SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHOD
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

Objective

Supportive close relationships are important for health. Mutuality, the reciprocal sharing of thoughts and feelings in close relationships, is linked with better outcomes for patients with rheumatoid arthritis (RA) in cross-sectional data. Hypothesizing that mutuality has a beneficial impact on inflammation, we tested potentially causal relations of couple mutuality with erythrocyte sedimentation rate (ESR) in prospective data.

Methods

Female patients with RA (n = 70; mean age 57 years, mean RA disease duration 5 years) completed questionnaires at baseline, 6 months, and 12 months, including measures of mutuality, RA flares, and negative affect. ESR laboratory values available near questionnaire dates were collected from medical charts. Using regression, we examined cross-lagged effects of mutuality and ESR over the two 6-month time spans (baseline to 6 months, 6 months to 12 months). We anticipated that mutuality would exert lagged inverse effects on subsequent ESR levels, and that ESR would have no effect on subsequent mutuality levels.

Results

After controlling for lagged effects of earlier inflammation, disease-modifying antirheumatic drugs, antiinflammatory drugs, RA flares, and negative affect, mutuality's lagged inverse effects over both time spans accounted for unique variance in subsequent levels of ESR, explaining 9% at 6 months and 12.5% at 12 months. Concomitantly, earlier ESR had no effect on subsequent mutuality.

Conclusion

Patients with RA reporting more mutuality had less inflammation at subsequent time points, but inflammation had no effect on subsequent reports of mutuality. This suggests that mutuality exerts a beneficial effect on inflammation. Clinical implications and potential applications are discussed.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHOD
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

Social relationships have long-term implications for health (1). In community samples, social isolation compared with known risk factors (smoking, obesity, and hypertension) in predicting increased morbidity and mortality (2). Similarly, in clinical samples, dependable nonhousehold social relationships predicted the survival of breast cancer patients at 9-year followup (3), and marital quality predicted 8-year survival among heart failure patients (4).

Social relationships, like other forms of social support, exert health-beneficial effects through biologic and behavioral pathways by influencing behaviors, cognitions, and affect (5). For example, social relationships may influence individuals to exercise, diet, and maintain medication regimens (behavioral pathways). Concomitantly, supportive relationships may influence cognitions and provide affective experiences that help regulate neuroendocrine, immune, and cardiovascular function (biologic pathways) (5–7).

Social relationships may have greater impact through biologic pathways for persons with rheumatoid arthritis (RA), a systemic inflammatory autoimmune disorder, due to their implications for the stress- and emotion-activated neuroendocrines relevant to RA disease activity. Zautra and colleagues (8) found that patients with RA are more psychologically and physiologically reactive to interpersonal stressors. Specifically, associations among interpersonal conflicts, depression, and the immune-stimulating hormones prolactin and estradiol were larger for patients with RA than for controls with osteoarthritis (8). Additionally, increased interpersonal stress resulted in increases from baseline in immunologic and clinical disease activity for women with RA (9). Similarly, chronic interpersonal stressors were linked with cellular inflammation, measured as in vitro production of the inflammatory cytokine interleukin-6 by the immune cells of patients with RA, and with resistance of this cytokine production to glucocorticoid inhibition (10).

Aspects of close, long-term social relationships such as marital status and couple relationship quality have also been linked with RA outcomes, leading to an increased focus on couple factors in RA coping (11, 12). For example, single patients with RA developed disability more rapidly than did married/partnered patients with RA over a 10-year period (13). Beyond simple marital status, the quality of spouse relationships appears to be important in the health of patients with RA. Spouses' negative reactions to RA patients' pain predicted depression at baseline and pain at 1-year followup (14). Similarly, spouse criticism and negative responses predicted increased depression and anxiety (15) and poorer coping (16), as well as larger increases in disease activity and anxiety from baseline to the occurrence of an interpersonally stressful period (17). In light of foreseeable long-term daily exposures to partners, such findings suggest that the couple relationship quality of patients with RA warrants attention.

As demonstrated above, couple relationship quality has often been measured as a stressor or risk factor in RA. However, the development of psychosocial interventions to enhance the health of patients with RA would be facilitated by clear descriptions of positive relational behaviors known to enhance health that could usefully serve as therapeutic targets.

Mutuality, a positive relationship quality of connectedness, is marked by the reciprocal sharing of thoughts and feelings in close relationships. Theorized as important for women's psychological development and health, mutuality is characterized by engaged, authentic, empathic exchanges that empower and broaden both partners (18, 19). Mutuality shares the features of responsiveness and empathy with two similar constructs. One is the model of intimacy, in which benefits of being encouraged to disclose accrue primarily to the discloser, who feels understood and valued by an empathic listener (20). The other is emotional responsiveness, defined as addressing the communications, needs, wishes, and actions of one's partner (21), thereby increasing that partner's feeling of being understood and valued. However, the conceptualization of mutuality differs from intimacy and emotional responsiveness in its emphasis on equal empowerment of both partners by their shared understanding resulting from empathic exchanges (18).

