The Clinical Journal of Pain
19:217–224 © 2003 Lippincott Williams & Wilkins, Inc., Philadelphia
Spousal Responses Are Differentially Associated With Clinical
Variables in Women and Men With Chronic Pain
*Roger B. Fillingim, PhD, †Daniel M. Doleys, PhD, ‡Robert R. Edwards, MA, MSPH, and
‡Daniel Lowery, MA
*University of Florida College of Dentistry and North Florida South Georgia VA Health System, Gainesville, Florida; †Pain and
Rehabilitation Institute, Birmingham, Alabama; and ‡University of Alabama at Birmingham, Birmingham, Alabama
Abstract:
Objectives: Spousal responses have been related to clinical variables in patients
with chronic pain. For example, solicitous responses from spouses have been associated with greater levels of pain and disability among patients with chronic pain.
However, few investigators have determined whether spousal solicitousness produces
different effects in women versus men with chronic pain. The present study examined
pain reports, medication use, psychosocial factors, functional measures, and pain tolerance in patients with chronic pain.
Methods: Subjects included 114 female and 213 male chronic pain patients, who
described their spouses as either high or low in solicitousness on the Multidimensional
Pain Inventory. Measures of pain severity, affective distress, physical function, medication use, and pain tolerance were examined in women and men with high versus low
scores on spousal solicitousness.
Results: Among males only, high spousal solicitousness was associated with greater
numerical ratings of pain and greater self-reported disability compared with patients
with low solicitous spouses. Among females only, the high spousal solicitousness
patients showed lower pain tolerance, greater pain-related interference, poorer performance on functional tasks (eg, timed walking, lifting, and carrying tasks), and greater
use of opioid medications. In both women and men, spousal solicitousness was associated with higher scores on the MPI pain severity scale.
Discussion: These results extend previous findings demonstrating a relationship
between spousal responses and patients’ adjustment to pain; however, the pattern of
these effects appears to be moderated by the sex of the patient. Implications for
assessment and treatment of chronic pain are discussed.
Key Words: sex differences, gender differences, spousal responses, pain tolerance,
pain-related disability
chronic pain. For example, spousal presence was associated with greater pain reports only for patients who described their spouses as solicitous.1 Relatedly, Gil et al2
found that higher patient ratings of satisfaction with social support were associated with higher levels of pain
behavior. Similar associations between spousal solicitousness and increased pain, disability, and pain behavior
have been reported by other investigators.3–5 More recently, solicitous spousal responses were associated with
higher levels of physical disability in patients with multiple sclerosis.6 These data are typically interpreted as supporting operant theories of pain behavior, which propose
The biopsychosocial model suggests that pain is influenced by social and environmental factors. In support of
this, spousal responses to patients’ pain behavior have
been associated with adjustment among patients with
Received June 17, 2002; accepted June 17, 2002.
Supported by NIH/NINDS grant NS41670. This material is also the
result of work supported with resources and the use of facilities at the
Malcom Randall VA Medical Center, Gainesville, FL.
Corresponding author: Roger B. Fillingim, PhD, University of
Florida College of Dentistry Public Health Services and Research, 1600
SW Archer Road, Room D8-44A, P.O. Box 100404, Gainesville, FL
32610. E-mail: [email protected].
217
218
Fillingim et al
that spousal responses can serve as positive reinforcers
of pain behavior leading to increased pain behavior and
disability. There is evidence that the association of social
support with pain may vary across patient populations.
For example, one study reported that perceived spousal
solicitousness was correlated with higher pain intensity
among low back pain patients, but the reverse was true
for patients with sickle cell disease and arthritis.7
Another factor known to influence pain perception and
adjustment to pain is the sex of the patient. Multiple
studies indicate sex differences in pain perception and in
the experience of clinical pain, with women typically
demonstrating increased pain sensitivity and greater risk
for experiencing clinical pain.8–11 Given these sex differences, it is important to determine whether spousal
responses influence male and female pain patients differently. Romano and colleagues12 reported that wives of
male pain patients reported lower marital satisfaction,
which was related to increased patient depression, but
these relationships were not apparent among husbands of
female patients. Patient ratings of spouse and significant
other solicitousness were positively correlated with pain
and pain-related interference for male but not female
patients who were married and for female but not male
patients who were unmarried.13 Another study demonstrated that for male pain patients, higher perceived negative responses from spouses were associated with greater
pain severity and interference of pain in daily living,
while these relationships were not present in female
patients.14
Taken together, prior research indicates that both
spousal responses and sex can influence adjustment to
pain, and some evidence suggests that spousal responses
to chronic pain and the effects of these responses may
differ as a function of the patient’s sex. However, to date
no studies have determined whether spousal solicitousness is associated differentially with self-reported pain,
pain-related function, and pain tolerance among female
compared with male patients. This study examined the
influence of spousal solicitousness on these pain-related
variables in women and men experiencing chronic noncancer pain.
