Ò
PAIN 154 (2013) 1691–1698
www.elsevier.com/locate/pain
Anger differentially mediates the relationship between perceived
injustice and chronic pain outcomes
Whitney Scott a, Zina Trost b, Elena Bernier a, Michael J.L. Sullivan a,⇑
a
b
Department of Psychology, McGill University, Canada
Department of Psychology, University of North Texas, USA
Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.
a r t i c l e
i n f o
Article history:
Received 26 November 2012
Received in revised form 8 May 2013
Accepted 8 May 2013
Keywords:
Chronic pain
Perceived injustice
Anger
Disability
Depression
a b s t r a c t
Emerging evidence suggests that perceived injustice is a risk factor for adverse outcomes associated with
chronic pain. To date, however, the processes by which perceived injustice impacts on pain outcomes
remain speculative. Evidence from several lines of research suggests that anger may mediate the relationship between injustice and pain outcomes. However, this relationship has not been empirically tested in
patients with chronic pain. Thus, the purpose of this study was to examine whether anger mediates the
relationships between perceived injustice and pain intensity, depressive symptoms, and self-reported
disability. One hundred and seventy-three individuals with chronic musculoskeletal pain completed
self-report measures of perceived injustice, anger, pain intensity, depressive symptoms, and disability.
Consistent with previous research, high scores on a measure of perceived injustice were associated with
greater pain, more severe depressive symptoms, and more pronounced disability. Hierarchical regression
analyses indicated that anger variables completely mediated the relationship between perceived injustice and pain intensity, and partially mediated the relationship between perceived injustice and depressive symptoms. Anger did not mediate the relationship between perceived injustice and self-reported
disability. The Discussion addresses the theoretical and clinical implications of the findings.
Ó 2013 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
1. Introduction
Biopsychosocial models suggest that the meaning individuals
impart on their pain influences pain-related adjustment. Much research has examined meaning in the form of appraisals of the
threat value of pain, or individuals’ perceived ability to cope with
the threat of pain [16,60]. Emerging research also suggests that
the construal of pain in terms of justice-related themes may be
an important determinant of pain outcomes.
Discourses of justice and injustice appear inherent to the
chronic pain experience [39]. The experience of undeserved suffering or of multiple losses (eg, loss of function, financial security,
identity, etc.) might give rise to perceptions of injustice among
individuals with chronic pain [36,45]. Individuals with chronic
pain may ascribe external blame for this suffering [18,40], which
may increase the likelihood that pain is experienced with an elevated sense of injustice [43,45].
⇑ Corresponding author. Address: Department of Psychology, McGill University,
1205 Doctor Penfield Avenue, Montréal, Québec, Canada H3A1B1. Tel.: +1 514 398
5677; fax: +1 514 398 4896.
E-mail address: [email protected] (M.J.L. Sullivan).
Perceived injustice in the chronic pain context has been operationally defined as an appraisal reflecting the severity and irreparability of pain-related loss, blame, and unfairness [56]. Mounting
evidence indicates that perceived injustice contributes to problematic outcomes associated with persistent musculoskeletal pain.
Perceived injustice has been associated with greater pain severity,
pain behavior, and mental health difficulties, reduced physical
function, and prolonged work disability [53,54,56,58,59]. Perceived
injustice predicts adverse pain outcomes even when controlling for
other pain-related psychosocial constructs, such as pain catastrophizing and fear of movement [52,53,58,59].
There are indications that perceived injustice might be more
resistant to change than other psychosocial pain-related variables
[56]. In a study examining treatment-related changes in psychosocial risk factors following participation in a multidisciplinary
rehabilitation program, perceived injustice was the risk factor that
showed the least improvement [56]. It is unclear whether these
data indicate that perceived injustice is non-modifiable, or
whether they reflect a failure to effectively target perceived injustice during treatment. An understanding of the processes linking
perceived injustice to adverse pain outcomes may provide insight
into potential intervention strategies to mitigate its impact.
0304-3959/$36.00 Ó 2013 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.pain.2013.05.015
1692
Ò
W. Scott et al. / PAIN 154 (2013) 1691–1698
Social psychological research indicates that anger is the predominant emotional response to perceiving injustice [42,43]. Furthermore, numerous investigations have shown that elevated
levels of anger are associated with adverse pain outcomes [4,12].
Research suggests that both anger intensity (state/trait) and regulation style (inhibition/expression) negatively impact pain outcomes [4,12,32,47]. It has recently been proposed that perceived
injustice might be a cognitive antecedent to anger in chronic pain
patients [63]. To date, however, the inter-relations among perceived injustice, anger, and adverse pain outcomes have not been
empirically investigated. It is also not presently known which facets of the anger experience (ie, intensity vs regulation style) may
account for the negative impact of perceived injustice.
This study examined whether anger intensity and regulation
style mediated the relationship between perceived injustice and
pain outcomes. Consistent with previous research, it was hypothesized that perceived injustice would be associated with greater
pain intensity, depressive symptoms, and disability. It was hypothesized that anger intensity and regulation style would be associated with more negative pain outcomes and that these variables
would mediate the relationship between perceived injustice and
pain outcomes.
2. Methods
2.1. Participants
One hundred and seventy-three individuals (113 women, 60
men) with chronic musculoskeletal pain participated in this study.
Chronic low back pain was the most prevalent diagnosis in this
sample (approximately 42% of participants), followed by fibromyalgia/diffuse myofacial pain (approximately 26%), and chronic pain
in the cervical spine (approximately 15%). Participants had a mean
age of 49.67 years with a range of 21–65 years. The mean duration
of pain was 9.98 years with a range from 1 to 50 years. The majority of participants (68%) had completed at least 12 years of education. Approximately half the sample (54%) was married or living
with a common-law partner. At the time of assessment, the majority of participants (65%) were not working. Approximately 94% of
participants were taking at least one class of analgesic medication,
and approximately 58% of the sample were receiving at least one
non-pharmacological intervention (including psychological intervention). At the time of assessment, approximately 24% of the sample indicated they were receiving psychological treatment.
2.2. Procedure
The Quebec Pain Registry (QPR) was used to identify patients
with diagnoses of chronic musculoskeletal pain. The QPR is a database of over 3000 patients with chronic pain conditions who have
received treatment at 1 of 3 university-affiliated tertiary pain management clinics in the province of Québec. Patients with musculoskeletal pain represent the majority of patients in the QPR.