Measured as the frequency of partners' engaged, authentic, empathic responses during important conversations (22), couple mutuality has been linked with psychological health for women. Consistently associated with fewer symptoms of depression (23–27), mutuality also predicts more self-care agency in women coping with cancer (26) and discriminates between eating-disordered women and healthy controls (27). In patients with RA, couple mutuality is linked in cross-sectional data with better outcomes, including fewer symptoms of depression and anxiety, less physical disability, and less overall arthritis impact (28).

It is not known whether mutuality merely correlates with RA health or has a beneficial (causal) role. No experimental manipulations of mutuality have been conducted to provide evidence addressing this question. Lacking such experimental data, the exploration of potentially causal relations can usefully begin with an examination of cross-lagged effects in observational prospective data (29). The underlying logic and goal of such an examination is to provide some evidence of temporal precedence, and hence causal predominance, of one variable relative to the other. In essence, where A exerts lagged effects on B, and where B does not exert lagged effects on A, we may entertain an initial hypothesis that A is causally prior to and exerts influence on B. The absence of differences in cross-lagged correlations has been proposed as evidence of spurious associations between two variables that are actually driven by unlabeled intervening variables (30). Although confidence in causal orderings ultimately requires evidence from experimental data, an examination of cross-lagged effects can be considered a first stage in developing arguments for causal orderings (29).

Regressions examining cross-lagged effects have been used in preliminary analyses, revealing that the baseline mutuality of patients with RA predicted better outcomes at 6 months, including less arthritis impact, less anxiety, less pain and fatigue, better overall health, and less physical disability, after controlling baseline outcome levels. These analyses were consistent in finding lagged effects of mutuality on health outcomes, suggesting a beneficial effect of mutuality in RA health (31–35). However, these preliminary analyses captured only one 6-month period of prospective data. Confirmation of such results in a second time span would increase confidence in the evidence of causal relations.

In this study, we tested potentially causal relations of couple mutuality and an RA-relevant health indicator, inflammation, using prospective observational data obtained by questionnaire and medical chart review from female patients with RA that included 2 time spans for cross-lagged analysis. Utilizing erythrocyte sedimentation rate (ESR) as our measure of inflammation, we hypothesized that mutuality exerts a beneficial effect that is causally prior to subsequent levels of inflammation. As a corollary, we hypothesized that ESR is not causally prior to mutuality. Therefore, we expected that mutuality would exert lagged inverse effects on ESR, and in contrast that ESR would not exert lagged effects on mutuality.

MATERIALS AND METHOD

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHOD
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

Participants and design.

Data from a prospective survey study were used for the current analysis. The study was conducted in compliance with the Helsinki Declaration; ethical approval was granted by the University of Arizona Institutional Review Board. Female patients with RA recruited through rheumatology clinics gave informed consent and completed questionnaires at baseline, 6 months, and 12 months. Questionnaires included measures of demographics, medical history, current medications, disease flares, negative affect, and couple mutuality. In addition, participants' medical charts were reviewed for laboratory measures of inflammation, as available, near the time of the questionnaires. These inflammation measures were ordered by the participants' rheumatologists in connection with routine care using standard procedures: blood specimens obtained by venipuncture were sent to clinical laboratories for standard rheumatology assays of acute-phase reactants. We chose ESR as our target measure of inflammation because it is widely used clinically in tracking disease activity and treatment response in patients with RA (36, 37). Another acute-phase reactant, C-reactive protein (CRP), is also widely used to monitor RA disease activity (37) and has the advantage of being more chemically stable than ESR after the blood has been drawn. However, CRP presented disadvantages within the context of our study. Three times more variance in CRP levels is explained by quantity and distribution of body fat in women than in men (38). Likewise, in an RA sample, CRP level was associated with body fat in female but not male patients (39). Further, CRP levels are higher in women receiving hormone replacement therapy (40) or taking oral contraceptives (41). Such issues pose confounds when using an acute-phase reactant to index RA disease activity in female patients. In addition, factors affecting CRP assay calibration standards (42) can vary among laboratories, potentially obscuring between-patient comparisons when using laboratory values not generated by the same laboratory, as when collected by medical chart review. Therefore, although we controlled for CRP values in hypothesis tests, we did not examine them as dependent variables.

The characteristics of the participants who were included in any of the hypothesis tests for this study (n = 70) are reported in Table 1. Their mean ± SD age was 56.94 ± 11.16 years, and their mean ± SD disease duration was 5.01 ± 1.75 years. On average, participants had been in relationships with their spouse/partners for a mean ± SD of 23.7 ± 15.51 years. Approximately 70% had received some college or more extensive education, 81% were white, 19% were Hispanic, and 12% were receiving disability benefits.

