SEARCH

SEARCH BY CITATION

Keywords:

  • Rheumatoid arthritis;
  • Social support;
  • Disability;
  • Valued life activities;
  • Depression

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

Objective

To examine the impact of instrumental and emotional support on valued life activity (VLA) disability and depressive symptoms. Instrumental support was expected to affect VLA disability; emotional support was expected to be associated with depressive symptoms and moderate the impact of VLA disability on depressive symptoms.

Methods

Data were collected over 3 years through interviews with the University of California, San Francisco, Rheumatoid Arthritis Panel. Analyses assessed whether instrumental support predicted later VLA disability and whether emotional support predicted both concurrent and later depressive symptoms.

Results

Receiving adequate instrumental support was associated with less subsequent VLA disability. Strong associations were noted between both VLA disability and emotional support with concurrent depressive symptoms. No relationship was found between emotional support and later depression. No evidence was found for the hypothesis that emotional support moderated the impact of VLA disability on depressive symptoms.

Conclusion

Results highlight the need to assess different types of support and their unique impact on critical outcomes. Instrumental support is beneficial to the maintenance of valued activities, a critical factor in the psychological adjustment of individuals living with rheumatoid arthritis. Emotional support has a significant short-term impact on depression, although it may not buffer the impact of VLA disability on future depression.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

People with rheumatoid arthritis (RA) are at risk for experiencing depressive symptoms (1–3). Understanding factors that may benefit psychological adjustment of individuals with RA is critical to developing interventions that may prevent depression. Social support may be particularly significant in adjustment to RA given the limitations that physical disability may create for both informal social contacts and intimate relationships with family members (4, 5). Social support has been found to have beneficial effects on both physical health and psychological well being of individuals coping with chronic illness (6–9), and has specifically been found to minimize the effects of physical limitations resulting from RA on psychological adjustment (10–14).

Social support is a complex construct that has been defined and measured in a variety of ways. Distinctions have been made concerning the functional aspects of support (e.g., emotional, instrumental, or informational support), the structural aspects of support (e.g., size of social networks, diffuse versus intimate relationships), and the adequacy of support (e.g., amount of support received, perceived availability of support, satisfaction with support) (6, 7, 15, 16). Social support measures are often composite in nature (13, 17, 18), limiting the interpretations that can be made about the specific impact of different support functions on adjustment. When distinctions are made, research suggests that emotional support may be more strongly associated with depression and well being than instrumental support (11, 14), and satisfaction with support is consistently found to be more strongly associated with psychological well being than amount of support received or perceived availability of support (12, 13, 19).

Although consistently associated with adjustment, support measures have been consistently unrelated to health status measures or disease severity (10, 12–14, 19). However, research indicates that social support moderates the impact of disease severity on psychological adjustment to RA. For individuals experiencing greater disease severity or physical disability, greater social support is associated with better psychological adjustment (10, 11, 18).

This article assesses the unique impact of individuals' satisfaction with the amount of instrumental and emotional support on 2 outcomes: valued life activity (VLA) disability and depressive symptoms. We hypothesize that instrumental support will be associated with VLA disability, whereas emotional support will be associated with depression.

Disability in VLAs has been found to be more strongly associated with increases in depressive symptoms than are general functional limitations (5, 20). It is expected that individuals who report more satisfaction with instrumental support (i.e., help with daily tasks) will be better able to maintain the valued activities, and that receiving adequate instrumental support will moderate the impact of poor physical function on VLA disability (Figure 1).

thumbnail image

Figure 1. Study model. Definitions taken from Verbrugge and Jette's model of disablement (42).

Download figure to PowerPoint

Emotional support is expected to be associated with depressive symptoms, such that individuals reporting more emotional support are expected to report fewer depressive symptoms. Furthermore, it is expected that emotional support will lessen the impact of VLA disability on depressive symptoms (6). For individuals who experience greater VLA disability, greater satisfaction with emotional support will be associated with fewer depressive symptoms (see Figure 1). Establishing these relationships will allow for further examination of the pathways through which both social support and activity loss affect depressive symptoms. In addition, examining these relationships both cross-sectionally and longitudinally may help clarify the causal nature of the relationship between social support and psychological adjustment.

SUBJECTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

Subjects.

This study used data from the University of California, San Francisco, Rheumatoid Arthritis Panel study. Detailed descriptions of the RA Panel have been published previously (5, 20, 21). Briefly, in 1982, participating Northern California rheumatologists listed all patients with RA who presented to their offices over a specified 1-month period. Of the initial 847 patients who were identified, 822 (97%) agreed to participate. In 1989 and 1995, 203 and 131 participants, respectively, were added to the panel in the same manner, bringing the sample size to 560.

Each year an average of 93% of the panel members has been reinterviewed. Approximately equal numbers of participants have been lost from the panel each year due to death, refusal, and loss to followup. Subjects who were lost from the panel prior to year 1 of these analyses were more likely to be male (27.6% of those who were lost versus 18.1% of those who remained; P < 0.0001), nonwhite (21.9% versus 17.1%; P < 0.04), older at study entry (mean age 59.9 years versus 50.0 years; P < 0.0001), have lower education level (11.9 years versus 13.4 years; P < 0.0001), have RA of longer duration at study entry (11.3 years versus 8.0 years; P < 0.0001), have worse function as measured by Health Assessment Questionnaire (HAQ) (22) score at study entry (1.31 versus 1.01; P < 0.0001), and have higher depression scores at the first assessment of depressive symptoms (2.5 versus 3.2; P < 0.0001).

Structured telephone interviews were conducted annually with panel members; interviews covered such topics as symptoms, disease history, functional status, comorbid conditions, medications, sociodemographic characteristics, and psychological status. All recruitment and data collection protocols were approved by the University of California, San Francisco, Committee on Human Research.

Data from 3 years were used in the current study. Subjects who participated in all years of the study and had complete data for all study variables were eligible for analysis (n = 404). Data concerning participants' functional status, social support, and performance of valued activities were taken from years 1 and 2; data concerning depressive symptoms were taken from years 2 and 3. Participants' depressive symptom scores from years 1 and 2, as well as from the 2 years prior to year 1, were used to control for previous levels of depressive symptoms.

Variables.

Social support.

Single items were used to assess each type of social support (23). To assess instrumental (or emotional) support, individuals were asked whether during the past year they felt they could have used a lot more help with daily tasks (or emotional support), some more help (or emotional support), a little more help (or emotional support), or that they received enough help with daily tasks (or emotional support). The scale for each item was 1 (needed a lot more support) to 4 (received enough support). The validity of these items has been established and they have been widely used (24, 25).

Valued life activity disability.

The activity measure used in this study assessed the impact of RA on a broad range of life activities, beyond traditional functional capacity measurements. Participants were asked whether their RA had affected their ability to perform activities that were important to them in 16 activity domains (e.g., housework, taking care of family members, going to social events, recreational activities, and work) (4). Response options were “no problems performing activities within that domain,” “trouble performing activities but have continued to perform them,” “trouble performing activities but have changed the way the activity is performed,” “unable to perform activities in that domain,” “no longer perform activities within that domain but for reasons other than RA,” or “activities in that domain are not important.” If participants indicated that they were unable to perform important activities in a given domain or that they had trouble with activities but either continued to do them or changed the way they performed those activities, that domain was coded as affected by RA. All other responses were coded to indicate that a given domain was not affected by RA or was not important to the individual. The number of valued activities newly affected by RA from year 1 to year 2 was calculated. Activities were defined as “newly affected” if they were unaffected in year 1 but affected in year 2.

Depressive symptoms.

Assessing depressive symptoms among persons with chronic medical problems is inherently difficult because somatic indicators of depression, such as fatigue or insomnia, may also be symptoms of some health conditions. This overlap is a particular problem in RA (26, 27). Depressive symptoms were measured with the short form of the Geriatric Depression Scale (S-GDS) (28, 29), which contains fewer items concerning somatic symptoms than other commonly used screening instruments. The GDS is a valid and reliable measure of depressive symptoms among both younger and older adults and has been used in a wide range of settings (30, 31). Higher scores indicate more depressive symptoms.

Covariates.