MATERIALS AND METHODS
Subjects
A total of 317 (203 male, 114 female) patients from a
multidisciplinary pain program participated in the study.
These were predominantly industrially injured workers
who were disabled by pain. Primary sites of pain included: low back (54%), leg (11%), shoulder or arm
(11%), cervical (6%), and other (18%). The typical patient reported primary pain in the lower back, with raThe Clinical Journal of Pain, Vol. 19, No. 4, 2003
diation into one or both legs. All patients underwent a
comprehensive pretreatment evaluation, which consisted
of several questionnaires, an assessment of ischemic pain
tolerance, and measures of physical function (see below).
Psychologic and pain measures
Medical and demographic data
Medical and demographic information was obtained
through patient interview and review of the medical records that were available. The following data were collected: age, sex, race, duration of pain, primary and secondary sites of pain. Also self-reported use of opioid
medication for pain was determined and confirmed by
medical record review.
Multidimensional Pain Inventory (MPI)15
The MPI consists of 13 scales divided into 3 sections
assessing: the impact of pain on the patient’s life, responses of significant others to the communication of
pain, and the daily activity level of the patient. The reliability and validity of the MPI have been wellestablished.16,17 In the present study, the MPI was used
in two ways. First, groups high (High SR) versus low
(Low SR) on spousal solicitousness were created by performing a median split on the Solicitous Responses subscale. This scale has been a commonly used measure of
spousal solicitousness in previous research.3,4 Second,
the subscales of pain severity, pain interference, affective
distress, perceived life control, and general activity level
were used as indices of adjustment to chronic pain. MPI
data were scored using the MPI computerized scoring
program; subscale scores are presented as T-scores.
Beck Depression Inventory (BDI)18
The BDI is a widely used, 21-item, self-report measure assessing common cognitive, affective, and vegetative symptoms of depression. Research evaluating the
psychometric properties of the BDI suggests that it
shows excellent reliability and validity as an index of
depression.19 Because chronic pain patients often endorse somatic symptoms assessed by the BDI, which
may artificially inflate their scores,20 the BDI was separated into its 13-item cognitive affective subscale and its
8-item somatic-performance subscale.21
Oswestry Disability Questionnaire (ODQ)22
The ODQ is a 10-item, self-report scale assessing the
extent to which functional daily activities are restricted
by pain. The ODQ demonstrates adequate reliability and
validity, and is a frequently used measure of pain-related
disability.23
McGill Pain Questionnaire (MPQ)24
The MPQ consists of 20 groups of single-word pain
descriptors with the words in each group increasing in
Relationships Among Sex, Spousal Responses, and Pain
rank order intensity. The sum of the rank values for each
descriptor based on its position in the word set yields an
overall measure of pain termed the Pain Rating Index
(PRI), which was the summary score used in the present
study. The MPQ is among the most widely used measures for rating pain.
State-Trait Anxiety Inventory (STAI)25
Both the state and trait versions of the STAI are selfreport scales consisting of 20 statements evaluating levels of anxiety. Each item is scored on a 4-point Likert
scale, with higher scores representing increasing levels
of state anxiety (recent, situationally-derived symptoms
of anxiety) or trait anxiety (longer standing symptoms of
anxiety). The STAI has consistently demonstrated adequate psychometric properties, and is a commonly used
measure of anxiety in the context of pain research.26,27
Pain ratings
Numerical pain ratings were obtained by asking each
patient to record hourly ratings of pain severity on a 0
(“no pain”) to 10 (“the most severe pain imaginable”)
scale. For the purposes of statistical analysis, hourly pain
ratings for the first 2 days of treatment were averaged to
yield a single overall pain rating.
Functional measures
Though self-report measures of physical function are
more commonly used, behavioral measures of physical
function have been widely recommended for use in assessment of chronic pain patients.28,29 The following
tests of physical function were conducted on all patients.