Table 1. Sample descriptions*
VariablesParticipants, no.Value, mean ± SD
  • *

    RA = rheumatoid arthritis; ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; NSAID = nonsteroidal antiinflammatory drug; MTX/LEF = methotrexate or leflunomide; BRM = biologic response modifier.

  • Out of 70 total female patients with RA whose available data permitted their inclusion in any of the hypothesis-testing regressions.

Age, years6956.94 ± 11.16
RA disease duration, years675.01 ± 1.75
Relationship duration, years6823.72 ± 15.51
ESR, mm/hour  
 Baseline6320.32 ± 18.94
 6-month5917.92 ± 17.43
 12-month5013.32 ± 10.85
Mutuality  
 Baseline704.48 ± 0.62
 6-month604.53 ± 0.64
 12-month604.60 ± 0.63
CRP level, mg/liter  
 Baseline665.75 ± 7.12
 6-month618.52 ± 16.73
NSAID/steroid use  
 Baseline690.75 ± 0.63
 6-month570.68 ± 0.63
MTX/LEF use  
 Baseline690.43 ± 0.50
 6-month570.47 ± 0.50
BRM use  
 Baseline690.51 ± 0.50
 6-month570.51 ± 0.50
RA flares, prior 6 months  
 Baseline691.66 ± 2.35
 6-month571.13 ± 1.48
Negative affect  
 Baseline701.72 ± 0.54
 6-month581.61 ± 0.53

Measures.

Variables of interest.

ESR laboratory values (in mm/hour) available near the questionnaire dates were collected from medical charts. The mean interval between questionnaires and laboratory tests was 25 days.

Mutuality was measured using the Mutual Psychological Development Questionnaire (22), a valid and reliable 22-item scale. To capture mutuality's bidirectional nature, the questionnaire uses 2 stems with 11 items each to elicit the respondents' perceptions of the frequencies of their partners' and their own engaged, validating, empathic, and authentic responses during important couple conversations during the past month: “When we talk about things that are important to me, my spouse/partner is likely to …” and “When we talk about things that are important to my spouse/partner, I am likely to …” Response options ranging from 1 to 6 (where 1 = none of the time and 6 = all of the time) are used to rate frequencies of behaviors such as “pick up on my feelings,” “share similar experiences,” “show an interest,” “respect my point of view,” “feel moved,” “get involved,” and “express an opinion clearly.” After the reverse coding of indicated items, scores were taken as means, with possible scores ranging from 1 to 6. In the current sample, Cronbach's alpha internal consistency coefficients ranged from 0.91 to 0.93 across the 3 measurement occasions.

Control variables.

Beyond controlling for the baseline levels of dependent variables in the hypothesis tests, we measured additional individual differences to control their potentially intervening effects in associations of mutuality with ESR.

CRP laboratory values (in mg/liter) available near the questionnaire dates were collected from medical charts. The mean interval between questionnaires and laboratory tests was 25 days. Log transformation was used to improve the distribution normality of CRP laboratory values for use in analyses, but the raw values are reported in the sample description (Table 1).

Disease-modifying antirheumatic drugs (DMARDs) and nonsteroidal antiinflammatory drugs (NSAIDs)/steroidal antiinflammatory drugs were coded from medication lists for use as control variables. Uses of DMARDs, methotrexate/leflunomide (MTX/LEF), and biologic response modifiers (BRMs) were binary coded as 0 = none used and 1 = used. Use of NSAIDs/steroids was coded as 0 = none used, 1 = use of any NSAID, and 2 = use of any steroid.

RA disease flares were measured as a single free-entry item as an approximation of disease severity. Respondents were asked to read a definition of flares (episodes lasting ≥7 days with marked increases in painful/swollen joints, overall pain, morning stiffness, and fatigue) and to enter the number of flares meeting the definition that they had experienced during the prior 6 months.

Negative affect was measured using the negative affect subscale of the Positive and Negative Affect Scales (43), a valid and reliable 20-item scale asking respondents to rate the extent to which they experienced specified emotions. For the negative affect subscale, respondents rated the degree to which they experienced negative emotions (e.g., “nervous,” “hostile,” “ashamed”) during the past week using response options ranging from 1 to 5, where 1 = very little/not at all and 5 = extremely. Scores were taken as the mean of items. In the current sample, internal consistency coefficients for the negative affect subscale ranged from 0.85 at baseline to 0.86 at 6 months. Negative affect was included as an approximation of a relevant stable personality trait (44) with expected effects on both mutuality and inflammation.

Statistical analysis.