Recruitment letters describing the procedures of the present
study were mailed to patients with a diagnosis of chronic musculoskeletal pain, who were entered in the QPR at the time the present study began. The desired sample size was calculated to be
between 150 and 200 patients, assuming power = 0.80, a < 0.05,
and medium effects sizes for the associations among study variables [27]. Approximately 900 letters were mailed to patients registered in the QPR. In total, 183 individuals contacted the
laboratory following receipt of the recruitment letter, representing
a response rate of approximately 20%.
Interested participants contacted a research coordinator at
McGill University. Individuals were considered eligible for the
study if they had a current diagnosis of a musculoskeletal pain
condition that had been present for at least 3 months. Individuals
were included if they were capable of completing study measures
in English or French. Of the 183 individuals who contacted the laboratory, 4 individuals were ineligible or declined to participate. Of
the 179 eligible participants, 6 were excluded due to incomplete
data. Participants with and without complete data did not differ
significantly for any demographic or study variables. Data were
collected over a period of approximately 3 months.
Participants provided demographic information and completed
self-report measures of perceived injustice, anger, pain intensity,
depressive symptoms, and disability. Participants were given the
option to complete questionnaires online or to mail their questionnaire responses to the laboratory. Volunteers were invited to sign a
consent form as a condition of participating in the study. The research was approved by the Comité d’Éthique de la Recherche du
Centre Hospitalier de L’Université de Montréal.
2.3. Measures
2.3.1. Perceived injustice
The Injustice Experiences Questionnaire (IEQ) was used to measure pain-related perceptions of injustice [56]. Participants rated
the frequency with which they experienced each of 12 pain-related
thoughts on a 5-point scale, ranging from 0 (never) to 4 (all the
time). Previous findings suggest that the IEQ yields two correlated
factors, labeled ‘‘severity/irreparability of loss’’ and ‘‘blame/unfairness.’’ Examples of items loading onto the former factor include,
‘‘Most people don’t understand how severe my condition is,’’ and
‘‘My life will never be the same.’’ Examples of items loading onto
the latter factor include, ‘‘I am suffering because of someone else’s
negligence.’’ and ‘‘It all seems so unfair.’’ The IEQ has been shown
to have high internal and test–retest reliability, and to be valid
for use among individuals with persistent musculoskeletal pain
[52,56].
2.3.2. Anger
The State-Trait Anger Expression Inventory—II (STAEI) [3,55]
was used to assess anger. Participants were asked to rate 57 statements on a 4-point Likert scale. The following subscales of the
STAEI were examined in the present study: state anger (15 items);
trait anger (10 items); anger inhibition (8 items); and anger
expression (8 items). Anger inhibition items assess the frequency
with which participants attempt to suppress feelings of anger. Anger expression items assessed the frequency with which anger is
outwardly expressed. Previous research suggests the distinctiveness of the anger intensity and anger regulation subscales (ie, inhibition and expression) of the STAEI [12]. Research supports the
reliability and validity of these subscales for use with chronic pain
patients [6,15].
2.3.3. Pain intensity
Participants were asked to rate their present pain intensity on a
numerical rating scale ranging from 0 (no pain) to 10 (excruciating
pain). Previous research indicates that this is a reliable measure of
pain intensity [19]. Participants also indicated the location of their
pain on a body schematic.
2.3.4. Depressive symptoms
The Patient Health Questionnaire 9 (PHQ-9)[33] was used to
measure depressive symptom severity. The PHQ-9 is a 10-item
questionnaire that asks respondents to indicate the frequency with
which they experience each of the 9 symptoms considered in the
diagnostic criteria for Major Depression, and 1 item assessing the
difficulty with which they experience these symptoms. The PHQ9 has been shown to be a valid and reliable measure of depressive
symptoms in patients with a variety of medical conditions [14,33].
Ò
1693
W. Scott et al. / PAIN 154 (2013) 1691–1698
2.3.5. Self-reported disability
The Pain Disability Index (PDI) [28,61] was used to assess the degree to which patients perceive themselves to be disabled by pain in
7 different areas of daily living: home, social, recreational, occupational, sexual, self-care, and life support. For each life domain,
respondents were asked to provide perceived disability ratings on
11-point scales with the endpoints (0) no disability and (10) total
disability. The PDI has been shown to be internally reliable and significantly correlated with objective indices of disability [28].
2.3.6. Demographic and treatment variables
Participants responded to questions concerning their age, sex,
marital status, education, employment status, and pain onset. Participants indicated current medications taken for pain. The sum of
different medication classes was taken as an index of medication
use. Participants also indicated the types of non-pharmacological
treatments (eg, physical therapy, occupational therapy, psychological treatment, etc.) they were currently receiving. Participants’
pain diagnosis was ascertained from the QPR following their participation in the present study.
2.4. Data analytic approach
The following subscale scores were computed for the anger
measure: state anger, trait anger, anger inhibition, and anger
expression [15,55]. Total scores were computed for all other measures. Total and subscale scores were computed on the measure of
perceived injustice [56]. Means and standard deviations were computed for total and subscale scores. t Tests for independent samples
were used to compare men and women on study variables. Independent samples t tests were also computed to compare scores
on study variables for patients whose pain was and was not precipitated by an injury. Where reported degrees of freedom do not correspond to the total sample size (n = 173), it is due to missing data
on variables included in those analyses.
Pearson product–moment correlations were computed to examine zero-order associations among perceived injustice, anger intensity and expression style, pain severity, depressive symptoms, and
self-reported disability. Multiple regression analyses were used to
test for mediation. According to Baron and Kenny [2], the following
conditions are necessary to test for mediation: the independent variable (IV; ie, perceived injustice) must be significantly correlated
with the dependent variable (DV; ie, pain, depressive symptoms,
and disability); the IV must be significantly correlated with the
mediator (ie, anger variables); the mediator must be significantly
correlated with the DV; and, the presence of the mediator reduces
the effect of the IV on the DV to zero (complete mediation) or partially reduces the effect of the IV (partial mediation). Therefore, each
of the anger intensity and expression variables was only entered
into the mediation analysis if it showed a significant zero-order correlation with both perceived injustice and the dependent variable of
interest. A separate hierarchical multiple regression analysis was
conducted for each of the dependent variables to test for mediation
of perceived injustice by the candidate anger variables identified in
zero-order correlations. Mediation analyses were repeated for the
blame/unfairness and irreparability of loss subscales of the IEQ. In
all regression analyses, the tolerance and variance inflation factors
of the IVs fell within acceptable ranges [46]. Sobel tests were conducted to examine the significance of the mediation effects.
