The Suffering Pictogram: Measuring Suffering in Palliative Care

Abstract

Background:

Measuring suffering objectively presents a challenge because suffering is a unique and subjective experience. However, objective tools are of profound importance in the detection and management of suffering in clinical practice for optimal patient care.

Objective:

The objective of the study is to assess the psychometric properties of the Suffering Pictogram, a new suffering assessment instrument on a population of palliative care patients.

Design and Setting:

This is a validation study conducted at University of Malaya Medical Centre, Kuala Lumpur, Malaysia. Ninety one palliative care patients were recruited. Patients were interviewed with the Suffering Pictogram and FACIT-Sp.

Results:

The median completion time for the Suffering Pictogram was five minutes. The Suffering Pictogram showed good internal consistency, with a Cronbach's alpha of 0.836. The total scores of the Suffering Pictogram correlated strongly and negatively with FACIT-Sp total score (Spearman's Rho = −0.625, p < 0.001).

Conclusion:

The Suffering Pictogram is a brief, reliable, and valid instrument to measure suffering in palliative care. The instrument can be used as a screening tool to detect suffering directly.

Introduction

Alleviation of suffering is critically important in palliative care. Most palliative care patients with advanced cancer (88%) have reported suffering.¹ One-fourth of incurably ill patients have suffered unbearably.² Suffering remains pervasive despite the efforts of a palliative care movement for decades. To palliate suffering effectively, a deep understanding of suffering is needed. The word “suffering” in the Oxford dictionary means the state of undergoing pain, distress, or hardship.³ Suffering is a specific state of severe distress associated with events that threaten the intactness of a person.^4,5 It is a severe form of stress experienced by a person beyond the confines of the body, which occurs when there is perceived damage to the integrity of the self, helplessness in the face of the threat, and exhaustion of coping resources.^6,7 It is the result of an imbalance between perceived threats and regulatory processes, leading to exhaustion.⁸

Among the specific dimensions of suffering are physical, psychological, social, and spiritual.⁹ These dimensions underscore the all-encompassing nature of suffering, often described as “total pain” in palliative care.¹⁰ Through a different lens, suffering can be seen from an existential perspective that focuses on suffering events (termed existential suffering, such as loss of control, loss of body image, loss of dignity, burdening of others, relationship issues, social isolation, unfinished business, and facing death) and an experiential perspective that examines the actual suffering experiences (termed experiential suffering: sensory, emotional, cognitive, and spiritual experiencing of suffering events).^11–14 Separation of events and experiences enables the differentiation of existential interventions that focus on removing or improving the situations and experiential interventions that transform or restructure the psychological landscape.

Measuring suffering objectively presents a challenge because it is a unique and subjective experience. However, objective tools are of profound importance in the detection and management of suffering in clinical practice for optimal patient care.¹⁵ Although many instruments are available to assess symptoms and quality of life, few have been developed to specifically assess suffering. From a review of suffering assessment tools (SATs), available instruments included initial assessment of suffering, Pictorial Representation of Illness and Self Measure (PRISM), Mini-Suffering State Examination (MSSE), SAT, Structured Interviews for Symptoms and Concerns (SISC), State of Suffering-V, and Single-item Numeric Rating. Among all instruments reviewed, PRISM and SISC have the strongest psychometric properties.¹⁶ These instruments are not used regularly in the palliative setting due to various reasons. PRISM has been validated in multiple populations, but not in the palliative care context. SISC took a longer time to administer.¹⁶ Distress Thermometer, endorsed by the NCCN Distress Management Guidelines panel for rapid screening of distress, is a validated single-item screen that assesses how much distress patients are going through in the past week from a scale of 0 to 10.¹⁷ Although the terms suffering and distress frequently overlap, they represent different experiences.¹⁸

The aim of the study was to assess the psychometric properties of a new suffering assessment instrument, the Suffering Pictogram (Fig. 1), on a population of palliative care patients. It has been developed primarily to measure experiential suffering based on existing literature.^11,14,19 Instead of measuring suffering from the perspective of the multitude of “external” events that could happen at the end of lives, the Suffering Pictogram was developed to measure the “internal” sensory, emotional, cognitive, and spiritual experiences of these events.

FIG. 1.

The Suffering Pictogram.