Sample descriptions were calculated and all variables used in hypothesis tests were examined in a correlation matrix. Four hierarchical regressions were used to examine cross-lagged effects over 2 time spans: baseline to 6 months, and 6 months to 12 months. Each time span was examined for cross-lagged effects of central predictors on dependent variables: of mutuality on ESR and, conversely, of ESR on mutuality (i.e., the reverse-causality test). In all regressions, control variables (including prior levels of dependent variables) were entered in the first step, and then the central predictor was entered in the second step. Regression analyses used all cases with the specified measures available. Because we tested our hypothesis in 2 time spans, we used α = 0.025 for interpreting significance to adjust for alpha slippage.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHOD
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

Sample means and SDs for measures are reported in Table 1. Means for mutuality appeared fairly stable across baseline, 6 months, and 12 months. Means ± SDs for ESR appeared to decrease over time, ranging from 20.32 ± 18.94 mm/hour at baseline to 13.32 ± 10.85 mm/hour at 12 months.

Correlations among all variables are reported in Table 2. Measures of mutuality were highly correlated across time points (r values ranged from 0.80 to 0.87; for all values P < 0.001). ESR levels were moderately correlated across time points (r values ranged from 0.52 to 0.60; for all values P < 0.001). Use of MTX/LEF and BRMs test–retest correlations from baseline to 6 months were high (r = 0.75 and r = 0.82, respectively; P < 0.001), suggesting that most patients were stable on DMARD regimens.

Table 2. Correlations among baseline, 6-month, and 12-month variables (n = 70)*
Variables123456789101112131415161718
  • *

    See Table 1 for definitions.

  • P ≤ 0.001.

  • P ≤ 0.05.

  • §

    Missing data resulted in slightly smaller number of patients.

  • P ≤ 0.1.

  • #

    P ≤ 0.01.

1. Baseline mutuality10.800.82−0.07−0.32−0.180.250.130.160.04−0.02−0.02−0.080.00−0.27−0.02−0.40−0.13
2. 6-month mutuality§ 10.87−0.09−0.34−0.340.180.090.11−0.08−0.09−0.080.040.07−0.17−0.10−0.41−0.21
3. 12-month mutuality§  10.00−0.29−0.140.240.18−0.09−0.18−0.03−0.13−0.040.13−0.15−0.07−0.38#−0.31
4. Baseline ESR§   10.600.590.410.220.210.100.040.03−0.06−0.020.000.210.08−0.02
5. 6-month ESR§    10.520.220.590.210.100.080.10−0.19−0.10−0.080.210.060.25
6. 12-month ESR§     10.370.300.230.13−0.060.020.04−0.05−0.010.31−0.02−0.11
7. Baseline CRP§      10.570.250.11−0.13−0.170.040.100.180.28−0.04−0.06
8. 6-month CRP§       10.11−0.02−0.12−0.09−0.040.030.030.42#−0.060.02
9. Baseline NSAIDs/steroids§        10.74−0.08−0.090.030.13−0.090.23−0.010.17
10. 6-month NSAIDs/steroids§         1−0.21−0.050.150.09−0.150.120.040.21
11. Baseline MTX/LEF§          10.75−0.89−0.72−0.11−0.13−0.09−0.03
12. 6-month MTX/LEF§           1−0.86−0.97−0.03−0.19−0.030.19
13. Baseline BRMs§            10.820.140.160.12−0.04
14. 6-month BRMs§             10.030.170.04−0.17
15. Baseline RA flares§              10.330.36#0.03
16. 6-month RA flares§               10.190.16
17. Baseline negative affect                10.47
18. 6-month negative affect§                 1

Regressions of ESR on mutuality.

In step 1 of the regression predicting 6-month ESR, baseline controls explained 42.7% of the variance, with baseline ESR the only significant predictor in step 1 (β = 0.608, P < 0.001) (Table 3). Baseline MTX/LEF use was not a significant predictor (β = −0.337, P = 0.142), whereas BRM use was marginally significant (β = −0.420, P = 0.078). In step 2, baseline mutuality exerted an inverse lagged effect on 6-month ESR (β = −0.464, P = 0.008), explaining 9% of the additional variance in 6-month ESR. Also in step 2, the estimated effect of MTX/LEF use attained marginal significance (β = −0.411, P = 0.062), and the estimated effect of BRM use emerged as significant (β = −0.537, P = 0.037) (Table 3).

Table 3. Regressions of ESR on mutuality and control variables*
Steps/predictors6-month ESR on baseline predictors (n = 50)12-month ESR on 6-month predictors (n = 33)
ΔR2β step 1β step 2ΔR2β step 1β step 2
  • *

    See Table 1 for definitions.

  • P ≤ 0.001.

  • P < 0.05.

  • §

    P < 0.01.

  • P < 0.08.

  • #

    P ≤ 0.02.