All analyses controlled for sex, age, race (white versus nonwhite), education, and marital status (married versus not married). Previous research has reported that women, individuals of nonwhite race, individuals with low education, and unmarried individuals are more likely to be depressed or have high levels of depressive symptoms (32, 33). Analyses also controlled for duration of RA and basic functional status, as measured by the HAQ (22).

Statistical analyses.

Two series of hierarchical multiple regression analyses were conducted to test the central hypotheses of the study. To assess the impact of instrumental support on VLA disability, a regression analysis was conducted with the number of valued activities newly affected by RA from year 1 to year 2 as the dependent variable and instrumental social support reported in year 1 as the primary independent variable, controlling for demographic variables, disease duration, and physical function in year 1. Age, sex, race, education, marital status, and RA duration were entered on the first step of the model; HAQ on the second step; and instrumental support rating on the third step. An interaction term representing the interaction of HAQ and instrumental support was entered on the fourth step of the model to evaluate the moderating effect of instrumental support on the relationship between HAQ and newly affected activities. The interaction term was created by multiplying the deviation scores for affected valued activities and instrumental support, a method recommended to reduce multicollinearity of main effect terms and interaction terms (34, 35).

The second set of regression analyses was conducted to assess the impact of emotional support on depressive symptoms and the moderating effect of emotional support on the relationship between VLA disability and depressive symptoms. In the main analysis, S-GDS score from year 2 was the dependent variable. In the secondary analysis, S-GDS from 1 year later (year 3) was the dependent variable. Demographic (age, sex, race, education, marital status) and disease variables (RA duration and HAQ) from year 1 were entered on the first step of these models, instrumental support from year 2 on the second step, number of valued activities newly affected from year 1 to year 2 on the third step, and emotional support from year 2 on the fourth step. An interaction term representing the interaction of the number of affected valued activities and emotional support was entered on the fifth step of these models to evaluate the moderating effect of emotional support on the relationship between VLA disability and depressive symptoms.

Exclusion criteria.

Depression in a previous year is a strong predictor of future depression (36, 37), and was therefore controlled for in all analyses. Because the focus of the analysis was on the development of depression, subjects with high levels of depressive symptoms in year 1 and the 2 years prior to year 1 (i.e., those who had already developed depressive symptoms; S-GDS score ≥7; n = 45) were excluded from the main analyses, yielding a sample size of 359 (29). The descriptive statistics for the demographic, psychosocial, and clinical variables with and without this exclusion are shown in Table 1. For the secondary analysis using year 3 data as the outcome variable, participants with scores ≥7 on the S-GDS in any year prior to year 3 (n = 58) were excluded to create a sample size of 346.

Table 1. Demographic, clinical, and psychosocial characteristics of sample*
 Excluding individuals with high GDS (n = 359)Full sample with complete data (n = 404)
  • *

    Data presented as percentage, unless otherwise noted. GDS = Geriatric Depression Scale.

  • Functional status in year 1 as measured with the Health Assessment Questionnaire.

Demographics  
 Female8181
 White8483
 Married7069
 Age, mean ± SD (range) years60.56 ± 12.78 (27–90)60.75 ± 12.69 (27–90)
 Education, mean ± SD (range) years13.55 ± 2.48 (3–20)13.50 ± 2.67 (0–20)
Clinical characteristics  
 Disease duration, mean ± SD (range) years17.42 ± 9.94 (1–69)17.71 ± 10.23 (1–69)
 Functional status, mean ± SD (range)0.99 ± 0.69 (0–2.88)1.08 ± 0.72 (0–2.88)
Psychosocial characteristics  
 GDS score year 2, mean ± SD (range)1.88 ± 1.99 (0–12) 
 GDS score year 3, mean ± SD (range)1.87 ± 2.23 (0–13)2.29 ± 2.68 (0–14)
 No. affected activities year 1, mean ± SD (range)1.55 ± 1.74 (0–12)1.49 ± 1.69 (0–12)
Instrumental support year 1, needed:  
 A lot more help34
 Some more help1314
 A little more help1113
 Received enough help7369
Emotional support year 1, needed:  
 A lot more support35
 Some more support1112
 A little more support1313
 Received enough support7369

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

Impact of instrumental support on VLA disability.

