Women's Health Issues
Volume 17, Issue 1 , Pages 44-51, January 2007

Age differences in the pain–depression link for women with osteoarthritis:

Functional Impairment and Personal Control as Mediators

  • Jessica M. McIlvane, PhD

      Affiliations

    • University of South Florida, School of Aging Studies, University of South Florida, Tampa, Florida
    • Corresponding Author InformationCorrespondence to: Jessica M. McIlvane, PhD, University of South Florida, School of Aging Studies, University of South Florida, 4202 East Fowler Avenue, MHC1318, Tampa, FL 33620.
    • ,
  • Kathleen M. Schiaffino, PhD

      Affiliations

    • Fordham University, Department of Psychology, Dealy Hall, Bronx, New York
    • ,
  • Stephen A. Paget, MD

      Affiliations

    • Department of Medicine Division of Rheumatology, Hospital for Special Surgery, New York, New York

Received 19 December 2005; received in revised form 24 August 2006; accepted 4 October 2006.

Article Outline

Purpose

This study examines functional impairment and personal control as mediators between pain and depressive symptoms in middle-aged and older women with osteoarthritis (OA).

Method

Ninety-nine middle-aged and older women with OA completed face-to-face interviews to assess pain, functional impairment, personal control, depressive symptoms, and self-rated health.

Results

Controlling for self-rated health, functional impairment mediated the relationship between pain and depressive symptoms for middle-age women but not for older women. Alternately, personal control was a mediator for older women but not for middle-aged women.

Conclusions

Functional limitations at least partially explain the relationship between pain and depression for middle-aged women who are juggling many roles and do not expect trouble with daily activities. For older women, functional limitations are expected, but personal control becomes more important. Implications for intervention are discussed.

 

Arthritis is a growing and important public health concern. With the expected growth in the proportion of older persons in the United States in the years to come, chronic illness including arthritis is becoming a critical issue in terms of independent living and quality of life. Arthritis is one of the most common chronic conditions and the leading cause of disability for women (Centers for Disease Control and Prevention 1995, Centers for Disease Control and Prevention, 2001). In addition to the personal impact of arthritis, arthritis has significant economic costs for society at $86 billion dollars annually in lost wages and medical costs (Centers for Disease Control and Prevention, 2004). Because women are disproportionately affected, women bear the brunt of this illness in terms of economic costs, quality of life, and disability (Hootman, Sniezek, & Helmick, 2002). Arthritis has not received adequate attention in the past, probably because it is not likely to be life threatening and it is incorrectly viewed as both a normal, inevitable part of aging and untreatable. In light of the potential negative impact of arthritis on older adults, and women in particular, it is striking that Healthy People 2010 is the first of three 10-year national health plans that includes objectives for arthritis (Hootman et al., 2002). With the advent of several recent national and global initiatives, including the National Arthritis Action Plan (Arthritis Foundation, 1999) and the Bone and Joint Decade (www.boneandjointdecade.org), arthritis is just beginning to receive much needed attention as a public health priority.

Osteoarthritis (OA) is the most common type of arthritis and women over the age of 45 are more likely to have OA compared with men (National Institute of Arthritis and Musculoskeletal and Skin Diseases [NIAMS], 2002). OA is most prevalent in older women, but affects many women in middle age as well. OA involves degeneration of cartilage in the joints that leads to the slow progression of joint pain, stiffness, and limitation of motion. The hand, knee, hip, and spine are the joints most commonly affected by OA. By some estimates, x-ray evidence of OA would be present in more than half of all people age ≥65 in at least 1 joint (NIAMS, 2002). Although the risk of OA does increase with age, OA is not a normal part of aging and it is indeed treatable through such means as self-management and exercise.

OA is frequently accompanied by pain and functional impairment (NIAMS, 2002), which can impact quality of life, and contribute to poorer psychological well-being (e.g., Williamson and Schulz 1992, Zeiss et al 1996) as well as higher risk for depression compared with the general population (DeVellis, 1995). It is unclear, however, exactly how pain, functional impairment, and depression are linked; understanding this relationship would be important for the design of interventions to improve the quality of life for women with OA and to guide public health efforts to educate women about how to best manage their OA.

