Full text of DOI:10.1207/S15327558IJBM0704_03

INTERNATIONAL JOURNAL OF BEHAVIORAL MEDICINE, 7(4), 305–321
Copyright © 2000, Lawrence Erlbaum Associates, Inc.
Gender Differences in Coping
With Musculoskeletal Pain
Giorgio Grossi, Joaquim J. F. Soares, and Ulf Lundberg
Gender differences in coping with musculoskeletal pain were cross-sectionally inves-
tigated, using questionnaires (Coping Strategies Questionnaire), in 446 Swedish pa-
tients (mean age 46 years, 72% women) seeking care for their ailments. Compared to
male patients, women reported more disability, a larger consumption of analgesics,
more work strain, higher levels of posttraumatic stress reactions, a lower self-esteem,
and higher scores for the Coping Strategies Questionnaire indexes: diverting atten-
tion, praying/hoping, catastrophizing, increased behavioural activity, and pain be-
haviours. All gender differences in coping were ruled out in multivariate analyses, ex-
cept for the association between the interaction term Gender × Posttraumatic Stress
Reactions and Catastrophizing. Among women, catastrophizing was positively asso-
ciated with posttraumatic stress reactions, perceived disability, and the number of
previous treatments for pain. No such associations were found among men. Women’s
poorer capacity to cope with musculoskeletal pain is related to higher level of emo-
tional distress, greater disability, and a history of treatments for pain.
Keywords:Women, musculoskeletalpain, coping, workstrain, posttraumaticstress
Gender differences in the experience of pain are receiving increasing attention
(e.g., Unruh, 1996). Women are over-represented in certain pain conditions, for ex-
ample, headache, abdominal pain and facial pain (Rasmussen & Olesen, 1992; Von
Korff, Dworkin, LeResche, et al., 1988); osteoarthritis, rheumatoid arthritis, and
Giorgio Grossi, National Institute for Psychosocial Factors and Health, Stockholm, Sweden and De-
partment of Psychology, Division of Biological Psychology, Stockholm University, Stockholm, Swe-
den; Joaquim J. F. Soares, Centre for Mental Health, Department of Public Health Sciences, Karolinska
Institute, Stockholm, Sweden and Department of Psychology, Uppsala University, Uppsala, Sweden;
Ulf Lundberg, Department of Psychology, Division of Biological Psychology, Stockholm University,
Stockholm, Sweden.
Correspondence concerning this article should be sent to Giorgio Grossi, National Institute for
Psychosocial Factors and Health, Box 230, 171 77 Stockholm, Sweden.

306
GROSSI, SOARES, LUNDBERG
fibromyalgia (Cathey, Wolfe, Kleinheksel, & Hawley, 1986; Verbrugge,
Lepkowski, & Konkol, 1991; Wolfe, 1989), and musculoskeletal pain, particularly
in the neck, shoulder, upper limbs, and hips (e.g., Andersson, Ejlertsson, Leden, &
Rosenberg, 1993; Havsvold & Johnsen, 1993). This is also the case in Sweden,
where approximately 75% of patients seeking care for long-standing
musculoskeletal pain are women (Lagerlöf, 1993; Statistics Sweden, 1999).
Furthermore, in patient samples with pain, women seem to report a more nega-
tive clinical picture than do men. They tend to report pain of a greater intensity
(Andersson et al., 1993) and frequency (Havsvold Johnsen, 1993), tend to con-
sume more health care for pain (e.g., Taylor & Curran, 1985), and are on
sick-leave or in early retirement due to such conditions more often than men (e.g.,
Lagerlöf, 1993). Finally, female patients report higher levels of anxiety and de-
pression than men (Hyyppä, 1987; Jensen, Nygren, Gamberale, Goldie, &
Westerholm, 1994; Magni, Calderion, Rigatti-Luchini, & Merksey, 1990), a dif-
ference that is also seen in the general population (Craighead & Vajk, 1998;
McLean & Woody, 1998).
Coping strategies are known to influence patients’ perceptions of their pain
conditions (e.g., Rosenstiel & Keefe, 1983). Pain coping strategies refer to the
thoughts and actions patients adopt to manage pain and its effects (e.g., Jensen,
Turner, Romano, & Karoly, 1991; Katz, Ritvo, Irvine, & Jackson, 1996). Coping
with pain can be classified into cognitive, including pain management techniques
(e.g., distraction), and behavioural strategies, referring to actions for managing
pain (e.g., taking pain medication; Fernandez, 1986). Further, they can reflect ac-
tive or passive styles (e.g., Snow-Turek, Norris, & Tan, 1996). Active coping (e.g.,
problem solving), involves conscious attempts to relieve or control pain or to func-
tion in spite of it. Passive coping (e.g., praying/hoping) generally entails with-
drawal or giving up and relinquishing control of the pain to something or someone
else. Helplessness and catastrophizing thoughts also characterize the passive cop-
ing style (Nicholas et al., 1992). Catastrophizing refers to negative self-statements
about pain (Kröner-Herwig et al., 1996; Rosenstiel & Keefe, 1983) and has been
linked to medically incongruent back-pain, depression, severe pain, functional im-
pairment, and unfavorable treatment outcomes (Jensen et al., 1991; Keefe, Brown,
Wallstone, & Caldwell, 1989; Reesor & Craig, 1988; Sullivan & D’Eon, 1990).
One possible reason for women’s poorer adaptation to musculoskeletal disorders
may be related to gender differences in the use of coping strategies for pain. This
topic is largely unexplored, but one Swedish study (Jensen et al., 1994) showed
that female patients reported higher levels on the Coping Strategies Questionnaire
subscale catastrophizing than did men, after relevant confounders had been taken
into consideration.
The principle aim of this investigation was, therefore, to examine gender differ-
ences in coping with pain in a sample of patients seeking care for musculoskeletal
disorders. Women were expected to report higher levels of catastrophizing, also

