates of elevated distress thermometer scores in breast cancer patients. Support Care Cancer 2013;21:2125-2136. doi: 10.1007/s00520-013-1773-z.
29.
Monod S, Martin E, Spencer B, Rochat E, Büla C. Validation of the Spiritual Distress
RI PT
Assessment Tool in older hospitalized patients. BMC Geriatr 2012;12:13. doi: 10.1186/1471-2318-12-13. 30.
Selman L, Speck P, Gysels M, et al. ‘Peace’ and ‘life worthwhile’ as measures of
Quality of Life Outcomes 2013;11:94.
M AN U
Schultz M, Baddarni K, Bar-Sela G. Reflections on palliative care from the Jewish and Islamic tradition. Evid Based Complement Alternat Med 2012;2012:693092. doi:
EP
TE D
10.1155/2012/693092.
AC C
31.
SC
spiritual well-being in African palliative care: a mixed-methods study. Health and
Spiritual distress compared to other measures: 21
ACCEPTED MANUSCRIPT
Legend to Figure ROC curves, various measures and spiritual distress
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1:
Spiritual distress compared to other measures: 22
ACCEPTED MANUSCRIPT
Table 1. Sample characteristics of study participants (n=202)
No.
%
Female
121
60
Male
81
40
<50
28
14
50-59
53
26
60-69
73
36
70+
48
24
Jewish
165
Muslim
15
Christian
14
Druze
7
Cancer stage
TE D
Religion
82 7 7 3
12
6
26
13
41
21
118
60
Hebrew
114
56
Arabic
34
17
Russian
21
10
English
8
4
Other
24
12
2 3 4
AC C
Native tongue
EP
1
SC
M AN U
Age (years)
RI PT
Gender
Spiritual distress compared to other measures: 23
ACCEPTED MANUSCRIPT
P value
95% CI
0.67
0.001
0.58-0.76
Peace subscale
0.73
<0.001
0.65-0.82
Meaning subscale
0.63
0.008
0.54-0.72
Faith subscale
0.54
0.42
0.44-0.64
SIS
0.79
<0.001
Distress Thermometer
0.68
<0.001
AC C
EP
0.71-0.87
0.60-0.77
M AN U
TE D
Facit-Sp-12
RI PT
Area
SC
Table 2. ROC area under the curve – Predictive ability vis-a-vis spiritual distress
Spiritual distress compared to other measures: 24
ACCEPTED MANUSCRIPT
Table 3. Bivariate analysis of correlation with spiritual distress, sensitivity, and specificity
OR
95% CI
Facit-Sp-12
<0.001
3.5
1.8-6.9
<0.001
6.4
2.7-14.9
0.02
2.2
1.1-4.5
<0.001
8.3
4.0-17.3
Peace Subscale General Distress
Spiritual Pain Spiritual Injury Scale
<0.001
0 vs 2+ items
<0.001
EP AC C
72%
45%
85%
41%
76%
72%
76%
5.8
2.5-13.1
83%
54%
9.5
4.0-23.0
64%
79%
TE D
0 vs 1+ items
57%
M AN U
Thermometer
Sensitivity Specificity
RI PT
P value
SC
Tool/Item
Spiritual distress compared to other measures: 25
ACCEPTED MANUSCRIPT
Table 4. Multivariable model for predicting spiritual distress
P value
Adjusted OR
95% CI
I feel peaceful.
0.002
1.6
1.2-2.2
I am not able to accept that this is
0.006
3.1
1.4-7.1
0.026
2.3
1.1-4.9
happening to me.
AC C
EP
TE D
M AN U
SC
How serious do you think your illness is?
RI PT
Item
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0885-3924(17)30187-2
DOI:
10.1016/j.jpainsymman.2017.03.018
Reference:
JPS 9421
To appear in:
Journal of Pain and Symptom Management
Received Date: 6 October 2016 Revised Date:
21 February 2017
Accepted Date: 22 March 2017
Please cite this article as: Schultz M, Meged-Book T, Mashiach T, Bar-Sela G, Distinguishing between spiritual distress, general distress, spiritual well-being, and spiritual pain among cancer patients during oncology treatment, Journal of Pain and Symptom Management (2017), doi: 10.1016/ j.jpainsymman.2017.03.018. 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.
