Pain, 44 (1991) 35-43 0 I991 Elsevier Science Publishers ADONIS 030439599100053K

35 B.V. (Biomedical

Division)

0304-3959/91/$03.50

PAIN 01721

The contributions of interpersonal conflict to chronic pain in the presence or absence of organic pathology Julia A. Faucett ’ School

of Nursing,

a and Jon D. Levine

b

University of California, San Francisco, CA 94143-0608 (U.S.A.), and b Schools of Medicine and Dentistry University of California, San Francisco, CA 94143-0724 (U.S.A.) (Received

13 February

1990, revision received

17 July 1990, accepted

23 July 1990)

This study investigated the influences of social support and interpersonal conflict on chronic pain in patients with arthritis or with Summary myofascial disorders. Measures of social support, conflict, and pain were drawn from subscales of the McGill Pain Questionnaire, the Multidimensional Pain Inventory, the Family Environment Scale, and the Interpersonal Relationships Inventory. Patients with myofascial disorders reported significantly worse pain (sensory and affective), higher depression scores, more interpersonal conflict, and less support from others than patients with arthritis, but did not differ from them on personality traits. Also, the contributions of conflict to pain were found to depend on the nature of the chronic disorder and on the source of the conflict, i.e., significant other, family, or social network members. For patients with arthritis, less intense pain (sensory and affective) was associated with higher family conflict. Less intense sensory pain in arthritis was also associated with more punishing responses from the significant other to pain. For patients with myofascial disorders, more intense affective pain was associated with higher social network conflict. Social support did not significantly contribute to pain for either group. Thus, chronic painful disorders may differ on the influences that social relationships have on pain. The implications of these differences for treatment are discussed. Key words: Chronic

pain;

Arthritis;

Myofascial

disorders;

Social support;

Introduction

Social support has been found to enhance psychological well-being and quality of life for patients with chronic painful or disabling disorders [9,34,42,49], while the conflictual and problematic aspects of social relationships appear to increase depression and somatization [13,14,36,38,41]. Recently, investigators have studied social support and interpersonal conflict for their influences on the severity of chronic pain. For example, it has been shown that conflict in the family and lack of social support in the work environment contribute to increases in back pain severity [13]. These findings suggest that pain, like depression, worsens in a non-supportive social environment. In contrast, higher levels of pain have also been reported for patients who receive greater pain-specific support or who are more satisfied with the support they receive [7,15,17]. The association between pain-related support from others and pain intensity has also been found to vary

Correspondence to: Julia

University

of California,

Faucett, R.N., Ph.D., School of Nursing, San Francisco, CA 94143-0608, U.S.A.

Conflict

with the nature of the clinical disorder [l]. Interactions with others may differ qualitatively in disorders characterized by the lack of organic pathology or in which pathology is not consonant with the level of pain, compared to disorders in which physical findings clearly corroborate pain reports [26]. The absence of organic pathology may, in fact, increase uncertainty for others about the cause or intensity of pain or about the contribution of psychological factors. For spouses of patients with back pain lacking detectable pathology, the attribution of the patient’s difficulties with pain to psychological problems has been associated with increased spousal distress and dissatisfaction with the marriage [6]. Furthermore, Rowat and Knafl [39] proposed that uncertainty about pain is central to spousal distress and found that highly distressed spouses were more likely to respond supportively to chronic pain than to ignore it. Nurses, in response to vignettes about hypothetical patients, have been found to attribute less pain and more negative personal characteristics to patients when physical pathology was absent rather that present [47, see also 351. It is likely, therefore, that the presence or absence of physical evidence validating the patient’s pain influences how others interpret pain complaints and respond to them.

The effects of social support or conflict on pain may also differ depending on the level or type of social relationship under consideration. Family members. friends, and neighbors may differ in the kind of emotional or tangible support they provide as well as in the

affective intensity of the relationship [37,27]. Additionally. support from work peers has been shown to play a more prominent role than that from family members in predicting back pain severity [13]. The nature of effects specifically related to conflict from different types of relationships has been less well studied. In this study, we compared the contributions of social support and conflict from the patient’s significant other, family, and social network to pain intensity for patients with and without organic pathology underlying their chronic painful disorders.