GENDER DIFFERENCES AND PAIN
307
when relevant confounders were statistically controlled. A second aim was to
identify psychosocial (e.g., work strain) and clinical predictors (e.g., pain intensity)
of coping strategies for pain separately for men and women.
METHODS
Patients
Patients ages 18 to 64 years seeking care to alleviate musculoskeletal pain from
general practitioners (GPs) or physiotherapists (PHTs) were approached and asked
to participate in this study. Out of 780 consecutive patients approached, 586 (75%)
agreed to participate (for details, see Grossi, Soares, Ängeslevä, Perski, 1999).
Analyses of variance and χ2 tests revealed no significant differences between the
participants and nonparticipants with respect to age, gender, ethnic background, or
complaints. To make the sample homogeneous in terms of ethnic background, 205
patients of non-Swedish origin were excluded from the analyses. These patients
have been described in greater detail in other publications (Soares & Grossi, in
1999a, in 1999b). Thus, a total of 446 Swedish patients were studied.
The mean age of the sample was 46±11 years and the majority of the partici-
pants (72%) were women. Seventy-one percent were married, 26% had a high edu-
cational level, (university) and 65% had white-collar occupations. The most
common complaint was back pain (32%), followed by neck/shoulder pain (17%)
and complex pain, that is, pain at multiple sites (12%). Fibromyalgia, myalgia,
lower limb pain, hip pain, traumatic injury, pain due to tension, herniated disk, and
”other” together accounted for 25% of the diagnoses. A considerable proportion of
patients (14%) had not been given any clear diagnosis.
Assessment
The patients’ diagnoses were established by GPs in accordance with the Swedish
version of the International Classification of Diseases, Ninth Revision
(Socialstyrelsen, 1987). To avoid interfering with the GPs’ work, no reliability
checks were performed on the diagnoses. The degree of reliability of the diagnoses
is therefore unknown.
Measures
The patients completed a questionnaire designed to assess demographics (i.e. age,
gender, marital status, occupation, work hours per day, employment status, number
of children in the household) and ethnic background (i.e. whether the patient was of
an ethnical origin other than Swedish).

308
GROSSI, SOARES, LUNDBERG
Independent Variables
Clinical characteristics were assessed with the Pain Questionnaire (Arnér, 1984;
Carlsson, 1984), which is currently used as a diagnostic tool at the Pain Manage-
ment Clinic (Karolinska Hospital, Stockholm). It contains questions about pain du-
ration, pain intensity (visual analogue scale scored from 0 to 10), pain complexity
(one or several types of pain), pain frequency, previous somatic treatments and
their effects, use of medication and other morbidity than pain. There is also a dis-
ability index that consists of 15 items (yes–no answers) covering various aspects of
disability due to pain, such as downtime, mobility, and social life. High scores cor-
respond to high disability. Cronbach’s alpha for the disability index was .84.
Work strain. This instrument (Karasek & Theorell, 1990) contains 11 items
(scored from 1 to 4) concerning work demands and control. Indexes for work de-
mands and control were calculated. By dividing demands with control a measure of
work strain was obtained for each patient. In previous research these indexes have
been shown to have high reliability, as expressed by Cronbach alphas of .75 for de-
mands and .76 for control (Theorell, Michélsen, & Nordemar, 1993). Patients who
were not working at the time of data collection were instructed to rate their last em-
ployment with respect to work strain.
Burnout. This instrument consists of 22 items (scored from 1 to 7) that mea-
sure different facets of the burnout syndrome, as expressed by the subscales of
burnout, tension, listlessness, and cognitive difficulties (Melamed, Kushnir, &
Shirom, 1992). Lisspers and Setterlind (1999) validated the questionnaire in a
cross-sectional investigation among 268 Swedes in low and middle white-collar
occupations. They found significant positive associations between Melamed and
coworkers’ overall burnout index and the overall index for the Pines Burnout Mea-
sure (Pines, Aronson, & Kafry, 1981), r = .738, p < .01; the emotional exhaustion
subscale of the Maslach Burnout Inventory (Maslach & Jackson, 1981), r = .737, p
< .01; and the burnout subscale of the Stress Profile (Setterlind & Larsson, 1994), r
= .749, p < .01. For purposes of data reduction, an overall burnout index was calcu-
lated for each patient. High scores correspond to high burnout levels. Alpha coeffi-
cient for this index was .78.
General Health Questionnaire (GHQ–12). The GHQ–12 (Goldberg,
1972, 1985; Goldberg & Williams, 1988) was used to measure minor psychiatric
morbidity. It contains 12 items about symptoms of anxiety and depression. Scores
of 0 to 2 correspond to an absence of psychiatric morbidity (well-being) and scores

