Full text of DOI:10.1207/S15324796ABM2302_8

Distress and Disease Status Among Patients With Rheumatoid Arthritis:
Roles of Coping Styles and Perceived Responses From Support Providers
Kenneth W. Griffin, Ph.D., M.P.H.
Weill Medical College of Cornell University
Ronald Friend, Ph.D., Alan T. Kaell, M.D., and Ronald S. Bennett, M.D.
State University of New York at Stony Brook
ABSTRACT
in pain (5). Other research has shown that daily stressful events
can be strongly associated with next-day mood disturbance
among patients with RA that have lower levels of social support,
whereas daily events are less closely associated with subsequent
mood for those with higher levels of support (6).
Previous research has shown that social support can have a
beneficial impact on coping processes and psychological ad-
justment in patients with rheumatoid arthritis (RA). The associ-
ation of individual coping styles and perceived responses from
others to one’s pain episodes with patients’distress and disease
status over time was investigated. The sample consisted of 42
middle-aged patients with RA who were predominantly White
(98%), female (64%), and married (88%). Participants com-
pleted surveys and their rheumatologist completed clinical as-
sessments of patient disease status at 2 time points over a
9-month period. Although punishing responses from others
(e.g., getting irritated or angry when the patient is in pain) were
perceived as relatively infrequent, they were associated with a
patient coping style of focusing on and venting of negative emo-
tion as well as elevated negative affect (NA). Findings also indi-
cated that those who perceived punishing responses from close
others and coped by venting negative emotions reported in-
creased NA over time and were rated by their rheumatologist as
having more severe RA disease status over time. Implications
for psychosocial intervention and directions for future research
are discussed.
Social networks of patients with RA often consist of a small
number of intimate relationships with family and friends, with
few opportunities for new social contacts (7). Accordingly, pa-
tients with RA tend to rely heavily on a partner or other primary
support provider for daily assistance and regular social support.
These close others may include the patient’s spouse, unmarried
partner, daughter, son, or other support provider who provides
tangible and emotional support on a regular basis. Patients with
RA are likely to come to rely heavily on these support providers
for various types of informal helping. A number of studies have
examined the role of the spouse, partner, or other close support
provider in the lives of patients with RA (8,9). Patients living
with a spouse or partner have been found to report more social
support, less depression and anxiety, and greater psychological
well-being when compared to patients who are without partners
(4,10). Furthermore, married patients with RA have been found
to have lower rates of progression of functional disability when
compared to unmarried patients (11), which may reflect in part
the tangible and emotional support that a spouse or partner can
provide.
(Ann Behav Med
2001, 23(2):133–138)
INTRODUCTION
However, just as close interpersonal relationships can be
supportive and caring, they can also be characterized by con-
flict, criticism, and misunderstanding, and such interactions be-
tween patients with RA and close others can have a negative im-
pact on psychosocial functioning (8,12). For example, research
has shown that patients with RA that have highly critical
spouses are more likely to engage in maladaptive coping and re-
port poor psychological adjustment (8). Problematic interac-
tions in the context of little other support has been found to have
a particularly strong negative effect on psychological adjust-
ment in patients with RA (12). Furthermore, problematic inter-
actions between patient and primary support provider may oc-
cur more frequently when patients are more severely ill or are
more distressed by their illness. Revenson and Majerovitz (9)
found that spouses of patients with RA provided more problem-
atic support (e.g., making unhelpful or upsetting suggestions,
becoming annoyed when their advice is not taken) to patients
with more severe RA. It appears that under certain circum-
stances, an ongoing cycle of negative interactions between pa-
tient and primary support provider may develop whereby pa-
tients place many demands on caregivers, who may in turn
Previous research has shown that social support from fam-
ily members and friends can help those with rheumatoid arthritis
(RA) actively cope with their illness and is associated with good
psychological adjustment in these patients (1–3). Studies have
shown that in addition to being directly associated with less psy-
chological distress, social support has stress-buffering effects in
patients with RA, especially those that are most severely ill (4).
In terms of buffering effects, social support from others can
lessen the effects of both illness and nonillness-related stressors.
For example, patients with RA unsatisfied with the emotional
support they receive have been found to be more likely to be-
come depressed when experiencing pain, whereas those more
satisfied with their support are less likely to get depressed when
Reprint Address: K. W. Griffin, Ph.D., M.P.H., Department of Public
Health, Weill Medical College of Cornell University, 411 East 69th
Street, New York, NY 10021. E-mail: kgriffin@med.cornell.edu
© 2001 by The Society of Behavioral Medicine.