Several models have been proposed to explain the pain–depression association and we rely on 2 such perspectives to examine this relationship in women with OA: the activity restriction model and personal control. The activity restriction model (Williamson 1998, Williamson and Dooley 2001) suggests that health-related stressors such as pain are related to adverse mental health outcomes (e.g., depression) as a function of the level of interference with normal, everyday activities such as household and social activities. For women with arthritis, this model suggests that pain influences well-being directly, but that pain also influences well-being indirectly through restricting activities. That is, pain leads to the inability to perform daily activities, which in turn leads to depression. This model is applicable as a framework for studying depression in arthritis; previous research demonstrates that individuals with OA are more likely than healthy controls to give up activities of daily living, such as household chores, shopping and errands, and leisure-related activities (Yelin, Lubeck, Holman, & Epstein, 1987) and that the inability to perform daily activities negatively impacts well-being for many individuals with arthritis (Zeiss et al., 1996).

Contradictory findings exist about whether functional impairment mediates, or explains, the relationship between pain and depression in older adults. Some studies demonstrate the importance of functional impairment in at least partially explaining the relationship between pain and depression (Williamson and Schulz 1992, Zeiss et al 1996). Other studies suggest that although there is a direct effect of pain on depression, there is not an indirect effect of pain on depression through functional limitations for community-dwelling older adults with OA (Bookwala, Harralson, & Parmalee, 2003) or for those who are institutionalized (Parmalee, Katz, & Lawton, 1991). These studies mainly focus on older adults, who are most likely to experience illness and functional limitations. However, little research addresses these same issues in middle-age adults for whom “the specter of chronic illness is hardly considered ’normative’” (Schiaffino, 2001).

An alternate construct to consider as a mechanism for explaining the relationship between pain and depression is that of personal control, specifically personal control over illness. Personal control is especially salient in the context of chronic illness that is both stressful and involves losses that are gradual and progressive (Thompson & Kyle, 2000). Individuals with chronic illness are better able to adapt to the illness to the extent that they feel they can change their situation or exercise control (e.g., see their doctor on a regular basis to monitor and treat their symptoms; Thompson & Kyle, 2000). Personal responsibility (e.g., “My knees hurt because I spent too much time working in the garden today”), unpredictability, and dependence on others are characteristics of chronic illness that can undermine a sense of personal control (Thompson & Kyle, 2000). Although control is a complex construct that has been variously defined, the current study focuses on perceptions of personal control over daily symptoms, the future course of the illness, and medical care and treatment (Affleck, Tennen, Pfeiffer, & Fifield, 1987).

It is well established that high personal control is related to a variety of positive outcomes, including better perceived health and adaptation and possibly even longevity (Krause and Shaw 2000, Skinner 1996). Some argue that a sense of control declines with age (Wolinsky, Wyrwich, Babu, Kroenke, & Tierney, 2003) and that domain-specific elements of control (e.g., health-related) are especially important for older adults (Lachman, 1986). At the same time, studies suggest that adults who lack a sense of control are more likely to have poorer adaptation (e.g., higher depressive symptoms and negative affect; Kunzmann et al 2002, Lachman and Weaver 1998). In the context of physical illness, for older adults, depression may result when control strategies cannot address threats and losses associated with health (Wrosch, Schulz, & Heckhausen, 2004). Moreover, as Krause (1987) noted, the combination of the possible decline in control with age and the importance of control as a buffer against stress (e.g., chronic pain from OA) means that older adults without a strong sense of personal control may be particularly vulnerable.

In the context of a painful chronic illness like OA, personal control may be a psychological resource for coping with stress (Krause 1987, Thompson and Kyle 2000). As demonstrated for rheumatoid arthritis (RA), a more severe form of arthritis, low perceptions of control over illness combined with pessimistic attributions about events lead to more depression (Chaney et al 1996, Schiaffino and Revenson 1992). Also, perceived control has been found to either mediate or moderate between other stressors (low income, financial strain) and well-being/depressive symptoms for older adults (Krause 1987, Lachman and Weaver 1998).

Different explanatory or mediating variables may be necessary to explain the relationship between pain and depression for women in middle-age compared to older age (Turk, Okifuji, & Scharff, 1995). Functional impairment is not only unexpected, but also interferes with valued social and work roles in middle age and may therefore be especially important. Alternately, a sense of control over chronic health problems, which are expected but nonetheless stressful and threatening, may be important for older age. Therefore, the mechanisms through which pain influences well-being may be different for these 2 groups.