100-yard timed walking test
Patients were required to walk once around a 100-yard
long indoor circular track as fast as they could. Performance was measured as the number of seconds required
to complete the task, such that a longer time reflects
poorer performance.
Lift and carry tests
Under the supervision of a physical therapist, patients
were asked to lift a box filled with weights from the floor
to waist level. The amount of weight was gradually increased until patients reported being unable to perform
the task. Performance was measured as the greatest
weight lifted from floor to waist. The carry task was
conducted in similar fashion, except patients were instructed to carry the box at waist level across the room
(approx. 10 meters). Performance was measured as the
greatest weight carried.
Push and pull tests
For the push test, patients were instructed to gradually
apply force to their maximum using a dynamometer
219
(Chatillon, Greensboro, NC) that was placed waist high
against a door frame. For the pull test, the dynamometer
was hooked to an eye-bolt at waist height and patients
were instructed to pull with maximum force. Three trials
were performed for each test, and the average of these 3
trials was computed for data analysis.
To develop a composite functional measure based on
the 4 strength tasks, the results of the lift, carry, push, and
pull tests were standardized to the same metric (mean ⳱
50, SD ⳱ 10) and averaged for each patient, and this
overall measure was used in subsequent analyses.
Ischemic Pain Tolerance (IPTO)
IPTO was assessed using the modified submaximal
tourniquet procedure as described previously.30,31 This
procedure involves exercising the hand as blood flow to
the arm is occluded, evoking ischemic pain. Following
determination of subjects’ maximum grip strength, the
arm was exsanguinated by elevating it above heart level
for 30 seconds, after which a standard blood pressure
cuff was positioned proximal to the elbow of the dominant arm and inflated to 200 mm Hg. Subjects then performed 20 handgrip exercises of 2-second duration at
4-second intervals at 30% of maximum grip strength.
Subjects were instructed to continue until the perceived
pain became intolerable or until they reached the uninformed cutoff time of 20 minutes. IPTO was operationalized as the number of seconds to pain tolerance
following cuff inflation. Patients with a history of hypertension, coronary artery disease, or upper extremity pain
of unknown etiology were excluded from the test. Patients who were not excluded based on these criteria
completed the procedure prior to beginning the interdisciplinary treatment program.
Data reduction and analysis
Data are presented as means and standard errors. As
data points were missing for a number of subjects, the
number of individuals included in the analysis is indicated in each table. The significance of group differences
in continuous variables was determined using Analysis
of Variance (ANOVAs). Group differences on nominal
variables were assessed using ␹2 tests. Significance level
was set to P < 0.05 for each analysis. All analyses were
carried out using Statistical Analysis System (SAS)
software.
RESULTS
The median split yielded 60 women in the High Solicitous Response (High SR) and 54 in the Low Solicitous Response (Low SR) group, and 108 men in the High
SR and 95 men in the Low SR groups. Demographic and
descriptive variables are presented for each group in
The Clinical Journal of Pain, Vol. 19, No. 4, 2003
220
Fillingim et al
Table 1. As can be seen, no SR group differences
emerged on age, pain duration, primary pain location, or
number of pain sites. Women reported significantly more
pain sites than men, while men had more previous surgeries. The High SR group had a greater proportion of
non-white patients compared with the Low SR group;
therefore, race was used as a covariate in subsequent
analyses. No other demographic differences emerged. As
expected, the High SR group had significantly higher
scores on the Spousal Solicitousness scale of the MPI,
and High SR patients also reported more Distracting Responses but not more Punishing Responses relative to
their Low SR counterparts.
Self-report measures
Pain measures
Three different self-report measures of clinical pain
were obtained, and these are presented in Table 2. For
hourly pain ratings averaged over the first 2 days of
treatment with the pain diary, a main effect of SR Group
emerged (P < 0.005), such that High SR patients provided higher hourly pain ratings compared with low SR
patients. However, separate analyses by sex indicated
that this effect was significant only for men (P < 0.005).
Main effects of both sex (P < 0.05) and SR group (P <
0.001) were found for the Pain Severity subscale of the
MPI, with women and High SR patients of both sexes
reporting greater pain. Sex differences also emerged for
the McGill PRI (P < 0.005), with women scoring higher
than men, but no effect of SR group was observed. Thus,
depending on the pain measure, both SR Group and sex
were associated with pain reports.
