as below

Aging & Mental Health, January 2007; 11(1): 89–98

ORIGINAL ARTICLE

The relationship of optimism, pain and social support to well-being in

older adults with osteoarthritis

V. M. FERREIRA & A. M. SHERMAN

Brandeis University, Waltham, MA, US

(Received 30 August 2005; accepted 13 March 2006)

Abstract Improving the psychological well-being of individuals with osteoarthritis (OA) is an important concern because the condition is highly prevalent and has no known cure. Few studies have assessed the joint contribution of social, personality, and physical factors in relation to well-being for OA patients. In a cross-sectional sample of older adults with OA (n¼ 73, 73% female), we assessed the role of support perceptions, optimism and pain in depressive symptoms and life satisfaction. Greater optimism and support were significantly related to both greater life satisfaction and lower depressive symptoms. Further, optimism partially mediated the relationship of pain to life satisfaction, while support partially mediated the role of pain in depressive symptoms. The interplay of these variables in relation to well-being is discussed in the context of chronic illness and older adulthood.

Introduction

Many older adults (85% of those over 75) currently experience a painful and often disabling disease, osteoarthritis (OA), for which there is no known cure (Sack, 1995). Osteoarthritis can negatively impact many aspects of life, including both social and physical functioning (Bookwala, Harralson, & Parmalee, 2003) through pain, physical and psycho- social limitations on valued activities, as well as comorbid depressive symptoms (Penninx et al., 1998). It is important to investigate factors that relate to psychological health outcomes of OA patients in order to better understand how to improve well-being. As Keefe and his colleagues suggest (Keefe & Bonk, 1999; Keefe et al., 2002), there exists a complex interplay of symptomology, social and psychological factors in arthritis patients. In the following sections, we review the role of optimism, social support, and pain as important correlates of psychological well-being, particularly for older adults.

Pain

The common and persistent nature of pain asso- ciated with OA (Schumacker, 1988) may be a reason for the variability in the well-being of OA patients (e.g., de Vellis et al., 1986; Klinger, Spaulding, Polatajko, MacKinnon, & Miller, 1999). Pain has a strong relationship with many other health- related variables in older adults with arthritis

(Roberts, Matecjyck, & Anthony, 1996). In OA samples, greater pain is a stressor linked to lower social support and well-being (de Vellis et al., 1986). Pain is also associated with greater depressive symptoms (Bookwala et al., 2003; Klinger et al., 1999) and lower life satisfaction (Laborde & Powers, 1985) in patients with OA. However, Blixen and Kippes (1999) present contrasting evidence showing that pain from OA is unrelated to life satisfaction when social support is accounted for, suggesting that support mediates the negative impact of pain.

Such a finding suggests that the role of psycho- social factors in the well-being of OA patients may be important to investigate. While the pain associated with OA is difficult to control, often with negative effects on well-being, understanding factors that may serve as points of intervention would allow for a better understanding of how to improve patients’ well- being. Thus, in the following sections, we review optimism and social support, two factors commonly linked to psychological well-being outcomes such as life satisfaction and depressive symptoms.

Optimism

Optimism, defined as an individual’s positive outlook on life (Scheier, Carver, & Bridges, 1994), is a dispositional (i.e., temperament) variable related to adaptation to chronic illness (Lazarus & Folkman, 1984). Optimism is related to both improved

Correspondence: V. M. Ferreira, Brandeis University, Waltham, MA, US. E-mail: [email protected]

ISSN 1360-7863 print/ISSN 1364-6915 online/07/010089–98 � 2007 Taylor & Francis DOI: 10.1080/13607860600736166

psychosocial and physical health outcomes (Barnwell & Kavanaugh, 1997; Brenner, Melamed, & Panush, 1994), as well as less frequent harmful health behaviours (i.e., frequent and excessive drinking in college students; Harju & Bolen, 1998). Greater optimism is also related to greater self-reported quality of life in adult samples (Barnwell & Kavanaugh, 1997).

Scheier and Carver (1987) argue that optimism plays a protective role in chronic illness by promot- ing more beneficial health outcomes, health habits, and effective disease coping. For example, optimism is not only associated with less severe physical symptomology (Reker & Wong, 1985), but also with electing to follow health-promoting practices as well as coping in healthy, helpful ways with chronic illness (Fournier, deRidder, & Bensing, 2002). However, the beneficial role of optimism differs across chronic illnesses.

For example, Fournier et al. (2002) report differing beneficial effects of optimistic beliefs on various facets of coping with chronic diseases such as type I diabetes, rheumatoid arthritis (RA), and multiple sclerosis (MS) in adults. That is, in the chronic illnesses which are controllable (type I diabetes), or at least in part controllable (MS), positive outcome expectancies are associated with greater physical functioning. This is not the case with the much less controllable illness of RA. Further, in a study comparing patients with Parkinson’s disease to those with MS, greater optimism was related to greater physical adjustment to disease symptoms as well as greater use of effective coping strategies (deRidder, Scheier, & Bensing, 2000). However, the relationship between optimism and coping was stronger among MS patients, compared to patients experiencing Parkinson’s. The authors suggest the need to focus on a specific chronic illness in understanding the role of optimism in disease.