This paper applies the activity restriction model and the concept of personal control over illness to explain the relationship between pain and depression in middle-aged and older women with OA. We hypothesized that pain would have a direct relationship with depressive symptoms for both middle-aged and older women. We also hypothesized that functional impairment would mediate the relationship between pain and depression for middle-aged women only and that personal control would mediate this relationship for older women only.

Back to Article Outline

Methods 

Participants 

Participants were recruited from the Hospital for Special Surgery in New York City, which specializes in the treatment of rheumatic disease. Participants included 47 middle-aged (age 45–69) and 52 older (age 70–87) women with a doctor-confirmed diagnosis of OA. Patients were selected for the study if they reported having OA but not RA or other complicating rheumatic disease (e.g., lupus, psoriatic arthritis), diagnosed with OA ≤15 years ago, female, and age ≥45. Participants completed face-to-face interviews at the hospital or in their homes. Informed consent was obtained prior to participation and respondents were given a stipend of $10 for their participation. Institutional Review Board approval was received by both the hospital and the researcher’s university.

Middle-aged women were significantly more likely to be married, employed, and have higher income compared with older women. However, these variables were unrelated to depressive symptoms. There were no significant differences in ethnicity, education, self-reported health status, or number of other health problems between middle-aged and older women.

Measures 

Personal control 

Perceived personal control over illness was assessed with a 3-item scale that measures appraisals of personal control over daily symptoms, future course of disease, and medical care and treatment (Affleck et al., 1987). The 3 items were 1) How much personal control do you think you have over your daily symptoms, that is the amount of pain, fatigue, discomfort, and immobility you experience from day to day?; 2) How much personal control do you believe you have over the long-term course of your underlying disease, that is, whether it will improve or at least not worsen in the future?; and 3) How much personal control do you think you have over the medical care and treatment of your illness? Items were responded to on a scale from 0–10, where 0 is absolutely no control, and 10 is extreme amount of control (mean [M] = 18.6; standard deviation [SD] = 5.8). Using this measure of personal control, RA patients demonstrated a relationship between control and both predictability of illness and emotional adjustment (Affleck et al., 1987). Acceptable reliability of the personal control measure was found in the current study (alpha = .61). The small number of items may explain the marginal alpha obtained.

Functional impairment and pain 

The Arthritis Impact Management Scale 2 (AIMS2; Meenan, Mason, Anderson, Guccione, & Kazis, 1992) was used to measure self-reported functional impairment and pain in the past month. The AIMS2 28-item measure of functional status consists of 6 subscales: mobility; walking and bending; hand and finger function; arm function; self-care; and household tasks. Higher scores indicate poorer functional status (M = 15.1; SD = 6.0). Pain was assessed using the 5-item arthritis pain subscale from the AIMS2. Higher scores indicate greater pain (M = 4.4; SD= 2.4). Items on the AIMS2 are responded to on a 5-point scale, recoded and summed to produce a raw score, then normalized to produce standard scores ranging from 0–10 for all scales. The AIMS2 is widely used in arthritis research and has been shown to be a valid and reliable instrument with alpha coefficients ranging from .74–.95 and test–retest reliability coefficients ≥.81 for all subscales in a sample of adults with OA or RA (Meenan et al., 1992). The AIMS2 demonstrated good reliability in the current study for functional impairment (alpha = .84) and pain (alpha = .83).

Depressive symptoms 

Depressive symptoms were measured with the Center for Epidemiological Studies Depression Scale (CES-D; Radloff, 1977). The CES-D is a 20-item self-report scale that measures frequency of mood and behavioral symptoms that occurred during the previous week. Items are rated on a 4-point scale ranging from rarely/none of the time to most or all of the time. Scores range from 0–60 with high scores indicating higher depressive symptoms (M = 12.6; SD = 9.8). Examples of items from the CES-D include “I was bothered by things that usually don’t bother me,” “I felt depressed,” and “I enjoyed life.” Positively worded items were reverse coded. The CES-D has been used extensively in community samples. The CES-D demonstrates adequate reliability and validity in the general population (alpha = .85; Radloff, 1977), RA patients (average internal consistency = .92; Smith & Wallston, 1992), and a sample of older adults with either arthritis or no chronic disease (alpha = .87; Penninx et al., 1997). The CES-D demonstrated good reliability in the current study (alpha = .90).

