Religious and nonreligious coping methods among persons with rheumatoid arthritis




To examine religious and nonreligious coping methods among persons with rheumatoid arthritis (RA). To identify positive and negative religious coping methods and personal characteristics associated with them.


Persons with RA (n = 181) completed a religious coping questionnaire, 6 subscales from a nonreligious coping inventory, and a depression scale.


Religious and nonreligious coping were moderately correlated. The scores of all positive religious coping subscales were positively related to the importance persons attributed to religion. Scores of all negative religious coping subscales were positively associated with self-reported depressive symptoms.


Correlations of religious and nonreligious coping methods were neither completely independent of each other nor functionally redundant, suggesting that each made unique contributions to coping with RA. Persons with no (or few) depressive symptoms who reported that religion was important to them tended to make positive use of their religion as they coped with the emotional stress of RA. A significant number of self-reported depressive symptoms were correlated with a negative use of religion.


Medical and psychological research increasingly examines the contribution of religion when individuals struggle with difficult life events. Pargament (1), for example, summarized 79 studies, some of which associate religious coping methods with such positive outcomes as better physical and mental health and reduced mortality (2, 3). Other religious coping methods are associated with such negative results as poorer psychosocial competence, anxiety, negative mood, lower self esteem, and problematic personal adjustment (4).

These results suggest that studies of religious coping methods among persons with rheumatoid arthritis (RA) are potentially relevant to their medical care because the disease is chronic, can be painful, and compromises personal functioning. These illness characteristics often prompt religious coping (5–7).

The current literature concerning arthritis and religious coping is limited, usually combining patients with RA or osteoarthritis into a single sample. Additionally, these reports mention religious coping only briefly. For example, one study (8) examined poor urban minorities and commented that 92% of African Americans and 50% of Hispanics used prayer as one way of coping with their RA or osteoarthritis. Another study (9) observed that 38% of 105 poor Hispanic women with arthritis used prayer and religious beliefs or activities in their search for relief; this being the second most frequent coping strategy. A third study (10) examined coping among rural Southern arthritis patients and mentioned that 92% regularly prayed concerning their disease; prayer was the most widely used intervention by the sample.

We found only 2 articles (11, 12) that described religious coping among only persons with RA. Both described the same prayer and spiritual healing intervention project. The authors prospectively studied 40 white patients with class II or III RA, all of whom received 6 hours of education and 6 hours of prayer during which personnel from a healing center laid their hands on affected joints. The reports describe no control group or random assignment. They measured 10 arthritis-specific outcomes at baseline, every 3 months, and at 1 year, reporting that the patients continued to show significant overall improvement (11).

This literature demonstrates the weakness of RA studies concerning religious coping. Articles mention results from only a few items embedded within questionnaires constructed for other purposes (8–10) or focus on one aspect of religious coping, such as prayer (11, 12). These weaknesses are endemic in religious coping research because, until recently, the literature did not contain a theoretically based and comprehensive data-gathering questionnaire.

A recent article (13) provides a new comprehensive religious coping assessment instrument (RCOPE) with 17 subscales within 5 dimensions. Based on statistical results in their study, Pargament et al designated 10 subscales as positive religious coping and 7 subscales as negative. In the study reported here, we compare results from this new instrument to nonreligious coping scores and describe the relationship of personal characteristics and behaviors, including depressive symptoms, to these positive and negative religious coping methods. Our project did not include a clinical intervention nor did we attempt to document actual benefits (or lack thereof) associated with religious coping, leaving such efforts for future studies.



After the Institutional Review Board of the host hospital approved the project, 17 physicians in its Division of Rheumatology granted permission to contact rheumatoid arthritis outpatients under their care (n = 451). Using information from the Division's registry, we mailed a consent form, a questionnaire, a postage-paid return envelope, and a cover letter on Division of Rheumatology stationary signed by the physician author (SP). The letter promised $10 upon receipt of the signed consent form and the completed questionnaire. Two weeks later, a reminder was sent. Three weeks later, nonrespondents received a second copy of the questionnaire, another postage-paid return envelope, and a new cover letter encouraging them to respond.