Fournier et al. (2002) discuss the several types of optimism that have been established and that, when the illness is controllable (i.e., when symptoms or side effects can be controlled by patients’ behaviors), positive outcome expectancies are the most bene- ficial type of optimism for chronically ill patients. Holding positive outcome expectancies suggests that individuals are generally oriented towards positive expectations of future events (Scheier & Carver, 1992). Fournier et al. (2002) discuss the benefit of positive outcome expectancies in the cases of such illnesses as Type I diabetes and, to some extent, RA, because in such cases optimistic (i.e., positive outcome) beliefs may lead to effective-self care as well as physical and mental health. In the present study, positive outcome expectancies were assessed. Although OA is not necessarily controllable, the pain experienced from OA can often be controlled through self-management strategies such as exercise or the use of pain medication, behaviours closely

associated with optimistic beliefs, even though the occurrence of pain is often unpredictable (Sack, 1995), thus making positive outcome expectancies an appropriate choice for this sample.

Positive outcome expectancies, also known as dispositional optimism, are predictive of more positive affect and less negative affect over time in older adults (Isaacowitz & Seligman, 2002). Optimism is also predictive of psychological well- being. For example, greater optimism predicts less depression three years after baseline in menopausal women after controlling for initial depression (Bromberger & Matthews, 1996). Dispositional optimism is related to greater psychological well- being in older adults, including fewer depressive symptoms and greater life satisfaction (Peterson & Bossio, 1991). Optimism is believed to be related to greater psychological well-being because being optimistic is associated with effective coping strate- gies (e.g., Carver, Scheier, & Weintraub, 1989). For example, optimists are more likely to use problem-focused coping strategies (e.g., Scheier & Carver, 1987).

In Long and Sangster’s (1993) work on optimism and coping specifically in the context of arthritis, greater optimism influenced the use of active or problem-solving coping techniques, a type of effective coping beneficial to health and well-being. This active coping reduced negative symptoms such as depressive symptoms in RA patients. In looking specifically at an OA sample of older adults, greater dispositional optimism was related to use of pain-coping strategies more frequently, even when level of pain was controlled (Benyamini, 2005).

However, little research currently explores the relationship between optimism and the pain asso- ciated with OA, especially in later life. In one study that specifically addressed the role of dispositional optimism in the physical functioning of older adults with knee OA, Brenes and colleagues (2002) report that optimism was positively related to task perfor- mance in walking. However, optimism was not related to performance in lifting objects, climbing stairs, or getting in and out of cars in their sample. Brenes, Rapp, and Rejeski (2002) did not explore the role of optimism and physical factors such as pain, or psychological well-being in their investi- gation. These factors will be addressed in the present study.

Stability of optimism. While optimism is considered to be generally stable (Scheier & Carver, 1987), Benyamini (2005) suggests that reports of optimism, as measured by items on the LOT-R, the measure used in the present study, may be dependent on the time-frame and/or context within which the measure was administered. Benyamini (2005) argues that, in his sample of older adults experiencing OA, optimism may be more easily influenced than with other samples for two reasons; because participants

90 V. M. Ferreira & A. M. Sherman

are experiencing late life as well as a stressful health context. If this is the case, it is conceivable that perceived social support, reported to both affect and be affected by health stressors such as physical functioning (Reinhardt & Blieszner, 2000), may play a role in reports of optimism. In addition, adaptation to chronic illness (Gil, Keefe, Crisson, & Van Dalfsen, 1987) as well as pain (de Vellis et al., 1986) may both be contextual factors affecting reports of optimism. In an example of the less- stable nature of optimism in chronic illness samples, Symister and Friend (2003) report that in a sample of end-stage renal disease patients, social support was predictive of optimism. The role of optimism along with other psychosocial as well as symptomol- ogy variables such as pain in the well-being of older adults has not yet thoroughly been investigated. Understanding such relationships is important in that these variables together may play a more complex role in well-being than is currently believed to be the case. For example, health-related interven- tions may need to be sensitive to personality dimensions such as optimism. In addition, beliefs in optimistic expectancies do decrease in the context of significant, repetitive stress (e.g., Bandura, 1988; Carver et al., 1998) such as the disability and persistent pain associated with OA. While not yet thoroughly investigated, some literature does suggest that reports of optimism may be influenced by health context (Schulz, Tompkins, & Rau, 1988). Schulz et al. (1988) report that in stroke victims and their support persons, optimism scores dropped over the six month period after the stroke. In addition, Benyamini (2005) discusses the possibility that, particularly in older adults, reports of optimistic expectations may be dependent upon more short- term contextual factors, such as the health context of pain. While an individual’s disposition, including optimism, is an important predictor of well-being, so too is the social aspect of one’s life (Finch & Graziano, 2001). Thus, social support perceptions, one facet of social support, will be reviewed in the following section.

Social support

Social support is defined as the resources perceived as available from others in social networks (Antonucci, 1993; Berkman, Glass, Brissette, & Seeman, 2000; Bisconti & Bergeman, 1999). Support involves the beneficial receipt of provisions in relationships, including informational, emotional, or tangible aid. Support is often assessed subjectively as an individual’s perception of the support they perceive as available to them. Perceptions of support are related to health, including symptoms associated with chronic illness, in addition to well-being variables such as life satisfaction (Antonucci & Knipscheer, 1990; Cummins & Nistico, 2002). In older adults, perceived support is linked to

fewer depressive symptoms (Finch & Graziano, 2001; Russell & Cutrona, 1991) and greater life satisfaction (Walen & Lachman, 2000). Looking specifically at older adults with OA, social support is associated with fewer depressive symptoms (Sherman, 2003) as well as less arthritis pain (Evers, Kraaimaat, Geenen, & Bijlsma, 1998). Research also suggests that perceptions of social support may contribute to positive well-being (Blixen & Kippes, 1999) by positively affecting such facets of coping as support seeking as a resource (e.g., Scheier & Carver, 1987).