Results supported the hypothesis that receiving adequate instrumental support reduces the impact of RA on valued activities (Table 2). In step 1 of this model, demographic variables did not explain a significant amount of variability in newly affected activities. In step 2, HAQ was significantly associated with the number of newly affected activities, explaining an additional 4% of the variance: individuals with better physical function were able to maintain more of their valued activities (β = 0.206, P < 0.001). In step 3, instrumental support reported in year 1 was significantly associated with the number of valued activities newly affected from year 1 to year 2, explaining an additional 2% of the variability, a small but significant increase. Individuals who reported greater satisfaction with the level of help received with daily tasks in year 1 had fewer valued activities newly affected from year 1 to year 2 (β = −0.143, P < 0.01). Finally, in step 4, the interaction of HAQ and instrumental support was not significant. However, the data indicated a trend in the expected direction (β = 0.109, P = 0.06), suggesting a possible moderation effect, such that for individuals with worse physical function, instrumental support may be more closely associated with the maintenance of valued activities.

Table 2. Hierarchical multiple regression analysis: impact of instrumental support on the number of valued activities affected by RA from year 1 to year 2*
StepVariableβFR2Change in R2
  • *

    Beta weights (and their significance) are taken from the final step of the model. F values (and their significance) are taken from each step. Beta weights from each step are reported in the text. RA = rheumatoid arthritis; HAQ = Health Assessment Questionnaire.

  • P < 0.01.

  • P < 0.001.

  • §

    P = 0.06.

1Sex−0.0171.330.0222 
 Race−0.017   
 Education−0.018   
 Disease duration−0.034   
 Marital status−0.069   
 Age0.064   
2HAQ, year 10.1563.100.05820.036
3Instrumental support, year 1−0.1943.570.07550.017
4Moderation (HAQ × instrumental support)0.109§3.590.08460.009

Impact of emotional support on depressive symptoms.

Main analysis.

Results supported the hypothesis that emotional support was associated with depressive symptoms (Table 3). However, emotional support did not appear to moderate the impact of VLA disability on depressive symptoms. In step 1 of the model, demographic and disease variables from year 1 accounted for 14% of the variability in depressive symptoms in year 2, with HAQ the only significant predictor (β = 0.352, P < 0.001). In step 2, instrumental support in year 1 was significantly associated with depressive symptoms in year 2 (β = −0.159, P < 0.01), explaining an additional 2% of the variability. In step 3, the number of valued activities newly affected from year 1 to year 2 was significantly associated with depressive symptoms in year 2 (β = 0.168, P < 0.001), explaining an additional 3% of the variability, with instrumental support remaining a significant predictor (β = −0.135, P < 0.05). In step 4, emotional support in year 2 was also significantly associated with depressive symptoms (β = −0.272, P < 0.001), explaining an additional 6% of the variance. Once emotional support was included in the model, instrumental support was no longer a significant predictor (β = −0.057, P = 0.28), but VLA disability remained significant (β = 0.136, P < 0.01), indicating that emotional support accounts for the impact of instrumental support on depression. Finally, in step 5, the interaction of newly affected activities and emotional support was not a significant predictor of depressive symptoms, indicating that emotional support did not moderate the impact of activity loss on depressive symptoms.

Table 3. Hierarchical multiple regression analysis: impact of emotional support on depressive symptoms, year 2*
StepVariableβFR2Change in R2
  • *

    Beta weights (and their significance) are taken from the final step of the model. F values (and their significance) are taken from each step. Beta weights from each step are reported in the text. HAQ = Health Assessment Questionnaire.

  • P < 0.05.

  • P < 0.001.

  • §

    P < 0.01.

1Sex0.106   
 Race−0.030   
 Education−0.040   
 Disease duration−0.0188.050.1383 
 Marital status−0.051   
 Age0.028   
 HAQ year 10.288§   
2Instrument support year 1−0.0468.300.15950.021
3No. of affected valued activities, year 1 to year 20.143§8.830.18550.026
4Emotional support year 2−0.25111.590.24980.064
5Moderation (activity loss × emotional support)−0.06510.690.25320.003
Secondary analysis.