Forty percent (n = 181) returned usable questionnaires, reporting a mean age of 59 (SD 1.04) years. Most were married (56%), white (68%), and female (86%). Education attainment was reported as completion of high school or less by 34%, some college by 20%, and college graduate by 46%.

Roman Catholics comprised 55% of the sample; 22% were Protestants and 23% reported other religious heritages. Among these other heritages, 47% were Jewish (n = 19). Respondents reported their frequency of attendance at worship services and other religious meetings as a measure of public religious practices. Many (43%) participated weekly or more frequently. Others attended “about once per month” (15%), on “religious holidays only” (11%), “about once per year” (14%), or “never” (17%). The frequency of engagement in private religious practices, defined in the questionnaire as “private prayer, meditation, or reading scriptures,” varied widely; 26% said they used such activities “once per day or more,” 22% participated “weekly,” 24% participated “once per month to once per year,” and 28% did not use these coping methods at all. Fifty percent said that religion was “very important” to them, 35% said it was “somewhat important,” and 15% described religion as “not important.”

Respondents reported the length of time since an RA diagnosis (mean ± SD 18 ± 0.97 years) and the extent of their pain during the last 6 months on a 10-cm visual analog scale anchored by 0 (“no pain in the last six months”) and 10 (“the most pain you can imagine during the last six months”). The mean ± SD amount of pain reported was 3.15 ± 0.17.


The questionnaire contained 3 instruments: selected subscales of a nonreligious coping inventory (COPE) (14), the long version of the RCOPE, and the Brief Depression Scale for the Medically Ill (15, 16). We inserted the words “arthritis” or “this disease and its problems” in all COPE and RCOPE items.

The COPE inventory.

We selected 6 subscales (each with 4 items) from the COPE inventory (14) to determine correlations between nonreligious and religious coping methods. The planning subscale inquired about steps that best respond to a stressor (“I made a plan of action…”). Suppression of competing activities subscale described strategies that put aside daily activities to concentrate on a stressor (“I put aside…”). The third subscale, instrumental social support, concerned gathering advice, information, or assistance in dealing with a stressor (“I tried to get advice from someone about what to do.”). Fourth, the emotional social support subscale included soliciting moral support, sympathy, or understanding (“I talked to someone about how I felt”). The fifth subscale focused on venting emotions (“I let my feelings out.”). The final subscale, mental disengagement, described strategies that distracted respondents from a stressful situation (“I slept more than usual.”).


The 17 subscales of this theoretically based instrument (Table 1) provide extensive information concerning religious coping methods (13). The authors reviewed the research literature and identified 5 dimensions of religion, including the search for meaning (17), control (18), comfort (19), intimacy (20), and life transformation (3). They wrote questionnaire items for each dimension and tested them on 2 samples (n = 1,091). The items described active, passive, and interactive religious coping methods. They included problem-focused and emotion-focused approaches, giving attention to cognitive, behavioral, interpersonal, and spiritual domains. The authors believe that the instrument can assess religious coping within the broad mainstream of the American Judeo-Christian traditions.