Step 10.427  0.367  
 ESR 0.6080.464§ 0.4060.102
 CRP level −0.130−0.018 −0.1670.031
 NSAIDs/steroids 0.0630.135 0.0400.199
 MTX/LEF −0.337−0.411 0.2190.229
 BRMs −0.420−0.537# 
 RA flares 0.077−0.140 0.3830.237
 Negative affect 0.040−0.039 −0.169−0.297
Step 20.090§  0.125#  
 Mutuality  −0.373§  −0.480#

In step 1 of the regression predicting 12-month ESR, 6-month control variables explained 36.7% of the variance, with 6-month ESR and RA flares as marginally significant predictors of 12-month ESR (β = 0.406 and β = 0.383, respectively; for both, P < 0.08) (Table 3). In step 2, 6-month mutuality exerted an inverse lagged effect (β = −0.480, P = 0.02), explaining an additional 12.5% of the variance in 12-month ESR (Table 3).

The slopes of the 6- and 12-month ESRs, adjusted for control variables, on baseline and 6-month mutuality, respectively, are graphed to illustrate the comparable results of these two regressions in Figure 1.

thumbnail image

Figure 1. Slopes of 6-month (solid line with circles; n = 50) and 12-month (dotted line with squares; n = 33) erythrocyte sedimentation rates (ESRs) on baseline and 6-month mutuality scores, adjusted for earlier levels of C-reactive protein, nonsteroidal and steroidal antiinflammatory drugs, disease-modifying antirheumatic drug use (methotrexate/leflunomide and biologic response modifiers), rheumatoid arthritis disease flares, and negative affect.

Download figure to PowerPoint

Regressions of mutuality on ESR.

In the regression predicting 6-month mutuality, baseline mutuality was the only significant predictor in steps 1 and 2 of the model (β = 0.783 and β = 0.782, respectively; for both, P < 0.001), with negative affect marginally significant (β = −0.167, P < 0.08) (Table 4). In step 2, baseline ESR failed to exert any lagged effect on 6-month mutuality (β = −0.007, not significant).

Table 4. Regressions of mutuality on ESR and control variables*
Steps/predictors6-month mutuality on baseline predictors (n = 51)12-month mutuality on 6-month predictors (n = 38)
ΔR2β step 1β step 2ΔR2β step 1β step 2
  • *

    See Table 1 for definitions.

  • P < 0.001.

  • P < 0.08.

Step 10.725  0.803  
 Mutuality 0.7830.782 0.8660.901
 CRP level 0.0230.027 0.1460.107
 NSAIDs/steroids −0.112−0.111 −0.062−0.078
 MTX/LEF −0.046−0.045 −0.021−0.023
 BRMs 0.0670.068 
 RA flares 0.0140.013 −0.089−0.082
 Negative affect −0.167−0.167 −0.032−0.036
Step 2< 0.001  0.003  
 ESR  −0.007  0.080

In the regression predicting 12-month mutuality, baseline mutuality was the only significant predictor in steps 1 and 2 of the model (β = 0.866 and β = 0.901, respectively; for both, P < 0.001) (Table 4). In step 2, 6-month ESR failed to exert any lagged effect on 12-month mutuality (β = 0.080, not significant).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHOD
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

In this study, we found that couple mutuality exerted a beneficial effect on inflammation for female patients with RA over two 6-month time spans. Utilizing a cross-lagged model for our regression analyses, we determined that female patients with RA who reported more mutuality at baseline and 6 months had lower levels of ESR at 6 and 12 months, respectively. After controlling for lagged effects of earlier inflammation, DMARD regimens, antiinflammatory drugs, RA disease flares, and negative affect, mutuality's lagged inverse effects explained unique variance, accounting for 9% and 12.5% of ESR variance at 6 and 12 months, respectively. The only significant predictor of 12-month ESR was 6-month mutuality (β = −0.480, P = 0.020) (Table 3). Conversely, ESR at baseline and 6 months exerted no effects on subsequent measures of mutuality. These results support our hypothesis that mutuality is a beneficial factor, suggesting mutuality's clinical relevance for female patients with RA. Our findings are congruent with other studies in rheumatology populations that link physical and psychological benefits with couple relationship qualities (9, 14–17, 35), especially those conceptualized as mutuality or emotional responsiveness (28, 45).

It is worth noting that the observed effects of mutuality were not only statistically significant, but could also have clinical significance and potential applications. For women, normal ESR levels are not higher than half the sum of their age in years plus 10 (e.g., ESR ≤25 mm/hour for a 40-year-old woman) (46). A 1-point difference in the mutuality score, representing the difference between reporting that empathic, engaged responses occur most of the time versus all of the time, was associated with estimated differences in ESR levels of 10.7 mm/hour at 6 months and 7.15 mm/hour at 12 months. These estimates represent approximately two-thirds of an SD in ESR for each test. Therefore, modest differences in the mutuality score may denote clinically significant differences in the ESR levels of some patients with RA. Given foreseeable long-term daily exposures within couple relationships, the magnitude of these estimates suggests that asking patients with RA whether they can engage in mutual conversations with partners would have clinical prognostic value in routine care.