133

134
Griffin et al.
Annals of Behavioral Medicine
TABLE 1
become distressed and overwhelmed by the burden of
caregiving. The support provider may then react negatively to
expressions of pain by the patient with a variety of
nonsupportive or punishing behaviors, which may contribute to
maladaptive coping and poorer psychological and physical
functioning in the patient. Thus, relationships that are emotion-
ally volatile, with frequent verbal expressions of negative emo-
tion from both parties, may play a role in patients’psychological
and physical status. This hypothesis is consistent with a growing
literature on the important role of family influences, family co-
hesion, and family expectations on patient adjustment among
patients with a variety of chronic medical conditions (13–16).
In this study, we investigated how coping styles and per-
ceived responses from others to pain episodes in patients with
RA were associated with patients’ distress and disease status
over time. We examined the frequency and impact of perceived
punishing responses from the patient’s primary support provider
in addition to more helpful ones such as being attentive or dis-
tracting, with a focus on those perceived responses that occur af-
ter pain episodes in patients with RA. We examined whether
perceived negative responses are associated with the patient’s
tendency to cope in adaptive or maladaptive ways and how per-
ceived responses from close others are associated with patients’
psychological distress both concurrently and longitudinally. We
also examined whether perceived negative responses from the
primary support provider predicted changes in RA disease sta-
tus as rated by the patient’s rheumatologist. Our hypotheses
were that perceiving negative responses to pain episodes from a
primary support provider would be associated with maladaptive
coping and greater psychological distress in patients with RA
and that problematic interpersonal processes as perceived by the
patient may lead to higher rates of distress and poorer RA dis-
ease status over time.
Demographic Characteristics of the Sample and Baseline
Descriptive Statistics
%
Women
Married
White
Attended/graduated college
Currently employed
64
88
98
55
60
M
SD
10
Age
55
WHYMPI subscales
Attentive responses
Distracting responses
Punishing responses
COPE subscales
Active coping
Behavioral disengagement
Mental disengagement
Venting
3.7
2.2
1.7
1.5
1.5
1.6
11.8
6.5
8.5
9.2
19.1
3.2
2.1
2.6
2.7
6.9
Negative affect
Note. WHYMPI = The West Haven–Yale Multidimensional Pain
Inventory (20).
months later, or 75% of potential responses. For each participant
who completed the questionnaires, their rheumatologist com-
pleted ratings of disease status at each time point. Thus, com-
plete self-report and physician ratings from two time points
were available for 42 participants, and these patients were in-
cluded in these analyses.
Measures
METHOD
Questionnaire data. The self-report questionnaire included
several standardized measures with established psychometric
properties. The Positive and Negative Affect Scale (PANAS)
(18) was used to assess state negative affect (NA). This scale
consists of 10 single-word items (e.g., “distressed,” “scared,”
“irritable”), and respondents indicated on a 5-point scale the ex-
tent that they had felt this way over the past week (α = .89 in this
study).
To measure coping style, we used several subscales of the
COPE (19), a multidimensional coping inventory that assesses a
variety of dispositional coping techniques. For this study, we
chose four COPE subscales that were particularly relevant to
how one might cope with negative reactions from another per-
son, including active coping, focusing on and venting of nega-
tive emotion, behavioral disengagement, and mental disengage-
ment. We did not include those coping subscales related to
social support to avoid conceptual redundancy with close oth-
ers’ responses to pain. The following COPE subscales were
used: Active Coping is characterized by taking action and solv-
ing problems, Venting is characterized by focusing on and ex-
pressing negative emotions, Behavioral Disengagement is char-
acterized by stopping attempts to solve a problem by
Participants
Participants were recruited from a rheumatology medical
practice located in the suburbs of New York City. Patients were
eligible to participate if they met four of the seven criteria estab-
lished by the College of Rheumatology for a definite RA diag-
nosis (17), as determined by the patient’s rheumatologist. Pa-
tients were excluded if they had any obvious cognitive
impairment or known psychiatric diagnosis. Demographic char-
acteristics of the sample are shown in Table 1.
Procedure
Study packets were mailed to 70 eligible patients and in-
cluded a cover letter inviting patients to participate, a consent
form, an 18-page questionnaire, and a stamped return envelope.