GENDER DIFFERENCES AND PAIN
309
3 to 12 correspond to increasing levels of psychiatric morbidity (low well-being).
The Swedish version of the GHQ–12 has a good internal consistency, as expressed
by Cronbach alphas ranging from .86 to .90 (Brenner, Petterson, Levi, & Arnetz,
1988).
Posttraumatic Symptom Scale (PTSS–10). The PTSS–10 (Holen,
1990) was used to assess symptoms of posttraumatic stress. This instrument was
developed to cover Diagnostic and Statistical Manual of Mental Disorders (Amer-
ican Psychiatric Association, 1980) criteria for posttraumatic stress disorder, such
as nightmares, sleep difficulties, impaired memory, irritability, tendency to with-
draw (Holen, 1990). The respondent is instructed to rate the presence or absence of
such symptoms in a yes-or-no form. The PTSS–10 contains 10 items; scores 0 to 2
correspond to no stress reactions, and scores of 3 to 10 correspond to increasing lev-
els of stress reaction. The alpha coefficient in this study was .81.
Rosenberg’ Self-Esteem Scale. Rosenberg’s
Self-Esteem
Scale
(Rosenberg, 1965) was used to investigate the patients’ feelings about themselves
in such regards as self-confidence and intrinsic value. It contains 10 items and the
total score ranges from 10 to 40. For the purpose of this study, the higher the score,
the higher the level of self-esteem. Cronbach alpha for this sample was .84 .
Dependent Variables
Coping strategies for pain were measured with the Coping Strategies Questionnaire
(Rosenstiel & Keefe, 1983). This instrument contains 44 items (scored from 0 to 6)
about eight strategies for coping with pain: reinterpret pain sensations, coping
self-statements, ignore sensations, diverting attention, praying/hoping,
catastrophizing, increased behavioural activities, and pain behaviours. The per-
ceived effectiveness of the coping efforts was rated with two items: control over
pain and ability to decrease pain (self-efficacy beliefs). To present a detailed de-
scription of the gender differences in the use of various coping strategies, male and
female patients were compared with regard to individual subscales rather than
composite measures. The Swedish version of the Coping Strategies Questionnaire
has been shown to have a satisfactory internal consistency, that is, Cronbach alphas
ranging from .70 to .80 (Jensen & Linton, 1993).
Procedure
The study design was cross-sectional. The research was conducted at health care
and physiotherapy centres serving a catchment area of about 250,000 people, situ-

310
GROSSI, SOARES, LUNDBERG
ated in the southwestern part of Stockholm. The participation of the GPs and the
PHTs in the study was based on informed consent. The patients were identified
when visiting the GPs and the PHTs during a period of 15 consecutive days. At this
time, patients were examined by GPs and PHTs in accordance with the Swedish In-
ternational Classification of Diseases, Ninth Revision (Socialstyrelsen, 1987), pro-
vided with the questionnaire, and informed about the study. They were also in-
structed to return the questionnaire by mail after completing it at home. All patients
were volunteers and gave their consent. Confidentiality was guaranteed. The study
was approved by the local Ethical Committee.
Statistical Analyses
In a first set of analyses, factorial analyses of variance and chi-square tests (χ2) were
used to assess differences between male and female patients in terms of
sociodemographic variables, demands and control at work, emotional distress, pain
parameters, and coping strategies. Gender differences in pain duration, de-
mands–control ratio, and the Coping Strategies Questionnaire subscale of reinter-
pret, pain sensations were analyzed by means of nonparametric tests (Mann–Whit-
ney U test), due to excessive skewness in these variables. Thereafter, the
associations between gender and coping strategies were analyzed with multivariate
linear regressions controlling for variables that were unequally distributed between
the sexes, that is, occupation, marital status, working hours, work strain,
posttraumatic stress reactions, pain complexity, perceived disability, use of medi-
cation, and mean number of previous treatments. In addition, the interaction terms
Gender × Pain Complexity, Gender × Number of Treatments, Gender × Analge-
sics, Gender × Perceived Disability, Gender × Work Strain, and Gender ×
PosttraumaticStressReactionswerecalculatedandtreatedasindependentvariables.
Finally, multivariate linear regression analyses were computed separately for
men and women to examine the association between the same set of independent
variables, with the exception of the interaction terms and coping. Statistical signif-
icance was set at an alpha level of .05. Single data were lost for a number of instru-
ments, as indicated by the n values and the degrees of freedom. The loss of data
was possibly due to the fact that patients completed the questionnaire at home and
did not feel compelled to answer each question.
RESULTS
Sociodemographic Variables
As illustrated in Table 1, men were significantly more often blue-collar workers,
were more often employed full-time, and spent more weekly hours in paid work.