There has been some speculation about the combined role of optimism and social support in psychological functioning. For example, Finch and Graziano (2001) reported that greater optimism and social support were each uniquely related to lower depressive symptoms in a general sample of older adults. That is, more optimistic older adults, and those reporting higher social support, reported fewer depressive symptoms. However, physical stressors, such as pain, have not been investigated jointly with support and optimism in relation to well-being, particularly for OA. This is a gap that should not be ignored because physical stressors are strongly related to poor psychological outcomes such as greater depressive symptoms (e.g., Evers et al., 1998).

We suggest that experiencing more pain may be related to individuals’ tendencies not to view their world in an optimistic, positive light and thus, these individuals may report less well-being. While this hypothesis has not yet been tested in the literature, less pain, like greater optimism in older adults, is associated with better psychological well-being in OA samples (de Vellis et al., 1986). In addition, we suggest that experiencing more pain may be related to individuals’ tendencies not to view their relationships as positive and supportive and thus, these individuals may report less well-being.

Thus, in the present study, the stressor of pain, a common symptom of OA, as well as social support and dispositional optimism were investigated in an effort to understand the role that these different factors play in the depressive symptoms and life satisfaction reported by older adults with OA.

Hypotheses

1. Greater pain is hypothesized to be associated with less life satisfaction and greater depressive symptoms.

2. Greater optimism and social support are expected to relate to less depressive symptoms and greater life satisfaction after accounting for pain.

3. Two possible mediation models will be tested. The first will assess the role of optimism in mediating the relationship of pain to life satisfac- tion and depressive symptoms. The second will

Optimism, pain, social support in well-being 91

assess the role of support in mediating the relationship of pain to life satisfaction and depressive symptoms.

Method

Participants and procedure

Ninety-five potential participants were contacted by telephone from a list of participants from a previous OA study (see Ferreira & Sherman, 2006). The original sample had been recruited through newspaper advertisements, fliers, and recruitment through a continuing education program for older adults. Of the original participants, 72 older adults (52 women and 20 men, ranging in age from 60 to 84, mean age¼ 72.4 years) volunteered to participate in the present study (response rate of 76%). Participants completed and returned the mailed survey. Follow-up data were used rather than baseline data because optimism was not measured at baseline.

There were no significant demographic differences between responders and non-responders. Participants all resided in the greater Boston area and spoke fluent English. The majority (92.6%) of participants considered themselves Caucasian. Nearly half of the participants (45.2%, n¼ 33) were married, 27.4% (n¼20) were widowed, and 26% (n¼ 19) were single, divorced or separated. See Table I for additional demographic information. There were no significant differences in depressive symptoms (t¼�0.80, ns) or pain (t¼ 1.28, ns) in those who did or did not respond.

Measures

Optimism was assessed using the Life Orientation Test-Revised (LOT-R; Scheier, Carver, & Bridges, 1994), which examines outcome expectancies. The measure consists of 12 items, including four filler items not included in the analyses, for a total of

eight items. Great debate has surrounded the use of the LOT in its entirety as a measure of optimism (with negatively worded items being reverse coded) or as a measure of both optimism and pessimism (with two subscales being conceptualized as separate constructs. While Scheier and Carver (1987) have suggested the scale may be used as a bipolar measure, research has suggested that optimism and pessimism are very different constructs (e.g., Brenes, Rapp, Rejeski, & Miller, 2002; Lai, 1994; Marshall, Wortman, Kusulas, Hervig, & Vickers, 1992). In light of this controversy, the four positively worded items assessing optimism were used. Response options utilized a five-point Likert scale. Higher scores indicate greater optimism (M¼ 13.26, SD¼2.89, high score¼20). Reliability was 0.76 with this sample.

Social support was assessed using the Medical Outcomes Study (MOS) Social Support Survey (Sherbourne & Stewart, 1991) which examines perceived social support (M¼ 65.06, SD¼ 18.44, high score¼ 95). Higher scores indicate greater perceived support. Sherbourne & Stewart (1991) report high internal consistency (Cronbach’s alpha¼ 0.97), which was replicated in this sample (Cronbach’s alpha¼ 0.97).

Pain was assessed using the pain subscale of the Medical Outcomes Study (MOS) 36-item short- form survey (SF-36; Ware & Sherbourne, 1992). The SF-36 includes self-report items of health perceptions, physical functioning, role limitations and pain. The pain subscale includes two items, which are summed so that a higher score indicates more pain (M¼ 5.85, SD¼ 1.91, high score¼ 10).

Depressive symptoms were assessed using the 20-item Centre for Epidemiological Studies Depression scale (CES-D; Radloff, 1977). The CES-D includes 20 items assessing emotional (‘I could not stop crying’) and somatic symptoms (‘I had trouble sleeping’) characteristic of depressed mood. Response options are on a four-point Likert type scale with possible responses ranging from ‘all of the time’ to ‘rarely’ (M¼ 26.25, SD¼ 8.8, high score¼ 40). Reliability was 0.84 with this sample. Higher scores indicate more depressive symptomatology.