Methods Subjects

Subjects were recruited from 3 university-affiliated rheumatology clinics, private rheumatology practices, and a community-based arthritis self-management course. All subjects were currently suffering pain either from myofascial disorders or from rheumatoid or osteoarthritis as diagnosed by their referring rheumatologists. To be included in the study, patients with chronic myofascial disorders had to meet all of the following criteria: (1) diffuse and persistent muscular aching for 3 months or more, (2) deep tenderness at localized sites (“ tender” or “trigger” points), and (3) little or no observable, laboratory or radiologic evidence of organic pathology [l&43,45]. The diagnosis of arthritis was based on evidence of joint inflammation or destruction or abnormal laboratory findings. Ambulatory, non-institutionalized, English-speaking adults between the ages of 21 and 75 who were not currently suffering pain from any other type of acute or chronic disorder were considered for the study. Eligible patients (n = 177) were asked to participate in a structured telephone interview; 26 of whom elected not to take part in the study. Of the resulting sample of 151 patients, 84 had arthritis and 67 had myofascial disorders. To gather data on the responses of a significant other to the patient’s pain, patients were asked to report on a spouse or other adult with whom they lived or, if they lived alone, on an adult they felt close to, whom they saw frequently (preferably more than twice a week) and who might be helpful to them when their chronic disorder flared. All patients were able to report on the responses of a significant other to their pain. Of the total patient sample, 107 were living with others (arthritis, n = 52; myofascial disorders, n = 55). Data from these 107 patients were used in the analyses of relationships in the family environment.

T.4Bl.E

1

DEMOGRAPHICS

OF PATIENT

Arthritis

GROUPS

(n = 84)

Myofascial

(n = 67)

Mean

(SD.)

Min/max

Mean

(SD.)

Min/max

58.3

(13.4)

24/

47.8

(12.0) **

73/

($lOOOs) Education (years) Pain duration

36.8

(25.4)

39.0

(31.4)

10/150

14.8

(2.5)

(years) Predictability

17.0 4.3

Age (years) Income

(l-10)

75

5/120 lO/

20

13.8

(13.0)

I,’ 65

16.0

(13.3)

(5.6)

O/ 10

5.1

(2.71

Significant t tests results for the 2 groups: Min = lowest value reported: max = highest

(2.8) *

I,,‘

73

20

1,’ 59 o/

10

* P i 0.05, * * P c 0.01. value reported.

Patients with myofascial disorders were compared to those with arthritis to determine the similarity of the groups on demographics. Patients with myofascial disorders differed significantly from those with arthritis on age and education, but not on household income (Table I). Both patient groups were primarily comprised of women, and did not differ significantly in gender ratios (arthritis = 82% vs. myofascial disorders = 92% women; P > 0.10); although they differed on marital status (arthritis = 43% vs. myofascial disorders = 66% married; P < 0.01). Additionally, the 2 patient groups did not differ on the duration of their pain or on the degree to which they found pain exacerbations to be unpredictable (Table I). Measures

The following instruments were included in the structured telephone interviews. Pain. The sensory and affective subscales of the McGill Pain Questionnaire (MPQ) were both used as measures of pain since support and conflict from interpersonal relationships may influence different aspects of pain [31]. Patients reported on “today’s pain.” Social support and conjlict. Perceptions by the patients of the availability of social support and the presence of conflict in their social relationships were measured using subscales of the (a) Multidimensional Pain Inventory (MPI) [24], (b) Family Environment Scale (FES) [22,32], and (c) Interpersonal Relationships Inventory (IPRI) [48]. Family support and conflict were measured using the Cohesion and Conflict subscales of the FES; while relationships with the social network were evaluated with the Support and Conflict subscales of the IPRI. The behavioral responses of the significant other to pain were evaluated using the Solicitous Responses and Punishing Responses subscales of the MPI. Personality. Four measures of personality were included to control for the influence of stable traits on

37

patient self-reports. Extraversion and neuroticism, which have previously demonstrated associations with reports of both pain [3,8,23,29] and social relationships [21], were evaluated using the short form of the Eysenck Personality Inventory (EPI) [12]. Traits of positive and negative affectivity, although frequently correlated with extraversion and neuroticism, were also included to specifically control for the influence of persistent affective styles and were measured using the Positive and Negative Affectivity Scale (PANAS) [52]. Negativity in particular has been associated with lower levels of social support [51] and increases in somatic symptoms [50,53]. Depression. Depression was measured using the short form of the Beck Depression Inventory (BDI) [5]. Depression was included to evaluate the influences of affect on pain independent of stable traits or social relationships. Statistical Analyses Group comparisons. Patients with myofascial disorders were compared to those with arthritis on the study variables using multivariate and then, univariate analyses. Additionally, univariate correlations between the pain-related responses of the significant other and the family and social network measures of support and conflict were computed for each of the 2 patient groups. The resulting independent correlations were then tested for equality. Multiple regression analyses. Four regression equations were determined to consider the relationships of support and conflict to pain. The contribution to pain of support and conflict within the family was determined using sensory pain and affective pain as separate dependent variables. Then, the contribution to pain of support and conflict in the social network context was examined, also using sensory pain and affective pain as separate dependent variables. In all 4 of the regressions, main effect variables were forced into the regression analyses in sets in the following order: (1) demographics including age, sex, income, and education, (2) patient group, a dichotomous variable of arthritis vs. myofascial disorder, (3) the 4 personality variables of extraversion, neuroticism, and positive and negative affectivity, (4) the selected set of support and conflict variables (see below), and (5) depression. Interaction terms, constructed to test the hypothesis that the relationship of support or conflict to pain varies with the type of chronic disorder, were then entered at step 6 using the forward procedure [33]. In the 2 regression analyses of pain within the family context, the set of support and conflict variables consisted of family cohesion and conflict as measured by the FES and the 2 subscales from the MPI measuring the responses of the significant other to pain. In the 2 regression analyses of pain for the social network context, the set of support and conflict variables consisted