Spiritual distress compared to other measures: 1
ACCEPTED MANUSCRIPT Original Article 16-00636R1
Distinguishing between spiritual distress, general distress, spiritual well-being, and
RI PT
spiritual pain among cancer patients during oncology treatment
Michael Schultz MCC1, Tehilah Meged-Book MD2,
Division of Oncology, 3Statistical Department, Rambam Health Care Campus 2
Rappaport Faculty of Medicine, Technion-Israel Institute of Technology
M AN U
1
SC
Tanya Mashiach MA3, Gil Bar-Sela MD1,2
Haifa, Israel
TE D
Running title: Spiritual distress compared to other measures
Address for correspondence: Prof. Gil Bar-Sela
EP
Director, Integrated Oncology and Palliative Care Unit
AC C
Rambam Health Care Campus
POB 9602, Haifa 31096, Israel Tel: 972-4-7773128; Fax: 972-4-7773168 Email: [email protected] Accepted for publication: March 22, 2017. Number of tables: 4; figures: 1; references: 31 Word count: 3752
Spiritual distress compared to other measures: 2
ACCEPTED MANUSCRIPT
Abstract Background: Spiritual distress is present in ~25% of oncology patients. We examined the extent to which this measure is identical to a variety of other measures, such as spiritual well-
RI PT
being, spiritual injury, spiritual pain, and general distress. Methods: Structured interview of oncology outpatients over 12 months, approached nonselectively. The presence or absence of spiritual distress was compared against spiritual pain
SC
and two spiritual well-being tools: FACIT-Sp-12 and the Spiritual Injury Scale (SIS). We also examined whether a general distress Visual Analogue Scale sufficed to identify spiritual
M AN U
distress. Other questions concerned demographic and clinical data.
Results: Of 416 patients approached, 202 completed the interview, of whom 23% reported spiritual distress. All measures showed significant correlation (Receiver Operating Characteristic, area under the curve 0.79, SIS; 0.68, distress thermometer; 0.67, Facit-Sp-12),
TE D
yet none were identical with spiritual distress (sensitivity/specificity 64%/79%, SIS; 72%/76%, spiritual pain; 41%/76%, distress thermometer; 57%/72%, Facit-Sp-12). Of the FACIT-Sp-12 subscales, only Peace correlated with spiritual distress. A significant predictor
EP
of spiritual distress was patients’ self-evaluation of grave clinical condition (odds ratio 3.3, 95% confidence intervals 1.1-9.5). Multivariable analysis of individual measure items
AC C
suggests an alternative three-parameter model for spiritual distress: not feeling peaceful, feeling unable to accept that this is happening, and perceived severity of one's illness. Conclusions: The distress thermometer is not sufficient to identify spiritual distress. The Peace subscale of Facit-Sp-12 is a better match than the measure as a whole. The SIS is the best match for spiritual distress, although an imperfect one.
Spiritual distress compared to other measures: 3
ACCEPTED MANUSCRIPT Key words: spiritual distress; spiritual well-being; spiritual pain; oncology; distress thermometer; spiritual screening Introduction Spiritual distress and spiritual well-being can have a significant impact on key health
and wish to die, and improves quality of life (QoL)(2-6).
RI PT
measures. Among advanced cancer patients, spiritual well-being reduces despair, depression,
Spiritual distress has a nursing diagnosis (NANDA International) of "Impaired ability to
SC
experience and integrate meaning and purpose in life through the individual’s connectedness with self, others, art, music, literature, nature, or a power greater than oneself" (7). This
M AN U
definition corresponds well with the consensus definition of spirituality: "Spirituality is a dynamic and intrinsic aspect of humanity through which persons seek ultimate meaning, purpose, and transcendence, and experience relationship to self, family, others, community, society, nature, and the significant or sacred. Spirituality is expressed through beliefs, values,
TE D
traditions, and practices" (8).
Spiritual well-being, on the other hand, is not formally defined. Although studies have found a strong inverse correlation between a diagnosis of spiritual distress and scores on varying
EP
measures of spiritual well-being (9-12), since lack of spiritual well-being is not a diagnosis we cannot say that spiritual distress and spiritual well-being are precise inverses of each other.
AC C
Various tools have been developed to measure spiritual well-being, examining differing elements of spirituality. Ideally, measures of spiritual well-being should examine a range of elements relating to intrapersonal, interpersonal, and transpersonal spirituality (13). It is worth examining each measure to see the extent to which it functions as a precise inverse correlate of spiritual distress. Facit-Sp-12, one of the most commonly used measures of spiritual well-being, has three subscales: Peace, Meaning, and Faith, focusing largely on the intrapersonal and somewhat on
Spiritual distress compared to other measures: 4
ACCEPTED MANUSCRIPT the transpersonal (14), and has shown validity in cross-cultural settings (15,16), including the Middle East (17). The Spiritual Injury Scale (SIS) has not been widely utilized but benefits from directly examining potential expressions of spiritual distress rather than indicating spiritual distress
RI PT
through low well-being scores. These stand-alone items relate to guilt, anger or resentment, sadness or mourning, lack of meaning, despair, unfairness, being troubled by lack of faith, and death fixation. The SIS correlates with depression and PTSD (18).
SC
Another measure which may or may not be identical to spiritual distress is spiritual pain.
Defined as "a pain deep in your soul (being) that is not physical," spiritual pain is common
M AN U
among advanced cancer patients and correlates with depression, anxiety, and lower QoL (1921). Mako's original study found three major components to spiritual pain, broken down into intrapersonal (despair and isolation), interpersonal (isolation), and transpersonal (despair and anxiety)(19).