Insofar as mutuality and other forms of social support represent unique sources of variance in RA outcomes that reflect clinically significant differences, reliance on exclusively pharmacologic therapies may fall short of optimal care. The fact that many DMARD treatments for RA carry potentially serious side effects yet are not universally effective underscores the desirability of developing psychosocial interventions to promote the health-enhancing relational behaviors that arise naturally in interpersonal relationships. The current study provides initial evidence that the relational behaviors measured as mutuality (engaged, authentic, empathic, and validating responses) exert a beneficial effect relative to inflammation for female patients with RA, suggesting their potential usefulness as therapeutic targets.

In a discussion of social relationships and issues for measurement and intervention, Cohen et al described some successful and unsuccessful attempts to harness social support to benefit chronically ill patients. They concluded that interventions targeting supportive groups and dyads hold great potential, because these formats provide for more tight control over the matching of participants, key maneuvers, and type of support delivered (6). Of these relational formats, we note that couple dyads present a unique therapeutic opportunity in terms of long-term daily exposure to a potentially health-enhancing relationship. Those authors further stated that ultimately the greatest benefits were found for patients who formed new relationships characterized by mutual exchanges (6). Other authors have also concluded that the meanings of giving and receiving support for interpersonal relationships have important implications for the acceptability and value of such support (47, 48). Revenson and colleagues found that problematic support (i.e., support that was not requested or was not responsive to the needs of the recipient) was associated with depression and was itself a source of stress for patients with RA (49). Congruent with those findings, Fekete and colleagues found that spouses' emotional responsiveness mediated the effects of the support they provided on their partners' psychological well-being in a study of couples in which the wife was experiencing a lupus disease flare (45). Such emotional responsiveness is consistent with mutuality.

As conceptualized and measured in the current study, mutuality involves a bidirectional flow of thoughts and feelings in which both partners are empowered by their shared understanding and develop psychologically through engaged, authentic, empathic exchanges (18). For the purposes of designing an intervention to enhance couple mutuality, the notion of mutual benefit to partners is consistent with the observation noted above that the most health benefit was found for patients who experienced mutual exchanges with their helpers (6). These considerations suggest that developing an intervention to enhance couple mutuality could promote sustained health benefits for patients with RA.

Important strengths of the study include its prospective design and the replication of directional results over two time spans. Further, our results are congruent with other studies. In addition, the linkage of a self-report measure of relationship quality with an objective medical measure of inflammation argues persuasively for the importance of psychosocial factors for physical health.

Some important limitations inherent in our nonexperimental design warrant caution in the interpretation and generality of the findings. In the absence of experimental data, an examination of cross-lagged effects provides a useful starting point for causal hypotheses (29). However, confidence in causal orderings should ultimately be based on evidence from an experimental design, such as testing the effect of mutuality enhancement on outcomes. Concomitantly, our nonexperimental design cannot exclude the possibility that observed effects of mutuality on health were due to intervening variables that we did not measure and control, such as individual differences (personality traits, disease severity, etc.) that might influence both mutuality and inflammation. We attempted to account for this type of intervening variable by controlling for negative affect, RA disease flares, and medications as approximations of relevant individual differences (44). We also tested the reverse-causality hypothesis to help rule out other potential intervening variables (Table 4) (30). Another limitation is that only 1 outcome, ESR, was used here to test our hypothesis; confidence in our finding would be increased by replicating this finding in other outcomes, such as physical disability, pain, and fatigue. Additionally, most of our sample of female patients with RA identified themselves as well-educated and white, limiting the generality of our findings to less-educated and more ethnically diverse populations. Finally, the average 23-year couple relationship duration and average 5-year RA disease duration may reflect couple stability and patient experience in coping with RA, forming a context in which couple relationship quality could emerge as a beneficial factor for RA health. It is quite conceivable that we would have found the opposite result in a sample of newly coupled patients with RA dealing with the initial impact of an RA diagnosis, i.e., that disease is a risk factor for couple relationship quality.

In conclusion, our findings join an accumulating body of evidence linking social relationships with health. The current findings suggest that mutuality exerts prospective beneficial effects on inflammation in women with RA. Given the effect sizes and foreseeable ongoing daily contact with partners, couple mutuality could have long-term clinical significance for some patients. Further elucidation of mutuality's role in RA health could inform new interventions targeting health-enhancing relational behaviors to improve outcomes and quality of life for individuals living with arthritis.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHOD
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Kasle had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Kasle, Wilhelm, Zautra.

Acquisition of data. Kasle, Sheikh.