Methods

The study was approved by the Ethics Committee at the University of Malaya Medical Centre (UMMC), a tertiary hospital in Kuala Lumpur, Malaysia. UMMC palliative care inpatients were recruited through convenience sampling from August 2014 to July 2015. Inclusion criteria were age >18 years, no known psychiatric disorder, not confused based on Confusion Assessment Method score, and able to communicate verbally. Written consents were obtained. Demographics data were collected. Patients were interviewed face-to-face by a trained research assistant with the Suffering Pictogram and FACIT-Sp (Version 4).

The Suffering Pictogram is a novel eight-item pictogram (Fig. 1) drawn to represent the burning flames of a fire. It measures inner experiences of suffering with a Likert scale (0 = none, 1 = a little bit, 2 = somewhat, 3 = quite a bit, 4 = a lot). It comprises one sensory scale (discomfort), four emotional scales (worry, fear, anger, and sadness), two cognitive scales (hopelessness and difficulty in acceptance), and one spiritual scale (emptiness). In addition, patients are required to rate their overall suffering score with a numerical scale of 0 to 10 at the center of the pictogram (0 = no suffering, 10 = worst possible suffering).

A minimally clinically important difference (MCID) was determined with the anchor question: If you were to receive treatment to reduce your suffering, what is the minimum reduction of suffering score that is acceptable to you? The answer to this anchor question was a change in the overall suffering score at the center of the pictogram. Interview time for the Suffering Pictogram was recorded.

The pictogram development involved the following phases: first, a qualitative study was conducted with semistructured interviews to explore the experiences of 20 individual palliative care patients.¹¹ Second, eight most common suffering experiences were extracted from 456 codes of suffering experiences and presented to five palliative care physicians, two psychiatrists, and two psychologists for feedback on appropriateness of content and breadth of coverage (content validity). Next, the preliminary eight-item pictogram was constructed and administered to five palliative care inpatients to assess their understanding of individual items and the extent to which they have experienced these problems. To examine the concurrent validity of the Suffering Pictogram, the FACIT-Sp, a valid quality-of-life instrument assessing five domains (physical, functional, emotional, social, and spiritual) was used. Compared with FACIT-Pal, FACIT-Sp was chosen because of its more comprehensive spiritual domain.²⁰

Data analyses were conducted by using SPSS (Version 22, Chicago). Demographics were computed using descriptive statistics. Internal reliability consistency of the pictogram was assessed by Cronbach's alpha and was considered acceptable if the coefficient exceeded 0.70.²¹ Cronbach's alpha tests whether the items in the pictogram measure the same construct of interest. Principal axis factor analysis with promax rotation was used to explore the factor structure of the Suffering Pictogram. All items were loaded of at least 0.5 and a difference of at least 0.25 in absolute value from all other observed factor loadings. Homogeneity of the pictogram was assessed by calculating correlation coefficients between items and total scores, if an item was deleted. In establishing concurrent validity, correlations between the suffering pictogram and FACIT-Sp were examined with Spearman's test.

Results

Ninety-four patients participated in the study. Three patients were excluded from the analysis because they were unable to complete ≥50% of the questionnaires. Demographic features of the 91 patients are presented in Table 1. Sample size was adequate based on a subject-to-item ratio of 10:1, given that the suffering pictogram has eight items.²² The pictogram was understandable for all patients. Median completion time for the Suffering Pictogram was 5 minutes. Scoring of the pictogram was easy. It involved summing up of the eight items. The median of MCID, the smallest difference in the overall suffering score that patients perceived as beneficial and that would mandate a change in treatment, at the center of the pictogram was 2.

Table 1.

Demographic Features

Variables	N (%)
Total number of patients	91
Age (years)
<50	22 (24)
50–59	26 (29)
60–69	24 (26)
≥70	19 (21)
Gender
Male	39 (43)
Female	52 (57)
Marital status
Married	68 (75)
Unmarried	23 (25)
Religion
Buddhism	47 (52)
Islam	19 (21)
Christianity	13 (14)
Hinduism	5 (5)
Others	7 (8)
Race
Chinese	60 (66)
Malay	18 (20)
Indian	11 (12)
Others	2 (2)
Eastern Cooperative Oncology Group (ECOG)
1	2 (2)
2	36 (40)
3	44 (48)
4	9 (10)
Types of disease
Breast cancer	26 (29)
Gastrointestinal and hepatobiliary cancer	17 (19)
Lung cancer	13 (14)
Gynecological cancer	6 (7)
Head and neck cancer	6 (7)
Motor neuron disease	3 (3)
Miscellaneous	20 (21)
Duration of sickness in months
<6	23 (25)
6–12	16 (18)
>12	52 (57)