TE D
All three measures share significant content elements with the diagnosis of spiritual distress, thus providing the conceptual basis that prompted our in-depth comparison. The key elements of the diagnosis of spiritual distress are a loss of meaning and a breakdown in relationships to
EP
oneself, others, and the world/the sacred. Despair and isolation dovetail with a loss of meaning and a breakdown of relationships. Faith ties in with the sacred. Peace speaks to one's
AC C
relationship with self. Unfairness and resentment reflect a break in one's relationship with the world or the sacred.
Finally, it is worth examining the extent to which spiritual distress is the same as general distress, both to understand the nature of spiritual distress and as an aid in screening. Various methods of screening for spiritual distress have been proposed (22). Several recent studies have found near-exact correspondence between the diagnosis of spiritual distress made by a trained clinician and the patient's own self-diagnosis (10-12), suggesting that
Spiritual distress compared to other measures: 5
ACCEPTED MANUSCRIPT screening for spiritual distress can be accomplished simply by asking the patient if he/she is experiencing spiritual distress and providing its definition. This became our "gold standard" for diagnosing spiritual distress. This study aimed to develop the precision of our terminology in studies of spiritual distress,
RI PT
by examining the extent to which a diagnosis of spiritual distress is identical with four
measures: Facit-Sp-12, SIS, spiritual pain, and the general distress thermometer. These four measures describe the state of a patient's spiritual well-being from substantially different
SC
directions, representing a wide range of approaches to identifying or describing spiritual
M AN U
distress.
Methods Protocol
The study protocol was approved by our institutional Ethics Committee. Research staff
TE D
underwent a training session in the protocol and the scripted interview procedure. Enrollment ran between April 2014 and April 2015 in our Oncology Day Care Clinic. The benefits of this study population include diversity of oncologic condition and of demographics. Eligibility
EP
criteria included currently receiving oncologic treatment, ability to complete a 30 minute interview, and Hebrew comprehension.
AC C
Sample
Minimum sample size necessary was estimated using events per potential predictors calculation. Patients were approached non-selectively during times of research staff availability, for 95 periods of 1-2 hours each. All participants provided informed consent. With an average of 25 patients present and receiving treatment at any given time, and estimating that each patient was present on average during three periods of study administration, approximately 790 patients were present during periods of study
Spiritual distress compared to other measures: 6
ACCEPTED MANUSCRIPT administration, of whom 412 (52%) were approached. Of these, 202 participated; 115 were excluded for reasons of language comprehension; 37 were too ill to complete the interview; 17 were excluded for other reasons, including physical or cognitive impairment; 41 chose not to participate without specifying why.
RI PT
Measures Distress Thermometer
Participants responded to a general distress visual analogue scale from 0-10 ("distress
SC
thermometer"), in common use and with widespread validity (23-24), before being asked any
clinical indicator of distress. Facit-Sp-12
M AN U
questions relating specifically to spirituality or spiritual distress. Scores of 8 or higher are a
The Functional Assessment of Chronic Illness Therapy – Spiritual Well-Being 12 Item Scale (Facit-Sp) has been validated cross-culturally(16) and has a validated Hebrew translation. The
TE D
12 items are divided into three subscales: Peace, Meaning, and Faith (25). The 5-option Likert-type scale yields potential scores of 0-48. There is no established clinical cut-off score for poor spiritual wellbeing.
EP
SIS
The SIS, developed by Chaplain Gary Berg (18), contains 8 stand-alone indicators of spiritual
AC C
injury as described above. Of the four response options, responses of "often" or "very often" are considered by the tool's developer to be clinical indicators of spiritual injury. Positive responses to at least one item indicate a diagnosis of spiritual injury. Both the Facit-Sp and the SIS were used with the permission of their owners. New Israeli Items Three new items were hypothesized to be culturally sensitive, strong expressions of spiritual distress in Israel: being unable to accept what is happening, feeling like one is on his/her own
Spiritual distress compared to other measures: 7
ACCEPTED MANUSCRIPT in coping, feeling cursed. These items have not been validated and were included to expand our analysis beyond the American-developed tools included here. Spiritual Pain Patients were asked whether or not they were experiencing spiritual pain, defined as "a pain
RI PT
deep in your soul (being) that is not physical." We limited ourselves to a yes/no screening question, rather than the full measure (19), for purposes of comparison to our primary, dichotomous spiritual distress item.
SC
Spiritual Distress
As discussed above, it is possible to rely on a patient’s self-diagnosis regarding the presence
M AN U
or absence of spiritual distress(10-12), since self-diagnosis accords almost perfectly with professional diagnosis. Following the protocol of these studies, patients were asked whether or not they were experiencing spiritual distress and the definition of spiritual distress was read to them.
TE D
Medical Information
Medical information was both self-reported and abstracted from the medical file. Selfreported items included time since diagnosis, severity of illness, and purpose of current
EP
oncologic treatment. Abstracted medical data included diagnosis, staging, type of treatment, and purpose of treatment.
AC C
Statistical Analysis
We used receiver operator characteristics (ROC) analysis to assess the predictive ability of the various measures and to determine the optimal cut-off for quantitative parameters of various measures. The area under the ROC curve was used as a measure of multivariable model discrimination. Two-tailed p values of 0.05 or less were considered as statistically significant.