Analysis and interpretation of data. Kasle, Wilhelm, McKnight, Zautra.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHOD
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

The authors thank Kyaw Swe, MD, for his assistance collecting inflammation data from medical charts early in the study.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHOD
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES
  • 1
    Berkman LF. The role of social relations in health promotion. Psychosom Med 1995; 57: 24554.
  • 2
    House JS, Landis KR, Umberson D. Social relationships and health. Science 1988; 241: 5405.
  • 3
    Weihs KL, Simmens SJ, Mizrahi J, Enright TM, Hunt ME, Siegel RS. Dependable social relationships predict overall survival in stages II and III breast carcinoma patients. J Psychosom Res 2005; 59: 299306.
  • 4
    Rohrbaugh MJ, Shoham V, Coyne JC. Effect of marital quality on eight-year survival of patients with heart failure. Am J Cardiol 2006; 98: 106972.
  • 5
    Cohen S, Wills TA. Stress, social support, and the buffering hypothesis. Psychol Bull 1985; 98: 31057.
  • 6
    Cohen S, Gottlieb BH, Underwood LG. Social relationships and health: challenges for measurement and intervention. Adv Mind Body Med 2001; 17: 12941.
  • 7
    Uchino BN, Cacioppo JT, Kiecolt-Glaser JK. The relationship between social support and physiological processes: a review with emphasis on underlying mechanisms and implications for health. Psychol Bull 1996; 119: 488531.
  • 8
    Zautra AJ, Burleson MH, Matt KS, Roth S. Interpersonal stress, depression, and disease activity in rheumatoid arthritis and osteoarthritis patients. Health Psychol 1994; 3: 13948.
  • 9
    Zautra AJ, Hoffman J, Potter P, Matt KS, Yocum D, Castro L. Examination of changes in interpersonal stress as a factor in disease exacerbations among women with rheumatoid arthritis. Ann Behav Med 1997; 19: 27986.
  • 10
    Davis MC, Zautra AJ, Younger J, Motivala SJ, Attrep J, Irwin MR. Chronic stress and regulation of cellular markers of inflammation in rheumatoid arthritis: implications for fatigue. Brain Behav Immun 2008; 22: 2432.
  • 11
    Revenson TA, Abraido-Lanza AF, Majerovitz SD, Jordan C. Couples coping with chronic illness: what's gender got to do with it? Washington, DC: American Psychological Association; 2005.
  • 12
    Sterba KR, DeVellis RF, Lewis MA, DeVellis BM, Jordan JM, Baucom DH. Effect of couple illness perception congruence on psychological adjustment in women with rheumatoid arthritis. Health Psychol 2008; 27: 2219.
  • 13
    Ward MM, Leigh JP. Marital status and the progression of functional disability in patients with rheumatoid arthritis. Arthritis Rheum 1993; 36: 5818.
  • 14
    Waltz M, Kriegel W, van't Pad Bosch P. The social environment and health in rheumatoid arthritis: marital quality predicts individual variability in pain severity. Arthritis Care Res 1998; 11: 35674.
  • 15
    Kraaimaat FW, van Dam-Baggen RM, Bijlsma JW. Association of social support and the spouse's reaction with psychological distress in male and female patients with rheumatoid arthritis. J Rheumatol 1995; 22: 6448.
  • 16
    Manne SL, Zautra AJ. Spouse criticism and support: their association with coping and psychological adjustment among women with rheumatoid arthritis. J Pers Soc Psychol 1989; 56: 60817.
  • 17
    Zautra AJ, Hoffman JM, Matt KS, Yocum D, Potter PT, Castro WL, et al. An examination of individual differences in the relationship between interpersonal stress and disease activity among women with rheumatoid arthritis. Arthritis Care Res 1998; 11: 2719.
  • 18
    Surrey JL. Self-in-relations: a theory of women's development. Adv Dev 1993; 5: 111.
  • 19
    Jordan JV. A relational perspective for understanding women's development. New York: Guilford Press; 1997.
  • 20
    Reis HT, Shaver P. Intimacy as an interpersonal process. Oxford (England): John Wiley & Sons; 1988.
  • 21
    Miller LC, Berg J. Selectivity and urgency in interpersonal exchange. In: DerlegaVJ, editor. Communication, intimacy, and close relationships. Orlando (FL): Academic Press; 1984. p. 161206.
  • 22
    Genero NP, Miller JB, Surrey J. Project report. In: The Mutual Psychological Development Questionnaire. Wellesley (MA): Wellesley College; 1992.
  • 23
    Genero NP, Miller JB, Surrey J, Baldwin LM. Measuring perceived mutuality in close relationships: validation of the Mutual Psychological Development Questionnaire. J Family Psychol 1992; 6: 3648.
  • 24
    Lippes TE. Mutuality in marriage: the self-esteem connection. A self-in-relation analysis [dissertation]. Ann Arbor (MI): ProQuest; 1999.