For reliability, the Suffering Pictogram exhibited good internal consistency, with a Cronbach's alpha of 0.836. For factor analysis, Bartlett's test of sphericity was significant (p < 0.01) and the Kaiser–Mayer–Olkin measure of sampling adequacy was 0.84, showing that the data were suitable for factor analysis.²³ Two factors were extracted with principal axis factor analysis and promax rotation with Kaiser normalization (eigenvalue >1.000), which accounted for 52.03% of total variance. Two factors that corresponded to the Suffering Pictogram items referred to Emotions and Cognitions (Table 2). Discomfort did not load to any factor. Corrected item-total correlations and Cronbach's alpha values, if an item is deleted in the Suffering Pictogram, are shown in Table 3. The deletion of discomfort from the factors increased the internal consistency of the total score.

Table 2.

Factor Analysis and the Reliability of the Suffering Pictogram

	Factors
	Emotions	Cognitions
Worry	0.89
Fear	0.81
Sadness	0.73
Hopelessness		0.80
Emptiness		0.65
Difficulty in acceptance		0.61
Anger		0.57
% variance	42.2	9.83
Cronbach's alpha	0.72	0.79

Loading below 0.50 is suppressed. Extraction method: principal axis factoring. Rotation method: promax with Kaiser normalization.

Table 3.

Corrected Item-Total Correlations and Cronbach's Alpha Values if an Item Is Deleted in the Suffering Pictogram

Items	Scale mean if item deleted	Scale variance if item deleted	Corrected item-total correlation	Cronbach's alpha if item deleted
Discomfort	8.81	45.97	0.37	0.84
Worry	8.97	43.46	0.56	0.82
Fear	9.27	41.17	0.61	0.81
Anger	9.72	44.64	0.56	0.82
Sadness	9.25	40.37	0.73	0.80
Hopelessness	9.70	43.20	0.65	0.81
Difficulty in acceptance	9.33	43.58	0.57	0.82
Emptiness	9.65	43.84	0.51	0.82

The results of the Spearman correlation analysis that was conducted between the Suffering Pictogram and FACIT-Sp are shown in Table 4. The strongest negative correlations were found between emotional suffering (worry, fear, anger, and sadness) and emotional well-being in FACIT-Sp. There were moderate negative correlations between discomfort and physical well-being (Spearman's Rho = −0.483, p < 0.001). Hopelessness and difficulty in acceptance were moderately and negatively correlated with emotional well-being (Spearman's Rho = −0.520 to 0.547, p < 0.001). There was weak negative correlation between emptiness and spiritual well-being (Spearman's Rho = −0.352, p = 0.001). There was no significant correlation between any items of the Suffering Pictogram and social well-being of FACIT-Sp.

Table 4.

Spearman Rank Correlation Coefficients between Scores on Different Aspects of the Suffering Pictogram and FACIT-Sp

	Discomfort	Worry	Fear	Anger	Sadness	Hopelessness	Difficulty in acceptance	Emptiness	Overall suffering
PWB	−0.483^**	−0.317^**	−0.286^**	−0.304^**	−0.311^**	−0.278^**	−0.206^*	−0.146	−0.370^**
SWB	0.145	0.001	0.071	0.004	0.097	0.041	0.106	−0.014	−0.011
EWB	−0.290^**	−0.724^**	−0.640^**	−0.533^**	−0.717^**	−0.520^**	−0.547^**	−0.311^**	−0.456^**
FWB	−0.387^**	−0.239^*	−0.293^**	−0.338^**	−0.372^**	−0.363^**	−0.329^**	−0.321^**	−0.322^**
SPWB	−0.170	−0.309^**	−0.372^**	−0.381^**	−0.337^**	−0.357^**	−0.278^**	−0.352^**	−0.369^**

N = 91.

EWB, emotional well-being; FWB, functional well-being; SPWB, spiritual well-being; SWB, social well-being; PWB, physical well-being.

p < 0.05, ^**p < 0.001.

Table 5 shows that overall suffering score correlated moderately and negatively with FACIT-Sp total score (Spearman's Rho = −0.448, p < 0.001). Total scores of the Suffering Pictogram correlated strongly and negatively with FACIT-Sp total score (Spearman's Rho = −0.625, p < 0.001). These indicated the higher overall suffering score (at the centre of pictogram) and higher total score of the Suffering Pictogram were both associated with poorer quality of life. There was also a moderate negative correlation between the total number of suffering symptoms and total score of FACIT-Sp (Spearman's Rho =−0.562, p < 0.001). This demonstrated that a higher number of suffering symptoms was associated with poorer quality of life.