of network support and conflict as measured by the IPRI and, again, the 2 subscales from the MPI measuring the responses of the significant other to pain. For the interaction terms, therefore, the support and conflict variables of each set were cross-multiplied with the patient group value, resulting in 4 possible interaction terms for each regression equation. For interaction terms that explained significant amounts of the variance in sensory or affective pain, the relationship of each independent variable to pain (holding demographics, personality, depression, and other social relationship variables constant) was plotted for each patient group.

Results Group comparisons Patients with myofascial disorders differed significantly from those with arthritis in the multivariate analysis of the study variables (Hotelling’s test, F (12, 138) = 5.7, P -C0.01). Family cohesion and family conflict were not included in the multivariate analyses in order to maximize the sample size. Subsequent univariate comparisons with Student’s t test were employed to determine the specific variables on which the 2 groups differed (Table II). The family variables were included in the univariate analyses. Due to the number of group comparisons, the level for significance was conservatively set at P < 0.01. Patients with myofascial disorders reported significantly more intense sensory and affective pain and higher levels of depression than those with arthritis. Arthritis patients reported more family and network support to be available than patients with myofascial disorders. However, the 2 patient groups did not differ on their reports of how supportive or solicitous the significant other was in response to their pain. Patients with myofascial disorders reported significantly more conflict from the social network than arthritis patients perceived. Patient reports additionally indicated that pain in myofascial disorders was more likely than pain in arthritis to be met with punishing responses from the significant other. The myofascial disorder group also reported more family conflict than the arthritis group; however, this was not a significant difference. There were no significant differences between the 2 patient groups on any of the personality measures. Each of the 2 groups was significantly higher in negative affectivity (P < 0.01) and lower in positive affectivity (P < 0.05) than norms reported by Watson et al. [52] for the PANAS. However, neither of the 2 groups differed significantly from norms for extraversion or neuroticism [12]. Thus, patients with myofascial disorders consistently reported their situation to be worse

ix than the arthritis patients, attributed to predisposing

but this could not be directly traits.

port, family cohesion, or the solicitous responses of the significant other) did not add to the explained variance of sensory pain for either group. Myofascial disorder and higher depression scores were both associated with higher sensory pain. The sets of demographics and personality also explained significant amounts of the variance in sensory pain. Of the demographics, only income had a significant unique contribution, with lower income predicting higher pain (P -c 0.01). None of the individual personality traits provided a significant unique contribution. Affective pain. Affective pain was also associated with interpersonal conflict. Again, the type of chronic disorder determined the magnitude and the direction of the association between affective pain and conflict (Table IV). For patients with arthritis, greater family conflict predicted less intense affective pain, while conflict with social network members had little relationship to

Regression analyses Sensory pain. The negative or conflictual rather than the supportive aspects of close personal relationships contributed to reports of sensory pain (Table III). The strength and direction of those contributions differed depending on the patient group (Fig. 1). Greater conflict with family members was associated with less intense arthritis pain, but was not strongly associated with pain in myofascial disorders (Fig. 1A). Likewise, more punishing responses from the significant other predicted lower arthritis pain, but not lower pain in myofascial disorders (Fig. 1B). Conflict with social network members, on the other hand, did not contribute to sensory pain as a main effect or for either group. With the conflict variables entered, social support (network sup-

TABLE

II

PATIENT

GROUP

COMPARISONS

ON PAIN, SUPPORT, Arthritis Mean

CONFLICT,

PERSONALITY,

(n = 84)

Myofascial

AND

DEPRESSION

MEASURES P

(n = 67)

(S.D.)

Mean

(S.D.)