Affective measures
Women had significantly higher scores on the Affective Distress subscale of the MPI (P < 0.05), and slightly
but not significantly higher scores on the CognitiveAffective subscale of the BDI (P ⳱ 0.07) and the Trait
Anxiety scale (P ⳱ 0.06). No sex differences emerged
for State Anxiety or the BDI-Somatic scale, and no effect
of SR Group was found for any of the affective measures
(Ps > 0.10).
Self-Report disability measures
Analysis of the ODQ revealed no significant effects of
sex or SR group, but there was a marginal sex X group
interaction (P ⳱ 0.07). Separate analyses by sex revealed that among males the High SR group reported
significantly greater disability than the low SR group (P
< 0.05), but no effect was seen in females. Overall, there
was a main effect of SR Group on the Interference and
General Activity scales of the MPI, such that High SR
groups reported higher interference and lower activity (P
< 0.005). However, sex-specific analysis revealed that
these effects were significant only among women (Ps ⱕ
0.05), though the effect on interference approached significance in men (P ⳱ 0.08).
Pain tolerance
For ischemic pain tolerance time, a main effect of SR
group emerged (P ⳱ 0.05); however, there was also a
significant group X sex interaction (Fig. 1) (P < 0.05).
Separate analyses by sex indicated that High SR females
had significantly lower pain tolerance than Low SR females (P < 0.01), but no effect of SR Group was present
for males (P > 0.5). There was no main effect of sex
(P > 0.2).
Functional testing
For the Walking Time measure, the sex X SR group
interaction approached significance (P < 0.07). Sexspecific analyses revealed that among females the High
SR group required significantly longer to complete the
walk than the Low SR group (P < 0.05), but for males no
effect of SR group was evident (Ps > 0.5). Similarly, for
the composite strength measure, High SR females displayed poorer performance than Low SR females (Ps ⱕ
0.05), while no SR group differences emerged for males
(Ps > 0.5) (Fig. 2).
TABLE 1. Demographic and descriptive characteristics by sex and SR group. Results are
presented as means and standard deviations unless otherwise indicated
Women
Age
Duration of pain (years)
Race (% white)*
Pain site (%LBP)
Solicitous responses score*
Distracting responses score*
Punishing responses score
High SR
(n ⳱ 60)
Low SR
(n ⳱ 54)
High SR
(n ⳱ 108)
Low SR
(n ⳱ 95)
41.0 (8.9)
2.4 (2.4)
66.2
53.3
58.2 (4.4)
56.6 (9.6)
49.8 (8.3)
40.5 (7.9)
1.9 (1.6)
83.3
44.4
43.9 (5.5)
50.2 (9.5)
50.8 (9.7)
39.7 (8.8)
2.2 (2.1)
74
60
60.0 (3.6)
58.3 (8.8)
50.0 (9.0)
40.3 (8.7)
2.5 (2.0)
92
57.5
44.8 (7.0)
51.7 (9.3)
51.4 (8.5)
*High vs. Low SR groups differ, P < 0.05.
The Clinical Journal of Pain, Vol. 19, No. 4, 2003
Men
Relationships Among Sex, Spousal Responses, and Pain
221
TABLE 2. Pain severity, psychosocial variables and self-reported disability measures by sex
and SR group. Results are presented as means and standard deviations unless
otherwise indicated
Women
Pain diary§
MPI pain severity*†
McGill PRI†
BDI-somatic
BDI-affective
MPI-affective distress†
Trait anxiety
State anxiety
MPI interference‡
MPI general activity‡
Oswestry§
Narcotic use (%)‡
Men
High SR
(n ⳱ 60)
Low SR
(n ⳱ 54)
High SR
(n ⳱ 108)
Low SR
(n ⳱ 95)
6.8 (1.4)
57.2 (6.1)
40.7 (14.0)
10.1 (3.6)
14.3 (7.7)
54.2 (6.7)
53.2 (10.4)
53.2 (12.0)
56.7 (5.7)
43.3 (6.9)
56.1 (12.6)
50
6.5 (1.3)
53.9 (7.5)
38.8 (12.3)
9.5 (3.6)
13.9 (8.3)
54.0 (8.2)
54.8 (11.4)
56.5 (11.1)
53.5 (7.9)
47.5 (9.2)
56.3 (11.0)
25
6.9 (1.3)
55.4 (7.2)
35.6 (12.0)
9.8 (3.2)
12.1 (7.8)
52.7 (7.9)
52.1 (10.4)
53.8 (10.6)
56.1 (5.9)
42.1 (8.1)
56.5 (11.7)
48
6.2 (1.6)
51.3 (9.6)
34.8 (11.0)
9.3 (3.5)
12.5 (7.6)
51.2 (7.9)
51.5 (10.7)
54.8 (12.2)
54.6 (5.4)
43.8 (7.5)
51.8 (12.7)
43
*Effect of SR group for both sexes; †Overall effect of sex, ‡SR group effect for women only; §SR group effect
for men only.