Spiritual distress compared to other measures: 8
ACCEPTED MANUSCRIPT We used bivariate logistic regression to calculate the odds ratios (OR) with 95% confidence intervals (95% CI), and p values for determining the associations of the various measures with spiritual distress. All variables from among patient characteristics and items within the larger measures which
RI PT
were found to be significantly associated with spiritual distress were selected as candidates for multivariable analysis. Multivariable forward stepwise logistic regression analysis was then performed.
SC
Statistical analyses were performed with SPSS (Statistics Products Solutions Services) 21.0
M AN U
software for Windows.
Results
Sample Characteristics and Spiritual Distress
Characteristics of the 202 study participants are summarized in Table 1. Most of the
TE D
participants were Jewish (82%). Participants were 60% female, 60% over age 60, and 60% had metastatic disease (stage IV). Forty-seven (23%) reported spiritual distress. Relationship Between Spiritual Distress and Four Measures
EP
One key means of seeing the relationship between each measure and spiritual distress is the area under the ROC curve, where an area of 0.5 indicates randomality and 1.0 indicates
AC C
identity. As seen in Table 2, Facit-Sp as a whole is significantly related to spiritual distress. However, when examining the Facit-Sp subscales independently, the Peace subscale shows high predictive ability without being identical to spiritual distress (p<0.001) and the Meaning subscale is statistically significant (p=0.008), whereas the Faith subscale shows no significant predictive ability of spiritual distress (p=0.42).
Spiritual distress compared to other measures: 9
ACCEPTED MANUSCRIPT To enable ROC analysis, we treated the SIS as a continuous variable, from 0-8 items showing the presence of spiritual injury. As seen in Figure 1, SIS was the measure most closely related to spiritual distress (p<0.001). The strength of the relationship between the general distress thermometer and spiritual
RI PT
distress is similar to that of the whole Facit-Sp. The distress thermometer identifies some elements of distress common to the spiritual distress measure, yet they are clearly not
identical. SIS, discussed below, was excluded from ROC analysis because it is a dichotomous
Spiritual Distress and the General Distress Thermometer
SC
measure. Figure 1 shows the comparison between the various measures.
M AN U
Fifty-six of 202 (28%) patients were experiencing general distress, as measured by the distress thermometer (score of 8+), 34% of whom were experiencing spiritual distress. This was a statistically significant predictor of spiritual distress (OR 2.2, 95% CI 1.1-4.5, p<0.03). However, although the absence of distress as measured by the thermometer is a fairly good
TE D
predictor of there not being spiritual distress (specificity=76%), the distress thermometer failed to identify spiritual distress in a majority of individuals. As a screening tool, it would have identified only 19/46 instances of spiritual distress (sensitivity=41%).
EP
Spiritual Distress and its Correlation with Patient Medical Characteristics In a bivariate analysis comparing spiritual distress and patient medical characteristics, only
AC C
two items significantly correlated with the presence of spiritual distress. These were: patients' perception of their illness' severity as being "quite serious" (OR 3.3, 95% CI 1.1-9.5, p<0.03) and patients' perception of their illness' severity as being worse than it actually was (patient report of "quite serious" together with localized-only documented disease (OR 6.3, 95% CI 1.1-35.7, p<0.04). No other medical or demographic item showed significant correlation.
Spiritual distress compared to other measures: 10
ACCEPTED MANUSCRIPT
Spiritual Distress and Facit-Sp-12 Because Facit-Sp-12 does not have clinical cutoff scores, we used ROC analysis to determine what they would be for our population. Using these scores, we found that 35% of patients had
RI PT
low scores for spiritual well-being, of whom 38.6% reported spiritual distress. However, the tool is still far from identical with spiritual distress, with sensitivity=57%, meaning that low scores on spiritual well-being using the Facit-Sp failed to identify nearly half of those
SC
experiencing spiritual distress. In the bivariate analysis, Facit-Sp as a whole and the Peace subscale each strongly correlated with spiritual distress, while the Faith subscale continued
M AN U
not to be significant. The meaning subscale, however, which in the ROC analysis had been significantly but more weakly associated with spiritual distress, showed no significant correlation in the bivariate analysis. Table 3 summarizes the bivariate significance, sensitivity, and specificity of the various measures examined.
TE D
Spiritual Distress and Spiritual Injury Scale (SIS)
Because each individual item of the SIS is a potential indicator of spiritual distress, we first examined them independently in the bivariate analysis and found that 6/8 were significantly
EP
correlated with spiritual distress (items regarding resentment, sadness/grief, lack of meaning, despair, unfairness, and frequent thoughts about death).