3. Results
3.1. Sample characteristics
Demographic information of the sample appears in Table 1.
Mean scores on the IEQ are comparable to mean scores on this
measure in previous samples of patients with chronic musculoskeletal pain [56]. Mean scores on other study variables are also
comparable to previous studies of patients with chronic pain
[15,58]. Mean scores for perceived injustice, pain intensity, depressive symptoms, and self-reported disability are above the established clinical cut offs for these measures [25,33,54,57]. Thus, on
average, participants in the present sample appeared to have clinically meaningful levels of perceived injustice, pain, depressive
symptoms, and self-reported disability.
Men (M = 6.35, SD = 1.89) rated their pain as significantly more
intense than women (M = 5.54, SD = 2.15), t (171) = 2.46, P < 0.05.
Men and women did not differ significantly on any other study variable. Participants whose pain was precipitated by an injury had
significantly higher IEQ total scores (M = 30.61, SD = 11.08) and
‘blame/unfairness’ scores (M = 13.15, SD = 6.87) than participants
whose pain was not precipitated by an injury (M = 26.36,
SD = 10.71; M = 10.21, SD = 6.43), t (170) = 2.52, P < 0.05 and t
(170) = 2.84, P < 0.01, respectively. The self-reported disability
scores of participants whose pain was precipitated by an injury
(M = 41.58, SD = 13.27) were significantly higher than the disability
scores of participants whose pain was not precipitated by an injury
(M = 37.13, SD = 14.84), t (170) = 2.07, P < 0.05. Participants with
and without precipitating injuries did not differ significantly on
any other study variable.
3.2. Zero-order correlations among study variables
Table 2 displays the zero-order correlations among study
variables. IEQ total and subscale scores were each significantly correlated with state anger, trait anger, and anger inhibition. IEQ total
and subscale scores were each significantly correlated with all of
the pain outcome variables. State anger and anger inhibition were
significantly correlated with pain intensity, depressive symptoms,
and self-reported disability. Trait anger and anger expression were
significantly correlated with depressive symptoms. All of the anger
subscales were significantly intercorrelated. Pain intensity,
Table 1
Sample demographics and descriptive statistics.
Variable
M (SD) or N (%)
Age
Duration (years)
49.67 (9.66)
9.98 (8.20)
Pain site (categories not mutually exclusive)
Neck
Back
Upper extremity
Lower extremity
99 (57.23%)
147 (84.97%)
81 (46.82%)
122 (70.52%)
Precipitating injury
No
Yes
Work accident
Motor vehicle accident
Post-surgical
Other/injury type not indicated
72 (41.86%)
100 (58.14)
39 (39.00%)
30 (30.00%)
21 (21.00%)
10 (10.00%)
Number of pain medications
Number non-pharmacological treatments
2.72 (1.53)
1.85 (2.54)
IEQ
STAEI-State
STAEI-Trait
STAEI-Expression
STAEI-Inhibition
PPI
PHQ-9
PDI
28.91 (11.11)
22.88 (9.49)
19.86 (6.52)
14.76 (3.97)
18.21 (5.58)
5.82 (2.09)
15.65 (7.06)
39.82 (14.10)
Abbreviations: IEQ, Injustice Experiences Questionnaire; STAEI, State Trait Anger
Expression Inventory—II; PPI, Present Pain Intensity; PHQ-9, Patient Health Questionnaire-9; PDI, Pain Disability Index.
Ò
1694
W. Scott et al. / PAIN 154 (2013) 1691–1698
Table 2
Zero-order correlations among study variables.
1. IEQ Total
2. IEQ-Loss
3. IEQ-Blame
4. STAEI-State
5. STAEI-Trait
6. STAEI-Expression
7. STAEI-Inhibition
8. PPI
9. PHQ-9
10. PDI
1
2
3
4
5
6
7
8
9
0.91**
0.95**
0.42**
0.27**
0.10
0.33 **
0.26**
0.56**
0.51**
0.75**
0.37**
0.24**
0.09
0.29**
0.22**
0.51**
0.47**
0.42**
0.26**
0.10
0.33**
0.26**
0.53**
0.49**
0.43**
0.35**
0.43**
0.38**
0.58**
0.31**
0.67**
0.39**
0.08
0.41**
0.05
0.18*
0.07
0.19*
0.07
0.17*
0.50**
0.16*
0.37**
0.44**
0.49**
Abbreviations: IEQ, Injustice Experiences Questionnaire; STAEI, State Trait Anger Expression Inventory—II; PPI, Present Pain Intensity; PHQ-9, Patient Health Questionnaire-9;
PDI, Pain Disability Index.
*
P < 0.05.
**
P < 0.01
depressive symptoms, and self-reported disability were all significantly inter-correlated.
3.3. Anger as a mediator of the relationship between perceived
injustice and pain intensity
Based on zero-order analyses, state anger and anger inhibition
were significantly correlated with both perceived injustice and
pain intensity, and were thus candidate variables for mediation.
Table 3 displays the results of regression analyses examining the
mediating role of state anger and anger inhibition in the relationship between perceived injustice and pain intensity. In the first
regression analysis, perceived injustice accounted for a significant
proportion of the variance in pain intensity (b = 0.26). In the second
analysis, state anger and anger inhibition entered in the first step
contributed significant variance to the prediction of pain intensity.
Perceived injustice was entered in the second step, but no longer
contributed significant unique variance to the prediction of pain
intensity, beyond that accounted for by anger variables. State anger
(Sobel test = 3.45, P < 0.01) mediated between perceived injustice
and pain intensity (b = 0.13). Examination of the beta weights in
the final regression equation indicated that only state anger contributed significant unique variance to the prediction of pain intensity, b = 0.33, t (171) = 3.94, P < 0.001.