Those who did not respond to the initial mailing were sent a re-
minder postcard after 2 weeks and were telephoned after 4
weeks and asked again to participate in the study. The final re-
sponse rate for the Time 1 mailing was over 85%, and there was
complete questionnaire and physician ratings data available for
56 patients. Of these, 42 patients completed Time 2 surveys 9

Volume 23, Number 2, 2001
Distress and Disease Status
RESULTS
135
withdrawing, and Mental Disengagement is characterized by
shifting one’s attention away from the problem. Response
choices for the COPE are on a 4-point scale ranging from 1 (I
usually don’t do this at all) to 4 (I usually do this a lot).
Cronbach’s alphas for the four COPE subscales ranged from .65
to .85.
The West Haven–Yale Multidimensional Pain Inventory
(20) was used to assess the patient’s perception of support pro-
viders’responses to the patient’s pain. The Significant Other Re-
sponse scale of this inventory asks the patient “How does your
spouse or significant other respond to you when he or she knows
that you are in pain?,” and 14 items tap three dimensions of re-
sponses to pain: punishing responses (α = .87 in our study), such
as “Gets angry with me”; distracting responses (α = .83), such as
“Tries to involve me in some activity”; and attentive responses
(α = .76), such as “Asks me what he/she can do to help.” Partici-
pants respond on a 7-point scale ranging from 0 (never) to 6
(very often). Those participants who were not currently married
responded to this scale using a primary support provider with
whom they had daily contact (e.g., partner, daughter, son).
Levels of NA were somewhat high in the sample, with base-
line scores (M = 19.1, SD = 6.9) higher than published norms for
healthy samples. This is consistent with previous reports of ele-
vated rates of psychological distress in RA samples (22,23).
From Time 1 to Time 2, there was little change in the NA scores.
The duration of illness in our sample ranged from 8 months to
more than 25 years, with a median of 12 years. Duration of ill-
ness and age of participants were unrelated to the outcomes in
this study. In terms of disease status, rheumatologist ratings in-
dicated substantial morbidity in the sample. At baseline, most
patients were classified on the Steinbrocker scale as in the mod-
erate (Stage II, 31%) or severe (Stage III, 62%) stages of RA,
with an additional 7% classified as in the early stages (Stage I).
No patients were classified as very severely ill (Stage IV) at
baseline. From Time 1 to Time 2, the distributions for the
Steinbrocker score changed: For 7 participants, or approxi-
mately 17% of our sample, their disease progressed into the
most severe stage over the time period of this study—a some-
what unexpected finding.
Perceived Responses to Pain and
Patient Coping Styles
Rheumatologist ratings. Each participant’s rheumatologist
completed a questionnaire assessing the patient’s overall disease
status and duration of illness. To measure disease status, we
used the Steinbrocker scale (21), a standardized RA classifica-
tion scheme with specific clinical diagnostic criteria. Using this
scale, the rheumatologist classified the patient into 1 of 4 cate-
gories of disease status: Stage I (early RA), with criteria for clas-
sification including “No destructive changes on
roentgenographic examination”; Stage II (moderate RA), with
criteria of “Roentgenographic evidence of osteoporosis; No
joint deformities”; Stage III (severe RA), with criteria of
“Roentgenographic evidence of cartilage and bone destruction;
Joint deformity”; and Stage IV (very severe RA) with criteria in-
cluding presence of “Fibrous or bony ankylosis.”
Rheumatologists also indicated how many months it had been
since the patient was diagnosed with RA.
First, we looked at the frequency of perceived attentive,
punishing, and distracting responses from the primary sup-
port provider. As shown in Table 1, perceived responses fell
in order of helpfulness, with attentive responses as most com-
mon, followed by distracting and punishing responses. Fur-
ther analyses indicated that patients with the longest duration
of illness reported less perceived attentive responses to pain
from the primary support provider (r = –.30, p < .05), al-
though the perception of negative responses did not increase.