Self-rated health 

Self-rated health was used as a control variable. Self-rated health was assessed with a 1-item question that asked, “How would you rate your health at the present time?” Respondents answered on a 5-point scale and choices ranged from excellent (1) to very poor (5). This measure of self-rated health is widely used and accepted as a strong predictor of health and functioning. This single item measure of self-rated health has been shown to be a strong predictor of morbidity and mortality, as well as functional status, in past research (Idler and Benyamini 1997, Idler and Kasl 1995).

Statistical Analyses 

We first examined correlation coefficients to assess the strength of the bivariate associations between self-rated health, pain, functional impairment, personal control, and depressive symptoms among middle-aged and older women.

Next, we examined whether functional impairment and personal control mediated the relationship between pain and depressive symptoms. A mediator is a variable that accounts for the relationship between a predictor and criterion variable, at least in part (Baron & Kenny, 1986). Following the mediation guidelines set forth by Baron and Kenny (1986), a series of regression analyses were conducted to determine if functional impairment and personal control mediated the relationship between pain and depressive symptoms, while controlling for self-rated health. Analyses were examined separately for each age group. Four sets of regression analyses were used to assess whether functional impairment mediated the relationship between pain and depressive symptoms and an additional set of 4 regression analyses were used to determine if personal control mediated the association between pain and depressive symptoms for middle-aged and older women independently.

According to Baron and Kenny (1986), the following conditions must be met to establish mediation: first, the independent variable (pain) must be related to the dependent variable (depressive symptoms). Second, the independent variable should be related to the mediator (functional impairment or personal control). Third, the mediator should be related to the dependent variable while controlling for the independent variable. Finally, the relationship between the independent variable and the dependent variable should be reduced to zero when controlling for the mediator. The Sobel test (Preacher and Leonardelli 2001, Sobel 1982) was used to confirm partial mediation. This occurs when the relationship between the independent and dependent variables is decreased, but not reduced to zero (full mediation only occurs when the effect is reduced to zero). Further confirmation of mediation is then established by using Mackinnon and Dwyer’s (1993) formula (αβ/(αβ + τ′) to determine the percentage of the total effect mediated (αβ equals the indirect effect and τ′ equals the direct effect). This statistic determines the percentage of the relationship between pain and depressive symptoms that is accounted for by the mediator, thus allowing comparisons between the 2 age groups.

Back to Article Outline

Results 

For older women, with the exception of self-rated health in relation to personal control, all other study variables were moderately and significantly correlated (Table 1). For middle-aged women, again most variables were significantly correlated, but there was no relationship between pain or functional impairment and personal control. Because self-rated health was related in most cases to the variables used in the mediation analyses for both groups, we controlled for self-rated health in all analyses.

Table 1. Bivariate Correlations Among Study Variables
Variables12345
Self-rated health—.31⁎.33⁎−.37⁎⁎.28
Pain.46⁎⁎—.64⁎⁎⁎−.18.42⁎⁎
Functional impairment.36⁎⁎.47⁎⁎⁎—−.15.51⁎⁎⁎
Personal control−.17−.37⁎⁎−.47⁎⁎⁎—−.31⁎
CES-D.38⁎⁎.41⁎⁎.32⁎−.45⁎⁎⁎—

Note: Correlations for middle-aged women are presented above the axis and correlations for older women are presented below the axis.

CES-D, Center for Epidemiological Studies Depression Scale.

⁎p < .05

⁎⁎p < .01

⁎⁎⁎p < .001.

Next, to examine whether functional impairment and personal control act as mediators in the relationship between pain and depressive symptoms, we followed the procedure suggested by Baron and Kenny (1986). In 4 sets of regression analyses, we examined whether 1) functional impairment was a mediator for middle-aged women, 2) functional impairment was a mediator for older women, 3) personal control was a mediator for middle-aged women, and 4) personal control was a mediator for older women. Standardized betas are reported below.