A subsequent regression analysis was conducted to examine the
independent contributions of the ‘‘severity/irreparability of loss’’
and ‘‘blame/unfairness’’ subscales of the IEQ to pain intensity. Only
the blame/unfairness subscale contributed significant unique variance to the prediction of pain intensity, b = 0.23, t (171) = 2.05,
P < 0.05. An additional hierarchical regression analysis indicated
Table 3
Mediating role of state anger and anger inhibition in the relationship between
perceived injustice and pain intensity.
2
R change
Regression 1: Dependent = PPI
Step 1
0.07
IEQ Total
Regression 2: Dependent = PPI
Step 1
0.14
STAEI-State
STAEI-Inhibition
Step 2
IEQ Total
0.01
F change
P
12.37
= 0.001
b
<0.001
0.33**
0.01
2.47
3.4. Anger as a mediator of the relationship between perceived
injustice and depressive symptoms
State anger, trait anger, and anger inhibition were each significantly correlated with perceived injustice and depressive symptoms, and were thus candidate mediators. Table 4 displays the
results of regression analyses examining the mediating role of state
anger, trait anger, and anger inhibition in the relationship between
perceived injustice and depressive symptoms. In the first regression analysis, perceived injustice accounted for a significant proportion of the variance in depressive symptoms (b = 0.49), above
and beyond that accounted for by pain intensity. In the second
analysis, state anger, trait anger, and anger inhibition together contributed significant unique variance to the prediction of depressive
symptoms, controlling for the influence of pain intensity. Perceived
injustice was entered in the third step of this analysis, and contributed significant unique variance to the prediction of depressive
Table 4
Mediating role of anger variables in the relationship between perceived injustice and
depressive symptoms.
R2 change
Regression 1: Dependent = PHQ-9
Step 1
0.14
PPI
Step 2
IEQ Total
0.23
>0.05
0.13
Abbreviations: IEQ, Injustice Experiences Questionnaire; STAEI, State Trait Anger
Expression Inventory—II; PPI, Present Pain Intensity; PHQ-9, Patient Health Questionnaire-9; PDI, Pain Disability Index.
**
P < 0.001.
F change
P
27.83
<0.001
b
0.25**
61.14
<0.001
0.49**
Regression 2: Dependent = PHQ-9
Step 1
0.14
PPI
0.26**
13.91
that state anger (Sobel test = 3.44, P < 0.001) mediated between
the blame/unfairness component of perceived injustice and pain
intensity. Examination of the beta weights in the final regression
equation indicated that only state anger contributed significant unique variance to the prediction of pain intensity, b = 0.33, t
(171) = 3.94, P < 0.001.
Step 2
STAEI-State
STAEI-Trait
STAEI-Inhibition
0.32
Step 3
IEQ Total
0.07
27.83
<0.001
0.16*
33.11
<0.001
0.25**
0.12*
0.22**
26.08
<0.001
0.31**
Abbreviations: IEQ, Injustice Experiences Questionnaire; STAEI, State Trait Anger
Expression Inventory—II; PPI, Present Pain Intensity; PHQ-9, Patient Health Questionnaire-9; PDI, Pain Disability Index.
*
P < 0.05.
**
P < 0.001.
Ò
1695
W. Scott et al. / PAIN 154 (2013) 1691–1698
symptoms, beyond that accounted for by pain intensity and anger
variables. State anger (Sobel test = 3.81, P < 0.001), trait anger (Sobel test = 2.76, P < 0.01), and anger inhibition (Sobel test = 3.37,
P < 0.001) partially mediated between perceived injustice and
depressive symptoms. Examination of the beta weights in the final
regression equation indicated that pain intensity (b = 0.16, t
(171) = 2.68, P < 0.01), state anger (b = 0.25, t (171) = 3.62,
P < 0.001), trait anger (b = 0.12, t (171) = 1.94, P = 0.05), anger inhibition (b = 0.22, t (171) = 3.65, P < 0.001), and perceived injustice
(b = 0.31, t (171) = 5.12, P < 0.001), each contributed significant unique variance to the prediction of depressive symptoms.
A subsequent regression analysis was conducted to examine the
independent contributions of the IEQ subscales to the prediction of
depressive symptoms. The severity/irreparability of loss and
blame/unfairness subscales each contributed significant unique
variance to the prediction of depressive symptoms, b = 0.24, t
(171) = 2.64, P < 0.05 and b = 0.29, t (171) = 3.07, P < 0.05, respectively. An additional hierarchical regression analysis indicated that
state anger (Sobel test = 3.80, P < 0.001), trait anger, (Sobel
test = 2.74, P < 0.01), and anger inhibition (Sobel test = 3.35,
P = 0.001) mediated between the blame/unfairness subscale and
depressive symptoms. State anger (Sobel test = 3.57, P < 0.001),
trait anger (Sobel test = 2.56, P = 0.01), and anger inhibition (Sobel
test = 3.01, P < 0.01) partially mediated between the severity/irreparability of loss subscale and depressive symptoms. Examination of
the beta weights in the final regression equation indicated that
pain intensity (b = 0.16, t (171) = 2.69, P < 0.01), state anger
(b = 0.25, t (171) = 3.63, P < 0.0001), anger inhibition (b = 0.23, t
(171) = 3.66, P < 0.0001), and the severity/irreparability of loss subscale (b = 0.19, t (171) = 2.32, P < 0.05), each contributed significant
unique variance to the prediction of depressive symptoms.
3.5. Anger as a mediator of the relationship between perceived
injustice and self-reported disability
State anger and anger inhibition were significantly correlated
with perceived injustice and self-reported disability in zero-order
analyses, and were thus candidate mediators. Table 5 displays
the results of regression analyses examining the mediating role
of state anger and anger inhibition in the relationship between perceived injustice and self-reported disability. In the first regression
analysis, perceived injustice accounted for a significant proportion
of the variance in disability (b = 0.43), above and beyond that accounted for by pain intensity. In the second analysis, state anger
and anger inhibition did not contribute significant unique variance
to the prediction of disability, above and beyond that accounted for
by pain intensity. Perceived injustice was entered in the third step
of this analysis, and contributed significant unique variance to the
prediction of disability, beyond that accounted for by pain intensity and anger variables. State anger (Sobel test = 0.04, P > 0.05)
and anger inhibition (Sobel test = 0.61, P > 0.05) did not mediate
between perceived injustice and disability. Examination of the beta
weights in the final regression equation indicated that pain intensity (b = 0.33, t (171) = 4.98, P < 0.001) and perceived injustice
(b = 0.43, t (171) = 6.26, P < 0.001) each contributed significant unique variance to the prediction of self-reported disability.