Emotional support received in year 2 was not significantly associated with later (year 3) depressive symptoms (β = −0.037, P = 0.49), nor was the interaction of emotional support and VLA disability significantly associated with depressive symptoms in year 3 (β = 0.011, P = 0.84). Therefore, results failed to support the hypothesis that emotional support would be predictive of depressive symptoms and moderate the impact of VLA disability on depressive symptoms at a later point in time.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

This study examined the impact of instrumental and emotional support on valued activity disability and depressive symptoms. Results indicate that instrumental support is associated with development of less disability in valued activities. Individuals who received adequate help with daily tasks during year 1 reported fewer valued activities newly affected by RA from year 1 to year 2. Although not significant, a statistical trend suggests that instrumental support may moderate the impact of basic function on VLA disability. In other words, the worse an individual's function, the more important receiving adequate help with daily tasks may be to minimize VLA disability.

In addition, analyses revealed that both VLA disability and emotional support were strongly associated with concurrent depressive symptoms. Individuals who had fewer newly affected activities from year 1 to year 2 and who received adequate emotional support in year 2 reported fewer depressive symptoms in the same year.

A significant relationship between emotional support and depressive symptoms was only found in the primary (short-term) analyses. Results failed to support the hypothesis that emotional support would be associated with future depression. Limitations in this study's measurement of social support may explain the insignificant longitudinal relationship between emotional support and depressive symptoms. Each type of social support was assessed with a single item, and the variability in responses to these items was limited. For example, in the analysis sample, 73% of participants responded that they received adequate support, both instrumental and emotional. Therefore, additional research is needed using a more sensitive multidimensional social support scale designed to independently assess both instrumental and emotional support, as well as different aspects of support (e.g., quantity of support and satisfaction with the quality of support) (38, 39). In addition, some research has shown negative effects of some types of support; future measures should consider this as well (19, 40).

In addition to the limitations posed by the social support measure, other limitations are also noted for this study. There were differences between the panel members who remained in the panel and those who did not. It is possible that attrition of panel members biased the results of this study by differential inclusion of persons with better function or psychological status. However, the individuals who have remained in the study have experienced a variety of outcomes over the course of the study, including progression of functional impairment (e.g., the mean HAQ in year 1 was 0.99, and ranged from 0 to 2.88), so by no means could all of the panel members be considered to have experienced a favorable disease course. It is possible that individuals who visit rheumatologists for their RA may be systematically different from those who visit general internists. It is also possible the persons treated by the study rheumatologists did not represent the full spectrum of persons with RA, treatments, or other independent variables. The average duration of RA among these individuals is quite long, primarily due to the length of time they have been in the panel. Although this panel includes persons as young as age 27 years, the mean age of this group is 61 years, again primarily due to the duration of the panel study. In general, though, this study sample is similar to other large panel studies of persons with RA (41) and reflects the population of persons with RA as a whole in many ways (e.g., demographic characteristics). Finally, because these analyses excluded individuals who were already depressed, these results may not be relevant to this group.

The interpretations and implications from this study concerning the impact of instrumental and emotional support on both VLA disability and depressive symptoms are clear. Receiving help with daily tasks enables individuals to maintain more of their valued activities over time. VLA disability has been identified as a significant predictor of future depressive symptoms (5, 20). Therefore, clinical interventions geared at improving the instrumental support received by individuals with RA should help them maintain more of their valued activities and thus benefit their long-term psychological adjustment. In contrast to emotional support, it may be easier to increase the amount of instrumental support individuals receive by providing them with access to resources and services to help with such daily tasks as cleaning, shopping, and cooking. Such access would then provide individuals more opportunity to stay involved in activities that more directly affect their psychological well being, such as involvement in recreational activities and social interactions (20).

The lack of a relationship between emotional support and later depressive symptoms may simply indicate that the benefit of receiving adequate emotional support is a short-term effect and may not have long-term implications for adjustment. Other longitudinal studies have failed to find a relationship between emotional support and depression (11), although consistent evidence of a cross-sectional association exists (11, 14). Therefore, interventions targeting emotional support may need to focus on the immediate impact of such efforts rather than concentrating on long-term implications. Otherwise, the significant benefits of receiving adequate emotional support may be overlooked.