Standardized and unstandardized beta coefficients for the series of hierarchical regressions involving functional impairment can be found in Table 2. Figure 1 provides a visual representation of the mediation effects. For middle-aged women, after finding a significant relationship between pain and depressive symptoms (β = .37, p < .05), we established that 1) pain was related to functional impairment (β = .59, p < .001), 2) functional impairment was related to depressive symptoms controlling for pain (β = .37, p < .05), and 3) while controlling for functional limitations, the relationship between pain and depressive symptoms was reduced (β = .15, p = NS). Using the Sobel test, the reduction due to functional impairment is statistically significant (Z = 2.01, p < .05). Thus, the results meet the conditions for mediation set forth by Baron and Kenny (1986). Using Mackinnon and Dwyer’s formula, 59% of the effect of pain on depressive symptoms is mediated by functional impairment.

Table 2. Summary of Regression Analyses for Functional Impairment and Personal Control as Mediators for Pain and Depressive Symptoms among Middle-Aged and Older Women
MediatorPain to Depressive Symptoms Without MediatorPain to MediatorMediator to Depressive Symptoms With PainPain to Depressive Symptoms With Mediator
BSE BβBSE BβBSE BβBSE Bβ
Middle-aged women
Functional impairment.21.08.37⁎.04.01.59⁎⁎⁎3.221.47.37⁎.08.09.15
Personal control.21.08.37⁎⁎−.17.33−.08−.05.04−.21.21.08.35⁎
Older women
Functional impairment.22.10.31⁎.03.01.39⁎⁎1.191.49.12.18.11.26
Personal control.22.10.31⁎−.95.39−.37⁎−.09.04−.35⁎⁎.13.10.18

Note. Self-rated health was controlled for in every analysis.

⁎p <.05;

⁎⁎p <.01;

⁎⁎⁎p <.001.

  • View full-size image.
  • Figure 1. 

    Functional impairment as a mediator between pain and depressive symptoms for middle-age and older women. Standardized betas are reported. The standardized beta indicating the initial path between pain and depressive symptoms is shown on top of the line between these variables and the standardized beta including functional impairment as a mediator is shown below the line. *p < .05; **p < .01; ***p < .001.

For older adults, following the same steps, we found that this set of analyses did not meet the criteria for mediation because functional impairment was not related to depressive symptoms. Therefore, functional impairment did not mediate the relationship between pain and depressive symptoms for older women.

We next examined personal control as a mediator of the relationship between pain and depression. Standardized and unstandardized beta coefficients for the series of hierarchical regressions involving personal control can be found in Table 2 and Figure 2 provides a visual representation of the mediation effects. Personal control did not meet the criteria for mediation for middle-aged women because pain was not related to personal control and personal control was not related to depressive symptoms. However, we did find evidence to suggest that personal control is a mediator for older women. We established that 1) pain was related to personal control (β = −.37, p < .05), 2) while controlling for pain, personal control was related to depressive symptoms (β = −.35, p < .01), and 3) while controlling for personal control, the relationship between pain and depressive symptoms was reduced (from β = .31, p < .05 to β = .18, p = NS). The Sobel test (Z = 1.81) was marginally significant at p = .07, which we interpret as evidence of mediation given the conservative nature of the Sobel test as well as the small sample size. To further validate the mediation effect, we again employed MacKinnon and Dwyer’s formula to demonstrate that 42% of the effect of pain on depressive symptoms is mediated by personal control.

  • View full-size image.
  • Figure 2. 

    Personal control as a mediator between pain and depressive symptoms for middle-aged and older women. Standardized betas are reported. The standardized beta indicating the initial path between pain and depressive symptoms is shown on top of the line between these variables and the standardized beta including personal control as a mediator is shown below the line. *p < .05; **p < .01; ***p < .001.

Back to Article Outline

Discussion 

The current study examined the possibility that the consistently observed relationship between pain and depressive symptoms in women with OA may be explained by different variables in older versus middle-aged women. As expected, we found a direct relationship between pain and depression for both groups. However, we found that functional impairment was a mediator between pain and depressive symptoms for middle-aged women whereas personal control over illness mediated that same relationship for older women with OA. It is possible that, for middle-aged women, functional impairment is unexpected and interferes with valued social and work roles that exemplify life in middle age. It has been found that women with rheumatic disease report high levels of psychological demands both at paid work and family work and that autonomy was important in reducing family demands and social support reduced the effects of work demands (Reisine & Fifield, 1995).