13.62 1.87

(7.09) (2.15)

17.21 3.36

(6.19) (2.70)

0.001 0.000

Significant other (MPI) Solicitous responses (O-6)

3.68

(1.37)

3.34

(1.72)

0.183

Family (FES) Cohesion a (O-9)

8.02

(1.50)

7.14

(1.91)

0.010

Social network (IPRI) support (l-5)

4.17

(0.58)

3.85

(0.80)

0.005

Significant other (MPI) Punishing responses (O-6)

1.00

(1.06)

1.97

(2.01)

0.000

Family (FES) Conflict a (O-9)

2.48

(1.74)

3.09

(2.44)

0.138

Social network (IPRI) Conflict (l-5)

2.74

(1.74)

3.11

(0.83)

0.002

2.64 -0.29

(3.26) (3.17)

2.03 0.30

(3.16) (3.25)

0.246 0.268

34.70 26.06

(7.59) (8.31)

32.67 27.49

(7.56) (9.12)

0.102 0.315

5.23

(4.59)

7.57

(5.06)

0.003

(arthritis,

n = 52; myofascial

Pain (MPQ) Sensory (O-42) Affective (O-14) Social support

Interpersonal

conflict

Pers0nalit.y EPI Extraversion Neuroticism

( - 6 to + 6) ( - 6 to + 6)

PANAS Positivity (10-50) Negativity (10-50) Depression

(BDI) (O-39)

a Only 107 subjects abbreviations.

lived

with

others

disorders,

n = 55). (Range

of scores

possible.)

See text

for instrument

39

TABLE

III

REGRESSIONS OF SENSORY CIAL VARIABLES Cum. R2

Step

PAIN

ON PERSONAL

AND

SG-

Change R2

P

F

(A) Regression of sensory pain using family environment variables (n = 107) 1. 2. 3. 4.

Demographics Group Personality Social relationships 5. Depression 6. Group x conflict

0.089 0.151 0.234

0.089 0.062 0.082

2.489 7.428 2.610

0.048 0.008 0.040

0.268 0.269 0.341

0.034 0.001 0.072

1.093 0.079 9.945

0.365 0.779 0.002

Demographics Group Personality Social relationships 5. Depression 6. Group x punish. responses

-

.B-

0

-lO-

Arthr,t,s

z iI

(B) Regression of sensory pain using social network variables (n = 151) 1. 2. 3. 4.

-6

i

,

0.048 0.065 0.058

1.861 10.657 2.449

0.120 0.001 0.049

0.184 0.221

0.013 0.037

0.546 6.485

0.702 0.012

0.246

0.025

4.473

0.036

.

-12 0

0.048 0.114 0.171

,\

2

E

4

6

8

Family conf IIct

::

affective pain (Fig. 2). For patients with myofascial disorders, family conflict had little relationship to affective pain, while greater conflict with social network members predicted more intense affective pain (Fig. 2). The pain-specific responses of the significant other did

TABLE

IV

REGRESSION OF AFFECTIVE CIAL VARIABLES

PAIN ON PERSONAL

AND SO0

Step

Cum. R2

R2

F

P

(A) Regression of affective pain using family enoironment variables (n = 107) 1. 2. 3. 4.

Demographics Group Personality Social

0.104 0.180 0.233

0.104 0.076 0.052

2.974 9.368 1.658

0.023 0.003 0.166

relationships 5. Depression 6. Group X conflict

0.274 0.311 0.371

0.041 0.037 0.060

1.311 4.927 8.712

0.272 0.029 0.004

(B) Regression of affective pain using social network variables (n = 151) 1. 2. 3. 4.

1

2

Punlshlng

Change

Demographics Group Personality Social

0.075 0.146 0.200

0.075 0.072 0.053

2.941 12.222 2.335

0.022 0.000 0.058

relationships 5. Depresson 6. Group x conflict

0.251 0.324 0.345

0.052 0.073 0.021

2.366 14.638 4.430

0.056 O.ooO 0.037

3

4

5

6

resr.m”ses

Fig. 1. The relationship between conflict and sensory pain (residual) plotted for each patient group. A: the relationship between family conflict and sensory pain (residual: with demographics, personality variables, cohesion, responses of the significant other, and depression held constant) for each patient group. B: the relationship between the punishing responses of the significant other and sensory pain (residual: with demographics, personality variables, network support and conflict, solicitous responses, and depression held constant) for each patient

group.

not provide a significant increase in the explained variance of pain. As with sensory pain, affective pain was more intense for patients with myofascial disorders or with higher depression. Among the demographic variables, only lower income and lower education significantly predicted greater pain intensity (P < 0.05). Personality variables did not significantly predict affective pain.