Opioid medication
A ␹2 analysis revealed that a greater proportion of
High SR females was using opioid medications compared with Low SR females (P < 0.05), while opioid use
was similar for High versus Low SR males (P > 0.40)
(Table 2).
related interference, lower general activity, and poorer
performance on functional measures (ie, walk time,
strength measures). Moreover, women who reported
high spousal solicitousness had significantly lower pain
tolerance compared with women reporting low solicitous
responses. This effect was not observed in men. Finally,
DISCUSSION
The results of this research suggest that patients’ perceptions of spousal responses are related to pain and
function in both female and male patients with chronic
pain; however, the associations differed for women and
men. For both sexes, spousal solicitousness was related
to greater clinical pain, but this effect appeared more
consistent for men. Among men but not women, spousal
solicitousness was also associated with greater selfreported disability on the Oswestry scale. For women,
spousal solicitousness was associated with greater pain-
FIGURE 1. Ischemic pain tolerance for women and men in the
opioid and non-opioid groups. Results are presented as means
and standard errors.
FIGURE 2. Functional measures for women and men in the High
versus Low SR groups. Means and standard errors for the composite strength index (top panel) and time to walk 100 meters
(bottom panel) are presented.
The Clinical Journal of Pain, Vol. 19, No. 4, 2003
222
Fillingim et al
spousal solicitousness was related to increased likelihood
of opioid use among women but not men. These results
support and extend previous research examining associations between spousal responses and adjustment to pain.
Several prior studies reported that greater perceived
spousal support was associated with greater pain, pain
behavior, or pain-related disability.1,3–5,32–35 In addition
to demonstrating similar associations between spousal
solicitousness and pain-related variables, our findings
suggest that the relationship between spousal responses
and adjustment to pain may differ for female and male
patients.
Several issues must be considered in interpreting these
findings. First, we have presented the data based on the
sex of the patient; however, this is inevitably confounded
with the sex of the spouse, since male and female patients had female and male spouses, respectively. Thus,
the observed relationships between spousal responses
and pain-related variables in female versus male patients
could be attributed to the sex of the spouse rather than the
sex of the patient. That is, it may be that the solicitous
responses emitted by women (ie, wives) affect patients
differently than the solicitous responses displayed by
men (ie, husbands). It is also important to remember that
the direction of effects cannot be determined in this
cross-sectional study. For instance, behavioral theories
suggest spouses’ solicitous responses produce increased
pain behavior and disability via positive reinforcement.
If this were the case, then female patients appear to be
more influenced by this process, since more differences
were observed between High SR and Low SR females
than males. These putative greater effects of reinforcement on pain-related behaviors in women could be due to
increased susceptibility to reinforcement in the female
patients or greater reinforcement value of the husbands’
responses.
On the other hand, it is plausible that higher spousal
solicitousness occurs in response to greater pain and disability exhibited by the patient. Applying this explanation to the present findings, female patients may need to
display greater pain-related disability to elicit supportive
responses from their husbands, while the wives of male
patients require less pain-related disability to provide increased support. Indeed, there is evidence that females
interpret nonverbal signals more accurately than males,36
and females show enhanced physiological and perceptual
responses to facial expressions relative to males.37–41 In
a recent study, females reported a significantly greater
number of familial pain models than males, suggesting
that females may be more aware of pain in others.42 If
women demonstrate greater awareness of nonverbal cues
and familial pain, then, compared with husbands of female patients, wives of male patients may provide inThe Clinical Journal of Pain, Vol. 19, No. 4, 2003
creased support in response to relatively subtle cues of
increased pain. In a recent study of spousal responses to
pain behavior in patients with osteoarthritis (OA), Keefe
and colleagues (unpublished data) found that, compared
with husbands of female OA patients, male OA patients’
wives were more likely to display facilitative behavior
(ie, approval, agreement, and support) before and after
the patients exhibited pain behavior. Thus, it seems quite
plausible that some of the findings described above could
be attributed to the different responses of husbands versus wives.