AC C
Taken as a whole, the SIS identifies spiritual distress if participants responded "often" or "very often" to one or more items, following the scoring system of the scale's creator. Of 202 patients, 110 (54.5%) showed spiritual injury, 39 (35.5%) of whom also expressed spiritual distress. We tested to see whether requiring two or more items showing injury would be a better measure of spiritual distress than just a single item and, indeed it was, yielding a very robust correlation with higher odds ratio (9.5) and higher combined sensitivity (64%) +
Spiritual distress compared to other measures: 11
ACCEPTED MANUSCRIPT specificity (79%). The SIS as a whole, using the preferable scoring system, shows both greater sensitivity and better specificity than Facit-Sp (Table 3). Spiritual Pain Spiritual pain, not surprisingly, shows very strong bivariate correlation with spiritual distress,
RI PT
but it is noteworthy that the two are not identical. Seventy-one (35%) patients reported
spiritual pain. Of those, 34 (47.9%) also reported spiritual distress. The balance of sensitivity (74%) and specificity (76%) in comparing spiritual pain with spiritual distress was greater
SC
than that of any of the other measures examined. Its OR, while higher than Facit-Sp or the distress thermometer, was still somewhat lower than that of SIS.
M AN U
New Israeli Items
All three of these items ("not able to accept that this is happening to me", p<0.001, OR 6.2, 95% CI 2.7-14.1; "on my own in dealing with this, "p=0.004, OR 4.2, 95% CI 1.6-10.9; "cursed" p<0.001, OR 6.1, 95% CI 2.4-15.4) were significant in the bivariate analysis.
TE D
Because these three items do not form a coherent tool, they are not included in Tables 2 and 3 where we present results from whole measures, but they were included in the multivariable analysis below. The item showing the best balance of sensitivity and specificity was being
EP
unable to accept what is happening (sensitivity 43%, specificity 86%). Multivariable Analysis
AC C
The analysis included patient characteristics (only one of which, perceived severity of illness, was significant in the bivariate analysis) and those items that were significant in the bivariate analysis from the SIS, the proposed Israeli items, and individual items from the Facit-Sp. We used the items individually in this analysis rather than as subscales to identify key elements that could form the basis of future study or of a new tool to be validated. Three items remained significant in this analysis: not feeling peaceful, not being able to accept that this is happening, and perceiving one's illness as being more serious (Table 4). This model is
Spiritual distress compared to other measures: 12
ACCEPTED MANUSCRIPT comparable to the best measures examined in terms of correlation with spiritual distress (ROC area under the curve 0.76, p<0.001, 95% CI 0.68-0.84).
Discussion
RI PT
Although it is likely that many of these results hold true universally, we acknowledge that there is significant cross-cultural variation in examining spirituality(16), which may limit the applicability of these findings to other cultures.
SC
One key question is whether or not spiritual distress and spiritual well-being are properly inverses of each other. This is complicated by the fact that spiritual distress is a diagnosis,
M AN U
while spiritual well-being is measured on a scale without set clinical cut-off points, and a variety of measures are used for spiritual well-being. Chaves compared three measures of spiritual well-being with diagnosed spiritual distress, and found that one did not match at all while the other two, although highly associated and showing very good sensitivity and
TE D
specificity, were not quite identical to spiritual distress (11). However, another study, using yet another measure for spiritual well-being, found a near-identical match between diagnosed
measure (12).
EP
spiritual distress and spiritual well-being, once the proper cutoff was established for that
We included Facit-Sp-12 as one measure of spiritual well-being because of its widespread use
AC C
in studies examining the medical impact of low spiritual well-being (2,3,26). Although the Facit-Sp was significantly associated with spiritual distress, it was much further from being identical to it than the measures mentioned above in other studies. Even after optimizing the cutoff score to ensure maximal similarity (following the standard approach of identifying the highest value for sensitivity plus specificity (27)), 43% of those in spiritual distress did not have low spiritual well-being as measured by Facit-Sp (i.e., 1–sensitivity). It would seem that there are other elements to a patient’s spiritual distress that this measure did not ask about.
Spiritual distress compared to other measures: 13
ACCEPTED MANUSCRIPT However, it is also not the case that all patients showing low spiritual well-being were found to be in spiritual distress. In fact, a majority of those with low spiritual well-being as measured by Facit-Sp were not in spiritual distress. Thus, both as an expression of spiritual
demonstrably far from being identical with spiritual distress.
RI PT
distress and as a predictor of spiritual distress, Facit-Sp is statistically significant yet is
Spiritual pain, which by its name sounds like it would be similar to spiritual distress, was indeed found to be more nearly identical with spiritual distress. In the bivariate analysis,
SC
spiritual pain was much more strongly associated with spiritual distress than either the FacitSp or the distress thermometer. We found similar levels of spiritual pain to those in other
M AN U
studies (20,21). However, here, too, the terms were far from identical. One-quarter of those in spiritual distress are not in spiritual pain (i.e., 1–sensitivity), and over half of those in spiritual pain did not report spiritual distress.