CORRELATIONS OF THE PAIN-SPECIFIC RESPONSES OF THE SIGNIFICANT OTHER WITH FAMILY AND NETWORK StJPPORT AND CONFLICT VARIABLES AND DEPRESSION Arthritis Solicitous responses Social support Family cohesion a Network support Solicitous responses

_~__ Myofascial (n = 67)

(n = X4) Punishing responses

Solicitous responses

0.223 0.398 * * 0.289 ** -0.109 h

Punishing responses

0.480 * * - 0.536 * * 0.472 ** -0.518 **

~0.106’

-0.471

Interpersonal conflict Family conflict a Network conflict

- 0.023 - 0.058

0.466 * * -0.304 0.346 * * -0.206

Depression

- 0.084

0.051 h

-0.249

*

*

0.489 ** 0.545 **

*

0.399 * *

than

those

---

1

05

smaller

for

those with arthritis. The associations of depression with the responses of the significant other are also higher for myofascial disorders than for arthritis. The associations of punishing responses with network support, solicitous responses, and depression are all significantly (P < 0.05) higher for the myofascial group than for the arthritis group.

,’ +

t

’ Arthritis n = 52; myofascial n = 55. h Correlations for arthritis are significantly myofascial disorders ( P i 0.05). * P cc 0.05: * *P < 0.01.

Discussion

-II

I

1 Soc\ol

I

I

I

I

2

3

4

5

ne!work

-

I

conflict

Fig. 2. The relationship between conflict and affective pain (residual) plotted for each patient group. A: the relationship between family conflict and affective pain (residual: with demographics, personality variables, cohesion, responses of the significant other, and depression held constant) for each patient group. B: the relationship between social network conflict and affective pain (residual: with demographics, personality variables, network support, responses of the significant other, and depression held constant) for each patient group.

In this study, we found that the contributions of social relationships to pain intensity depend on the nature of the chronic disorder and on the type or level of the interpersonal relationship. Namely, more conflict in close relationships such as with family members or a significant other predicts less intense sensory and affective arthritis pain; while in myofascial disorders, more conflict in relationships throughout the social network predicts more intense affective pain (Table Vi). Im-

TABLE

VI

THE CONTRIBUTIONS OF CONFLICT PAIN IN ARTHRITIS AND MYOFASCIAL

Univariate correlations with the significant otherS responses to pain Patient reports of the responses of the significant other to pain may be key determinants of their sense of support and emotional well-being. As Table V shows, univariate correlations between patients’ reports about the responses of the significant other to pain and the remaining support and conflict variables are uniformly higher for those with myofascial disorders than for

_ negative tion.

Signif. other

Family

Social network

Signif. other

Family

Social network

_

_

_

0

0

0

+

+

+

0

0

0

association;

TO

support

Conflict

Arthritis pain Myofascial pain

AND SUPPORT DISORDERS

+ positive

association;

0 no unique

associa-

41

portantly, differences in underlying personality traits, demographics, or the duration and predictability of pain as measured in this study do not account for these associations of pain with conflict. Conflict with family or significant others and pain When pain is consistent with organic findings. Arthritis pain in the presence of family conflict or punishing responses from the significant other is less intense, with individual differences held constant. There are at least 2 alternative explanations for these findings. Negative interactions with others may abate in the presence of flares of active arthritis or in severe debilitating disease. Such a decline in conflict might occur due to the withdrawal of the ill person from social interaction and/or due to the curtailment of conflict on the part of concerned others who believe that stress exacerbates arthritis. The presence of underlying pathophysiology would legitimize such care-giving responses to pain. Rowat and Knafl [39] reported that 25% of spouses in their study of chronic pain patients described “keeping tension levels down, as one way of trying to effect control over the pain” (p. 264). Similarly, others have noted that physical illness may mobilize support [4,19]. The negative correlation of family cohesion and conflict (r = -0.39, P < 0.05) found among patients with arthritis in this study is compatible with the suggestion that reductions in conflict may be perceived as acts of support. Alternatively, as arthritis worsens with increases in pain and disability, patients may minimize their complaints in an attempt to reduce the strain they place on those they care about and to forestall the distress that may come to accompany the burden of long-term family care-giving [2,46]. Thus, increases in pain-related negative responses from close supporters would result in the patient learning to withhold pain complaints. Pain in the absence of organic findings. In this study, significant univariate correlations of pain with the pain-specific responses of the significant other were found only with punishing responses and only in myofascial disorders (punishing responses with sensory pain, r = 0.30, P < 0.05; with affective pain, r = 0.38, P < 0.01). Once, however, personal characteristics are taken into consideration, family conflict and the punishing responses of the significant other only weakly explain variations in pain from myofascial disorders. These findings may reflect a lack of consistency across families in their responses to pain from myofascial disorders as opposed to arthritis. Alternatively, since the literature describes these patients as perfectionistic [44], nervous, restless, irritable [16], and with a need to be constantly busy [30], patients with myofascial disorders may continue to engage in routine patterns of stressful interaction despite exacerbations of pain; that is, they may be unable to withdraw from conflict in the family setting