A subsequent regression analysis was conducted to examine the
independent contributions of the IEQ subscales to the prediction of
self-reported disability. The severity/irreparability of loss and
blame/unfairness subscales each contributed significant unique
variance to the prediction of self-reported disability, b = 0.23, t
(171) = 2.51, P < 0.05 and b = 0.23, t (171) = 2.43, P < 0.05, respectively. An additional hierarchical regression analysis indicated that
state anger (Sobel test = 0.27, P > 0.05) and anger inhibition (Sobel
test = 0.44, P > 0.05) did not mediate between the blame/unfairness subscale and disability. State anger (Sobel test = 0.52,
Table 5
Mediating role of state anger and anger inhibition in the relationship between
perceived injustice and disability.
R2 change
Regression 1: Dependent = PDI
Step 1
0.20
PPI
Step 2
IEQ Total
0.17
F change
P
41.91
<0.001
b
0.33**
45.60
<0.001
0.43**
Regression 2: Dependent = PDI
Step 1
0.20
PPI
Step 2
STAEI-State
STAEI-Inhibition
0.02
Step 3
IEQ Total
0.15
41.91
<0.001
0.33**
2.64
>0.05
0.02
0.05
39.15
<0.001
0.43**
Abbreviations: IEQ, Injustice Experiences Questionnaire; STAEI, State Trait Anger
Expression Inventory—II; PPI, Present Pain Intensity; PHQ-9, Patient Health Questionnaire-9; PDI, Pain Disability Index.
**
P < 0.001
P > 0.05) and anger inhibition (Sobel test = 0.16, P > 0.05) also
did not mediate between the severity/irreparability of loss subscale and disability. Examination of the beta weights in the final
regression equation indicated that pain intensity (b = 0.33, t
(171) = 4.98, P < 0.001), the blame/unfairness subscale (b = 0.23, t
(171) = 2.42, P < 0.05), and the severity/irreparability of loss subscale (b = 0.23, t (171) = 2.50, P < 0.05) each contributed significant
unique variance to the prediction of self-reported disability.
In light of the finding that anger did not mediate the injustice–
disability relationship, an additional regression analysis was conducted to determine whether depressive symptoms mediated the
relationship between perceived injustice and disability. After
controlling for pain intensity, depressive symptoms entered in
the second step of the analysis contributed an additional 12% of
the variance to the prediction of disability (F (1, 170) = 30.29,
P < 0.001). Perceived injustice was entered in the third step, and
contributed an additional 7% of the variance to the prediction of
disability, above and beyond that accounted for by pain intensity
and depressive symptoms (F (1, 169) = 20.50, P < 0.001). Depressive
symptoms (Sobel test = 2.51, P = 0.01) partially mediated between
perceived injustice and disability. Examination of the beta weights
from the final regression equation revealed that pain intensity
(b = 0.28, t (171) = 4.38, P < 0.001), depressive symptoms
(b = 0.20, t (171) = 2.65, P < 0.01), and perceived injustice
(b = 0.33, t (171) = 4.53, P < 0.001) were all significant unique predictors of disability.
4. Discussion
The present study adds to a growing body of research examining the contribution of perceived injustice to chronic pain-related
adjustment. Consistent with previous research, perceived injustice
was significantly correlated with pain intensity, depressive symptoms, and self-reported disability [53,56,58]. This study provides
an important extension of previous research by showing that anger
intensity and inhibition differentially mediate the association between perceived injustice and pain, depressive symptoms, and disability. To the authors’ knowledge, this is the first study to provide
support for perceived injustice as a cognitive correlate of anger
intensity and anger inhibition in chronic pain.
The present study replicated previous findings showing that
perceived injustice, state anger, and anger inhibition are significantly correlated with pain intensity [5,15,56,58]. Research has
not previously investigated the processes by which perceived
1696
Ò
W. Scott et al. / PAIN 154 (2013) 1691–1698
injustice contributes to increased pain intensity. The present results indicated that perceived injustice was significantly correlated
with state anger and anger inhibition. Regression analyses suggested that state anger and anger inhibition completely mediated
the relationship between perceived injustice and pain intensity.
Thus, anger appears to be the vehicle through which perceived
injustice impacts pain intensity.
Subscale analyses indicated that only the ‘blame/unfairness’
subscale of the IEQ contributed significant unique variance to the
prediction of pain intensity. This association was completely mediated by state anger. Conceptual models highlight the attribution of
blame as an antecedent of injustice perceptions and anger reactions [1,41,51]. Experimental and clinical research likewise suggests that the identification of external sources of blame
contributes to greater levels of anger [35,42,47]. It is possible that
chronic pain patients’ attributions of blame might give rise to anger reactions that, in turn, trigger a cascade of physiological responses that ultimately result in more intense pain experience.
Increasing evidence points to the mechanisms by which anger
may exacerbate pain. In both patients and healthy controls, acute
induction of anger is associated with endogenous opioid dysfunction in response to painful stimuli [5,7]. Increased muscle tension
and systolic blood pressure have been shown to follow acute
induction of anger, which may also increase pain sensitivity
[9,10]. Although not yet empirically examined, it has been suggested that activation in the anterior cingulate cortex and dorsal
lateral prefrontal cortex may subserve the relationship between
anger and pain [12].
The present results indicated that perceived injustice and anger
variables were each significantly correlated with depressive symptoms, which is in line with previous research [11,20,47,53,56]. Previous work has not examined mechanisms linking perceived
injustice to depressive symptoms. In the present study, regression
analyses revealed that state and trait anger and anger inhibition
partially explained the association between perceived injustice
and depressive symptoms. This suggests that anger may be one
mechanism through which perceived injustice influences depressive symptoms.
Analyses of the IEQ subscales indicated that anger variables
completely mediated the relationship between blame/unfairness
and depressive symptoms, and partially mediated between severity/irreparability of loss and depressive symptoms. The severity/
irreparability of loss subscale remained a significant unique predictor of depressive symptoms in the final equation. The direct effect
of this subscale is unsurprising in light of numerous reports highlighting the central role of loss experiences in the precipitation of
depressive symptoms [26,31,49].