Results from this study highlight the need to assess the relationship between different types of support and various aspects of adjustment to RA. Receiving adequate instrumental support is beneficial to the maintenance of valued activities over time, a critical factor in the psychological adjustment of individuals with RA. This support may be particularly important for individuals experiencing poorer physical function. Instrumental support is not directly associated with depressive symptoms. However, both receiving adequate emotional support and maintaining valued activities are beneficial to the short-term psychological adjustment of individuals with RA. This study thus illustrates that instrumental and emotional social support serve different functions in the lives of individuals with RA and that these different functions have unique effects on both psychological and physical outcomes.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES
  • 1
    Katz PP, Yelin EH. Prevalence and correlates of depressive symptoms among persons with rheumatoid arthritis. J Rheumatol 1993; 20: 7906.
  • 2
    Anderson KO, Bradley LA, Yound MD, McDaniel LK, Wise CM. Rheumatoid arthritis: review of psychological factors related to etiology, effects, and treatment. Psychol Bull 1984; 98: 35887.
  • 3
    Frank RG, Beck NC, Parker JC, Kashani JH, Elliott TR, Haut AE, et al. Depression in rheumatoid arthritis. J Rheumatol 1988; 15: 9205.
  • 4
    Katz PP, Yelin EH. Life activities of persons with rheumatoid arthritis with and without depressive symptoms. Arthritis Care Res 1994; 7: 6977.
  • 5
    Katz PP, Yelin EH. The development of depressive symptoms among women with rheumatoid arthritis: the role of function. Arthritis Rheum 1995; 38: 4956.
  • 6
    Cohen S, Wills TA. Stress, social support, and the buffering hypothesis. Psychol Bull 1985; 98: 31057.
  • 7
    Wortman CB, Conway TL. The role of social support in adaptation and recovery from physical illness. In Cohen L, Syme, SL, editors. Social support and health. New York: Academic Press; 1985. p. 281302.
  • 8
    Dunkel-Schetter C. Social support and cancer: findings based on patient interviews and their implications. J Soc Issues 1984; 40: 7798.
  • 9
    Cohen L, Syme SL. Issues in the study and application of social support. In Cohen L, Syme SL, editors. Social support and health. New York: Academic Press; 1985. p. 322.
  • 10
    Affleck G, Pfeiffer CA, Tennen J, Fifield J. Social support and psychological adjustment to rheumatoid arthritis: quantitative and qualitative findings. Arthritis Care Res 1988; 1: 717.
  • 11
    Brown GK, Wallston KA, Nicassio PM. Social support and depression in rheumatoid arthritis: a one-year prospective study. J Appl Soc Psychol 1989; 19: 114681.
  • 12
    Fitzpatrick R, Newman S, Lamb R, Shipley M. Social relationships and psychological well-being in rheumatoid arthritis. Soc Sci Med 1988; 27: 399403.
  • 13
    Goodenow C, Reisine ST, Grady KE. Quality of social support and associated social and psychological functioning in women with rheumatoid arthritis. Health Psychol 1990; 9: 26684.
  • 14
    Doeglas D, Suurmeijer T, Krol B, Sanderman R, van Rijswijk M, van Leeuwen M. Social support, social disability, and psychological well-being in rheumatoid arthritis. Arthritis Care Res 1994; 7: 105.
  • 15
    Barrera M. Distinctions between social support concepts, measures, and models. Am J Commun Psychol 1986; 14: 41345.
  • 16
    Schaefer C, Coyne J, Lazarus R. The health-related functions of social support. J Behav Med 1981; 4: 381406.
  • 17
    Schiaffino KM, Revenson TA. Relative contributions of spousal support and illness appraisals to depressed mood in arthritis patients. Arthritis Care Res 1995; 8: 807.
  • 18
    Penninx BW, van Tilburg T, Boeke AJ, Deeg DJ, Kriegsman DM, van Eijk JT. Effects of social support and personal coping resources on depressive symptoms: different for various chronic diseases? Health Psychol 1998; 17: 5518.