Another interesting aspect of these findings is the nature of the variables that differed as a function of SR
Group in women and men. Specifically, high spousal
solicitousness was associated with greater self-reported
disability only for men, and the association between
spousal solicitousness and reported pain severity appeared to be more consistent for men. For women, however, self-reported activity level, functional performance,
pain tolerance, and opioid use were associated with spousal responses, but not in men. Thus, solicitous responding appears to be more strongly associated with selfreport measures in men and with more objective
functional or behavioral measures in women. These latter
measures may reflect more observable signs of pain,
which might be particularly important in eliciting solicitous responses from husbands, who may be less cognizant of more subtle manifestations of pain, as discussed
above. Similar to the present findings related to pain
tolerance, an association between spousal solicitousness
and experimental pain responses has been previously reported. Flor and colleagues43 found that chronic pain
patients who rated their spouses as high in solicitousness
showed lower cold pressor pain tolerance when the
spouse was present versus absent, while spouse presence
had no impact on pain tolerance in patients with low
solicitous spouses. Interestingly, in that study 76%
(13/17) of the patients were female. The present results
extend these findings, in that pain tolerance was related
to solicitousness only among female patients, and in our
study the spouses were not present during testing. Inspection of the Flor et al data43 suggests that the pain
tolerance of the high solicitousness group was substantially higher than for the low solicitousness group, when
collapsed across spouse conditions. This would contradict the present results; however, the authors do not mention this finding.
Interesting, too, in the present study is the absence of
an association between affective measures and spousal
solicitousness in women or men. This is somewhat surprising, given that the other clinical variables associated
with spousal responses in women and men are frequently
accompanied by emotional distress. This may be due to
Relationships Among Sex, Spousal Responses, and Pain
the nature of the scale used to assess spousal solicitousness, which asks patients to describe their spouses’ responses specifically to the patients’ pain.
Some prior work has indicated sex differences in
spousal responses to chronic pain. A stronger relationship has been reported between patient and spouse dysfunction when patients are male (and spouses female)
than when patients are female (and spouses male),44 suggesting that wives of male patients are more affected by
their husbands’ pain and dysfunction than are husbands
of female patients. In addition, marital dissatisfaction
was associated with greater pain, disability, and psychologic distress in female but not male chronic pain patients.45 Relatedly, punishing spousal responses were associated with increased pain and interference among
male patients, but this association was not present for
female patients.14 In contrast, spousal criticism contributed to increased affective distress in both women and
men with rheumatoid arthritis.46 Because punishing responses have been associated with pain-related variables
in these prior studies, we re-ran our analyses using the
Punishing Responses scale from the MPI as a covariate,
and the results were not affected. Thus, solicitous, or
excessively positive, responses were related to pain adjustment in a sex-dependent manner, independent of
negative spousal responses.
The potential practical implications of these findings
deserve mention. First, these results serve as a reminder
that chronic pain influences and is influenced by the
social environment, and it should not be surprising that
the social correlates of pain-related adjustment may be
different for women and men. For example, a recent
population survey found that women reported seeking
social support as a pain coping response more frequently
than men.47 Thus, social factors may be particularly important moderators of women’s’ pain experiences. Based
on behavioral theory, these findings suggest that spouse
retraining may be more beneficial for female than male
patients, although it is unclear whether the spouse or the
patient is most in need of retraining. Regardless, instruction in providing the type of spousal support associated
with improved function and reduced pain-related disability may be particularly helpful. Previous research has
demonstrated that spouse-assisted cognitive behavioral
treatment is effective for reducing pain and psychologic
disability in patients with arthritis.48 Whether such treatment produces differential effects in women versus men
remains to be determined. Also, prior research on the
association between spousal responses and pain has generally not reported data separately for women and men;
therefore, it is difficult to know whether any of those
effects were sex-related.