Why, however, would personal control be more important than impairment in older women? We know that chronic illness, especially OA, tends to be more common in older age. To the extent that we value control because it makes life understandable and predictable, loss of control is perceived as a major threat. Perceived control has been associated with competence on the one hand versus helplessness on the other (Smith, Dobbins, & Wallston, 1991). Perhaps the experience of control over what one does is, for older women, more important than actual levels of functioning.

There are several limitations of the current study that need to be addressed. First, the results are preliminary and need to be interpreted with caution. It must be acknowledged that the sample in this study is small, very select, and not representative of an increasingly diverse aging population. The majority of participants are white, middle- to upper-class women and results may vary based on race and socioeconomic status. Moreover, this is a “specialty clinic”-based sample indicating that participants are a select group of women who are already actively seeking help and treatment and the results may not apply to women in different circumstances. Given these issues regarding sample size and selectivity, the results may not be generalizable to all middle-aged and older women with OA and the conclusions that can be drawn from this study may be limited. Future research using large, representative samples is needed to replicate these findings and future studies should include samples that reflect the diversity of women in terms of age, race, socioeconomic status, access to health care and other services, and life circumstances. The results are, however, useful for guiding future research with larger, representative samples and for providing preliminary evidence that different factors may explain the relationship between pain and depression for middle-aged versus older women. However, these issues should be kept in mind as we discuss the potential implications for intervention below.

In addition, because our study is cross-sectional, we cannot be certain that our findings are due to a developmental change versus a cohort effect. Although we assumed that pain preceded depression in our analysis and conceptualization, it is possible that depression precedes pain. However, it is also important to note that cross-sectional findings suggesting that functional impairment mediates the relationship between pain and depressive symptoms were confirmed longitudinally in a recent study on pain in cancer patients suggesting that chronic pain is experienced first and then leads to functional impairment and subsequent poor well-being (Williamson & Schulz, 1995).

Our results suggest the value of a longitudinal examination of the relationships between these variables in these groups, which could contribute to a more sophisticated understanding of cohort and developmental differences in adjustment to chronic illness. Results from this study also have implications for interventions designed to increase the likelihood of successful adaptation to arthritis. It may be useful for interventions to target areas that are important based on life stage rather than applying the same intervention to all women with OA. The results of this study suggest, as noted by Turk & Okifuji (2002), that 1 intervention does not fit all. Turk and Okifuji (2002) emphasize that given the same diagnosis individuals may respond differently to the same intervention and may have different important psychosocial characteristics. Moreover, they explain that broad multifaceted interventions may not be cost effective in that certain components will be effective and other components will be ineffective for subgroups of patients. Our results suggest that it may be important for interventions to target middle-aged women’s perceptions about their functional limitations. It may be useful to teach middle-aged women strategies for functioning in their social and work worlds effectively even when they have pain. As has been noted by Williamson and Dooley (2001), the common perception that if one has pain and functional limitations they should “limit their activities” may actually be maladaptive. They suggest that improvement in activity restrictions will very likely be accompanied by improvement in psychological well-being.

On the other hand, interventions for older women with OA might be tailored to increase their sense of personal control over their illness. Perhaps having less personal control over a painful, chronic illness leads to poor emotional adjustment for older women. Interventions might be designed to enhance older womens’ perceptions of personal control over their health. Whether interventions target functional limitations or sense of personal control, the main point is that women do not need to suffer in silence with pain from arthritis. We know that there are things that can be done to treat OA through self-management techniques, exercising and staying active, and improving sense of personal control over illness. Effective management strategies can be taught to women who have OA; however, this information is not common knowledge. Public health efforts need to educate women that something can be done about OA and about how to best manage this chronic illness.

Back to Article Outline

Acknowledgments 

The authors gratefully acknowledge Alexandra Barsdorf for assistance with data collection and data cleaning.

