Participants and Procedures
This study used cross-sectional baseline data from adults with RA from the greater Southern California area who participated in a clinical trial that compared behavioral treatments for RA. This study received approval from the Institutional Review Board at the study site. Participants were recruited from clinic offices in the divisions of rheumatology at an academic medical center and a private hospital system and through newspaper advertisements and flyers posted in the community. After obtaining participants’ informed consent, the study rheumatologist at completed a physical examination during which the diagnosis of RA was confirmed. Participants were then referred to the academic medical center for an evaluation of their clinical functioning and psychiatric status. Reports of medication use were also obtained, including analgesics/nonsteroidal anti-inflammatory drugs, biologic agents, disease-modifying anti-rheumatic drugs (DMARDs), and “other” medication (drugs for other medical conditions, including psychotropic agents). Details on the participant recruitment and evaluation process have been reported previously (24).
Eligible participants had to (a) be 18 years of age or older, (b) meet the American College of Rheumatology revised criteria for RA, (c) have a stable disease-modifying drug regimen for 3 months prior to study entry, (d) have a stable disease course for 3 months, (e) be free of serious co-morbid medical conditions such as diabetes, congestive heart failure, renal failure, cancer, or fibromyalgia, that would confound interpretations of health status, (f) not be pregnant, (g) not have a serious psychiatric condition such as bipolar disorder, psychosis, or post-traumatic stress disorder, (h) not be suicidal, and (i) not have previous experience with cognitive behavioral therapy. All participants underwent a psychiatric evaluation using the Structured Clinical Interview for DSM Disorders, (25) under the direction of the project psychologist and psychiatrist.
Measures
The structural model evaluated in this study (Fig. 1) was comprised of the constructs of SES, reserve capacity, negative emotions, and pain. Educational attainment and household income were used as indicators of the latent variable, socioeconomic status. Specifically, participants were asked to indicate their number of years of education and their total annual household income.
Identified as a limitation in research demonstrating the mediators of the relationship between SES and health outcomes, our analyses distinguished reserve capacity mediators as the aggregate of resilience factors that have been previously identified as comprising aspects of reserve capacity (20, 23). Reserve capacity was included as a latent variable with 3 indicators representing the Personal Mastery Scale (PMS; (26)), the Social Provisions Scale (SPS; (27)), and the Arthritis Helplessness Index (AHI; (28)). The PMS is a 7-item scale that measures the extent to which an individual perceives a sense of optimism, personal control, or mastery over life outcomes. Responses are measured on a 4-point scale and total scores may range from 7–28, with higher scores reflecting a greater sense of personal mastery. The SPS assesses 6 functions or “provisions” that may be derived from social relationships (i.e., attachment, social integration, opportunity for nurturance, reassurance of worth, reliable alliance, and guidance). Items are rated on a 4-point scale; the total sum score may range from 24 to 96, with higher scores signifying greater perceived social support. The AHI is a 15-item questionnaire designed to measure participants’ perceptions helplessness and loss of control focusing on a patient’s external, health-related locus of control in association with their arthritis symptoms and pain. Items are rated on a 6-point scale and the sum score may range from 15 to 90. For analysis purposes, AHI items were reverse scored so that higher scores on this measure reflected less external, health-related locus of control and helplessness, and therefore higher levels of reserve capacity (19).
Negative emotion was included as a latent variable with 3 indicators representing the Perceived Stress Scale (PSS; (29)), the Hamilton Depression Rating Scale (HDRS; (30, 31)) and the Negative Affect Scale of the Positive Negative and Affect Schedule (PANAS; (32)). The PSS is a 10-item scale that assesses the degree to which participants find their lives to be unpredictable, uncontrollable, and overwhelming. Responses are measured on a 4-point scale; the sum score may range from 0 to 40, with higher scores indicating greater perceived stress. The HDRS is an observer-rated assessment of the nature and severity of mood, anxiety, neurovegetative, and cognitive symptoms associated with depression. The 17 items are rated on a 0–4 or 0–2 scale, and total scores may range from 0 to 50, with higher scores signifying the presence of more severe depressive symptoms. A trained project research assistant completed the HDRS on each research participant. The Negative Affect Scale of the PANAS contains a list of 10 mood adjectives and measures the extent to which participants experience negative affective states (e.g., anger, guilt, and nervousness). Items are rated on a 5-point scale, and higher scores represent greater subjective distress and negative affectivity.
Two indicators were used to measure the latent variable, arthritis pain: the total joint score from the Rapid Assessment of Disease Activity in Rheumatology (RADAR; (33)) and a pain visual analogue scale (VAS). The RADAR total joint score assess joint pain/tenderness in 10 joints on the right and left sides of the body. Items are rated on a 4-point scale and total scores may range from 0 to 60, with higher scores indicating more severe joint pain. On the pain VAS, participants indicated the severity of their arthritis pain by placing a mark on a 10.0 cm horizontal line anchored by no pain (0 cm) and severe pain (10.0 cm). The pain VAS score measured the distance from the scale origin (0 cm) to point on the line marked by the participant.
Data Analyses
Structural equation modeling (SEM) was used to examine the proposed model and analyses were conducted using EQS 6.1 (34). The associations between medication use (i.e., analgesics/nonsteroidal anti-inflammatory drugs, biologic agents, DMARDs, and other medications) and the model indicator variables were assessed to determine their potential impact on model findings. If statistically significant, covariates of medication use related to RA disease and pain were partitioned from relevant indicators prior to analyses. The rule of a minimum of 10 cases to the number of measured variables was used in determining the adequacy of the data for testing the model. Additionally, > 80% power for the regression coefficients among latent variables in the model required 85 cases. The SEM model was evaluated using multiple fit criteria: χ2 goodness-of-fit statistic, the comparative fit index (CFI), the standardized root mean square residual (SRMR), and the root mean square error of approximation (RMSEA). A statistically nonsignificant χ2 (p > .05) is suggestive of a good match between the data and the hypothesized model. A CFI value greater than .95 is considered evidence of a good fitting model (35). For SRMR, a value < .08 is considered acceptable (36). A RMSEA < .08 may also be indicative of good fit (37). Model modifications were performed based on results from the Wald test and Lagrange multiplier (LM) test, along with theoretical considerations.
Mediation analyses examined the extent to which reserve capacity and negative emotions mediated the effect of SES on pain. First, the preconditions for mediation were assessed to confirm that SES was significantly related to pain and the hypothesized mediators (i.e., reserve capacity and negative emotions) (38). Then, single mediator models were assessed to discern the mediating effects of reserve capacity and negative emotions separately. The 3-path mediated effect was also examined (i.e. the indirect effect from SES to pain mediated by reserve capacity and negative emotions). Statistical significance of the indirect effect, reflective of a significant decrease in the direct influence of SES on pain, was taken as evidence of mediation (38). The significance of indirect effect estimates was calculated by EQS, based on the Sobel method (39). Elimination of the initially significant direct effect of SES on pain indicated full mediation; partial mediation was established if the strength of this association was diminished but still significant (38). The 3-path mediated effect was also evaluated with the joint significance test, which offers evidence of mediation provided all paths involved in the collective indirect effect are significantly non-zero (40).