or they may not associate withdrawal from conflict with pain prevention or relief. Habitual patterns of reacting to interpersonal conflict around pain, its causes or its treatments may be a factor in the persistence of conflict in close relationships across all levels of pain in myofascial disorders. Further research will be needed to clarify the role of individual personality vs. family patterns of interaction in myofascial disorders.

Broad social network vs. intimate relationships When the source of conflict is the entire social network, conflict appears to play a stronger role in affective pain for patients with myofascial disorders than for those with arthritis. Conflict as measured for the social network represents not only interpersonal discord but also the stress of meeting obligations and maintaining relationships with a wide variety of others such as in-laws, ‘grown children, neighbors, and employers, as well as with those who live with the patient. It is this cumulative social network stress rather than conflict solely in close relationships that is associated with increasing myofascial pain identified as wretched, tiring, fearful, sickening, and punishing. While such stress may exacerbate pain, increasing pain may also make it more difficult to manage social demands. It is also possible that patients with myofascial disorders experience greater frustration in legitimizing or explaining their pain to social contacts who are not within the family or most intimate circle of relationships. Correspondingly, feelings of stigmatization, among patients with pain lacking organic findings, have been associated with greater reliance on household members for support [26]. Arthritis pain and related disability, on the other hand, are widely validated and accepted, engendering less conflict and securing more support across the social network. Further research, however, will be required to determine whether specific types of relationships such as with co-workers and or employers account for social network findings. Although patients with myofascial disorders were found to be no more neurotic or negative than patients with arthritis, they reported greater suffering in the form of pain, depression, and poorer social relationships. Some investigators have suggested that affective distress amplifies reports of pain [25]; while others have suggested that pain increases depression through generating interference with daily life [40]. Pain-related interpersonal conflict may moderate the relationships of pain and depression by heightening associated distress or life interference. Correlations of the pain-related responses of the significant other (Table V) suggest that conflict about pain figures more prominently in the social relationships of patients with myofascial disorders than those with arthritis. Stigmatization, related to the lack of validation for pain from organic findings,

42

may be a source of difficulty in social relationships for these patients [26]. The presence of a chronic disorder unquestionably affects social relationships [ 11,20,28]. The findings of this study suggest that social relationships may differ from one painful chronic disorder to another. Given the increasing number of treatment programs available for patients with chronic pain which focus on the management of stress or pain behavior as well as on the involvement of family or spouse, the implications of these findings may be relevant to the type of treatment selected. Assessment of both intimate and social network relationships of the patient may be important. Research with family members and key others in the patient’s social network on their attitudes and responses towards the patient’s illness and pain will build our understanding of the contribution of social dynamics to the experience of pain. Longitudinal studies will also be required to distinguish between the influence of support and conflict on pain and the influence of pain on support and conflict.

Acknowledgements We would like to thank Jane Norbeck, R.N., D.N.Sc., for her comments on the manuscript and her continuing support. We would also like to thank Steven Paul, Ph.D., for his consultation on the data analysis and Kate Lorig, R.N., D.P.H., Christopher Lorish, Ph.D., Kenneth Sack, M.D., Paul Davidson, M.D., Kenneth Fye, M.D., Thomas Jam&on, M.D., Stephen Nimelstein, M.D., and William Seaman, M.D., for their help in finding subjects. Funding for the research was provided by the National Center for Nursing Research and the National Institute for Arthritis, Metabolic and Skin Diseases (AM 32634) NIH; the Regents of the University of California; Nurses’ Educational Funds, Inc.; the School of Nursing, University of California, San Francisco; and the Arthritis Foundation, No. California Chapter. Portions of this study were previously reported at the VIth World Congress on Pain.

References 1 Anderson, L.P. and Rehm, L.P., The relationship

between strategies of coping and perception of pain in three chronic pain groups, J. Clin. Psycho]., 40 (1984) 1170-1177. 2 Baillie, V., Norbeck, J. and Barnes, L.E., Stress, social support, and psychological distress of family caregivers of the elderly, Nurs. Res., 37 (1988) 217-222. 3 Barnes, G.E., Extraversion and pain, Br. J. Sot. Clin. Psycho]., 14 (1975) 303-308. 4 Barrera, M., Distinctions between social support concepts, measures, and models, Am. J. Commun. Psycho]., 14 (1986) 413445.