Life satisfaction was assessed with the Life Satisfaction Index (LSI; Neugarten, Havinghurst, & Tobin, 1961). The scale consists of 13 items using a five-point Likert-type format for response options (M¼ 41.31, SD¼ 10.14, high score¼ 61). A higher score means more life satisfaction. Reliability was 0.89 with this sample.

Data analysis

In order to investigate the role of pain, optimism and perceived social support in depressive symptoms and life satisfaction, hierarchical multiple regression analyses were conducted. In the first step of each

Table I. Demographic characteristics, n¼ 73.

No. (%)

Marital status Married 33 (45.2) Widowed 20 (27.4) Single/divorced/separated 19 (26.0) No response 1 (1.4)

Education Bachelor’s degree or greater 42 (57.5) Some college 13 (17.8) Associate’s degree 4 (5.5) High school diploma or less 10 (13.7) No response 4 (5.5)

Income (annual) �20,000 23 (31.5) >20,001 48 (65.8) No response 2 (2.7)

92 V. M. Ferreira & A. M. Sherman

regression, age was entered. In the second step for each regression, pain was entered.

To assess the potentially mediating roles of optimism and support to pain, two multiple regres- sion models were developed. In these two regression analyses, life satisfaction and depressive symptoms were each assessed as an outcome. In each of these models, after age and pain were entered in steps one and two, support was entered at the third step of the model, followed by optimism at the fourth step. This model investigated the role of either optimism or support as mediators of pain in relation to the outcomes (see Figure 1).

In order to assess the role of either support or optimism as a mediator of pain, three aspects to the mediation model must be assessed (Baron & Kenny, 1986). In the first leg of the model, the association of pain to the outcome must be significant. Next, pain must significantly predict the mediator (support or optimism). Finally, in the third leg of the model, the relationship of pain to the outcome must be reduced to non-significance once the mediator (support or optimism) is entered into the model. Only when all three of these criteria are met will mediation be demonstrated (see Figure 1;

Baron & Kenny, 1986). Follow-up Sobel tests will then be conducted to test for partial mediation (Preacher & Leonardelli, 2003).

Results

Bivariate analyses

Correlations between the variables measured in the present study are reported in Table II. Gender, marital status, and income were excluded from further multivariate analyses to conserve power because they were not related to any of the other measures. Age was used as a control variable because older age was correlated with lower life satisfaction. Pain was significantly correlated with greater depres- sive symptoms and less life satisfaction. Support and optimism were both significantly correlated with fewer depressive symptoms and greater life satisfaction.

Multivariate analyses

Life satisfaction. In the first hierarchical multiple regression model with life satisfaction as an outcome,

(a)

(b)

Pain

Social support

Depressive symptoms Life satisfaction

Pain

Optimism

Depressive symptoms Life satisfaction

Figure 1. (a) Role of optimism as a mediator; (b) Role of social support as a mediator.

Table II. Correlations of measures.

1 2 3 4 5 6 7 8 9

Age – �0.12 0.04 �0.04 0.14 �0.14 �0.04 0.10 �0.29* Gender – – �0.13 0.09 �0.17 0.03 0.01 0.02 �0.09 Income – – – �0.18 �0.06 0.08 0.10 �0.16 0.12 Marital status – – – – �0.01 �0.14 0.01 0.22 0.03 Pain – – – – – �0.28* �0.30* 0.43** �0.46** Social support – – – – – – 0.07 �0.44** 0.41** Optimism – – – – – – – �0.53** 0.64** Depressive symptoms – – – – – – – – �0.64** Life satisfaction – – – – – – – – –

*p50.05; **p50.01.

Optimism, pain, social support in well-being 93

we assessed the role of pain, perceptions of social support, and optimism, with optimism as a possible mediator of the relationship of pain to life satisfaction. The final model was significant (F (4,63)¼ 20.13, p50.01, adjusted R2

¼0.53; see Table III). Greater age was significantly related to higher life satisfaction at the first step and throughout the model (R2

¼ 0.08, F (1,66)¼5.36, p50.05). In the second step, less pain was related to greater life satisfaction (�R2

¼0.12, F (1,65)¼9.69, p50.01). Perceived social support explained addi- tional variance (�R2

¼ 0.06, F (1,64)¼4.98, p50.01) when entered at the third step. Greater support was related to more life satisfaction, while pain remained significant. Support remained sig- nificant at the fourth step (t¼2.75, p50.01). However, the relationship of pain to life satisfaction was reduced to non-significance once optimism was entered in the fourth step of the model (t¼ 1.46, ns). Greater optimism was significantly related to life satisfaction at this step (�R2

¼0.31, F (1,63)¼ 45.08, p50.01). This model suggests that optimism should be tested as a partial mediator of pain (Baron & Kenny, 1986). The relationship of pain as a predictor of optimism was significant (F (1, 70)¼ 7.69, p50.01, adjusted R2

¼ 0.09). This finding suggests that pain may be partially mediated by optimism in relation to life satisfaction. A mediation analysis was conducted assessing opti- mism as a mediator of pain. The Sobel mediation test was significant (t¼�2.54, p¼ 0.01), suggesting that optimism does partially mediate the relationship

of pain to life satisfaction (Preacher & Leonardelli, 2003). Full mediation is not demonstrated because the beta values were not reduced to zero (Baron & Kenny, 1986). There was no evidence that support mediated pain in the analysis.