223
This study has a number of strengths, including a relatively large sample size, and the assessment of a variety
of pain-related variables, including functional performance and pain tolerance. However, several limitations
should also be noted. First, we did not collect data on the
psychosocial status of the spouses, which may be an
important moderating factor. Second, while spousal solicitousness was assessed with a commonly used scale,
more sophisticated measurement of multiple dimensions
of social support would have been ideal. Third, some of
the significant findings are of fairly low magnitude (eg,
differences in diary ratings of pain severity) and may not
be clinically meaningful; the large number of statistical
tests increases the likelihood of spurious findings. Nonetheless, these results indicate that spousal solicitousness
may be more strongly associated with decreased function, lower pain tolerance, and increased use of opioids
in women and with pain severity and self-reported disability in men. Future research examining social influences on chronic pain should consider sex as a potential
moderating variable.
REFERENCES
1. Block AR, Kremer EF, Gaylor M. Behavioral treatment of chronic
pain: the spouse as a discriminative cue for pain behavior. Pain.
1980;9:243–252.
2. Gil KM, Keefe FJ, Crisson JE, et al. Social support and pain
behavior. Pain. 1987;29:209–217.
3. Flor H, Kerns RD, Turk DC. The role of spouse reinforcement,
perceived pain, and activity levels of chronic pain patients. J Psychosom Res. 1987;31:251–259.
4. Lousberg R, Schmidt AJ, Groenman NH. The relationship between
spouse solicitousness and pain behavior: searching for more experimental evidence. Pain. 1992;51:75–79.
5. Williamson D, Robinson ME, Melamed B. Pain behavior, spouse
responsiveness, and marital satisfaction in patients with rheumatoid arthritis. Behav Modification. 1997;21:97–118.
6. Schwartz L, Kraft GH. The role of spouse responses to disability
and family environment in multiple sclerosis. Am J Phys Med
Rehab. 1999;78:525–532.
7. Anderson LP, Rehm LP. The relationship between strategies of
coping and perception of pain in three chronic pain groups. J Clin
Psychol. 1984;40:1170–1177.
8. Fillingim RB. Sex, Gender, and Pain. Seattle: IASP Press;
2000.
9. Berkley KJ. Sex differences in pain. Behav Brain Sci. 1997;20:
371–380.
10. Fillingim RB. Sex, gender and pain: women and men really are
different. Curr Rev Pain. 2000;4:24–30.
11. Unruh AM. Gender variations in clinical pain experience. Pain.
1996;65:123–167.
12. Romano JM, Turner JA, Clancy SL. Sex differences in the relationship of pain patient dysfunction to spouse adjustment. Pain.
1989;39:289–295.
13. Flor H, Turk DC, Rudy TE. Relationship of pain impact and significant other reinforcement of pain behaviors: the mediating role
of gender, marital status and marital satisfaction. Pain. 1989;38:
45–50.
14. Burns JW, Johnson BJ, Mahoney N, et al. Anger management
style, hostility and spouse responses: gender differences in predictors of adjustment among chronic pain patients. Pain. 1996;64:
445–453.
The Clinical Journal of Pain, Vol. 19, No. 4, 2003
224
Fillingim et al
15. Kerns RD, Turk DC, Rudy TE. The West Haven-Yale Multidimensional Pain Inventory (WHYMPI). Pain. 1985;23:345–356.
16. Rudy TE, Turk DC, Zaki HS, et al. An empirical taxometric alternative to traditional classification of temporomandibular disorders. Pain. 1989;36:311–320.
17. Turk DC, Rudy TE. The robustness of an empirically derived
taxonomy of chronic pain patients. Pain. 1990;43:27–35.
18. Beck AT. Ward C. H. Mendelson M. et al. An inventory for measuring depression. Arch Gen Psychiatry. 1961;4:561–571.
19. Beck AT. Steer R. A. Garbin M. G. Psychometric properties of the
Beck Depression Inventory: twenty-five years of evaluation. Clin
Psychol Rev. 1988;8:77–100.
20. Novy DM, Nelson DV, Berry LA, et al. What does the Beck
Depression Inventory measure in chronic pain?: a reappraisal.
Pain. 1995;61:261–270.
21. Beck AT, Steer RA. Beck Depression Inventory Manual. San Antonio, TX: Psychological Corporation; 1987.
22. Fairbank JC, Couper J, Davies JB, et al. The Oswestry low back
pain disability questionnaire. Physiotherapy. 1980;66:271–273.
23. Fairbank JC, Pynsent PB. The Oswestry disability index. Spine.
2000;25:2940–2953.