Table 1. The 5 religious coping questionnaire (RCOPE) dimensions and their 17 subscales
Dimension 1: religious coping methods to find meaning (23 items)
  1. Benevolent religious reappraisal: redefining stressor situations as potentially beneficial (9 items)
  2. Punishing God reappraisal: redefining stressor situations as punishment from God (5 items)
  3. Demonic Reappraisal: redefining stressor situations as acts of the devil (5 items)
  4. Reappraisal of God's power: redefining God's power to influence stressor situations (4 items)
Dimension 2: religious coping methods to gain control (23 items)
  5. Collaborative religious coping: seeking control through a partnership with God (8 items)
  6. Active religious surrender: actively giving up of control to God who provides control (5 items)
  7. Passive religious deferral: a passive waiting for God to control the situation (5 items)
  8. Pleading for direct intercession: seeking a miracle or other direct divine interventions (5 items)
Dimension 3: religious coping methods to gain comfort and closeness to God (28 items)
  9. Religious focus: engaging in religious activities to shift the focus from stressor situations (5 items)
 10. Religious purification and forgiveness: using religious acts for spiritual cleansing (10 items)
 11. Spiritual connection: seeking connection with forces that transcend the individual (3 items)
 12. Spiritual discontent: expressing confusion/dissatisfaction to the self-God relationship (6 items)
 13. Marking religious boundaries: remaining within the boundaries of acceptable religious behaviors (4 items)
Dimension 4: religious coping methods to gain intimacy with others and closeness to God (16 items)
 14. Seeking closeness with clergy and/or congregational members (5 items)
 15. Religious helping: coping by providing support and comfort to others (6 items)
 16. Interpersonal religious discontent: expressing confusion/dissatisfaction with relationships (5 items)
Dimension 5: religious coping methods to achieve a life transformation (10 items)
 17. Seeking religious direction/conversion: looking for a new life direction after old ways failed (10 items)

When completing the COPE and RCOPE, respondents described their coping methods during the last 6 months using a 4-point Likert scale (0 = not at all like me; 3 = a great deal like me). Anticipating a culturally and religiously diverse sample, we also included a “cannot answer” category, instructing respondents to chose it only if an item “does not pertain to your religious beliefs or traditions.”

The brief depression scale for the medically ill.

Depressive symptoms are linked to religious coping (12–23), and their increased prevalence among persons with RA is well documented (24–26). DeVellis (24) summarized previous studies, noting that obtaining trustworthy data concerning depressive symptoms has been difficult because scores were inflated by somatic difficulties associated with rheumatic disease. In light of this difficulty, we used the Brief Depression Scale for the Medically Ill (15, 16), consisting of 11 dichotomous items that minimize the tendency to confound depression scores with symptoms of medical illness (e.g., appetite changes). Its items concerned mood, suicidal intent, guilt or worthlessness, concentration, and psychomotor agitation. When respondents acknowledged symptoms on 3 or more items, the authors reported acceptable sensitivity (83%) and specificity (77%) for detection of major depression.

Statistical analysis.

Data analyses consisted of Pearson correlations between COPE and RCOPE subscales. A multivariate analysis (General Linear Model; SPSS version 10.1; Chicago, IL) examined the relationship between RCOPE scores (dependent variables) and 10 personal characteristics that functioned as covariates. These characteristics were age, sex, marital status (1 = single or widowed, 2 = married or in a committed relationship), ethnic group (1 = nonwhite, 2 = white), education level completed (1 = elementary or high school, 2 = some college experience; 3 = college or university graduate), religious heritage (1 = Roman Catholic, 2 = Protestant, 3 = other), attendance frequency at worship and other religious meetings (1 = weekly or more frequently, 2 = about once per month, 3 = on religious holidays only, 4 = about once per year, 5 = never), private religious practices (1 = once per day or more, 2 = weekly, 3 = once per month to once per year, 4 = never), importance of religion (1 = not important, 2 = somewhat important, 3 = very important), and scores from the Brief Depression Scale for the Medically Ill. We excluded the length of time since diagnosis and the pain report from these analyses because a questionnaire error made the data available for only 155 respondents. Neither of the 2 variables remained in initial multivariate analyses using this smaller sample.


The correlations between COPE and RCOPE scores (Table 2) suggest that religious and nonreligious coping are moderately related but separate. Most of the scores were positively related; only 6 of the 102 correlations reported in Table 2 were negative.

Table 2. Pearson correlations between 6 COPE and 17 RCOPE subscales*
RCOPE dimensions and subscalesCOPE subscales
PlanningSuppressionInstrumental social supportEmotional social supportFocus on or venting emotionsMental disengagement
  • *

    COPE = nonreligious coping inventory; RCOPE = religious coping questionnaire.