Second, we examined how perceived responses to pain from
the support provider were associated with various indicators
of coping and NA in patients. As shown in Table 2, attentive
responses to pain were associated with an active coping style
and less behavioral disengagement in patients. However,
TABLE 2
Correlations Among Disease Status, Perceived Support Provider’s Response to Pain, Patient Coping Style
and Psychological Distress Variables at Time 1
PR
1
DR
2
AR
3
AC
4
VC
5
BDC
6
MDC
7
NA
8
DS
9
—
1. Punishing responses
2. Distracting responses
3. Attentive responses
4. Active coping
–.39
–.52
–.19
.45
—
.51
.27
—
.33
—
.23
–.21
.28
–.01
.24
5. Venting coping
–.18
–.19
–.04
–.07
.04
–.06
–.37
–.10
–.04
.06
—
.27
.34
.31
.31
6. Behavioral disengagement coping
7. Mental disengagement coping
8. Negative affect
.35
.25
.32
.04
—
.36
.40
—
.56
.17
—
–.06
9. Disease status
–.30
—
Note. |r| ≥ .31, p < .05. PR = punishing responses; DR = distracting responses; AR = attentive responses; AC = active coping; VC = venting coping; BDC =
behavioral disengagement coping; MDC = mental disengagement coping; NA = negative affect; DS = disease status.

136
Griffin et al.
Annals of Behavioral Medicine
most notable, the strongest correlation between the perceived
responses to pain from the support provider and the patient
coping styles was between punishing responses and focusing
on/venting negative emotion. Furthermore, perceived punish-
ing responses to pain were associated with greater levels of
NA. From these cross-sectional analyses at baseline, how-
ever, we cannot determine whether perceived negative re-
sponses to pain led to greater NA, or whether distress tended
to elicit more punishing responses (perceived or actual) from
the primary support provider.
Long-Term Effects of Patient Coping Styles
and Perceived Responses to Pain
Next we examined to what extent patient coping and per-
ceived responses to pain predicted Time 2 NA and disease status
in a series of multiple regression analyses, controlling for base-
line NA and disease status. These analyses indicated the follow-
ing significant main effects. In terms of patient coping styles,
mental disengagement coping predicted an increase in NA from
Time 1 to Time 2 (β = .80, p < .05). In terms of perceived re-
sponses to pain, attentive responses from a close support pro-
vider predicted a significant decrease in disease status at Time 2
(β = –.19, p < .05).
FIGURE 1 Adjusted means for Time 2 negative affect (NA) scores.
Covariate = Time 1 NA and Steinbrocker disease stage; Independent
Variables = perceived punishing responses from support provider, pa-
tient venting of emotion, and the Punish × Venting interaction term,
F(1, 36) = 4.0, p < .05.
To examine potential interactive effects of perceived re-
sponses to pain and coping styles on subsequent NA and disease
status, a series of analysis of covariance models were tested.
Participants were categorized as either high or low on each pre-
dictor variable based on median splits, and the main effects and
interactions of the predictors were examined, controlling for
baseline NA and disease status. In the first set of analyses, the
Time 2 NA score was the dependent variable, and in the second
set of analyses, the Time 2 disease status score was the depend-
ent variable. For all analyses the Time 1 NA and disease status
scores were covariates, and the independent variables were the
perceived response score and coping score at Time 1 and their
interaction term.
From these analyses, the combination of perceived punish-
ing responses to pain and the coping style of venting emotions
was revealed as a significant predictor of both outcome variables.
First, the Punish × Venting interaction term had a significant ef-
fect on Time 2 NA, F(1, 36) = 4.0, p < .05, although the relevant
main effects were not significant. These predictor variables, along
with baseline NA and disease stage, accounted for a large propor-
tion of variance in Time 2 NA (R2 = .73). As shown in Figure 1,
those patients in potentially emotionally volatile relationships
(i.e., the patient copes by venting negative emotion and perceived
the support provider to exhibit punishing responses to pain epi-
sodes) showed increased NA over time. Post hoc comparisons of
adjusted means showed that the greatest difference in mean NA
scores across groups was between the high-venting, low-punish
group (M = 16.9, standard error [SE] = 1.8) and the high-venting,
high-punish group (M = 22.4, SE = 1.3).
FIGURE 2 Adjusted means for Time 2 Steinbrocker rheumatoid ar-
thritis disease progression scores. Covariate = Time 1 negative affect
and Steinbrocker disease stage; Independent Variables = perceived
punishing responses from support provider, patient venting of emotion,
and the Punish × Venting interaction term, F(1, 36) = 5.2, p < .03).
ishing responses to pain from the support provider occurred
within relationships in which the patient tends to cope by vent-
ing his or her emotions, this led to poorer disease status over
time. These predictor variables, along with baseline NA and dis-
ease stage, accounted for a moderate proportion of variance in
Time 2 disease status (R2 = .22). Post hoc comparisons of ad-
justed means showed that the greatest difference in disease sta-
tus was between the high-venting, low-punish group (M = 2.4,
SE = 0.3) and the high-venting, high-punish group (M = 3.0, SE
= 0.2). Taken together, these analyses revealed that perceived
punishing responses to pain from the support provider affected
the patients’psychological distress levels and disease status over
time for patients who cope by focusing on and venting emotion.