TABLE 1
Gender Differences in Demographic Characteristics and Types of Complaints
Among Patients With Musculoskeletal Disorders
Mena
Womenb
Test of Significance
Age
Mean±SD
45±12 years
46±11 years
ns
Marital Status
Single
Married/cohabit
Divorced
28 (23%)
87 (70%)
9 (7%)
0
47 (14%)
231 (72%)
38 (12%)
6 (2%)
ns
ns
ns
ns
Widow/er
Education
Mandatory
High school
University
33 (27%)
56 (45%)
28 (23%)
6 (5%)
84 (26%)
120 (37%)
88 (28%)
30 (9%)
ns
ns
ns
ns
Other
Occupation
Blue-collar
χ2 (1)= 9.59, p < .01
52 (45%)
21 (18%)
37 (32%)
5 (5%)
86 (28%)
115 (37%)
101 (33%)
5 (2%)
χ2 (1)= 13.44, p < .001
Low white-collar
Intermediate/high white-collar
Own business
Working time
Full-time
ns
ns
χ2 (1)= 10.68, p < .01
χ2 (1)= 14.04, p < .001
ns
77 (63%)
7 (6%)
40 (31%)
144 (45%)
68 (21%)
110 (34%)
Part-time
Not working
Working hours/week
Mean±SD
41±11
36±11
F(1, 431)=20.55, p < .0001
Complaints
Back pain
44 (35%)
19 (15%)
13 (11%)
0
2 (2%)
5 (4%)
3 (2%)
4 (3%)
1 (1%)
0
98 (30%)
59 (18%)
39 (12%)
5 (2%)
11 (4%)
11 (4%)
10 (3%)
9 (3%)
3 (1%)
8 (2%)
3 (1%)
20 (6%)
46 (14%)
ns
ns
ns
Neck/shoulder pain
Complex pain
Fibromyalgia
Myalgia
Lower-limb pain
Hip pain
Traumatic injury
Strain injury
Tension pain
Herniated disk
Other
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
ns
5 (4%)
10 (8%)
18 (15%)
No clear diagnoses
ns
an = 124. bn = 322.
311

312
GROSSI, SOARES, LUNDBERG
Women were more often in low-status white-collar occupations and more often
employed part-time. There were no other gender differences in sociodemographic
variables.
Complaints
Analyses performed on complaints recorded by the GPs and PHTs (Table 1) re-
vealed no gender differences in terms of back pain, neck/shoulder pain, complex
pain, other or unclear complaints. Comparisons between the sexes regarding other
complaints were not performed due to the small sample size in each category.
Pain Parameters
On the Visual Analog Scale (Table 2), women did not rate their pain as significantly
more intense than did men. However, they reported higher scores for perceived dis-
ability due to pain and a higher consumption of analgesics. Analyses performed on
other pain parameters indicated that female patients experienced their pain as more
complex, that is, had different types of pain and had undergone more somatic treat-
ments, such as transentaneous nerve stimulation (TNS) and acupuncture. There
were no significant gender differences in pain duration or in the use of sedatives.
Work Strain
Analyses with factorial analyses of variance (Table 2) revealed that women re-
ported lower levels of control in the work setting, but comparable levels of de-
mands. A Mann–Whitney U test also revealed higher scores for the demand–con-
trol ratio work strain among women.
Emotional Distress
Compared to male patients, women reported significantly higher scores for
posttraumatic stress reactions and lower self-esteem (Table 2). There were no reli-
ablegenderdifferencesintermsofanxietyanddepressionorsymptomsofburnout.
Coping
As shown in Table 3, the assumption that women would report higher scores for
catastrophizing was supported by the data. Furthermore, female patients reported
higher scores for diverting attention, praying and hoping, increased behavioural ac-
tivity, and pain behaviours. There were no significant gender differences in the use
of the strategies coping self-statements, ignore sensations, and reinterpret pain sen-
sations, or in the ability to decrease or control pain.

GENDER DIFFERENCES AND PAIN
313
TABLE 2
Gender Differences in Clinical Characteristics, Work Strain, and Emotional Distress
Among Patients with Musculoskeletal Disorders
Mena
Womenb
Test of Significance
Pain intensity (0–10)
Mean±SD
Disability (0–15)
Mean±SD
Use of analgesics
Regularly
Occasionally
Never
7.14±1.94
4.46±3.28
7.43±2.01
5.35±3.78
ns
F(1, 444) = 5.34, p <.05
χ2 (2) = 7.65, p < .0001
14 (11%)
49 (40%)
61 (49%)
52 (16%)
157 (49%)
113 (35%)
Use of sedatives
Regularly
Occasionally
Never
6 (5%)
7 (6%)
111 (89%)
10 (3%)
23 (7%)
289 (90%)
ns
Pain duration (months)
Median
11.5
12
ns
Range
0–431
0–468
Pain complexity (types)
One type
Several types
Number of previous treatments
Mean±SD
χ2 (1) = 9.10, p < .01
66 (53%)
58 (47%)
119 (37%)
203 (63%)
1.50±1.20
2.72±.56
3.22±.47
1.98±1.38
2.75±.63
3.04±.48
F(1, 444 = 11.34, p < .001
ns
F(1, 413) = 12.23, p <.001
Demands
Control
Work strain
Median
.84
.91
U = 5.13, p <.001
Range
PTSS
Self-esteem
Burnout
.36–1.7
.25–2.42
3.46±2.65
3.13±.67
3.09±1.23
2.49±2.44
3.31±.55
3.04±1.21
F(1, 444) = 12.44, p <.001
F(1, 441) = 7.74, p <.01
ns
GHQ
<3
≥3
84 (68%)
40 (32%)
215 (67%)
107 (33%)
ns
Note. PTSS = Posttraumatic Symptoms Scale; GHQ = General Health Questionnaire.
an = 124. bn = 322.
Multivariate Analyses
A set of multivariate regression was computed to analyze the associations between
gender and coping strategies while controlling for variables that were unequally
distributed between the sexes, that is, marital status, occupation, working hours,
work strain, emotional distress, pain complexity, perceived disability, use of medi-
cation, and mean number of previous treatments. All categorical variables with