  • P ≤ 0.01.

  • P ≤ 0.05.

Religious coping to find meaning      
 Benevolent religious reappraisal0.
 Punishing God reappraisals0.010.25−
 Demonic reappraisals0.
 Reappraisal of God's power0.150.360.
Religious coping to gain control      
 Collaborative religious coping0.
 Active religious surrender0.
 Passive religious deferral−
 Pleading for direct intercession0.
Religious coping to gain comfort      
 Religious focus0.
 Religious purification and forgiveness0.
 Spiritual connection0.
 Spiritual discontent−0.020.20−
 Marking religious boundaries0.
Religious coping to gain intimacy      
 Seeking closeness with clergy/congregation members0.*
 Religious helping0.
 Interpersonal religious discontent−0.030.11−
Religious coping to achieve life transformation      
 Seeking religious direction and conversion0.

In Table 3, we report results of the 10 positive RCOPE subscales. The independent variable, “How important is religion to you personally?” was positively associated with each subscale, except for a negative relationship with collaborative religious coping. The frequency of attendance at worship and other religious meetings was positively associated with 8 subscales; the private practice of religion was positively associated with 6 subscales. Nonwhite respondents had significantly higher scores in 8 subscales. Significantly higher depression scores occurred in relation to the religious focus and religious purification/forgiveness subscales.

Table 3. Means, SEM, and results from a multivariate analysis of 10 positive RCOPE subscales*
Positive religious coping subscalesMeanSEMβ95% CI
  • *

    Means and SEM are derived from a Likert-type scale (0 = not at all like me; 3 = a great deal like me). Standardized betas are from a multivariate analysis of variance. RCOPE = religious coping questionnaire; 95% CI = 95% confidence interval.

Benevolent religious reappraisal1.200.07  
 Age  −0.008−0.015, −0.000
 Worship attendance frequency  0.1020.021, 0.183
 Private religious practices  0.1710.061, 0.281
 Importance of religion  0.3960.215, 0.578
Collaborative religious coping1.030.04  
 Importance of religion  −0.148−0.289, −0.006
Active religious surrender1.530.08  
 Ethnic background  −0.352−0.654, −0.050
 Private religious practices  0.1690.025, 0.314
 Importance of religion  0.6190.381, 0.858
Religious focus0.740.06  
 Ethnic background  −0.392−0.548, −0.109
 Education  −0.235−0.359, −0.111
 Worship attendance frequency  0.0970.020, 0.175
 Private religious practices  0.1350.030, 0.240
 Importance of religion  0.2210.048, 0.394
 Depressive symptoms  0.0560.019, 0.093
Religious purification and forgiveness0.870.06  
 Ethnic background  −0.401−0.629, −0.173
 Worship attendance frequency  0.1690.088, 0.250
 Importance of religion  0.2760.095, 0.456
 Depressive symptoms  0.0630.024, 0.101
Spiritual connections0.990.07  
 Ethnic background  −0.417−0.692, −0.142
 Worship attendance frequency  0.1370.039, 0.234
 Private religious practices  0.1390.007, 0.270
 Importance of religion  0.3380.121, 0.556
Marking religious boundaries0.900.05  
 Age  0.0080.001, 0.014
 Ethnic background  −0.267−0.464, −0.071
 Worship attendance frequency  0.1030.034, 0.173
 Private religious practices  0.2200.127, 0.314
 Importance of religion  0.1870.032, 0.342
Religious support from clergy and members0.690.06  
 Ethnic background  −0.253−0.497, −0.008
 Education  −0.161−0.300, −0.023
 Worship attendance frequency  0.1570.070, 0.243
 Importance of religion  0.2390.046, 0.432
Religious helping1.040.07  
 Ethnic background  −0.333−0.590, −0.077
 Education  −0.148−0.293, −0.003
 Worship attendance frequency  0.1040.013, 0.194
 Private religious practices  0.1770.055, 0.300
 Importance of religion  0.3090.107, 0.512
Seeking religious direction0.760.06  
 Marital status  −0.235−0.468, −0.002
 Ethnic background  −0.547−0.785, 0.309
 Worship attendance frequency  0.1060.021, 0.190
 Importance of religion  0.2240.036, 0.413