Second, analyses showed that the Punish × Venting interac-
tion term had a significant effect on Time 2 Steinbrocker disease
status scores F(1, 36) = 5.2, p < .03, although the main effects
were not significant. As shown in Figure 2, when perceived pun-

Volume 23, Number 2, 2001
DISCUSSION
Distress and Disease Status
137
volatile close interpersonal relationships that are harmful to the
patient over time, rather than the simple main effects of the pa-
tient’s maladaptive coping or perceived negative responses from
others. Our findings suggest that patients with RA in volatile
close relationships may have increased NA and poorer disease
status over time, and this is consistent with previous studies
showing that patients with RA are particularly reactive to inter-
personal stressors, both psychologically and physiologically. A
study by Zautra et al. (25) found that interpersonal conflict
events in patients with RA were associated with greater depres-
sion and elevated levels of the immune-stimulating hormones
prolactin and estradiol in patients. Similar physiological pro-
cesses may contribute to increased disease progression among
patients with RA in emotionally volatile relationships. Another
interesting finding is that patients who reported low venting of
negative emotions and perceived low punishing from a support
provider were rated by their physician as having poorer levels of
disease status (see Figure 2). This suggests that the lack of ver-
bal expression in both patient and support provider may be asso-
ciated with the patient’s disease status, which is consistent with
research showing the negative effects of social isolation and lack
of social support on physical health (26). Although a number of
factors can contribute to elevated levels of psychological dis-
tress in persons with RA, it should be kept in mind that the rates
of depression are similar to those for persons with other serious
chronic conditions (27) and many of the interpersonal phenom-
ena described here may generalize to patients living with other
chronic conditions.
In this study, we examined RA patients’perceptions of how
close others respond to pain episodes and how these perceived
responses were associated with patients’ coping styles, psycho-
logical distress, and disease status over a 9-month period. Find-
ings indicated that the most common perceived reaction from
the primary support provider was to be attentive when the pa-
tient was in pain, followed in frequency by distracting and pun-
ishing responses. Although punishing responses (such as getting
irritated or angry when the patient is in pain) were perceived as
relatively infrequent, these responses were associated with
maladaptive coping and greater NA in patients. Contrary to ex-
pectation, punishing responses were not perceived as more com-
mon in the most severely ill patients, although patients with the
longest duration of illness reported less perceived attentive re-
sponses from support providers. This latter finding may reflect a
decreased involvement of the support provider as patients gain
more experience with their illness and become increasingly
self-reliant, or it may reflect the fact that supportive partners
may become tired with the long-lasting burden of caring for the
patient and eventually lose energy for attentiveness.
An important finding of this study was that perceived pun-
ishing behaviors from the primary support provider in response
to pain episodes in patients with RA were associated with poten-
tially maladaptive coping in patients, particularly in terms of fo-
cusing on and venting of negative emotion. This finding sug-
gests that for some patients with RA, certain close relationships
may become emotional volatile, with frequent expressions of
negative emotions from both parties. Furthermore, these rela-
tionships are likely to be a major source of interpersonal stress
for patients with RA. We also found that perceived punishing re-
sponses from the support provider were associated with greater
behavioral disengagement in patients. Thus, in the face of per-
ceived negative or punishing behavior from close others, pa-
tients in this study tended to either exhibit potentially explosive
behaviors (venting of negative emotions) or attempt to withdraw
(behavioral disengagement) in response. This is consistent with
previous findings that patients with RA often fail to use active
coping techniques when dealing with problems involving social
relationships, although they may use active coping techniques
for problems related to daily activities, leisure, and work (24).