Depressive symptoms. In the hierarchical multiple regression model assessing the role of optimism, pain and support in their relationship to depressive symptoms, we assessed the possible role of optimism as a mediator of pain. The final model was significant (F (4,62)¼14.68, p50.01, adjusted R2

¼0.45; see Table IV). Age was not significantly related to depressive symptoms at the first step, or throughout the model. In the second step, more pain was related to greater depressive symptoms (�R2

¼ 0.16, F (1,64)¼ 12.13, p50.01). Perceived social support explained additional variance (�R2

¼ 0.12, F (1,63)¼10.85, p50.01) when entered at the third step. Less support was related to greater depressive symptoms, while pain remained significant though was reduced in its statistical relationship. The relationship of pain to depression was reduced to non-significance once optimism was entered in the fourth step of the model, where lower optimism and support were both related to higher depressive symptoms (�R2

¼ 0.45, F (1,62)¼ 24.06, p50.01). This model suggests that both optimism and pain should be tested as possible mediators of pain in relation to depressive symptoms (Baron & Kenny, 1986).

Table III. Summary of hierarchical regression analysis for life satisfaction showing optimism as a mediator (n¼ 70).

1 2 3 4

Step B SE B � B SE B � B SE B � B SE B �

Age 0.46 0.20 0.27* 0.37 0.19 0.22* 0.34 0.14 0.20* 0.33 0.14 0.20* Pain – – – �1.62 0.52 �0.35** �1.34 0.52 �0.29* �0.61 0.42 �0.13 Support – – – – – – 0.13 0.06 0.25* 0.12 0.05 0.24** Optimism – – – – – – – – – 1.93 0.29 0.58** Adjusted R2 – – – 0.06* – – – 0.17** – – – 0.22* – – – 0.53**

*p50.05; **p50.01.

Table IV. Summary of hierarchical regression analysis for depressive symptoms showing optimism as a mediator (n¼ 70).

1 2 3 4

Step B SE B � B SE B � B SE B � B SE B �

Age �0.12 0.19 �0.08 �0.30 0.18 �0.02 0.02 0.17 0.01 0.02 0.14 0.01 Pain – – – 1.74 0.50 0.40** 1.35 0.48 0.31 0.81 0.43 0.19 Support – – – – – – �0.18 0.05 �0.36** �0.17 0.05 �0.35** Optimism – – – – – – – – – �1.45 0.30 �0.46** Adjusted R2 – – – 0.01* – – – 0.14** – – – 0.25* – – – 0.45**

*p50.05; **p50.01.

94 V. M. Ferreira & A. M. Sherman

Support as a mediator. The relationship of pain to support was significant (F (1, 70)¼ 5.16, p50.05, adjusted R2

¼0.06). Finally, because the first two legs of the mediation model were significant, a Sobel test was used to assess the final leg of the mediation model. The mediation test was significant (t¼1.98, p¼0.05). This finding suggests that pain is partially mediated by support in relation to depressive symptoms. There was no evidence that optimism mediated pain in this analysis.

Optimism as a mediator. The relationship of pain to optimism was assessed to be significant (F (1, 70)¼ 7.69, p50.01, adjusted R2

¼ 0.09). As was done in the previous analysis, because the first two legs of the mediation model were significant, a Sobel test was used to assess the final leg of the mediation model. The mediation test was not significant (t¼ 0.06, ns). This finding suggests that pain is not mediated by optimism in relation to depressive symptoms.

Discussion

In the present study, we investigated the role of social relations, optimism and the common symp- tom of pain in the well-being of older adults experiencing OA. The findings suggest that pain, social support and optimism, are related in complex ways for this sample. Furthermore, these findings suggest that these factors relate in unique ways depending upon the outcome being investigated.

Life satisfaction

In investigating life satisfaction, pain was partially mediated by optimism, while support was indepen- dently related to life satisfaction. These results suggest that optimism is partially responsible for the relationship between pain and life satisfaction, possibly by negatively influencing the optimistic outlook that is associated with life satisfaction. This finding that optimism partially mediated pain has important implications for our understanding of the process by which pain is related to psychological well-being (Bookwala et al., 2003; Klinger et al., 1999; Labourde & Powers, 1985; Robert et al., 1996). An optimistic disposition appears to at least be partially diminished by reports of pain, a contextual factor which OA patients may struggle with on a daily basis through the rest of their lives (Schumacker, 1988). As discussed previously in our literature review, although considered to be generally stable (Scheier & Carver, 1987), research suggests that optimism may be dependent upon the context within which it was assessed (Benyamini, 2005). Optimism may be salient for well-being in older adults because, similar to personality constructs such as hardiness (Wallace, Bisconti, & Bergeman, 2001), an optimistic personality may act as a resource in

helping older adults to compensate for other losses in resources as they age (Baltes, 1996).

Support did not mediate the relationship of pain to life satisfaction. Instead, support was directly linked to this outcome. This finding is consistent with much research suggesting the strong relation- ship of support to greater well-being (e.g., Cummins & Nistico, 2002; Russell & Cutrona, 1991; Walen & Lachman, 2000). As discussed in the introduction, support has been demonstrated to play an important role in both physical health as well as well-being (Antonucci & Knipscheer, 1990; Cummins & Nistico, 2002).

Depressive symptoms

In contrast to the findings for life satisfaction, for depressive symptoms, there was no evidence of mediation of pain by optimism. However, support partially mediated pain in relation to depressive symptoms. That is, the relationship of greater pain to greater depressive symptoms appears to at least in part be accounted for by the relationship of less support to greater depressive symptoms where older adults experiencing more pain may perceive their social networks as less supportive and thus, report greater depressive symptoms.