24. Melzack R. The McGill Pain Questionnaire: major properties and
scoring methods. Pain. 1975;1:277–299.
25. Spielberger CD, Gorusch RL, Lushene R, et al. Manual for the
State-Trait Anxiety Inventory (Form Y1). Palo Alto, CA: Consulting Psychologists Press; 1983.
26. Campbell LC, Riley JL III, Kashikar-Zuck S, et al. Somatic, affective, and pain characteristics of chronic TMD patients with
sexual versus physical abuse histories. J Orofac Pain. 2000;14:
112–119.
27. Caumo W, Schmidt AP, Schneider CN, et al. Risk factors for
preoperative anxiety in adults. Acta Anaesthesiol Scand. 2001;45:
298–307.
28. Simmonds MJ, Olson SL, Jones S, et al. Psychometric characteristics and clinical usefulness of physical performance tests in patients with low back pain. Spine. 1998;23:2412–2421.
29. Harding VR, Williams AC, Richardson PH, et al. The development
of a battery of measures for assessing physical functioning of
chronic pain patients. Pain. 1994;58:367–375.
30. Maixner W, Gracely RH, Zuniga JR, et al. Cardiovascular and
sensory responses to forearm ischemia and dynamic hand exercise.
Am J Physiol. 1990;259:R1156–R1163.
31. Moore PA, Duncan GH, Scott DS, et al. The submaximal effort
tourniquet test: its use in evaluating experimental and chronic pain.
Pain. 1979;6:375–382.
The Clinical Journal of Pain, Vol. 19, No. 4, 2003
32. Block AR. Investigation of the response of the spouse to chronic
pain behavior. Psychosom Med. 1981;43:415–422.
33. Gil KM, Keefe FJ, Crisson JE, et al. Social support and pain
behavior. Pain. 1987;29:209–217.
34. Paulsen JS, Altmaier EM. The effects of perceived versus enacted
social support on the discriminative cue function of spouses for
pain behaviors. Pain. 1995;60:103–110.
35. Romano JM, Turner JA, Jensen MP, et al. Chronic pain patientspouse behavioral interactions predict patient disability. Pain.
1995;63:353–360.
36. Hall JA. Gender effects in decoding nonverbal cues. Psych Bull.
1978;85:845–857.
37. Orozco S, Ehlers CL. Gender differences in electrophysiological
responses to facial stimuli. Biol Psychiatry. 1998;44:281–289.
38. Otta E, Folladore AF, Hoshino RL. Reading a smiling face: messages conveyed by various forms of smiling. Perceptual Mot Skills.
1996;82:1111–1121.
39. Lundqvist LO. Facial EMG reactions to facial expressions: a case
of facial emotional contagion? Scand J Psychol. 1995;36:130–141.
40. Boyatzis CJ, Chazan E, Ting CZ. Preschool children’s decoding of
facial emotions. J Genetic Psychol. 1993;154:375–382.
41. Dimberg U, Lundquist LO. Gender differences in facial reactions
to facial expressions. Biol Psychol. 1990;30:151–159.
42. Koutantji M, Pearce SA, Oakley DA. The relationship between
gender and family history of pain with current pain experience and
awareness of pain in others. Pain. 1998;77:25–31.
43. Flor H, Breitenstein C, Birbaumer N, et al. A psychophysiological
analysis of spouse solicitousness towards pain behaviors, spouse
interaction, and pain perception. Beha Ther. 1995;26:255–272.
44. Romano JM, Turner JA, Clancy SL. Sex differences in the relationship of pain patient dysfunction to spouse adjustment. Pain.
1989;39:289–295.
45. Saarijarvi S, Rytokoski U, Karppi SL. Marital satisfaction and
distress in chronic low-back pain patients and their spouses. Clin J
Pain. 1990;6:148–152.
46. Kraaimaat FW, Van Dam-Baggen RM, Bijlsma JW. Association of
social support and the spouse’s reaction with psychological distress
in male and female patients with rheumatoid arthritis. J Rheumatol.
1995;22:644–648.
47. Unruh AM, Ritchie J, Merskey H. Does gender affect appraisal of
pain and pain coping strategies? Clin J Pain. 1999;15:31–40.
48. Keefe FJ, Caldwell DS, Baucom D, et al. Spouse-assisted coping
skills training in the management of osteoarthritic knee pain. Arthritis Care Res. 1996;9:279–291.