Back to Article Outline

References 

  1. Affleck G, Tennen H, Pfeiffer C, Fifield J. Appraisals of control and predictability in adapting to a chronic disease. Journal of Personality and Social Psychology. 1987;53:273–279
  2. Arthritis FoundationAssociation of State and Territorial Health OfficersCenters for Disease Control and Prevention. National arthritis action plan: A public health strategy. Atlanta: Arthritis Foundation; 1999;
  3. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology. 1986;51:1173–1182
  4. Bookwala J, Harralson TL, Parmalee PA. Effects of pain on functioning and well-being in older adults with osteoarthritis of the knee. Psychology and Aging. 2003;18:844–850
  5. Centers for Disease Control and Prevention. Prevalence and impact of arthritis among women—United States, 1989–1991. Morbidity Mortality Weekly Report. 1995;44:329–352
  6. Centers for Disease Control and Prevention. Prevalence of disabilities and associated health conditions among adults—United States, 1999. Morbidity Mortality Weekly Report. 2001;50:120–125
  7. Centers for Disease Control and Prevention. Update: Direct and indirect costs of arthritis and other rheumatic conditions—United States, 1997. Morbidity Mortality Weekly Report. 2004;53:388–389
  8. Chaney JM, Mullins LL, Uretsky DL, Doppler MJ, Palmer WR, Wees SJ, et al. Attributional style and depression in rheumatoid arthritis: The moderating role of perceived illness control. Rehabilitation Psychology. 1996;41:205–223
  9. DeVellis BM. The psychological impact of arthritis: Prevalence of depression. Arthritis Care and Research. 1995;8:284–289
  10. Hootman JM, Sniezek JE, Helmick CG. Women and arthritis: Burden, impact, and prevention programs. Journal of Women’s Health & Gender-Based Medicine. 2002;11:407–416
  11. Idler EL, Benyamini Y. Self-rated health and mortality: A review of twenty-seven community studies. Journal of Health and Social Behavior. 1997;38:21–37
  12. Idler EL, Kasl SV. Self-ratings of health: Do they predict change in functional ability?. Journal of Gerontology: Social Sciences. 1995;50B:S344–S353
  13. Krause N. Chronic strain, locus of control, and distress in older adults. Psychology and Aging. 1987;2:375–382
  14. Krause N, Shaw BA. Role-specific feelings of control and mortality. Psychology and Aging. 2000;15:617–626
  15. Kunzmann U, Little T, Smith J. Perceiving control: A double-edged sword in old age. Journal of Gerontology: Psychological Sciences. 2002;57B:P484–P491
  16. Lachman ME. Locus of control in aging research: A case for multidimensional and domain-specific assessment. Journal of Psychology and Aging. 1986;1:34–40
  17. Lachman ME, Weaver SL. The sense of control as a moderator of social class differences in health and well-being. Journal of Personality and Social Psychology. 1998;74:763–773
  18. Mackinnon DP, Dwyer JH. Estimating mediated effects in prevention studies. Evaluation Review. 1993;17:144–158
  19. Meenan RF, Mason JH, Anderson JJ, Guccione AA, Kazis LE. The content and properties of a revised and expanded arthritis impact measurement scales health status questionnaire. Arthritis and Rheumatism. 1992;35:1–10
  20. National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS). Osteoarthritis. 2002;Available: www.niams.nih.gov. Accessed March 6, 2005.
  21. Parmalee PA, Katz IR, Lawton MP. The relation of pain to depression among institutionalized aged. Journal of Gerontology: Psychological Sciences. 1991;46:P15–P21
  22. Penninx BWJH, Van Tilburg T, Deeg DJH, Kriegsman DMW, Boeke AJP, Van Eijk JTM. Direct and buffer effects of social support and personal coping resources in individuals with arthritis. Social Science & Medicine. 1997;44:393–402
  23. Preacher KJ, Leonardelli GJ. (March). Calculation for the Sobel test: An interactive calculation tool for mediation tests. 2001;Available: www.unc.edu. Accessed October 4, 2005.
  24. Radloff LS. The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement. 1977;1:385–401
  25. Reisine S, Fifield J. Family work demands, employment demands and depressive symptoms in women with rheumatoid arthritis. Women’s Health. 1995;22:25–45
  26. Schiaffino KM. Human development in rheumatic diseases. In:  Robbins L,  Burckhardt CS,  Hannan MT,  DeHoratius RJ editor. Primer for clinical care in rheumatic disease. 2nd ed.. Atlanta: American College of Rheumatology; 2001;p. 25–29
  27. Schiaffino KM, Revenson TA. The role of perceived self-efficacy, perceived control, and causal attributions in adaptation to rheumatoid arthritis: Distinguishing mediator from moderator effects. Personality and Social Psychology Bulletin. 1992;18:709–718
  28. Skinner EA. A guide to constructs of control. Journal of Personality and Social Psychology. 1996;71:549–570
  29. Smith CA, Dobbins CJ, Wallston KA. The mediational role of perceived competence in psychological adjustment to arthritis. Journal of Applied Social Psychology. 1991;15:1218–1247
  30. Smith CA, Wallston KA. Adaptation in patients with chronic rheumatoid arthritis: Application of a general model. Health Psychology. 1992;11:151–162
  31. Sobel ME. Asymptotic intervals for indirect effects in structural equations models. In:  Leinhart S editors. Sociological methodology 1982. San Francisco: Jossey-Bass; 1982;p. 290–312
  32. Thompson SC, Kyle DJ. The role of perceived control in coping with the losses associated with chronic illness. In:  Harvey JH,  Miller ED editor. Loss and trauma: General and close relationship perspectives. Philadelphia: Brunner-Routledge; 2000;p. 131–145
  33. Turk DC, Okifuji A. Psychological factors in chronic pain: Evolution and revolution. Journal of Consulting and Clinical Psychology. 2002;70:678–690
  34. Turk DC, Okifuji A, Scharff L. Chronic pain and depression: Role of perceived impact and perceived control in different age cohorts. Pain. 1995;61:93–101
  35. Williamson GM. The central role of restricted normal activities in adjustment to illness and disability: A model of depressed affect. Rehabilitation Psychology. 1998;43:327–347
  36. Williamson GM, Dooley WK. Aging and coping: The activity solution. In:  Snyder CR editors. Coping with stress: Effective people and processes. London: Oxford University Press; 2001;p. 240–258
  37. Williamson GM, Schulz R. Pain, activity restriction, and symptoms of depression among community-residing elderly adults. Journal of Gerontology: Psychological Sciences. 1992;47:P367–P372
  38. Williamson GM, Schulz R. Activity restriction mediates the association between pain and depressed affect: A study of younger and older adult cancer patients. Psychology and Aging. 1995;10:369–378
  39. Wolinsky FD, Wyrwich KW, Babu AN, Kroenke K, Tierney WM. Age, aging, and the sense of control among older adults: A longitudinal reconsideration. Journal of Gerontology: Social Sciences. 2003;58B:S212–S220
  40. Wrosch C, Schulz R, Heckhausen J. Health stresses and depressive symptomatology in the elderly: A control-process approach. Current Directions in Psychological Science. 2004;13:17–20
  41. Yelin E, Lubeck D, Holman H, Epstein W. The impact of rheumatoid arthritis and osteoarthritis: The activities of patients with rheumatoid arthritis and osteoarthritis compared to controls. The Journal of Rheumatology. 1987;14:710–717
  42. Zeiss AM, Lewinsohn PM, Rohde P, Seeley JR. Relationship of physical disease and functional impairment to depression in older people. Psychology and Aging. 1996;11:572–581