The 7 negative subscales (Table 4) were all positively related to depression scores. Frequency of attendance at worship and other religious meetings increased with higher coping scores in 3 subscales: punishing God reappraisals, passive religious deferral, and spiritual discontent. Younger respondents had higher scores in the punishing God reappraisal, spiritual discontent, and interpersonal religious discontent subscales. Nonwhite respondents had significantly higher scores on 2 subscales: demonic reappraisal and pleading for direct intercession.

Table 4. Means, SEM, and results from a multivariate analysis of 7 negative RCOPE subscales*
Negative religious coping subscalesMeanSEMβ95% CI
  • *

    Means and SEM are derived from a Likert-type scale (0 = not at all like me; 3 = a great deal like me). Standardized betas are from a multivariate analysis of variance. RCOPE = religious coping questionnaire; 95% CI = 95% confidence interval.

Punishing God reappraisal0.270.04  
 Age  −0.006−0.011, −0.001
 Worship attendance frequency  0.0620.003, 0.120
 Depressive symptoms  0.1110.083, 0.139
Demonic reappraisal0.250.05  
 Ethnic background  −0.390−0.591, −0.188
 Depressive symptoms  0.0350.001, 0.070
Reappraisal of God's power0.620.06  
 Depressive symptoms  0.0890.046, 0.132
Passive religious deferral0.330.05  
 Worship attendance frequency  0.0880.014, 0.161
 Depressive symptoms  0.0520.016, 0.087
Pleading for direct intercession0.870.07  
 Ethnic background  −0.519−0.769, −0.268
 Importance of religion  0.2830.085, 0.481
 Depressive symptoms  0.0490.006, 0.091
Spiritual discontent0.260.04  
 Age  −0.006−0.011, −0.001
 Marital status  −0.165−0.311, −0.019
 Worship attendance frequency  0.0550.003, 0.108
 Depressive symptoms  0.0750.050, 0.100
Interpersonal religious discontent0.190.03  
 Age  −0.004−0.009, −0.000
 Depressive symptoms  0.0490.027, 0.070

The depression scores suggested many depressive symptoms. Using the suggested cutoff score of 3 or more positive responses, 46% of respondents registered significant depressive symptoms. Six percent agreed with the item, “I often wish I were dead.”


The moderate correlations between religious and nonreligious coping methods (Table 2) suggest that they are not independent of each other or functionally redundant. Further, the results in Table 2 suggest that religious coping tends to have an emotional rather than a problem-solving focus. In the Table columns, more than 70% of the correlations between 3 COPE subscales that describe emotional or mental methods (emotional social support, focus on/venting emotions, and mental disengagement) and the RCOPE are significant; only 47% of the correlations between the more problem-oriented COPE subscales (planning, suppression, and instrumental social support) and RCOPE are significant.

Significant correlations in the rows of Table 2 suggest commonality between COPE subscales and 6 RCOPE subscales (benevolent religious reappraisal, reappraisal of God's power, collaborative religious coping, active religious surrender, spiritual connection, and religious helping). The other RCOPE subscales, such as passive religious deferral, demonic reappraisal, marking religious boundaries, and interpersonal religious discontent, display only 1 or 2 significant correlations with COPE subscales, suggesting that these are more unique coping strategies.

Multivariate results of RCOPE subscales associated with positive outcomes (Table 3) consistently produced significant positive associations with the item, “How important is religion/spirituality to you?” This suggests that when religion is viewed as important, it is experienced as relevant and available during coping efforts. This relevance and availability is demonstrated by 2 behaviors also frequently found in these multivariate results. Respondents demonstrate the importance of religion by frequent worship attendance (significant in 8 of the subscale results) and engaging in private religious practice (significant in 6 subscales). Furthermore, the results suggest that nonwhite respondents tend to find religion more important.