314
GROSSI, SOARES, LUNDBERG
more than two categories were transformed into dummies. To avoid problems of
multicollinearity, only one measure of emotional distress, posttraumatic stress re-
actions, was entered in the analyses. Furthermore, interaction terms were calcu-
lated between gender and work strain, posttraumatic stress reactions, number of
previous treatments, pain complexity, use of analgesics, and perceived disability
and used as independent variables. The results are shown in Table 4. The associa-
tions between gender and coping strategies were ruled out by the introduction of
confounders. However, the interaction term Gender × Posttraumatic Stress Reac-
tions was found to be positively associated with catastrophizing, indicating that
women with high PTSS scores made more frequent use of catastrophizing.
Separate Multivariate Analyses
In the separate multivariate regression analyses, the same set of independent vari-
ables was employed, with the exception of the interaction terms between gender
and other variables, in the statistical prediction of catastrophizing. Among women
(Table 5), catastrophizing was positively associated with posttraumatic stress reac-
tions, perceived disability, and the number of previous treatments. A virtually iden-
tical pattern of results was obtained in secondary analyses, in which posttraumatic
stress reactions were substituted with scores for self-esteem, anxiety/depression,
and burnout, respectively. The results thus indicate a more frequent use of
catastrophizing among women with high levels of emotional distress, great disabil-
ity, and a history of treatments for pain. Among male patients, no significant associ-
ations were observed between the independent variables and catastrophizing.
TABLE 3
Gender Differences in Coping (Coping Strategies Questionnaire)
Among Patients With Musculoskeletal Disorders
Mena
Womenb
Test of Significance
Diverting attention
Reinterpret pain sensations
Coping self-statements
Ignore sensations
Praying & hoping
Catastrophizing
Increasing activities
Pain behaviors
Control over pain
1.39±1.14
.59±.73
1.89±1.12
.78±.89
F (1, 444) = 17.81, p < .0001
ns
ns
ns
2.82±1.40
2.08±1.31
1.20±1.05
1.20±1.01
1.78±1.25
2.13±1.08
3.73±1.69
2.84±1.40
3.08±1.30
2.24±1.26
1.48±1.11
1.67±1.25
2.61±1.27
2.62±1.05
3.61±1.36
2.97±1.16
F(1, 444) = 5.93, p < .05
F(1, 444) = 13.61, p < .001
F(1, 444)=38.46, p < .0001
F(1, 444) = 19.57, p < .0001
ns
Ability to decrease pain
ns
an = 124. bn = 322.

315

316
GROSSI, SOARES, LUNDBERG
TABLE 5
Multivariate Regression Analyses (Standardized Betas) of the Associations Among
Sociodemographics, Pain Parameters, Work Strain, Emotional Distress, and
Catastrophizing Among Male and Female Patients With Musculoskeletal Pain
Catastrophizing Men
Catastrophizing Women
Single
Married
Divorced
Blue-collar
.171
–.036
–.001
.102
.076
.113
–.026
.192
–.004
.105
.159
.056
.178
.077
.037
–.041
.008
–.078
.156
.170
.176
.124
.061
–.029
.056
.115*
.009
Low white-collar
Intermediate/high white-collar
Full-time employed
Part-time employed
Working hours/week
Pain complexity
Number of previous treatments
Use of analgesics
Perceived disability
Work strain
.361***
.081
.267***
PTSS
Note. PTSS = Posttraumatic Symptoms Scale.
*p < .05. **p < .01. ***p < .001.
DISCUSSION
This cross-sectional study examined gender differences in coping with
musculoskeletal pain. A second aim was to identify statistical predictors of men’s
and women’s coping strategies for pain. Compared to male patients, women re-
ported (a) a more negative clinical picture, including a greater disability; (b) less
control in the work setting and thus more work strain; (c) higher levels of
posttraumatic stress reactions and a lower self-esteem; (d) a broader spectrum of
coping strategies for pain, including catastrophizing, but equal levels of self-effi-
cacy beliefs. All gender differences in coping variables were ruled out when rele-
vant confounders were taken into consideration, with the exception of the associa-
tion between catastrophizing and the interaction term Gender × Posttraumatic
Stress Reactions. In further analyses it was found that women’s levels of
catastrophizing were significantly related to higher levels of posttraumatic stress
reactions, to a greater disability, and to a history of health care consumption. Such
associations were not seen among men.
In line with our findings, Jensen et al. (1994) observed that female patients with
long-standing musculoskeletal pain used catastrophizing more frequently than did
male patients and proposed that this difference may account for women’s gener-
ally poorer adaptation to pain. These results expand our knowledge by showing