This mediational role of perceived support con- tributes support to the work of Antonucci, Langfahl, and Akiyama (2004) who suggest that new research should investigate predictors of support perceptions as many contextual factors may play a role in perceptions of social relations. In the case of our findings, the context of consistent pain associated with OA is demonstrated as a contextual fact playing a role in support perceptions.

Support seems to be a powerful resource for this sample, with pain working indirectly through sup- port to influence depressive symptoms. This finding is consistent with a wealth of research suggesting not only that perceived social support cyclically affects and is affected by health stressors (e.g., Gil et al., 1987; Reinhardt & Blieszner, 2000), but also that social support plays a complex role in the relation- ship of health variables to psychological well-being outcomes (e.g., Bisconti & Bergeman, 1999).

In summary, in this sample, optimism partially mediated pain perceptions in relation to life satisfac- tion while support partially mediated pain percep- tions in relation to depressive symptoms. Taken together, the findings of the present study are difficult to explain. Why does optimism play a mediational role in the relationship of pain to life satisfaction while support mediates the relationship of pain to depressive symptoms? The reasons remain unclear. Follow-up analyses with different samples of older adults experiencing OA are necessary for two reasons. First, such analyses would either replicate or refute these findings. And second, if the present findings are supported, differences in the

Optimism, pain, social support in well-being 95

sample may shed light on the mechanisms by which these unique and interesting findings have occurred. Longitudinal, as opposed to the cross-sectional data used in the present study, are suggested in order to provide a more rigorous test of mediation in follow- up research. In addition to these limitations, other limitations to the present study do exist, a few of which will be highlighted next.

Limitations and future research

The small, non-random nature of our sample may result in somewhat limited generalizability. Replication of this study with a more representative sample may provide stronger statistical evidence for the role of support and optimism as mediators of pain in well-being in samples of older adults with OA. However, our demonstration of significant patterns of association in such a small sample indicates that the interplay of these variables should be a fruitful arena for further investigation.

In addition, the self-report nature of the measures means that shared method variance could be a problem. Future research could replicate this study with additional observed measures of physical functioning in order to assess the role of physical impairment and symptoms in well-being while still taking into account personality and social relations factors.

In addition, the findings suggest that future research might examine whether other health related stressors such as physical limitations, account for more of the variance in well-being than pain. In this way, the physical symptoms of OA might be better understood in terms of how they affect the overall experience of the patient.

Additional suggestions for future research include investigating the role of other dispositional variables in the well-being outcomes of OA patients, in an effort to provide insight into how to better improve the well-being of individuals suffering from pain. For example, the personality variable of hardiness is negatively linked to both self-reported and objective measures of health in older adults (Wallace et al., 2001). This may indicate the importance of person- ality as a resource that allows older adults to compensate for losses and changes as they age (Baltes, 1996). Finally, longitudinal work is sug- gested to look at the significance of all three areas of life (social, personality, and physical) in their relationships to well-being. As mentioned above, optimism and support may be a powerful factor in contributing to well-being due to its value in compensating for other losses in later life.

Interventions

Although follow-up research is necessary, the pre- sent findings lead us to suggest further research into two approaches to intervention aimed at improving

the well-being of older adults experiencing OA. While pain is a symptom of OA that is difficult to control entirely, our findings of support acting as a mediator of pain to depressive symptoms and of optimism mediating the role of pain in life satisfac- tion suggests that focusing on these two psychosocial facets may be an efficacious way to approach intervention strategies. For example, a focus might be placed on working with OA patients and their partners or close family members to strengthen relationships as well as raise awareness in the patient of the social support available to them. Zautra, Hamilton, and Yocum (2000) report that, in their sample of older women experiencing rheumatoid arthritis, interventions designed to enhance social engagement not only improved psychological well-being immediately, but also increased illness coping capacity when assessed four months later. In addition, interventions might focus on increasing positive outcome expectancies, a psychosocial factor considered to be less stable in samples experiencing a chronic condition (Benyamini, 2005).

Conclusion

In summary, our results indicate that pain has a complex relationship with facets of well-being, both life satisfaction and freedom from depressive symp- toms, in older adults experiencing OA. That is, the effect of pain on perceptions of well-being is influenced by both personality and social relations variables. Being optimistic appears to mediate the relationship of pain in life satisfaction while support appears to mediate pain in relation to depressive symptoms. While the mediational role of support in health context and well-being relationships is well- documented, the mediational role of optimism is much more recent. Thus, investigation of these findings in additional samples of older adults experiencing OA, as well as other chronic illness samples is suggested in order to better understand these findings.

This is one of the only studies to look at the role of support, pain, and optimism together in an OA sample. This investigation has afforded us some insight into some aspects of life that allow older adults to maintain a sense of well-being in the face of the challenging disease of OA. Although all three variables had been established as individually important to well-being, the contribution of the present results lies in showing the importance of both optimism and support in well-being independent of other factors, as well as serving as a mediator of physical factors in relation to depressive symptoms and life satisfaction. We have demon- strated the joint contribution of both positive social relationships as well as a positive outlook on life

96 V. M. Ferreira & A. M. Sherman

in contributing to the well-being of older adults with OA.