Several processes may account for the strong association between anger and depressive symptoms. It has been suggested that
serotonergic and dopaminergic dysfunction may be common physiological substrates for both anger and depression [44,48,64]. Reduced activation in the orbitofrontal cortex also appears common
in states of both anger and depression [34]. Behaviorally, characteristic suppression of anger may contribute to withdrawal from
anger-provoking social environments [13,17,22]. Consequently, social isolation and the associated loss of positive reinforcement from
interpersonal interactions may exacerbate depressive symptoms
[30,67].
Perceived injustice, state anger, and anger inhibition were each
significantly correlated with self-reported disability, which replicates previous findings [20,47,58]. Despite significant zero-order
associations between perceived injustice, anger intensity and inhibition, and disability, the current results did not support anger as a
mediator between perceived injustice and disability. Subsequent
analyses also did not support depressive symptoms as a mediator
between perceived injustice and disability. Subscale analyses
revealed that both the blame/unfairness and severity/irreparability
of loss subscales of the IEQ each uniquely predicted disability.
These results highlight the complexity of the injustice construct,
and suggest that different processes may underlie its impact on
various pain outcomes.
Previous research suggests that heightened displays of pain
behavior partially account for the impact of perceived injustice
on disability [58]. However, reasons for the link between perceived
injustice, pain behavior, and disability remain speculative. One
explanation may be that pain behaviors represent a public display
of pain and suffering and, by association, the injustice experienced.
Consequently, pain behavior may be a means to receive validation
for the injustice. It has also been suggested that injustice perceptions contribute to retribution motives, which might compromise
engagement in or response to rehabilitation interventions
[43,62]. In the rehabilitation context, failure to functionally improve may be a powerful means to punish the perceived unjust
behavior of the insurer or the treating clinician. Future research
is needed to determine the motivational factors that underlie increased pain behavior and disability among individuals with
heightened injustice perceptions, and the degree to which these
factors are consciously represented [58].
The present results suggest that interventions targeting anger
may reduce the impact of perceived injustice on pain intensity
and depressive symptoms. At present, limited research has examined the effectiveness of interventions designed to manage anger
reactions in individuals with chronic pain [63]. One study examined the effects of written anger expression in patients with pain
[29]. During 2 sessions (20 minutes each, over 2.5 weeks) participants wrote letters describing their anger, what made them angry,
and a resolution that would reduce their anger. Nine weeks later,
participants who received the anger expression intervention experienced greater improvements on depression and pain compared to
control participants [29]. Future research is needed to replicate this
finding and to examine the long-term effectiveness of written anger expression.
Several additional intervention strategies may be useful. Forgiveness interventions attempt to reduce anger by increasing
empathy and compassion towards perceived offenders, and have
been shown to benefit crime and accident victims [38,66]. However, ongoing pain may serve as an ongoing reminder of the offence, which may impede the forgiveness process [23,24].
Acceptance-based interventions have also been advocated for
problematic anger [21]. Acceptance interventions use metaphor
and experiential exercises to demonstrate how the struggle to control anger may exacerbate suffering. Acceptance-based interventions direct behavior towards achieving valued life goals, rather
than control and avoidance of difficult experiences, such as pain
and anger [37]. Interventions may also need to address the social
context within which perceptions of injustice and anger arise. Recent research suggests that validation techniques might be effective for reducing anger in patients with pain [65]. Future
research is needed to investigate the utility of these interventions
to reduce the impact of perceived injustice and anger among patients with pain.
Several limitations warrant consideration. First, the cross-sectional, correlational design of this study limits the nature of conclusions that can be drawn about the causal and sequential
relations among perceived injustice, anger, and pain outcomes. Future research using paradigms to experimentally induce blame
attributions (eg, [8,10,50]) will be necessary to clarify the antecedent status of perceptions of injustice in contributing to problematic
health and mental health outcomes in individuals with chronic
pain. The reliance on self-report measures is also a limitation, as
shared method variance may partially account for the observed
relationships.
Ò
W. Scott et al. / PAIN 154 (2013) 1691–1698
The sample consisted of patients with musculoskeletal pain of
longstanding duration that were attending or had previously attended a tertiary care outpatient pain management clinic. Furthermore, the response rate of individuals who were contacted and
agreed to participate was low (20%). This may give rise to concerns
over selection bias and the generalizability of the present results.
However, mean scores on the IEQ in the present study are comparable (ie, within one standard deviation) to mean scores reported
in a sample of patients with persistent pain following whiplash injury attending a secondary care rehabilitation program [54]. Mean
IEQ scores in the present study are likewise comparable to those
reported in a sample of fibromyalgia patients recruited consecutively from primary care [52]. Taken together, these findings increase confidence that the present results are not simply due to
selection bias. The generalizability of the present results to patients with varying pain diagnoses and with pain of less chronic
duration remains an important question for future research.
Despite these limitations, this study represents an important
step forward in understanding the association between perceived
injustice and chronic pain outcomes. Results suggest that anger
intensity and inhibition differentially explain the impact of perceived injustice on pain, depressive symptoms, and disability.
The results highlight the multi-faceted nature of the injustice experience in chronic pain, and suggest that multiple processes may
underlie its impact. Future experimental research may provide insight into underlying mechanisms and intervention approaches to
reduce the impact of perceived injustice in patients with pain.
Conflict of interest statement
The authors have no conflicts of interest.
Acknowledgements
The authors thank Dr. Manon Choinière, Dr. Mark Ware, and
Annie Trépanier for their assistance in accessing the Quebec Pain
Registry for participant recruitment. This research was supported
by funds from the Canadian Institutes for Health Research, le Fonds
de la Recherche en Santé du Québec, and l’Institut de Recherche
Robert-Sauvé en Santé et en Sécurité du Travail.
References
[1] Averill JR. Studies on anger and aggression: implications for theories of
emotion. Am Psychol 1983;38:1145.
[2] Baron RM, Kenny DA. The moderator–mediator variable distinction in social
psychological research: conceptual, strategic, and statistical considerations. J
Pers Soc Psychol 1986;51:1173–82.
[3] Borteyrou X, Bruchon-schweitzer M, Spielberger C-D. L’inventaire de colère
trait et de colère état de cd spielberger, le staxi-2. Une adaptation française.
L’Encéphale 2008;34:249–55.