The Spiritual Injury Scale, given that it directly asks about items that would seem to be strong
TE D
indicators of distress (as indicated by the name "injury"), was expected to show high predictive ability. That is to say, a patient showing spiritual injury certainly ought to show spiritual distress. This is measured by specificity, which can be logically interpreted as the
EP
extent to which, if the secondary measure is present, then the primary measure will also be present. This hypothesis was fairly well supported by our findings, especially when using the
AC C
alternate scoring system suggested by our statistical analysis. In Berg's scoring system, as long as even one item shows injury, the patient is considered to be experiencing spiritual injury (18). Using this system, spiritual injury was found to be a very common expression of spiritual distress, but a relatively poor predictor of spiritual distress because of the cumulative effect of the "false positives" contained within each item. However, if we raise the bar for diagnosing spiritual injury to two items showing injury, this tool remains sufficiently good at
Spiritual distress compared to other measures: 14
ACCEPTED MANUSCRIPT capturing the expression of spiritual distress (64% sensitivity) while becoming a very good predictor of spiritual distress (79% specificity). Finally, we examined the similarity of spiritual distress and general distress. As a screening tool, the distress thermometer (DT) is fairly accurate (i.e., high scores on the DT correlate
RI PT
with spiritual distress most of the time, similar to results reported elsewhere (28)), but it would still fail to identify those in spiritual distress over half the time (i.e., only 41% of
people in spiritual distress scored high on the DT). It could be used in conjunction with other
SC
approaches but seems not to suffice by itself. Where it is already in routine use, a high score on the DT could be used as a strong indicator for referring to spiritual care.
M AN U
Given that neither spiritual well-being as measured by Facit-Sp, spiritual pain, spiritual injury, nor general distress were sufficiently identical to spiritual distress, we performed a multivariable analysis to see if we could suggest the beginnings of an alternate measure that would be a better stand-in for spiritual distress, which could be further studied and validated.
TE D
The model generated by the multivariable analysis included three items: peace ("I feel peaceful"), ability to accept that this is happening, and patient perceived severity of illness, each of which we will now discuss.
EP
Inner peace or serenity is increasingly proving significant in examining the impact of illness on patients' well-being. It is particularly noteworthy that, in the present study, of the three
AC C
subscales of the Facit-Sp – Peace, Meaning, and Faith – only Peace was significant in all tests. In several studies showing medical benefit of spiritual well-being for cancer patients and the elderly, only the Peace and Meaning subscales were significant, but not the Faith subscale (2,3,15). In a Middle Eastern cancer study of health related QoL, Peace was the strongest factor followed by Meaning, while Faith was not significant. A number of studies have begun focusing specifically on Peace, using variants of the single-item Steinhauser measure ("I feel at peace") to demonstrate significant associations with improved medical condition (1,6) and
Spiritual distress compared to other measures: 15
ACCEPTED MANUSCRIPT to measure the validity of assessment tools (29). Other studies show a correlation between peace and spiritual well-being (30). The present study buttresses this approach of using peacefulness as the single most medically important measure of spiritual well-being. The second item persisting in the multivariable model, an inability to accept that this is
RI PT
happening, was a new item hypothesized to be significant in Israel. In Middle Eastern,
Islamic-influenced societies, a commonly heard theme in the face of illness is the belief that whatever happens is God's will and that man's job is to accept God's will (31). Thus, in this
SC
religious cultural context, an inability to accept what is happening would seem to be an
expression of spiritual distress, of not finding comfort within commonly shared spiritual
M AN U
beliefs.
The final item in the multivariable model was patients' perception of the severity of their illness. The data from the medical file, by contrast, did not significantly correlate with spiritual distress, in line with previous findings (12). This finding also supports the connection
TE D
between spiritual well-being and patients' subjective feelings regarding their physical wellbeing (2,17). Patients who see their situation as being worse than it actually is are, unsurprisingly, more likely to despair of the possibility of a continued life of purpose and
EP
connectedness, and vice versa.
AC C
Limitations
Given that cultural influence on spiritual distress' expression is central to this study, ideally we would have liked to study the major subgroups in Israel separately, including Arabs (21% of the national population) and Russian immigrants (15%). Unfortunately, the sample size was not large enough to enable us to draw significant conclusions regarding the expression of spiritual distress within these subgroups examined separately.
Spiritual distress compared to other measures: 16
ACCEPTED MANUSCRIPT
Conclusion Spiritual distress is a significant source of suffering for many patients with cancer, especially those perceiving their situation as grave. It is closely related to but distinct from general
RI PT
distress, spiritual pain, and two common measures of spiritual well-being. Feeling at peace is a key element of not being in spiritual distress and continues to show promise, in this as in
SC
other recent studies, as a simple research measure.
Disclosures and Acknowledgements
M AN U
Thank you to the UJA/Federation of New York for their ongoing support of the spiritual care service and research efforts at Rambam Health Care Campus. Thank you to Professor George
AC C
EP
TE D
Fitchett of Rush University for his assistance in thinking through the study design.
Spiritual distress compared to other measures: 17
ACCEPTED MANUSCRIPT
References 1. Park C, Aldwin C, Choun S, et al. Spiritual peace predicts 5-year mortality in congestive heart failure patients. Health Psychol 2016;35:203-210. doi: 10.1037/hea0000271.