References

Antonucci, T. C. (1993). Attachment: From youth through adulthood. In R. Kastenbaum (Ed.), Encyclopedia of adult

development (pp. 40–44). Phoenix, AZ: Oryx Press. Antonucci, T. C., & Knipscheer, C. P. M. (1990). Social network

research: Review and perspectives. In C. P. M. Knipsheer & T. C. Antonucci (Eds.), Social network research (pp. 161–174). Amsterdam: Swets & Zeitlinger.

Antonucci, T. C., Langfahl, E. S., & Akiyama, H. (2004). Relationships as outcomes and contexts. In F. R. Lang & K. L. Fingerman (Eds.), Growing together: Personal relationships

across the lifespan (pp. 24–44). Cambridge: Cambridge University Press.

Arthritis Foundation (Retrieved May 17, 2005). www.arthritis.org

Baltes, M. M. (1996). The many faces of dependency in old age. New York: Cambridge University Press.

Bandura, A. (1988). Self-regulation of motivation and action through goal systems. In V. Hamilton, G. H. Bower & N. H. Frijda (Eds.), Cognitive perspectives on emotion and

motivation (pp. 37–61). Dordrecht: Kluwer Academic Publishers.

Barnwell, A. M., & Kavanaugh, D. J. (1997). Prediction of psychological adjustment to multiple sclerosis. Social Science

and Medicine, 45, 411–418. Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator

variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173–1182.

Benyamini, Y. (2005). Can high optimism and high pessimism co-exist? Findings from arthritis patients coping with pain. Personality & Individual Differences, 38, 1463–1473.

Berkman, L. F., Glass, T., Brissette, I., & Seeman, T. E. (2000). From social integration to health: Durkheim in the new millennium. Social Science and Medicine, 51, 843–857.

Bisconti, T. L., & Bergeman, C. S. (1999). Perceived social control as a mediator of the relationships among social support, psychological well-being, and perceived health. The

Gerontologist, 39, 94–103. Blixen, C. E., & Kippes, C. (1999). Depression, social support,

and quality of life in older adults with osteoarthritis. Journal of Nursing Scholarship, 31, 221–226.

Bookwala, J., Harralson, T. L., & Parmalee, P. A. (2003). Effects of pain and well-being in older adults with osteoarthritis of the knee. Psychology and Aging, 18, 844–850.

Brenes, G. A., Rapp, S. R., Rajeski, W. J., & Miller, M. E. (2002). Do optimism and pessimism predict physical functioning? Journal of Behavioral Medicine, 25, 219–231.

Brenner, G. F., Melamed, B. G., & Panush, R. S. (1994). Optimism and coping as determinants of psychosocial adjust- ment to rheumatoid arthritis. Journal of Clinical Psychology in

Medical Settings, 1, 115–134. Bromberger, J. T., & Matthews, K. A. (1996). A longitudinal

study of the effects of pessimism, trait anxiety, and life stress on depressive symptoms in middle-aged women. Psychology and

Aging, 11, 207–213. Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989).

Assessing coping strategies: A theoretically based approach. Journal of Personality and Social Psychology, 56, 267–283.

Cummins, R. A., & Nistico, H. (2002). Maintaining life satisfaction: The role of positive cognitive bias. Journal of

Happiness Studies, 3, 37–69. deRidder, D., Scheier, K., & Bensing, J. (2000). The relative

benefits of being optimistic: Optimism as a coping resource in

multiple sclerosis and Parkinson’s disease. British Journal of

Health Psychology, 5, 141–155. de Vellis, R. F., McEvoy, de vellis, B., Sauter, S. V. H.,

Harring, K., & Cohen, J. L. (1986). Predictors of pain and functioning in arthritis. Health Education Research, 1, 61–67.

Evers, A. M. W., Kraaimaat, F. W., Geenen, R., & Bijlsma, J. W. J. (1998). Psychosocial predictors of functional change in recently diagnosed rheumatoid arthritis patients. Behavior Research and Therapy, 36, 179–193.

Ferreira, V. M., & Sherman, A. M. (2006). Control beliefs and social relations: Their relationship to well-being outcomes in older adults with Osteoarthritis. International Journal of Aging and Human Development, 62, 255–274.

Finch, J. F., & Graziano, W. G. (2001). Predicting depression from temperament, personality, and patterns of social relations. Journal of Personality, 69, 27–52.

Fournier, M., de Ridder, D. T. D., & Bensing, J. (2002). Optimism and adaptation to chronic disease: The role of optimism in relation to self-care options of type 1 diabetes mellitus, rheumatoid arthritis and multiple sclerosis. British

Journal of Health Psychology, 7, 409–432. Gil, K. M., Keefe, F. J., Crisson, J. E., & Van Dalfsen, P. J.

(1987). Social support and pain behavior. Pain, 29, 209–217. Harju, B. L., & Bolen, L. M. (1998). The effects of optimism

on coping and perceived quality of life of college students. Journal of Social Behavior and Personality, 13, 185–200.

Isaacowitz, D. M., & Seligman, M. E. P. (2002). Cognitive style predictors of affect change in older adults. International Journal of Aging and Human Development, 54, 233–253.