GENDER DIFFERENCES AND PAIN
317
that a greater use of catastrophizing is not related to female gender per se, but is
contingent rather on female patients’ lack of mental well-being, perceptions of dis-
ability, and a history of treatments for pain.
The study has several limitations, which need to be considered before a more
detailed discussion of the results. Firstly, the cross-sectional design does not allow
issues of causality to be ascertained. The results may, however, generate hypothe-
ses to be tested in prospective studies. Secondly, the extent to which differential
selection mechanisms related to health care seeking may explain the observed gen-
der differences is unknown. The reversed pattern of associations, that is, a more
negative clinical picture among men, would have been plausible because the data
suggest that men are less prone to seek medical care for their musculoskeletal ail-
ments. However, the analyses of the patients’ clinical characteristics indicate that
women not only suffer a greater dysfunction due to pain, but are also greater con-
sumers of analgesics and other health care (e.g., TNS, acupuncture) than men.
Women’s greater health care consumption is also reflected by the greater percent-
age of women in the study sample. Similar observations have been made by others
(e.g., Andersson et al., 1993; Jensen et al., 1994; Lagerlöf, 1993).
Thirdly, patient’s self-reports could not be validated with objective measures,
and the reliability and validity of the diagnoses given by the GPs and PHTs are un-
known. However, physical diagnoses are often vague and seldom corroborated by
objective findings (e.g., Boden, Davis, Dina, et al., 1990; Borenstein & Wiesel,
1989; Rothman, 1984; Von Korff et al., 1988). Because no structured interview
was employed for diagnosing posttraumatic stress disorder, we cannot be certain
about whether the posttraumatic stress reactions reported by the patients in the
study sample truly reflect the sequelae of previous trauma or, rather, a general di-
mension of emotional distress.
The assumption that physical abuse, sexual abuse, or both may be prevalent in
the sample is corroborated by findings showing that 38% of women and 10% of
men with long-standing musculoskeletal pain reported having been exposed to
some form of sexual abuse, mostly in adulthood (Linton, 1997; Linton, Lardén, &
Gillow, 1996). A possible interpretation of the results is thus that women suffering
from posttraumatic stress reactions, due to sexual or physical abuse, develop
highly debilitating pain conditions, which are not readily ameliorated by such
treatments as acupuncture, TNS, or physiotherapy. The failure to relieve symp-
toms in spite of repeated treatments may, in turn, not only increase levels of emo-
tional distress but also lead to an increasing use of catastrophizing cognitions
about pain.
All measures of emotional distress employed in the study are highly
intercorrelated (Grossi et al., 1999) and share a common dimension of poor mental
well-being. The regressions yielding practically identical results, independently of
whether scores for the PTSS, self-esteem, GHQ, or burnout were included among
the independent variables, supports the notion that it is a general dimension of

318
GROSSI, SOARES, LUNDBERG
emotional distress, which is prevalent among women scoring high on
catastrophizing. Longitudinal data (Keefe et al., 1989) show that catastrophizing
predicts depression at 6 months’ follow-up, independently of initial levels of de-
pression, pain, and functional impairment. Also, reduction of catastrophizing is
one of the key change mechanisms for positive adaptation to pain (Kendall, 1992;
Newton & Barbaree, 1987; Turk & Rudy, 1992; Turner & Clancy, 1986). An alter-
native explanation to our results is thus that female patients with a propensity to-
ward catastrophizing suffer a deeper impact from their pain conditions, are less
responsive to treatment, and develop high levels of emotional distress.
Finally, it could be hypothesized that women’s higher levels of emotional dis-
tress and catastrophizing are both determined by their generally worse clinical pic-
ture, which, in turn, may be related to gender differences in nociceptive
mechanisms (Walker & Carmody, 1998); muscle fibre composition (Ng, Richard-
son, Kippers, & Parnianpour, 1998); biologic responses to pregnancy and child-
bearing (Biering-Sørensen, 1983a; Frymoyer et al., 1980; Kelsey, Greenberg,
Hardy, & Johnson, 1975); responses to stress (Biering-Sørensen, 1983b); total
workload (Lundberg, Märdberg, & Frankenhaeuser, 1994; Mårdberg, Lundberg,
& Frankenhaeuser, 1991); or a combination of these. All these plausible causal
mechanisms remain to be elucidated in future investigations.
In keeping with these data, clinicians need to make thorough assessments not
only of their patients’ physical status and past treatments, but also of their mental
health and cognitions toward pain. Interventions among women who manifest
great disability and have already undergone a number of somatic treatments
should not solely be targeted at physical symptoms, but also at decreasing emo-
tional distress and improving coping skills.
REFERENCES
American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders (3rd
ed.). Washington, DC: Author.
Andersson, H. I., Ejlertsson, G., Leden, I.,& Rosenberg, C. (1993). Chronic pain in a geographically de-
fined general population: Studies of differences in age, gender, social class, and pain localization.
Clinical Journal of Pain, 9, 174–182.
Arnér, S. (1984). Clinicalpainanalysis—somaticaspects. KlinikochTerapi(ASTRA)(Suppl.1), 9–12.
Biering-Sørensen, F. (1983a). A prospective study of low back pain in the general population: I. Occur-
rence, recurrence and aetiology. Scandinavian Journal of Rehabilitation in Medicine, 15, 71–79.
Biering-Sørensen, F. (1983b). A prospective study of low back pain in the general population: II. Loca-
tion, character, aggravating and relieving factors. Scandinavian Journal of Rehabilitation in Medi-
cine, 15, 81–88.
Boden, S. D., Davis, D. O., Dina, T. S., Patronas, N. J., & Wiesel, S. W. (1990). Abnormal magnetic-res-
onance scans of the lumbar spine in asymptomatic subjects. American Journal of Joint Bone Sur-
gery, 72, 403.