Table 5.

Spearman Rank Correlation Coefficients between Scores on the Suffering Pictogram and FACIT-Sp Total Score

	Overall suffering score	Total scores of the suffering pictogram	Total number of suffering symptoms
FACIT-Sp total score	−0.448^**	−0.625^**	−0.562^**

N = 91.

p < 0.001.

Discussion

The purpose of this study was to establish the reliability and validity of the Suffering Pictogram in the assessment of suffering in patients receiving palliative care. Compared with other SATs, the Suffering Pictogram has several advantages: (1) it was simple to administer and score. It required a median of 5 minutes to complete with a face-to-face interview. The length and ease of use of a questionnaire affect its usability and have a threshold effect on response rate.^24,25 The Suffering Pictogram did not present any difficulty for the patients, and 97% filled in all questions. The high response rate indicated that the length of the questionnaire would probably not be a significant limiting aspect of its clinical use in everyday practice.²⁶ Scoring of the pictogram was easy by summing up the eight items; hence, it would not increase the workload of medical or nursing staff. The median MCID obtained by using an anchor-based method in the overall suffering score of the Suffering Pictogram was 2. To date, none of the SATs reported calculations of MCID.¹⁶

Internal consistency is considered adequate if Cronbach's alpha exceeds 0.7.²¹ The Suffering Pictogram showed good reliability with internal consistency of 0.836, indicating that the proposed eight items measured the same general construct of suffering. Factor analysis revealed a two-factor structure for the eight items based on principal axis factor analysis with promax rotation. The two factors were emotions and cognitions, representing the two basic dimensions of experiential suffering. Discomfort was removed from the factors identified due to the low discrimination value with a 0.37 item-total correlation.

Face and content validity are of central importance, as they highlight the extent to which the measure captures the views of patients.²⁷ In developing the instrument, literature review, expert opinions, and opinions from the target population were used in selecting and reviewing the items for the Suffering Pictogram. The criterion validity of the Suffering Pictogram was verified by comparing it with the gold standard FACIT-Sp. The results showed that there was a significant negative correlation between all items and the physical, functional, emotional, and spiritual domains of FACIT-Sp except items of discomfort and emptiness, in which there were no significant correlations between discomfort and spiritual well-being and between emptiness and physical well-being. The strongest negative correlations were found between emotional and cognitive items of the Suffering Pictogram and emotional well-being of FACIT-Sp, indicating that the Suffering Pictogram gives more information on the emotional aspect of suffering. Not surprisingly, there was no significant correlation between any items of the Suffering Pictogram and social well-being of FACIT-Sp because the social aspect of suffering was not a component of experiential suffering (the experience component of suffering) but existential suffering (the event component of suffering).¹¹ Different social events give rise to similar types of emotions and cognitions.

The psychometric evaluation of the Suffering Pictogram provided evidence that the items of measurement appropriately reflect experiential suffering in palliative care patients, since the validity and reliability yielded consistent results. Nevertheless, the study has several limitations. Potential selection bias beyond the control of the researcher could not be excluded due to convenience sampling. Confused or noncommunicative patients could not be included. The pediatric population was not included. The study sample concerns an Asian population of patients in a specific setting of care, which limits generalizability. To reduce cognitive burden and improve item completion, the pictograms were administered via face-to-face interviews rather than self-administration, so additional human resources are needed for routine use. Language posed another limitation. English is the second language in the country. Misinterpretation could happen for some patients who sought clarification of terms such as “suffering” and “emptiness” in their mother tongue.

In conclusion, the Suffering Pictogram is a brief, reliable, and valid instrument to measure experiential suffering in palliative care. The instrument offers a potential screening tool to detect suffering directly. It could also be used to explore the experiences of suffering further, establish the baselines of suffering, monitor change, and evaluate the effects of suffering interventions in the clinical setting.

Footnotes

Acknowledgments

We express our heartfelt gratitude to all patients who have participated in the study. The authors disclosed receipt of the following financial support for the research, authorship, and publication of this article: High Impact Research Grant (Cycle 3), University of Malaya, Ministry of Higher Education, Malaysia.

Author Disclosure Statement

No competing financial interests exist.