This study showed that close relationships characterized by
perceived punishing responses from the support provider and
venting of negative emotion by the patient led to increased nega-
tive affectivity in the patient over time. In addition, these pa-
tients had poorer disease status over time, as indicated by
rheumatologist ratings on the Steinbrocker scale. It should be
noted that, on the Steinbrocker scale, categories are based on
clinical observation by physician, with specific criteria required
for classification into a particular stage. Thus, it is unlikely that
the rheumatologist ratings of disease status were unduly influ-
enced by factors associated with the patient’s psychological
state or mood. Also, our findings indicated that the interaction of
patient venting and perceived punishing from support provider
significantly predicted greater distress and poorer disease status.
This suggests that it is the negative qualities of these potentially
There are several strengths and limitations to this study.
Among the strengths are the longitudinal design and the use of
physician ratings rather than patient self-reports to assess dis-
ease status. This is the first study we know of that provides evi-
dence that emotionally volatile close relationships among pa-
tients with RA can predict change over time in two important
outcomes (patient NA and disease status). Furthermore, a simi-
lar pattern was observed for outcomes assessed by two inde-
pendent sources (patient self-reports of NA and physician rat-
ings of disease status). Limitations include the small sample size
and that physicians making ratings were not masked to the pa-
tient’s disease status. One limitation is that our measure of reac-
tions to pain episodes were based on the patient’s perception
rather than reports by significant others themselves. We also did
not measure the psychological burden of caregiving that may
have been experienced by the support providers in this study,
and we did not do a comprehensive assessment of patients’pain
levels. Measuring the actual behavior and experiences of the
support provider would have helped in determining whether a
negative interactional style in fact occurs when close others be-
come emotionally overwhelmed by caregiving. Another limita-
tion is that the sample was largely married, White, and middle
class, and findings may have limited generalizability to patients
with other demographic characteristics; the sample was small,
and some of the findings would be nonsignificant if Bonferroni
corrections were applied. Future research should examine
interactional styles using a variety of methods (e.g., interview
assessments of both parties, direct or videotaped observation,

138
Griffin et al.
Annals of Behavioral Medicine
(10) Kraaimaat FW, Van Dam-Baggen CM, Bijlsma JW: Depres-
sion, anxiety, and social support in rheumatoid arthritic women
with and without a spouse. Psychology and Health. 1995,
10:387–396.
(11) Ward MM, Leigh JP: Marital status and the progression of func-
tional disability in patients with rheumatoid arthritis. Arthritis
and Rheumatism. 1993, 36:581–588.
(12) Revenson TA, Schiaffino KN, Majerovitz S, Gibofsky A: Social
support as a double-edged sword: The relation of positive and
problematic support to depression among rheumatoid arthritis
patients. Social Science and Medicine. 1991, 33:807–813.
(13) Hatchett L, Friend R, Symister P, Wadhwa N: Interpersonal ex-
pectations, social support, and adjustment to chronic illness.
Journal of Personality and Social Psychology. 1997,
73:560–573.
(14) Kerns RD, Haythornthwaite J, Southwick S, Giller EL: The role
of marital interaction in chronic pain and depressive symptom
severity. Journal of Psychosomatic Research. 1990,
34:401–408.
(15) Kerns RD, Weiss LH: Family influences on the course of
chronic illness: A cognitive-behavioral transactional model.
Annals of Behavioral Medicine. 1994, 16:116–121.
(16) Nicassio PM, Radojevic V, Schoenfeld-Smith K, Dwyer K: The
contribution of family cohesion and the pain-coping process to
depressive symptoms in fibromyalgia. Annals of Behavioral
Medicine. 1995, 17:349–356.
(17) Arnett FC, Edworthy S, Block DA, et al.: The 1987 revised
ARA criteria for rheumatoid arthritis. Arthritis and Rheuma-
tism. 1987, 30:S17.
etc.) and further explore the role of caregiving burden in the de-
velopment of emotionally volatile close relationships in those
with chronic illness. Future work should further examine the
role of interpersonal dynamics in the patient’s experience of
rheumatoid arthritis and investigate the potential positive effects
that might be obtained from interventions that focus on interper-
sonal dynamics.
A recent study (28) found that self-disclosure of emotions
concerning a stressful event (through writing exercises) led to a
reduction in disease activity in patients with RA. Our findings
suggest that venting of negative emotions within a volatile close
relationship in which the partner exhibits punishing behavior
may increase disease activity. Thus, not all emotional expres-
sion appears to be equally beneficial in terms of disease activity
in patients with RA. In terms of implications for intervention,
these findings indicate that spouses and other support providers
should be encouraged to help the patient and avoid reacting neg-
atively to the patient’s pain whenever possible. To facilitate this,
support providers should be encouraged to seek support them-
selves in dealing with the demands of caregiving, especially
when the patient’s illness is more severe. An important goal of
future research should be to understand how close relationships
among those with chronic illness can be utilized to promote
mental and physical functioning in patients.