  • 19
    Revenson TA, Schiaffino KM, Majerovitz SD, 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.
  • 20
    Katz PP, Yelin EH. Activity loss and the onset of depressive symptoms: do some activities matter more than others? Arthritis Rheum 2001; 44: 1194202.
  • 21
    Yelin EH, Henke CJ, Kramer JS, Nevitt MC, Shearn M, Epstein WV. A comparison of the treatment of rheumatoid arthritis in health maintenance organizations and fee-for-service practices. N Engl J Med 1985; 312: 9627.
  • 22
    Fries JF, Spitz P, Kraines RG, Holman H. Measurement of patient outcomes in arthritis. Arthritis Rheum 1980; 23: 13745.
  • 23
    Berkman L. Social networks, host resistance, and mortality: a follow-up study of Alameda County residents [doctoral dissertation]. University of California, Berkeley; 1977.
  • 24
    Berkman LF, Syme SL. Social networks, host resistance, and mortality: a nine-year follow-up of Alameda County residents. Am J Epidemiol 1979; 109: 186204.
  • 25
    Seeman TE, Berkman LF. Structural characteristics of social networks and their relationship with social support in the elderly: who provides support. Soc Sci Med 1988; 26: 73749.
  • 26
    Callahan LF, Kaplan MR, Pincus T. The Beck Depression Inventory, Center for Epidemiological Studies Depression Scale (CES-D), and General Well-Being Schedule Depression Subscale in rheumatoid arthritis. Arthritis Care Res 1991; 4: 311.
  • 27
    Pincus T, Callahan LF, Bradley LA, Vaughn WK, Wolfe F. Elevated MMPI scores for hypochondriasis, depression, and hysteria in patients with rheumatoid arthritis reflect disease rather than psychosocial status. Arthritis Rheum 1986; 29: 145666.
  • 28
    Sheikh JI, Yesavage JA. Geriatric Depression Scale (GDS): recent evidence and development of a shorter version. Clin Gerontol 1986; 5: 16573.
  • 29
    Cwikel J, Ritchie K. Screening for depression among the elderly in Israel: an assessment of the Short Geriatric Depression Scale (S-GDS). Isr J Med Sci 1989; 25: 1317.
  • 30
    Olin JT, Schneider LK, Eaton EM, Zemansky MF, Pollock VE. The Geriatric Depression Scale and the Beck Depression Inventory as screening instruments in an older adult outpatient population. Psychol Assess 1992; 4: 1902.
  • 31
    Rule BG, Harvey HZ, Dobbs AR. Reliability of the Geriatric Depression Scale for younger adults. Clin Gerontol 1989; 9: 3743.
  • 32
    Kaplan GA, Roberts RE, Camacho TC, Coyne JC. Psychosocial predictors of depression: prospective evidence from the Human Population Laboratory studies. Am J Epidemiol 1987; 125: 20620.
  • 33
    Kessler R, Foster C, Webster P, House J. The relationship between age and depressive symptoms in two national surveys. Psychol Aging 1992; 7: 11926.
  • 34
    Cronbach LJ. Statistical tests for moderator variables: flaws in the analyses recently proposed. Psychol Bull 1987; 102: 4147.
  • 35
    Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychology: conceptual, strategic, and statistical considerations. J Pers Soc Psychol 1986; 51: 117382.
  • 36
    Katz PP. The impact of rheumatoid arthritis on life activities. Arthritis Care Res 1995; 8: 2728.
  • 37
    Lewinsohn PM, Hoberman HM, Rosenbaum M. A prospective study of risk factors for unipolar depression. J Abnormal Psychol 1988; 97: 25164.
  • 38
    Barrera M, Ainsley SL. The structure of social support: a conceptual and empirical analysis. J Community Psychol 1983; 11: 13343.
  • 39
    Zimet GD, Dahlem W, Zimet SC, Farley GK. The Multidimensional Scale of Perceived Social Support. J Pers Assess 1988; 52: 3041.
  • 40
    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.
  • 41
    Fifield J, Reisine ST, Grady K. Work disability and the experience of pain and depression in rheumatoid arthritis. Soc Sci Med 1991; 33: 57985.
  • 42
    Verbrugge LM, Jette AM. The disablement process. Soc Sci Med 1994; 38: 114.