References

Al-Shahri

, Eldali

, Al-Zahrani

: Prevalence and severity of suffering among patients with advanced cancer. Support Care Cancer, 2012; 20:3137–3140.

Ruijis

, Kerkhof

, van der Wal

, et al.: The broad spectrum of unbearable suffering in end-of-life cancer studied in Dutch primary care. BMC Palliat Care, 2012; 11:12.

Stevenson

: Oxford Dictionary of English, 3rd ed. Oxford University Press, NY, 2010.

Cassell

: The Nature of Suffering and the Goals of Medicine. Oxford, UK: Oxford University Press, 1991, pp. 3–33.

Cassell

: The nature of suffering and the goals of medicine. N Engl J Med, 1982; 307:758–760.

Chapman

, Gavrin

: Suffering and its relationship to pain. J Palliat Care, 1993; 9:5–13.

Chapman

, Gavrin

: Suffering: the contributions of persistent pain. Lancet, 1999; 353:2233–2237.

Krikorian

, Limonero

: An integrated view of suffering in palliative care. J Palliat Care, 2012; 28:41–49.

Engel

: The need for a new medical model: A challenge for biomedicine. Science, 1977; 196:129–136.

10.

Saunders

, Baines

: Living with Dying: The Management of Terminal Disease, 1st ed. Oxford, UK: Oxford University Press, 1983, p. 74.

11.

Beng

, Guan

, Seang

, et al.: The experiences of suffering of palliative care patients in Malaysia: A thematic analysis. Am J Hosp Palliat Care, 2014; 31:45–56.

12.

Morita

: Existential concerns of terminally ill cancer patients receiving specialized palliative care in Japan. Support Care Cancer, 2004; 12:137–140.

13.

Blinderman

, Cherny

: Existential issues do not necessarily result in existential suffering: Lesion from cancer patients in Israel. Palliat Med, 2005; 19:371–380.

14.

Tan

, Loh

, Lam

, et al.: Psychological processes of suffering of palliative care patients in Malaysia: A thematic analysis. BMJ Support Palliat Care, 2016. [Epub ahead of print]; DOI: 10.1136/bmjspcare-2015-001064.

15.

Aminoff

, Purits

, Noy

, et al.: Measuring the suffering in end-stage dementia: Reliability and validity of the Mini Suffering State Examination. Arch Gerontl Geriatr, 2004; 38:123–130.

16.

Krikorian

, Limonero

, Corey

: Suffering assessment: A review of available instruments for use in palliative care. J Palliat Med, 2013; 16:130–142.

17.

The National Comprehensive Cancer Network: Distress Management Guidelines. Published (Version 2). 2014. www.nccn.org (last accessed January 11, 2017).

18.

Krikorian

, Limonero

, Maté J: Suffering and distress at the end-of-life. Psychooncology, 2012; 21:799–808.

19.

Beng

, Chin

, Guan

, et al.: Mindfulness-based supportive therapy (MBST): Proposing a palliative psychotherapy from a conceptual perspective to address suffering in palliative care. Am J Hosp Palliat Care, 2015; 32:144–160.

20.

Peterman

, Fitchett

, Brady

, et al.: Measuring spiritual well-being in people with cancer: The functional assessment of chronic illness therapy–Spiritual Well-being Scale (FACIT-SP). Ann Behav Med, 2002; 24:49–58.

21.

Bland

, Altman

: Cronbach's alpha. BMJ, 1997; 314:572.

22.

DeVellis

: Guidelines in scale development. In: Scale Development: Theory and Applications, Bickman

and Rog

(eds.). Newbury, CA: Sage, 1991, p. 5191.

23.

Kaiser

: An index of factorial simplicity. Psychometrika, 1974; 39:31–36.

24.

Lund

, Gram

: Response rate according to title and length of questionnaire. Scand J Soc Med, 1998; 26:154–160.

25.

Jepson

, Asch

, Hershey

, et al.: In a mailed physician survey, questionnaire length had a threshold effect on response rate. J Clin Epidemiol, 2005; 58:103–105.

26.

Manhalter

, Bozsik

, Palásti

, et al.: The validation of a new comprehensive headache-specific quality of life questionnaire. Cephalalgia, 2012; 32:668–682.

27.

Greenhalgh

, Long

, Brettle

, et al.: Reviewing and selecting outcome measures for use in routine practice. J Eval Clin Pract, 1998; 4:339–350.