5 Beck. A.T. and Beck, R.W., Screening depressed patients in family practice: a rapid technique, Postgrad. Med.. 52 (1972) 81-85. 6 Block. A.R. and Boyer. S.L., The spouse’s adjustment to chronic pain: cognitive and emotional factors. Sot. Sci. Med.. 19 (1984) 1713-1317. 7 Block, A.R., Kremer, E. and Gaylor. M., Behavioral treatment of chronic pain: the spouse as a discriminative cue for pain behavior, Pain, 9 (1980) 243-252. 8 Bond, M.R. and Pearson. 1.B.. Psychological aspects of pain in women with advanced cancer of the cervix, J. Psychosom. Res., 13 (1969) 13-19. 9 Burckhardt, C.S.. The impact of arthritis on quality of life. Nurs. Res., 34 (1985) 11-16. 10 Chapman, C.R., Sola, A.E. and Bonica, J.J., Illness behavior and depression compared in pain center and private practice patients, Pain. 6 (1979) l-7. 11 Corbin, J.M. and Strauss, A. Unending Work and Care: Managing Chronic Illness at Home, Jossey-Bass, San Francisco, CA. 1988, 358 pp. 12 Eysenck. H.J.. A short questionnaire for the measurement of two dimensions of personality. In: H.J. Eysenck (Ed.), Eysenck on Extraversion, Wiley, New York, 1973, pp. 31-37. 13 Feuerstein, M.. Sult, S. and Houle, M., Environmental stressors and chronic low back pain: life events, family and work environment, Pain, 22 (1985) 295-307. 14 Fiore. J.. Becker, J. and Coppel, D.. Social networks interactions: a buffer or a stress, Am. J. Commun. Psycho]., 11 (1983) 423-439. 15 Flor, H., Kerns, R.D. and Turk, D.C., The role of spouse reinforcement, perceived pain, and activity levels of chronic pain patients, J. Psychosom. Res., 31 (1987) 251-259. 16 Gaston-Johansson. F., Johansson, G.. Felldin, R. and Sanne, H., A comparative study of pain description, emotional discomfort and health perception in patients with chronic pain syndrome and rheumatoid arthritis, Stand. J. Rehab. Med., 17 (1985) 109-119. 17 Gil, K.M.. Keefe, F.J., Crisson, J.E. and Van Dalfsen. P.J.. Social support and pain behavior, Pain, 29 (1987) 209-217. 18 Goldenberg, D.L.. Fibromyalgia syndrome. JAMA, 257 (1987) 2782-2787. 19 Grant. I., Patterson, T.L. and Yager, J., Social supports in relation to physical health and symptom of depression in the elderly. Am. J. Psychiat., 145 (1988) 1254-1257. 20 Haffstrom, J.L. and Schram, V.R. Chronic illness in couples: selected characteristics, including wife’s satisfaction with and perception of marital relationships, Fam. Relat., 33 (1984) 195-203. 21 Henderson. S., Byrne, D. and Duncan-Jones, P., Neurosis and the Social Environment, Academic Press, San Francisco. CA, 1981. 279 pp. 22 Holahan. C.J. and Moos, R., Personality, coping, and family resources in stress resistance: a longitudinal analysis, J. Pers. Sot. Psycho]., 51 (1983) 389-395. 23 Jacox. A. and Stewart, M., Relation of psychosocial factors and type of pain, In: E.M. Jacobi and L.E. Notter (Eds.), Proc. Am. Nurses’ Ass. 9th Annual Research Conference, ANA, Kansas City. 1973. pp. 13-31. 24 Kerns, R.D., Turk, D.C. and Rudy, T.E., The West Haven-Yale Multidimensional Pain Inventory (WHYMPI), Pain, 23 (1985) 345-356. 25 Leavitt. F. and Garron, D.C., Demographic and affective covariates of pain. Psychosom. Med., 41 (1979) 525-533. 26 Lennon, M.C., Link, B.G., Marbach, J.J. and Dohrenwend. B.P., The stigma of chronic facial pain and its impact on social relationships, Sot. Probl., 36 (1989) 117-134. 27 Litwak, E. and Szelenyi, I., Primary group structures and their functions: kin, neighbors, and friends, Am. Social. Rev., 34 (1969) 4655481. 28 Locker, D., Disability and Disadvantage: the Consequences of Chronic Illness. Tavistock, London, 1983. 211 pp.