  • 25
    Sperberg ED, Stabb SD. Depression in women as related to anger and mutuality in relationships. Psychol Women Q 1998; 22: 22338.
  • 26
    Kayser K, Sormanti M, Strainchamps E. Women coping with cancer: the influence of relationship factors on psychosocial adjustment. Psychol Women Q 1999; 23: 72539.
  • 27
    Tantillo M, Sanftner J. The relationship between perceived mutuality and bulimic symptoms, depression, and therapeutic change in group. Eat Behav 2003; 3: 34964.
  • 28
    Kasle S, Wilhelm MS, Zautra AJ. Rheumatoid arthritis patients' perceptions of mutuality in conversations with spouse/partners and their links with psychological and physical health. Arthritis Rheum 2008; 59: 9218.
  • 29
    Locascio JJ. The cross-lagged correlation technique: reconsideration in terms of exploratory utility, assumption specification, and robustness. Educ Psychol Meas 1982; 42: 102336.
  • 30
    Kenny DA. Cross-lagged panel correlation: a test for spuriousness. Psychol Bull 1975; 82: 887903.
  • 31
    Kasle S, Yocum DE, Wilhelm MS. Relational mutuality at baseline predicts functional status 6 months later more strongly for women in a rheumatoid arthritis sample [abstract]. Arthritis Rheum 2003; 48 Suppl: S646.
  • 32
    Kasle S, Yocum DE, Wilhelm MS. Quality of partner relationships may be protective factor with regard to anxiety for patients with rheumatoid arthritis [abstract]. Arthritis Rheum 2003; 48 Suppl: S644.
  • 33
    Kasle S, Wilhelm MS, Yocum DE. Mutuality predicts pain and fatigue six months later for women with rheumatoid arthritis [abstract]. Arthritis Rheum 2004; 50 Suppl: S680.
  • 34
    Kasle S, Wilhelm MS, Yocum DE, Zautra AJ. Negative affect mediates mutuality's link with overall health for women with rheumatoid arthritis [abstract]. Arthritis Rheum 2005; 52 Suppl: S442.
  • 35
    Kasle S, Swe K, Hymson L, Lisse JR, Wilhelm MS, Yocum DE, et al. The protective role of relationship quality in physical disability for rheumatoid arthritis patients [abstract]. Arthritis Rheum 2006; 54 Suppl: S304.
  • 36
    Lane SK, Gravel JW Jr. Clinical utility of common serum rheumatologic tests. Am Fam Physician 2002; 65: 107380.
  • 37
    Donald F, Ward MM. Evaluative laboratory testing practices of United States rheumatologists. Arthritis Rheum 1998; 41: 7259.
  • 38
    Thorand B, Baumert J, Doring A, Herder C, Kolb H, Rathmann W, et al. Sex differences in the relation of body composition to markers of inflammation. Athersclerosis 2006; 184: 21624.
  • 39
    Giles JT, Bartlett SJ, Andersen R, Thompson R, Fontaine KR, Bathon JM. Association of body fat with C-reactive protein in rheumatoid arthritis. Arthritis Rheum 2008; 58: 263241.
  • 40
    Cushman M, Legault C, Barrett-Connor E, Stefanick ML, Kessler C, Judd HL, et al. Effect of postmenopausal hormones on inflammation-sensitive proteins: the Postmenopausal Estrogen/Progestin Interventions (PEPI) Study. Circulation 1999; 100: 71722.
  • 41
    Dreon DM, Slavin JL, Phinney SD. Oral contraceptive use and increased plasma concentration of C-reactive protein. Life Sci 2003; 73: 124552.
  • 42
    Johnson AM. Commutability of the CRM 470 C-reactive protein value in the Dade Behring N high sensitivity CRP assay. Clin Chem Lab Med 2003; 41: 17782.
  • 43
    Watson D, Clark LA, Tellegen A. Development and validation of brief measures of positive and negative affect: the PANAS scales. J Pers Soc Psychol 1988; 54: 106370.
  • 44
    Watson D, Walker LM. The long-term stability and predictive validity of trait measures of affect. J Pers Soc Psychol 1996; 70: 56777.
  • 45
    Fekete EM, Stephens MA, Mickelson KD, Druley JA. Couples' support provision during illness: the role of perceived emotional responsiveness. Fam Syst Health 2007; 25: 20417.
  • 46
    Wener MH. Autoantibodies and other laboratory tests. In: St.ClairEW, PisetskyDS, HaynesBF, editors. Rheumatoid arthritis. Illustrated ed. Hagerstown (MD): Lippincott Williams & Wilkins; 2004. p. 6479.
  • 47
    Fisher JD, Nadler A, Whitcher-Alagna S. Recipient reactions to aid. Psychol Bull 1982; 91: 2754.
  • 48
    Steinberg M, Gottlieb BH. The appraisal of spousal support by women facing conflicts between work and family. In: BurlesonBR, AlbrechtTL, SarasonIG, editors. Communication of social support: messages, interactions, relationships, and community. Thousand Oaks (CA): Sage; 1994. p. 15272.
  • 49
    Revenson TA, Schiaffino KM, Majerovitz DS, Gibofsky A. Social support as a double-edged sword: the relation of positive and problematic support to depression among rheumatoid arthritis patients. Soc Sci Med 1991; 33: 80713.