[4] Bruehl S, Chung OY, Burns JW. Anger expression and pain: an overview of
findings and possible mechanisms. J Behav Med 2006;29:593–606.
[5] Bruehl S, Chung OY, Burns JW, Diedrich L. Trait anger expressiveness and paininduced beta-endorphin release: support for the opioid dysfunction
hypothesis. PAINÒ 2007;130:208.
[6] Burns JW. Anger management style and hostility: predicting symptom-specific
physiological reactivity among chronic low back pain patients. J Behav Med
1997;20:505–22.
[7] Burns JW, Bruehl S. Anger management style, opioid analgesic use, and chronic
pain severity: a test of the opioid-deficit hypothesis. J Behav Med
2005;28:555–63.
[8] Burns JW, Bruehl S, Caceres C. Anger management style, blood pressure
reactivity, and acute pain sensitivity: evidence for ‘‘trait situation’’ models.
Ann Behav Med 2004;27:195–204.
[9] Burns JW, Bruehl S, Quartana PJ. Anger management style and hostility among
patients with chronic pain: effects on symptom-specific physiological
reactivity during anger-and sadness-recall interviews. Psychosom Med
2006;68:786–93.
[10] Burns JW, Holly A, Quartana P, Wolff B, Gray E, Bruehl S. Trait anger
management style moderates effects of actual (‘‘state’’) anger regulation on
symptom-specific reactivity and recovery among chronic low back pain
patients. Psychosom Med 2008;70:898–905.
1697
[11] Burns JW, Johnson BJ, Devine J, Mahoney N, Pawl R. Anger management style
and the prediction of treatment outcome among male and female chronic pain
patients. Behav Res Ther 1998;36:1051–62.
[12] Burns JW, Quartana PJ, Bruehl S. Anger inhibition and pain: conceptualizations,
evidence and new directions. J Behav Med 2008;31:259–79.
[13] Butler EA, Lee TL, Gross JJ. Emotion regulation and culture: are the social
consequences of emotion suppression culture-specific? Emotion 2007;7:30.
[14] Carballeira Y, Dumont P, Borgacci S, Rentsch D, Tonnac N, Archinard M,
Andreoli A. Criterion validity of the French version of patient health
questionnaire (PHQ) in a hospital department of internal medicine. Psychol
Psychother 2007;80:69–77.
[15] Carson JW, Keefe FJ, Goli V, Fras AM, Lynch TR, Thorp SR, Buechler JL.
Forgiveness and chronic low back pain: a preliminary study examining the
relationship of forgiveness to pain, anger, and psychological distress. J Pain
2005;6:84–91.
[16] Crombez G, Vlaeyen JWS, Heuts PHTG, Lysens R. Pain-related fear is more
disabling than pain itself: evidence on the role of pain-related fear in chronic
back pain disability. PAINÒ 1999;80:329–39.
[17] Dahlen ER, Martin RC. The experience, expression, and control of anger in
perceived social support. Pers Individ Dif 2005;39:391–401.
[18] DeGood DE, Kiernan B. Perception of fault in patients with chronic pain. PAINÒ
1996;64:153–9.
[19] Downie WW, Leatham PA, Rhind VM, Wright V, Branco JA, Anderson JA.
Studies with pain rating-scales. Ann Rheum Dis 1978;37:378–81.
[20] Duckro PN, Chibnall JT, Tomazic TJ. Anger, depression, and disability: a path
analysis of relationships in a sample of chronic posttraumatic headache
patients. Headache 2005;35:7–9.
[21] Eifert GH, McKay M, Forsyth JP. Act on life not on anger. Oakland: New
Harbinger Publications; 2006.
[22] Erwin BA, Heimberg RG, Schneier FR, Liebowitz MR. Anger experience and
expression in social anxiety disorder: pretreatment profile and predictors of
attrition and response to cognitive-behavioral treatment. Behav Ther
2003;34:331–50.
[23] Exline JJ, Baumeister RF. Expressing forgiveness and repentance. Forgiveness
2000:133–55.
[24] Exline JJ, Worthington Jr EL, Hill P, McCullough ME. Forgiveness and justice: a
research agenda for social and personality psychology. Pers Soc Psychol Rev
2003;7:337–48.
[25] Fejer R, Jordan A, Hartvigsen J. Categorising the severity of neck pain:
establishment of cut-points for use in clinical and epidemiological research.
PAINÒ 2005;119:176–82.
[26] Finlay-Jones R, Brown GW. Types of stressful life event and the onset of
anxiety and depressive-disorders. Psychol Med 1981;11:803–15.
[27] Fritz MS, MacKinnon DP. Required sample size to detect the mediated effect.
Psychol Sci 2007;18:233–9.
[28] Gauthier N, Thibault P, Adams H, Sullivan MJL. Validation of a French-Canadian
version of the pain disability index. Pain Res Manag 2008;13:327.
[29] Graham JE, Lobel M, Glass P, Lokshina I. Effects of written anger expression in
chronic pain patients: making meaning from pain. J Behav Med
2008;31:201–12.
[30] Hagerty BM, Williams A. The effects of sense of belonging, social support,
conflict, and loneliness on depression. Nurs Res 1999;48:215–9.
[31] Kendler KS, Hettema JM, Butera F, Gardner CO, Prescott CA. Life event
dimensions of loss, humiliation, entrapment, and danger in the prediction of
onsets of major depression and generalized anxiety. Arch Gen Psychiatry
2003;60:789–96.
[32] Kerns RD, Rosenberg R, Jacob MC. Anger expression and chronic pain. J Behav
Med 1994;17:57–67.
[33] Kroenke K, Spitzer RL, Williams JBW. The PHQ-9. J Gen Intern Med
2001;16:606–13.
[34] Lee B-T, Seok J-H, Lee B-C, Cho SW, Yoon B-J, Lee K-U, Chae J-H, Choi I-G, Ham
B-J. Neural correlates of affective processing in response to sad and angry facial
stimuli
in
patients
with
major
depressive
disorder.
Prog
Neuropsychopharmacol Biol Psychiatry 2008;32:778–85.
[35] León I, Hernández JA, Rodríguez S, Vila J. When head is tempered by heart:
heart rate variability modulates perception of other-blame reducing anger.
Motiv Emotion 2009;33:1–9.
[36] Lyons R, Sullivan M. Curbing loss in illness and disability. In: Harvey J, editor.