Jessica M. McIlvane, PhD, is Assistant Professor in the School of Aging Studies at the University of South Florida, Tampa, FL. Her research interests include coping with stress and chronic illness in adulthood, health disparities in arthritis, and social relationships.

Kathleen M. Schiaffino, PhD, is Associate Professor of Psychology at Fordham University, Bronx, NY. She is on the faculty of the Clinical Psychology and Applied Developmental Psychology doctoral programs and conducts research related to illness identity and illness beliefs.

Stephen A. Paget, MD, FACP, FACR, is the Physician-in-Chief, Chairman of the Division of Rheumatology and the Joseph P. Routh Professor of Medicine and Rheumatic Diseases at Hospital For Special Surgery, the Weill Medical College of Cornell University and the New York Presbyterian Hospital. He is the head of the largest academic division of Rheumatology in the country. His research and clinical interests include rheumatoid arthritis, systemic lupus erythematosus, osteoporosis, the vasculitides and osteoarthritis.

 Supported by an Arthritis Foundation Dissertation Grant (to J. M. M.) and a Fordham University Faculty Grant (to K. M. S.).

PII: S1049-3867(06)00115-0

doi:10.1016/j.whi.2006.10.005

Women's Health Issues
Volume 17, Issue 1 , Pages 44-51, January 2007

Article Tools

* Image Usage & Resolution