GENDER DIFFERENCES AND PAIN
319
Borenstein, D. G., & Wiesel, S. W. (1989). Low-back pain. Medical diagnosis and comprehensive man-
agement. Philadelphia: Saunders, Harcourt Brace Jovanovich.
Brenner, S. O., Petterson, I. L., Levi, L., & Arnetz, B. (1988). Stressreaktioner på hot om arbetslöshet
och upplevd arbetslöshet. Psykologiska, biokemiska och ekonomiska effekter av arbetslöshet och
otrygghet i arbete [Hysiological, psychological and economic effects of job loss and unemploy-
ment] (Stress Research Reports No. 210). Karolinska Institute.
Carlsson, A. M. (1984). Assessment of chronic pain: II. Problems in the selection of relevant question-
naire items for classifications of pain and evaluation and prediction of therapeutic effects. Pain, 19,
173–184.
Cathey, M. A., Wolfe, F., Kleinheksel, S. M., & Hawley, D. J. (1986). Socioeconomic impact of fibrosi-
tis:Astudyof81patientswithprimaryfibrositis. AmericanJournalofMedicine, 81(Suppl. 3A), 78.
Craighead, W. E., & Vajk, F. C. (1998). Depression and comorbid disorders. In E. A. Blechman & K. D.
Brownell (Eds.), Behavioral medicine and women—A comprehensive handbook (pp. 757–763).
New York: Guilford.
Fernandez, E. (1986). Aclassificationsystemofcognitivecopingstrategiesforpain. Pain, 26, 141–151.
Frymoyer, J. W., Pope, M. H., Costanza, M. C., Rosen, J. C., Goggin, J. E., & Wilder, D. G. (1980).
Epidemiologic studies of low back pain. Spine, 5, 419–423.
Goldberg, D. P. (1972). The detection of psychiatric illness by questionnaire. New York: Oxford Uni-
versity Press.
Goldberg, D. P. (1985). Identifying psychiatric illness among general medical patients. British Medical
Journal, 291, 161–163.
Goldberg, D. P., & Williams, P. S. (1988). A user’s guide to the General Health Questionnaire. Wind-
sor, England: NFER, Wilson.
Grossi, G., Soares, J., Ängeslevä, J., & Perski, A. (1999). Psychosocial correlates of long-term
sick-leave among patients with musculoskeletal pain. Pain, 80, 607–619.
Havsvold, T., & Johnsen, R. (1993). Headache and neck or shoulder pain—frequent and disabling com-
plaints in the general population. Scandinavian Journal of Primary Health Care, 11, 219–224.
Holen, A. (1990). A long-term outcome study of survivors from a disaster. The Alexander L. Kjelland
disaster in a perspective. Monograph. University of Oslo.
Hyyppä, M. T.(1987). The understanding and evaluation of chronic pain: Subjective vs objective deter-
minants of pain. International Journal of Rehabilitation Research, 10, 243–247.
Jensen, I., Nygren, Å., Gamberale, F., Goldie, I., & Westerholm, P. (1994). Coping with long-term
musculoskeletal pain and its consequences: Is gender a factor? Pain, 57, 167–172.
Jensen, I. B., & Linton, S. J. (1993). Coping strategies questionnaire (CSQ): Reliability of the Swedish
version of the CSQ. Scandinavian Journal of Behaviour Therapy, 22, 139–145.
Jensen, M. P., Turner, J. A., Romano, J. M., & Karoly, P. (1991). Coping with chronic pain: A critical re-
view of the literature. Pain, 47, 249–283.
Karasek, R. A., & Theorell, T. (1990). Healthy work. New York: Basic Books.
Katz, J., Ritvo, P., Irvine, J., & Jackson, M. (1996). Coping with chronic pain. In M. Zeidner & N. S.
Endler (Eds.). Handbook of coping: Theory, research, applications (pp. 252–278). New York:
Wiley.
Keefe, F. J., Brown, G. K., Wallstone, K. A., & Caldwell, D. S. (1989). Coping with rheumatoid arthritis
pain: Catastrophizing as a maladaptive strategy. Pain, 37, 51–56.
Kelsey, J. L., Greenberg, R. A., Hardy, R. J., & Johnson, M. F. (1975). Pregnancy and the syndrome of
herniated lumbar intervertebral disc: An epidemiological study. Yale Journal of Biological Medi-
cine, 48, 361–368.
Kendall, P. C. (1992). Healthy thinking. Behavior Therapy, 23, 1–11.
Kröner-Herwig, B., Jäkle, C., Frettlöh, J., Peters, K., Seeman, H., Franz, C., & Basler, H. D. (1996). Pre-
dicting subjective disability in chronic pain patients. International Journal of Behavioral Medicine,
3, 30–41.