Perspectives on personal and interpersonal loss. New York: Taylor & Francis;
1998. p. 579–605.
[37] McCracken LM. Contextual cognitive-behavioral therapy for chronic
pain. Seattle: IASP Press; 2005.
[38] McCullough ME. Forgiveness as human strength: theory, measurement, and
links to well-being. J Soc Clin Psychol 2000;19:43–55.
[39] McParland J, Hezseltine L, Serpell M, Eccleston C, Stenner P. An investigation of
constructions of justice and injustice in chronic pain: a q-methodology
approach. J Health Psychol 2011;16:873–83.
[40] McParland JL, Whyte A. A thematic analysis of attributions to others for the
origins and ongoing nature of pain in community pain sufferers. Psychol
Health Med 2008;13:610–20.
[41] Mikula G. Testing an attribution-of-blame model of judgments of injustice. Eur
J Soc Psychol 2003;33:793–811.
[42] Mikula G, Scherer KR, Athenstaedt U. The role of injustice in the elicitation of
differential emotional reactions. Pers Soc Psychol B 1998;24:769–83.
[43] Miller DT. Disrespect and the experience of injustice. Annu Rev Psychol
2001;52:527–53.
1698
Ò
W. Scott et al. / PAIN 154 (2013) 1691–1698
[44] Mischoulon D, Dougherty DD, Bottonari KA, Gresham RL, Sonawalla SB,
Fischman AJ, Fava M. An open pilot study of nefazodone in depression with
anger attacks: relationship between clinical response and receptor binding.
Psychiatry Res 2002;116:151–61.
[45] Montada L. Attribution of responsibility for losses and perceived injustice. In:
Montada L, Sigrun-Heide F, Lerner MJ, editors. Life crises and experiences of
loss in adulthood. Hillsdale: Lawrence Erlbaum and Associates; 1992. p.
133–61.
[46] Morrow-Howell N. The m-word—multicollinearity in multiple-regression. Soc
Work Res 1994;18:247–51.
[47] Okifuji A, Turk DC, Curran SL. Anger in chronic pain: investigations of anger
targets and intensity. J Psychosom Res 1999;47:1–12.
[48] Painuly N, Sharan P, Mattoo SK. Relationship of anger and anger attacks with
depression. Eur Arch Psychiatry Clin Neurosci 2005;255:215–22.
[49] Paykel ES, Myers JK, Dienelt MN, Klerman GL, Lindenthal JJ, Pepper MP. Life
events and depression. A controlled study. Arch Gen Psychiatry
1969;21:753–60.
[50] Quartana PJ, Yoon KL, Burns JW. Anger suppression, ironic processes and pain. J
Behav Med 2007;30:455–69.
[51] Quigley BM, Tedeschi JT. Mediating effects of blame attributions on feelings of
anger. Pers Soc Psychol B 1996;22:1280–8.
[52] Rodero B, Luciano JV, Montero-Marín J, Casanueva B, Palacin JC, Gili M, López
del Hoyo Y, Serrano-Blanco A, Garcia-Campayo J. Perceived injustice in
fibromyalgia: psychometric characteristics of the injustice experience
questionnaire and relationship with pain catastrophising and pain
acceptance. J Psychosom Res 2012;73:86–91.
[53] Scott W, Sullivan M. Perceived injustice moderates the relationship between
pain and depressive symptoms among individuals with persistent
musculoskeletal pain. Pain Res Manag 2012;17:335–40.
[54] Scott W, Trost Z, Milioto M, Sullivan M. Further validation of a measure of
injury-related injustice perceptions to identify risk for occupational disability:
a prospective study of individuals with whiplash injury. J Occup Rehabil 2013.
http://dx.doi.org/10.1007/s10926-013-9417-1.
[55] Spielberger CD. State-trait anger expression inventory: professional
manual. Odessa: Psychological Assessment Resources; 1988.
[56] Sullivan MJL, Adams H, Horan S, Maher D, Boland D, Gross R. The role of
perceived injustice in the experience of chronic pain and disability: scale
development and validation. J Occup Rehabil 2008;18:249–61.
[57] Sullivan MJL, Adams H, Rhodenizer T, Stanish WD. A psychosocial risk factor–
targeted intervention for the prevention of chronic pain and disability
following whiplash injury. Phys Ther 2006;86:8–18.
[58] Sullivan MJL, Davidson N, Garfinkel B, Siriapaipant N, Scott W. Perceived
injustice is associated with heightened pain behaviour and disability in
individuals with whiplash injuries. Psychol Inj Law 2009;2:238–47.
[59] Sullivan MJL, Thibault P, Simmonds MJ, Milioto M, Cantin AP, Velly AM. Pain,
perceived injustice and the persistence of post-traumatic stress symptoms
during the course of rehabilitation for whiplash injuries. PAINÒ
2009;145:325–31.
[60] Sullivan MJL, Thorn B, Haythornthwaite JA, Keefe F, Martin M, Bradley LA,
Lefebvre JC. Theoretical perspectives on the relation between catastrophizing
and pain. Clin J Pain 2001;17:52–64.
[61] Tait RC, Chibnall JT, Krause S. The pain disability index: psychometric
properties. PAINÒ 1990;40:171–82.
[62] Tripp TM, Bies RJ, Aquino K. A vigilante model of justice: revenge,
reconciliation, forgiveness, and avoidance. Soc Justice Res 2007;20:10–34.
[63] Trost Z, Vangronsveld K, Linton SJ, Quartana PJ, Sullivan MJL. Cognitive
dimensions of anger in chronic pain. PAINÒ 2012;153:515–7.
[64] Van Praag H. Serotonin-related, anxiety/aggression-driven, stressor-precipitated
depression. A psycho-biological hypothesis. Eur Psychiatry 1996;11:57–67.
[65] Vangronsveld KL, Linton SJ. The effect of validating and invalidating
communication on satisfaction, pain and affect in nurses suffering from low
back pain during a semi-structured interview. Eur J Pain 2011;16:239–46.
[66] Wade NG, Worthington EL. In search of a common core: a content analysis of
interventions to promote forgiveness. Psychotherapy 2005;42:160–77.
[67] Youngren MA, Lewinsohn PM. The functional relation between depression and
problematic interpersonal behavior. J Abnorm Psychol 1980;89:333.