Negative religious coping is significantly associated with depressive symptoms (Table 4). Fourth-six percent reported depressive symptoms above the cutoff score. Other studies report rates of depression and anxiety disorders ranging from 14% to 42%. Compared with our results, 1 study (27) used the Geriatric Depression Scale and reported that 15–17% of their RA respondents met the criteria for clinical depression. Another study (28) reported that 36% obtained scores that indicated possible clinical depression. We gathered no information concerning the use of new medications, but their potential depressive side effects merit investigation. The mechanism that links depressive symptoms and negative religious coping is unclear.

Negative religious coping is also positively associated with increased worship attendance—significant for 3 subscales (punishing God reappraisals, passive religious deferral, and coping to gain control and closeness to God). The content of these subscales suggest that worship attendance can have diverse implications.

Although the religious heritage data provided information concerning demographic characteristics of the sample, it is not significantly related to any of the RCOPE subscales. This lack of relationship is understandable because, as others have pointed out (29, 30), many persons are on a spiritual quest that is only minimally related to their religious heritage. Consequently, health care providers gain little helpful information concerning religious coping from religious heritage data.

This study's limitations must be considered when evaluating results. First, the 181 respondents comprise less than half of those invited to participate. Thus, these results may not represent all RA persons or even all those cared for by physicians at the host hospital. Additionally, even though this sample contained many Roman Catholics (55%), Protestants (22%), and Jews (10%), this East Coast urban sample from a tertiary care hospital may not be representative.

Second, the presence of religious items in the questionnaire may have encouraged some to respond and discouraged others. Many persons hold strong opinions concerning religion and this may have influenced whether they chose to respond, thus potentially biasing the sample.

Third, all data for the study were gathered by self-report questionnaires. This data gathering method is limited, particularly in regard to the individual's disease status, measured in this study by the self-reported experience of pain. Laboratory results constitute a more reliable measure. Additionally, the need to complete a questionnaire may have prohibited responses from those patients with RA severely affecting the hands.

Fourth, some respondents volunteered that the questionnaire items tended not to accurately reflect their religious beliefs or practices. Others commented that some items reflected Christian fundamentalism. Such concerns suggest limitations of the questionnaire and the difficulty of constructing one comprehensive religious coping instrument applicable to diverse traditions and beliefs. In fact, this concern suggests that results from the RCOPE may reflect how closely the religious beliefs of individuals correspond to those assumed in the questionnaire items.

Future studies should give attention to at least 3 considerations. First, future research should clearly examine the religious beliefs and assumptions within the RCOPE items. This will clarify the instrument's ability to gather valid and reliable comprehensive data.

Second, the RCOPE authors report initial validation of their instrument by university students in the Midwest and hospital inpatients in the US Bible belt. Validity and reliability studies from more diverse samples will increase confidence in the results.

Third, future studies should explore the relationship between RA, depressive symptoms, and negative religious coping. This can clarify what clinical interventions might be helpful.

How are these results helpful to health care providers? Persons with RA who engage in negative religious coping pose significant clinical challenges. How can providers learn about these negative coping methods? Inquiry about these sensitive matters requires time, skill, and often falls outside the usual focus of provider interviews.

Additionally, the tendency to use religious coping methods, whether negative or positive, is often deeply embedding in personal convictions and reinforced by family members, friends, and clergy. Attempts by health care providers to dissuade individuals from negative religious coping will likely be dismissed, damaging the professional relationship. Religiously trained counselors may be helpful if the individual will accept a referral.

In summary, the religious coping literature frequently makes assumptions that religious coping is a simple, singular process. Our findings, supported by other researchers (3), suggest that it is a complex phenomenon that merits detailed investigation in the practice of clinical medicine.