43

29 Lynn, R. and Eysenck, H.J., Tolerance for pain, extraversion and neuroticism, Percept. Motor Skills, 12 (1961) 161-162. 30 Marks, R.C., River, L., Kimball, C.P. and Medof, M.E., Fibrositis: psychological profile, Psychosom. Med., 45 (1983) 82. 31 Melzack, R., The McGill Pain Questionnaire. In: R. Melzack (Ed.), Pain Measurement and Assessment, Raven Press, New York, 1983, pp. 41-48. 32 Moos, R. and Moos, B., Family Environment Scale Manual (2nd Edn.), Consulting Psychologists’ Press, Palo Alto, CA, 1981, 63 pp. 33 Munro, B.H., Visintainer, M.A. and Page, E.B., Statistical Methods for Health Care Research, Lippincott, Philadelphia, PA, 1986, 381 pp. 34 Murphy, S., Creed, F. and Jayson, M., Psychiatric disorders and illness behavior in rheumatoid arthritis, Br. J. Rheumatol., 27 (1988) 357-363. 35 Oberst, M., Nurses’ inferences of suffering: the effects of nursepatient similarity and verbalizations of distress. In: M.J. Nelson (Ed.), Clinical Perspectives in Nursing Research, Teachers College Press, New York, 1976, pp. 39-60. 36 Pagel, M.D., Erdly, W.W. and Becker, J., Social networks: we get by with (and in spite of) a little help from our friends, J. Pers. Sot. Psychol., 53 (1987) 793-804. 37 Primomo, J., Yates, B.C. and Woods, N.F., Social support for women during chronic illness: the relationship among sources and types to adjustment, Res. Nurs. Health, 13 (1990) 153-161. 38 Rook, K.S., The negative side of social interaction: impact on psychological well-being, J. Pers. Sot. Psychol., 46 (1984) 10971108. 39 Rowat, K.M. and Knafl, K., Living with chronic pain: the spouse’s perspective, Pain, 23 (1985) 259-271. 40 Rudy, T.E., Kerns, R.D. and Turk, D.C., Chronic pain and depression: toward a cognitive-behavioral mediation model, Pain, 35 (1988) 129-140. 41 Sandier, I. and Barrera, M., Toward a multimethod approach to assessing the effects of social support, Am. J. Commun. Psychol., 12 (1984) 37-51. 42 Schultz, R. and Decker, S., Long-term adjustment to physical

43

44 45

46

47

48 49 50

51

52

53

disability: the role of social support, perceived control, and selfblame, J. Pers. Sot. Psychol., 48 (1985) 1162-1172. Simons, D.G., Myofascial pain syndromes due to trigger points. In: M. Osterweis, A. Kleinman and D. Mechanic (Eds.), Pain and Disability, National Academy Press, Washington, DC, 1987, pp. 285-292. Smythe, H.A., “Fibrositis” as a disorder of pain modulation, Clin. Rbeum. Dis., 5 (1979) 823-832. Smythe, H.A., “Fibrositis” and other diffuse musculoskeletal syndromes. In: W.N. Kelley, E.D. Harris, Jr., S. Ruddy and C.B. Sledge (Eds.), Textbook of Rheumatology, 2nd Edn., Saunders, Philadelphia, PA, 1985, pp. 481-489. Suls, J., Social support, interpersonal relations, and health: benefits and liabilities. In: G.S. Sanders and J. Suls (Eds.), Social Psychology of Health and Illness, Erlbaum, Hillsdale, NJ, 1982, pp. 255-277. Taylor, A.G., Skelton, J. and Butcher, J., Duration of pain condition and physical pathology as determinants of nurses’ assessments of patients in pain, Nurs. Res., 33 (1984) 4-8. Tilden, V., Nelson, C. and May, B., Use of qualitative methods to enhance content validity, Nurs. Res., 39 (1990) 172-175. Turner, R.J. and Noh, S., Physical disability and depression: a longitudinal analysis, J. Health Sot. Behav., 29 (1988) 23-37. Vassend, O., Dimensions of negative affectivity, self-reported somatic symptoms, and health-related behaviors, Sot. Sci. Med., 28 (1989) 29-36. Vinokur, A., Schul, S. and Caplan, R.D., Determinants of perceived social support: interpersonal transactions, personal outlook, and transient affective states, J. Pers. Sot. Psychol., 53 (1987) 1137-1145. Watson, D., Clark, L.A. and Tellegen, A., Development and validation of brief measures of positive and negative affect: the PANAS scales, J. Pers. Sot. Psychol., 54 (1988) 1063-1070. Watson, D. and Pennebaker, J.W., Health complaints, stress, and distress: exploring the central role of negative affectivity, Psychol. Rev., 96 (1989) 234-254.