REFERENCES
(1) Affleck G, Pfeiffer CA, Tennen J, Fifield J: Social support and
psychological adjustment to rheumatoid arthritis: Quantitative
and qualitative findings. Arthritis Care and Research. 1988,
1:71–77.
(18) Watson D, Clark LA, Tellegen A: Development and validation
of brief measures of positive and negative affect: The PANAS
Scales. Journal of Personality and Social Psychology. 1988,
54:1063–1070.
(2) Fitzpatrick R, Newman S, Lamb R, Shipley M: Social relation-
ships and psychological well-being in rheumatoid arthritis. So-
cial Science and Medicine. 1988, 27:399–403.
(19) Carver CS, Scheier MF, Weintraub JK: Assessing coping strate-
gies: A theoretically based approach. Journal of Personality
and Social Psychology. 1989, 56:267–283.
(3) Goodenow C, Reisine ST, Grady KE: Quality of social support
and associated social and psychological functioning in women
with rheumatoid arthritis. Health Psychology. 1990,
9:266–284.
(4) Penninx BW, van Tilburg T, Deeg DJ, et al.: Direct and buffer
effects of social support and personal coping resources in indi-
viduals with arthritis. Social Science and Medicine. 1997,
44:393–402.
(5) Brown GK, Wallston KA, Nicassio PM: Social support and de-
pression in rheumatoid arthritis: A one-year prospective study.
Journal of Applied Social Psychology. 1989, 19:1146–1181.
(6) Affleck G, Tennen H, Urrows S, Higgins P: Person and contex-
tual features of daily stress reactivity: Individual differences in
relations of undesirable daily events with mood disturbance and
chronic pain intensity. Journal of Personality and Social Psy-
chology. 1994, 66:329–340.
(7) Bolwijn PH, van Santen-Hoeufft MH, Baars HM, van der Lin-
den S: Social network characteristics in fibromyalgia or rheu-
matoid arthritis. Arthritis Care and Research. 1994, 7:46–49.
(8) Manne SL, Zautra AJ: Spouse criticism and support: Their as-
sociation with coping and psychological adjustment among
women with rheumatoid arthritis. Journal of Personality and
Social Psychology. 1989, 56:608–617.
(20) Kerns RD, Turk DC, Rudy TE: The West Haven–Yale Multidi-
mensional Pain Inventory (WHYMPI). Pain. 1985,
23:345–356.
(21) Steinbrocker O, Traeger CH, Balterman RC: Therapeutic crite-
ria for rheumatoid arthritis. Journal of the American Medical
Association. 1949, 140:659–662.
(22) Anderson KO, Bradley LA, Young LD, McDaniel LK, Wise
CM: Rheumatoid arthritis: Review of psychological factors re-
lated to etiology, effects, and treatment. Psychological Bulletin.
1985, 98:358–387.
(23) Young LD: Psychological factors in rheumatoid arthritis. Jour-
nal of Consulting and Clinical Psychology. 1992, 60:619–627.
(24) Blalock SJ, DeVellis BM, Holt K, Hahn PM: Coping with rheu-
matoid arthritis: Is one problem the same as another? Health
Education Quarterly. 1993, 20:119–132.
(25) Zautra AJ, Burleson MH, Matt KS, Roth S, Burrows L: Inter-
personal stress, depression, and disease activity in rheumatoid
arthritis and osteoarthritis patients. Health Psychology. 1994,
13:139–148.
(26) Seeman TE: Social ties and health: The benefits of social inte-
gration. Annals of Epidemiology. 1996, 6:442–451.
(27) DeVellis BM: The physiological impact of arthritis: Prevalence
of depression. Arthritis Care and Research. 1995, 8:284–289.
(28) Revenson TA, Majerovitz SD: The effects of chronic illness on
the spouse: Social resources as stress buffers. Arthritis Care
and Research. 1991, 4:63–72.
(9) Revenson TA, Majerovitz SD: Spouses’ support provision to
chronically ill patients. Journal of Social and Personal Rela-
tionships. 1990, 7:575–586.