RI PT
2. McClain CS, Rosenfeld B, Breitbart W. Effect of spiritual well-being on end-of-life despair in terminally-ill cancer patients. Lancet 2003;361:1603-1607.
3. Nelson CJ, Rosenfeld B, Breitbart W, Galietta M. Spirituality, religion, and depression in
SC
the terminally ill. Psychosomatics 2002;43:213-220.
4. Brady MJ, Peterman AH, Fitchett G, Mo M, Cella D. A case for including spirituality in
M AN U
quality of life measurement in oncology. Psychooncology 1999;8:417-428. 5. McCoubrie RC, Davies AN. Is there a correlation between spirituality and anxiety and depression in patients with advanced cancer? Support Care Cancer 2006;14:379-385. 6. Rabow MW, Knish SJ. Spiritual well-being among outpatients with cancer receiving
TE D
concurrent oncologic and palliative care. Support Care Cancer 2015;23:919-923. 7. Herdman T. (Ed.). North American Nursing Diagnosis Association International—Nursing Diagnoses: Definitions and classification 2009–2011. Oxford: Wiley-Blackwell, 2009.
EP
8. Puchalski C, Vitillo R, Hull S, Reller N. Improving the spiritual dimension of whole person care: reaching national and international consensus. J Palliat Med 2014;17:642-
AC C
656. doi: 10.1089/jpm.2014.9427. 9. Simão TP, Lopes Chaves Ede C, Campos de Carvalho E, et al. Cultural adaptation and analysis of the psychometric properties of the Brazilian version of the Spiritual Distress Scale. J Clin Nurs 2016;25:231-239. doi: 10.1111/jocn.13060. 10.
Caldeira S, Timmins F, de Carvalho EC, Vieira M. Clinical validation of the nursing diagnosis spiritual distress in cancer patients undergoing chemotherapy. Int J Nurs Knowl. 2015 Jun 11. doi: 10.1111/2047-3095.12105.
Spiritual distress compared to other measures: 18
ACCEPTED MANUSCRIPT 11.
Chaves Ede C, Carvalho EC, Beijo LA, Goyatá SLT, Pillon SC. Efficacy of different instruments for the identification of the nursing diagnosis spiritual distress. Rev Lat Am Enfermagem 2011;19:902-910.
12.
Caldeira S, Carvalho EC, Vieira M. Between spiritual wellbeing and spiritual distress:
RI PT
possible related factors in elderly patients with cancer. Rev Lat Am Enfermagem 2014;22:28-34. 13.
Benito E, Oliver A, Galiana L, et al. Development and validation of a new tool for the
SC
assessment and spiritual care of palliative care patients. J Pain Symptom Manage 2014;47:1008-1018.e1. doi: 10.1016/j.jpainsymman.2013.06.018.
Brady MJ, Peterman AH, Fitchett G, Mo M, Cella D. A case for including spirituality in
M AN U
14.
quality of life measurement in oncology. Psychooncology 1999;8:417-428. 15.
Agli O, Bailly N, Ferrand C. Validation of the Functional Assessment of Chronic Illness Therapy-Spiritual Well-being (FACIT-Sp12) on French Old People. J Relig Health
16.
TE D
2016 Mar 14. [Epub ahead of print]
Selman L, Harding R, Gysels M, Speck P, Higginson IJ. The measurement of spirituality in palliative care and the content of tools validated cross-culturally: a
EP
systematic review. J Pain Symptom Manage 2011;41:728-753. doi: 10.1016/j.jpainsymman.2010.06.023. Lazenby M, Khatib J, Al-Khair F, Neamat M. Psychometric properties of the Functional
AC C
17.
Assessment of Chronic Illness Therapy--Spiritual Well-being (FACIT-Sp) in an Arabicspeaking, predominantly Muslim population. Psychooncology 2013;22:220-227. doi: 10.1002/pon.2062. 18.
Berg G. The relationship between spiritual distress, PTSD and depression In Vietnam combat veterans. J Pastoral Care Counsel 2011;6:1-11.
Spiritual distress compared to other measures: 19
ACCEPTED MANUSCRIPT 19.
Mako C, Galek K, Poppito SR. Spiritual pain among patients with advanced cancer in palliative care. J Palliat Med 2006;9:1106-1113.
20.
Delgado-Guay MO, Hui D, Parsons HA, et al. Spirituality, religiosity, and spiritual pain in advanced cancer patients. J Pain Symptom Manage 2011;41:986-994. doi:
21.
RI PT
10.1016/j.jpainsymman.2010.09.017.
Delgado-Guay MO, Chisholm G, Williams J, et al. Frequency, intensity, and correlates of spiritual pain in advanced cancer patients assessed in a supportive/palliative care
22.
SC
clinic. Palliat Support Care 2016;14:341-348. doi: 10.1017/S147895151500108X.
Blanchard JH, Dunlap DA, Fitchett G. Screening for spiritual distress in the oncology
M AN U
inpatient: a quality improvement pilot project between nurses and chaplains. J Nurs Manag 2012;20:1076-1084. doi: 10.1111/jonm.12035. 23.