Keefe, F. J., & Bonk, V. (1999). Psychosocial assessment of pain in patients having rheumatic diseases. Rheumatic Diseases Clinics

of North America, 25, 81–103. Keefe, F. J., Smith, S. J., Buffington, A. L. H., Gibson, J.,

Studts, J. L., & Caldwell, D. S. (2002). Recent advances and future directions in the biopsychosocial assessment and treatment of arthritis. Journal of Consulting and Clinical

Psychology, 70, 640–655. Klinger, L., Spaulding, S. J., Polatajko, H. J., MacKinnon, J. R.,

& Miller, L. (1999). Chronic pain in the elderly: Occupational adaptation as a means of coping with osteoarthritis of the hip and/or knee. Clinical Journal of Pain, 15, 275–283.

Labourde, J. M., & Powers, M. J. (1985). Life satisfaction, health control orientation, and illness-related factors in persons with Osteoarthritis. Research in Nursing and Health, 8, 183–190.

Lai, J. C. L. (1994). Differential predictive power of the positively versus the negatively worded items of the Life Orientation Test. Psychological Reports, 75, 1507–1515.

Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer.

Long, B. C., & Sangster, J. L. (1993). Dispositional optimism/ pessimism and coping strategies: Predictors of psychosocial adjustment of rheumatoid and osteoarthritis patients. Journal of Applied Social Psychology, 23, 1069–1091.

Marshall, G. N., Wortman, C. B., Kusulas, J. W., Hervig, L. K., & Vickers, R. R. (1992). Distinguishing optimism from pessimism: Relations to fundamental dimensions of mood and personality. Journal of Personality & Social Psychology, 62, 1067–1074.

Neugarten, B. L., Havinghurst, R. J., & Tobin, S. S. (1961). The measurement of life satisfaction. Journal of Gerontology, 16, 134–143.

Penninx, B. W. J. H., van Tillburg, T., Boeke, A. J. P., Deeg, D. J. H., Kriegsman, D. M. W., & van Eijk, J. Th. M. (1998). Effects of social support and personal coping resources on depressive symptoms: Different for various chronic diseases? Health Psychology, 17, 551–558.

Peterson, C., & Bossio, L. M. (1991). Health and Optimism. New York: Free Press.

Preacher, K. J., & Leonardelli, G. J. (2003). Calculation for the

Sobel Test: An interactive calculation tool for mediation tests.

Optimism, pain, social support in well-being 97

Retrieved Sept. 24, 2005, from http://www.unc.edu/�preacher/ sobel/sobel.htm

Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied

Psychological Measurement, 3, 385–401. Reinhardt, J. P., & Blieszner, R. (2000). Predictors of perceived

support quality in visually impaired elders. The Journal of

Applied Gerontology, 19, 345–362. Reker, G. T., & Wong, P. T. P. (1985). Personal optimism,

physical and mental health: The triumph of successful aging. In J. E. Birren & J. Livingston (Eds.), Cognition, stress and aging

(pp. 134–173). New York: Prentice Hall. Roberts, B. L., Matecjyck, M., & Anthony, M. (1996). The

effects of social support on the relationship of functional limitations and pain to depression. Arthritis Care and Research, 9, 67–73.

Russell, D. W., & Cutrona, C. E. (1991). Social support, stress, and depressive symptoms among the elderly: Test of a process model. Psychology and Aging, 6, 190–201.

Sack, K. (1995). Osteoarthritis: A continuing challenge. Western

Journal of Medicine, 163, 579–586. Scheier, M. F., & Carver, C. S. (1987). Dispositional optimism

and physical well-being: The influence of generalized outcome expectancies on health. Journal of Personality, 55, 169–210.

Scheier, M. F., & Carver, C. S. (1992). Effects of optimism on psychological and physical well-being: Theoretical overview and empirical update. Cognitive Theory Research, 16, 201–228.

Scheier, M. F., Carver, C. S., & Bridges, M. W. (1994). Distinguishing optimism from neuroticism (and trait anxiety, self-mastery, and self-esteem): A re-evaluation of the Life

Orientation Test. Journal of Personality and Social Psychology, 67, 1063–1078.

Schulz, R., Tompkins, C. A., & Rau, M. T. (1988). A longitudinal study of the psychosocial impact of stroke on primary support person. Psychology & Aging, 3, 131–141.

Schumacker, H. R. (1988). Primer on the rheumatic diseases

(9th ed.). Atlanta, GA: Arthritis Foundation. Sherbourne, C. D., & Stewart, A. L. (1991). The MOS social

support survey. Social Science and Medicine, 32, 705–714. Sherman, A. (2003). Social relations and depressive symptoms in

older adults with osteoarthritis. Social Science and Medicine, 56, 247–257.

Symister, P., & Friend, R. (2003). The influence of social support and problematic support on optimism and depression in chronic illness: A prospective study evaluating self-esteem as a mediator. Health Psychology, 22, 123–129.

Walen, H. R., & Lachman, M. E. (2000). Social support and strain from partner, family, and friends: Costs and benefits for men and women in adulthood. Journal of Social and Personal

Relationships, 17, 5–30. Wallace, K. A., Bisconti, T. L., & Bergeman, C. S. (2001). The

mediational effect of hardiness on social support and optimal outcomes in later life. Basic and Applied Social Psychology, 23, 267–279.

Ware, J. E., & Sherbourne, C. D. (1992). The MOS 36-item short-form health survey (SF-36): Conceptual framework and item selection. Medical Care, 30, 473–481.

Zautra, A. J., Hamilton, N., & Yocum, D. (2000). Patterns of positive social engagement among women with rheumatoid arthritis. The Occupational Therapy Journal of Research, 20, 21s–40s.

98 V. M. Ferreira & A. M. Sherman

contentserver__4_.pdf