320
GROSSI, SOARES, LUNDBERG
Lagerlöf, E. (1993). OECD working party on the role of women in economy. Panel of experts on women,
work and health. National Report Sweden, Ministry of Health and Social Affairs, Ds. 98.
Linton, S. J. (1997). A population based study of the relationship between sexual abuse and back pain:
Establishing a link. Pain, 73, 47–53.
Linton, S. J., Lardén, M., & Gillow, A. M. (1996). Sexual abuse and chronic musculoskeletal pain: Prev-
alence and psychological factors. Clinical Journal of Pain, 12, 215–221.
Lisspers, J., & Setterlind, S. (1999). Burnout and hardiness: Some psychometric properties of different
assessment instruments. Manuscript in preparation.
Lundberg, U., Märdberg, B., & Frankenhaeuser, M. (1994). The total workload of male and female
white-collar workers as related to age, occupational level, and number of children. Scandinavian
Journal of Psychology, 35, 315–327.
Magni, C., Calderion, C., Rigatti-Luchini, S., & Merskey, H. (1990). Chronic musculoskeletal pain and
depressive symptoms in the general population. An analysis of the first National Health and Nutri-
tion Examination Survey data. Pain, 43, 299–307.
Maslash, C. & Jackson, S. (1981). Maslach Burnout Inventory (researchmanual, 1st ed.). PaloAlto, CA:
Consulting Psychologists Press.
Märdberg, B., Lundberg, U., & Frankenhaeuser, M. (1991). The total workload of parents employed in
white-collar jobs: Construction of a questionnaire and a scoring system. Scandinavian Journal of
Psychology, 32, 233–239.
McLean, P., & Woody, S. (1998). Depression. In E. A. Blechman & K. D. Brownell (Eds.), Behavioral
medicine and women—A comprehensive handbook (pp. 752–756). New York: Guilford.
Melamed, S., Kushnir, T., & Shirom, A. (1992). Burnout and risk factors for cardiovascular diseases.
Behavioral Medicine, 18, 53–60.
Newton, C. R., & Barbaree, H. E. (1987). Cognitive changes accompanying headache treatment. The
use of a thought sampling procedure. Cognitive Therapy and Research, 11, 635–651.
Ng, J. K., Richardson, C. A., Kippers, V., & Parnianpour, M. (1998). Relationship between muscle fiber
composition and functional capacity of back muscles in healthy subjects and patients with back pain.
Journal of Orthopaedic and Sports Physical Therapy, 6, 389–402.
Nicholas, M. K., Wilson, P. H., & Goyen, J. (1992). Comparison of cognitive-behavioral group treat-
mentandanalternative, non-psychologicaltreatmentforchroniclowbackpain. Pain, 48, 339–347.
Pines, A. M., Aronson, E., & Kafry, D. (1981). Burnout: From tedium to personal growth. New York:
Free Press.
Rasmussen, B. K., & Olesen, J. (1992). Migraine with aura and migraine without aura: An epidemiolog-
ical study. Cephalalgia, 12, 221–228.
Reesor, K. A., & Craig, K. D. (1988). Medically incongruent chronic back pain: Physical limitations,
suffering and ineffective coping. Pain, 32, 35–45.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University
Press.
Rosenstiel, A. K., & Keefe, F. J. (1983). The use of coping strategies in chronic low-back pain patients:
Relationship to patient characteristics and current adjustment. Pain, 17, 33–44.
Rothman, R. H. (1984). A study of computer-assisted tomography: Introduction. Spine, 9, 548.
Setterlind, S., & Larsson, G. (1994). The Stress Profile: A psychosocial approach to measuring stress.
Stress Medicine, 10, 183–190.
Sherry, D. D., McGuire, T., Mellins, E., Salmonson, K., Wallace, C. A., & Nepom, B. (1991). Psychoso-
matic musculoskeletal pain in childhood: Clinical and psychological analyses of 100 children. Pedi-
atrics, 88, 1093–1099.
Snow-Turek, A. L., Norris, M. P., & Tan, G. (1996). Active and passive coping in chronic pain patients.
Pain, 64, 455–462.

GENDER DIFFERENCES AND PAIN
321
Soares, J., & Grossi, G. (1999a). Psychosocial factors, pain parameters, mental health and coping among
Turkish and Swedish patients with musculoskeletal pain. Scandinavian Journal of Occupational
Therapy, 6, 174–183.
Soares, J., & Grossi, G. (1999b). The experience of musculoskeletal pain: A comparison of immigrant
and Swedish patients. Scandinavian Journal of Caring Sciences, 13, 254–266.
Socialstyrelsen. (1987). Klassifikation av sjukdomar. Systematisk förteckning. Stockholm: Liber,
Allmänna förlaget.
Statistics Sweden. (1999). Work related disorders 1999.
Sullivan, M. J. L., & D’eon, J. L. (1990). Relation between catastrophizing and depression in chronic
pain patients. Journal of Abnormal Psychology, 99, 260–263.
Taylor, H., & Curran, N. M. (1985). The Nuprin Pain Report. New York: Louis Harris.
Theorell, T., Michélsen, H., & Nordemar, R. (1993). Music I study group. Validitetsprövningar av
psykosociala indexbildningar. In M. Hagberg & C. Hogstedt (Eds.), Stockholmsundersökningen 1
(pp. 163–178). Stockholm: MUSIC Books.
Toomingas, A., Theorell, T., Michélsen, H., & Nordemar, R. (1997). Associations between self-rated
psychosocial work conditions and muskuloskeletal symptoms and signs. Scandinavian Journal of
Work, Environment and Health, 23, 130–139.
Turk, D. C., & Rudy, T. E. (1992). Cognitive factors and persistent pain: A glimpse into Pandora’s Box.
Cognitive Therapy and Research, 16, 99–122.
Turner, J. A., & Clancy, S. (1986). Strategies for coping with chronic low-back pain: Relationship to
pain and disability. Pain, 24, 355–364.
Unruh, A. M. (1996). Gender variations in clinical pain experience. Pain, 65, 123–167.
Verbrugge, L. M., Lepkowski, J. M., & Konkol, L. L. (1991). Levels of disability among U.S. adults
with arthritis. Journal of Gerontology and Social Science, 46, 71–83.
Von Korff, M., Dworkin, S. F., LeResche, L. A., & Kruger, A. (1988). An epidemiological comparison
of pain complaints. Pain, 32, 173–183.
Walker, J. S., & Carmody, J. J. (1998). Experimental pain in healthy human subjects: Gender differ-
ences in nociception and in response to ibuprofrin. Anasthesia and Analgesia, 86, 1257–1262..
Wolfe, F. (1989). Fibromyalgia: The clinical syndrome. Rheumatic Diseases Clinics of North America,
15, 1.