Vitek L, Rosenzweig MQ, Stollings S. Distress in patients with cancer: definition, assessment, and suggested interventions. Clin J Oncol Nurs 2007;11:413-418. Snowden A, White CA, Christie Z, et al. The clinical utility of the distress thermometer:
TE D
24.
a review. Br J Nurs 2011;20:220-227. 25.
Murphy PE, Canada AL, Fitchett G, et al. An examination of the 3-factor model and
EP
structural invariance across racial/ethnic groups for the FACIT-Sp: a report from the American Cancer Society's Study of Cancer Survivors-II (SCS-II). Psychooncology
26.
AC C
2010;19:264-272. doi: 10.1002/pon.1559. Wilson CS, Forchheimer M, Heinemann AW, Warren AM, McCullumsmith C. Assessment of the relationship of spiritual well-being to depression and quality of life for persons with spinal cord injury. Disabil Rehabil 2016;25:1-6. 27.
Hegel MT, Collins ED, Kearing S, et al. Sensitivity and specificity of the Distress Thermometer for depression in newly diagnosed breast cancer patients. Psychooncology 2008;17:556-560.
Spiritual distress compared to other measures: 20
ACCEPTED MANUSCRIPT 28.
Agarwal J, Powers K, Pappas L, et al. Correlates of elevated distress thermometer scores in breast cancer patients. Support Care Cancer 2013;21:2125-2136. doi: 10.1007/s00520-013-1773-z.
29.
Monod S, Martin E, Spencer B, Rochat E, Büla C. Validation of the Spiritual Distress
RI PT
Assessment Tool in older hospitalized patients. BMC Geriatr 2012;12:13. doi: 10.1186/1471-2318-12-13. 30.
Selman L, Speck P, Gysels M, et al. ‘Peace’ and ‘life worthwhile’ as measures of
Quality of Life Outcomes 2013;11:94.
M AN U
Schultz M, Baddarni K, Bar-Sela G. Reflections on palliative care from the Jewish and Islamic tradition. Evid Based Complement Alternat Med 2012;2012:693092. doi:
EP
TE D
10.1155/2012/693092.
AC C
31.
SC
spiritual well-being in African palliative care: a mixed-methods study. Health and
Spiritual distress compared to other measures: 21
ACCEPTED MANUSCRIPT
Legend to Figure ROC curves, various measures and spiritual distress
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1:
Spiritual distress compared to other measures: 22
ACCEPTED MANUSCRIPT
Table 1. Sample characteristics of study participants (n=202)
No.
%
Female
121
60
Male
81
40
<50
28
14
50-59
53
26
60-69
73
36
70+
48
24
Jewish
165
Muslim
15
Christian
14
Druze
7
Cancer stage
TE D
Religion
82 7 7 3
12
6
26
13
41
21
118
60
Hebrew
114
56
Arabic
34
17
Russian
21
10
English
8
4
Other
24
12
2 3 4
AC C
Native tongue
EP
1
SC
M AN U
Age (years)
RI PT
Gender
Spiritual distress compared to other measures: 23
ACCEPTED MANUSCRIPT
P value
95% CI
0.67
0.001
0.58-0.76
Peace subscale
0.73
<0.001
0.65-0.82
Meaning subscale
0.63
0.008
0.54-0.72
Faith subscale
0.54
0.42
0.44-0.64
SIS
0.79
<0.001
Distress Thermometer
0.68
<0.001
AC C
EP
0.71-0.87
0.60-0.77
M AN U
TE D
Facit-Sp-12
RI PT
Area
SC
Table 2. ROC area under the curve – Predictive ability vis-a-vis spiritual distress
Spiritual distress compared to other measures: 24
ACCEPTED MANUSCRIPT
Table 3. Bivariate analysis of correlation with spiritual distress, sensitivity, and specificity
OR
95% CI
Facit-Sp-12
<0.001
3.5
1.8-6.9
<0.001
6.4
2.7-14.9
0.02
2.2
1.1-4.5
<0.001
8.3
4.0-17.3
Peace Subscale General Distress
Spiritual Pain Spiritual Injury Scale
<0.001
0 vs 2+ items
<0.001
EP AC C
72%
45%
85%
41%
76%
72%
76%
5.8
2.5-13.1
83%
54%
9.5
4.0-23.0
64%
79%
TE D
0 vs 1+ items
57%
M AN U
Thermometer
Sensitivity Specificity
RI PT
P value
SC
Tool/Item
Spiritual distress compared to other measures: 25
ACCEPTED MANUSCRIPT
Table 4. Multivariable model for predicting spiritual distress
P value
Adjusted OR
95% CI
I feel peaceful.
0.002
1.6
1.2-2.2
I am not able to accept that this is
0.006
3.1
1.4-7.1
0.026
2.3
1.1-4.9
happening to me.
AC C
EP
TE D
M AN U
SC
How serious do you think your illness is?
RI PT
Item
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT