Blackwell Publishing IncMalden, USAPMEPain Medicine1526-2375American Academy of Pain Medicine? 20079811641172 ORIGINAL ARTICLEPain Catastrophizing in CFSNijs et al.

PA I N M E D I C I N E Volume 9 • Number 8 • 2008

PSYCHOLOGY, PSYCHIATRY AND BRAIN NEUROSCIENCE SECTION Original Research Article Exercise Performance and Chronic Pain in Chronic Fatigue Syndrome: The Role of Pain Catastrophizing Jo Nijs, PhD,*† Karen Van de Putte, MSc,* Fred Louckx, PhD,‡ Steven Truijen, PhD,† and Kenny De Meirleir, PhD* Department of *Human Physiology, Faculty of Physical Education and Physiotherapy, Vrije Universiteit, Brussels; †Division of Musculoskeletal Physiotherapy, Department of Health Care Sciences, University College Antwerp, Brussels, Belgium; ‡ Medical Sociology, Faculty of Medicine and Pharmacy, Vrije Universiteit, Brussels

ABSTRACT ABSTRACT

Objectives. This study aimed to examine the associations between bodily pain, pain catastrophizing, depression, activity limitations/participation restrictions, employment status, and exercise performance in female patients with chronic fatigue syndrome (CFS) who experience widespread pain. Design. Cross-sectional observational study. Setting. A university-based clinic. Patients. Thirty-six female CFS patients who experienced widespread pain. Outcome Measures. Patients filled in the Medical Outcomes Short-Form 36 Health Status Survey, the Chronic Fatigue Syndrome Activities and Participation Questionnaire, the Beck Depression Inventory, and the Pain Catastrophizing Scale, and underwent a maximal exercise stress test with continuous monitoring of electrocardiographic and ventilatory parameters. Results. Pain catastrophizing was related to bodily pain (r = −0.70), depression (r = 0.55), activity limitations/participation restrictions (r = 0.68), various aspects of quality of life (r varied between −0.51 and −0.64), and exercise capacity (r varied between −0.41 and −0.61). Based on hierarchical multiple regression analysis, pain catastrophizing accounted for 41% of the variance in bodily pain in female CFS patients who experience chronic widespread musculoskeletal pain. Among the three subscale scores of the Pain Catastrophizing Scale, helplessness and rumination rather than magnification were strongly related to bodily pain. Neither pain catastrophizing nor depression was related to employment status. Conclusions. These data provide evidence favoring a significant association between pain catastrophizing, bodily pain, exercise performance, and self-reported disability in female patients with CFS who experience widespread pain. Further prospective longitudinal studying of these variables is required.

Key Words. Pain, Catastrophizing, Fatigue Syndrome, Chronic, Exercise Capacity

Reprint requests to: Jo Nijs, PhD, Vrije Universiteit Brussel, Sport KRO-1, Laarbeeklaan 101, B-1090 Brussels, Belgium. Tel: 32-2477-4604, Fax: 32-2477-4607; E-mail: [email protected]. © American Academy of Pain Medicine 1526-2375/08/$15.00/1164 1164–1172

doi:10.1111/j.1526-4637.2007.00368.x

1165

Pain Catastrophizing in CFS Introduction

I

n essence, chronic fatigue has been arbitrarily put forward as the primary symptom of chronic fatigue syndrome (CFS) patients; however, 54– 75% of CFS patients experience chronic widespread pain [1]. As more homogeneous subgroups are less likely to reveal conflicting data among investigators, the international consensus that the CFS population should be subclassified is gaining interest [2]. Chronic fatigue with widespread muscle and joint pain has been suggested as an important subclass of CFS [2]. This notion has been supported by the observed associations between pain severity and disability (r between 0.51 and 0.58) in CFS patients, which was similar to the association between fatigue severity and activity limitations/participation restrictions (r = 0.50) [3,4]. Little is known about the nature of pain in patients with CFS. From two previous studies, it was concluded that neither pain-related fear of movement (“kinesiophobia”) [5], nor generalized joint hypermobility [6], was of clinical importance in CFS patients who experience chronic widespread pain. Whiteside et al. [7] found that, following graded exercise, the pain threshold increased in healthy controls, but decreased in CFS patients. Morriss and colleagues [8] were unable to find an association between selfreported pain and depression in those with CFS. The possible role for pain beliefs in chronic musculoskeletal disorders has been discussed at length in the scientific literature [9–12]. Catastrophic cognitions about the consequences of pain appear to play a crucial role in the chronic pain experience. Spinhoven et al. [10] defined beliefs as “stable thoughts patients have regarding their pain problem.” As pain catastrophizing concerns interpretations of the pain in terms of relevance and potential danger, it is classified as an attribution [10]. Pain catastrophizing is broadly defined as an exaggerated negative orientation toward actual or anticipated pain experiences [13]. Catastrophizing is currently viewed as a multidimensional construct comprising elements of rumination, magnification, and helplessness [13]. The present study aimed to examine whether catastrophic interpretations of pain are associated with pain severity and disability in CFS patients who experience chronic widespread pain. It was hypothesized that pain catastrophizing, independent from depression, predicts pain severity in patients with CFS who experience chronic widespread pain. Furthermore, from a

recent literature review it was concluded that unemployment among patients with CFS is high, and that future studies should try to determine patients’ characteristics that are associated with the inability to work [14]. Therefore, the present study was undertaken to examine whether pain catastrophizing is related to employment status in CFS patients who experience widespread pain. In addition, previous research has shown that patients with CFS present with an abnormal exercise response [7,15–17] and exacerbation of symptoms after physical activity [18,19]. Little is known about the etiology of impaired exercise capacity in CFS patients. Evidence regarding the role of pain cognitions, including pain catastrophizing, in determining physical performance in chronic low back pain has been provided [20]. It is hypothesized that both the aggravated painful experience during exercise [7] and pain catastrophizing limit exercise performance in patients with CFS. Therefore, the associations between pain catastrophizing, pain severity, and exercise capacity were explored. Methods

Subject Recruitment and Research Design To be included into the study, subjects had to fulfill the Center for Disease Control and Prevention (CDCP) criteria for CFS [18]. Any active medical condition that may explain the presence of chronic fatigue, including psychiatric illnesses, prohibits the diagnosis of CFS [18]. Therefore, all subjects underwent an extensive medical evaluation prior to study participation. The medical evaluation consisted of a standard physical examination, medical history, and routine laboratory tests. The laboratory tests included a complete blood cell count, determination of the erythrocyte sedimentation rate, serum electrolyte panel, and measures of renal, hepatic and thyroid function, as well as rheumatic and viral screens. In a number of cases, further neurological, psychiatric, gynecologic, endocrine, cardiac, and/or gastrointestinal evaluations were performed. The medical records were also reviewed to determine whether patients suffered from organic or psychiatric illnesses, which could explain their symptoms. This study focused on pain in CFS patients; thus all patients had to experience widespread pain, as defined in the American College of Rheumatology 1990 criteria for the classification of fibromyalgia [21]. Pain is considered widespread when all of the following are present: pain in both the left and the right side

1166 of the body, pain both above and below the waist, and axial skeletal pain [21]. All study participants had Dutch as their native language, and were within the age range of 18–65 years. Subjects not having Dutch as their native language were excluded, because no data documenting the psychometric properties of the French or the English version of two of the self-reported measures used in the present trial (the Pain Catastrophizing Scale, or PCS, and the Chronic Fatigue Syndrome Activities and Participation Questionnaire, or CFS-APQ) are currently available. Subjects had to be available for the labor market, which justifies the age limits. In addition, all patients who were or who had previously been self-employed were excluded. In preparation for this study, a number of CFS patients were questioned about their work status and employment status. It was concluded that subjects who were self-employed were unable to quantify their employment status. Likewise, all subjects working at home (housekeeping, childcare) were excluded. Finally, all men were excluded from the dataset to preclude bias originating from pooling of gender data [9,22]. Women report greater levels of catastrophizing [9], and it is well established that women have a lower maximal oxygen uptake than men. A total of 36 consecutive CDCP-defined female CFS patients, who experienced widespread pain, were (or used to be) gainfully employed, had Dutch as their native language, and were within the age range of 18–65 years, were recruited. The study was conducted between September 2004 and March 2005. An information leaflet was handed out to all participants, and they were instructed to read it carefully and, if applicable, to ask for additional clarification. Subjects who provided their written informed consent were then asked to fill in the Dutch versions of the following questionnaires: the Medical Outcomes ShortForm 36 Health Status Survey (SF-36), the CFSAPQ, a short questionnaire aiming at assessing occupational disability, the Beck Depression Inventory (BDI), and the PCS. As the assessment of catastrophizing should be controlled for depression [9], the BDI was included within the protocol. Afterwards, the CFS patients underwent a maximal exercise stress test on a bicycle ergometer with continuous monitoring of electrocardiographic and ventilatory parameters.

Self-Reported Measures The SF-36 assesses functional status and wellbeing or quality of life and contains eight sub-

Nijs et al. scales: physical, emotional, social and role functioning, bodily pain, mental health, vitality, and general health [23]. Bodily pain, as measured by the SF-36, is defined as the self-reported intensity of bodily pain or discomfort and the extent of interference with normal activities due to pain [23]. Subscale scores ranged between 0 and 100, with higher scores indicating better health and less body pain. The psychometric properties of the SF-36 are well characterized [23–25]. For each subscale, the median of the reliability coefficients across studies was ≥0.80, with the exception of the social functioning scale [23]. The reliability coefficients of the physical functioning subscale consistently exceeded 0.90 [23]. The CFS-APQ evaluates activity limitations and participation restrictions in patients with CFS. The scoring system of the CFS-APQ [3,26] generates two overall scores; the first one (CFS-APQ1) uses an importance verification to acknowledge that people value things differently, while the second total score (CFS-APQ2) does not take this importance verification into account. A CFS-APQ1 score of 1 indicates no activity limitations or participation restrictions, while 16 represents the maximum score; for CFS-APQ2, the scores range between 1 and 4. The internal consistency, expressed by the Cronbach alpha coefficient, of the items included in the CFS-APQ is 0.94, and the test–retest reliability coefficients of the overall scores were ≥0.95 (P < 0.001) [3,4]. In addition to the reliability statistics, various data documenting the psychometric properties of the Dutch CFS-APQ have been reported [3,4,27]. In order to assess employment status in a standardized manner, a short questionnaire consisting of five items (illness duration, disability compensation, activity status, employment status, and premorbid vs current employment rate in percentage) was constructed and can be requested from the corresponding author. The BDI was used for the assessment of depression; higher total scores reflect more severe depression (possible range 0– 63). The BDI appears to be a reliable and valid tool for the assessment of depressive symptoms in chronic pain patients [28]. The PCS is a questionnaire aiming at assessing pain catastrophizing both in clinical and in non-clinical populations [29]. Higher scores correspond to more severe catastrophic thoughts about pain (possible range 0–52). The total score and three subscale scores (rumination, magnification, and helplessness) were counted. Data supportive of the psychometric properties of the PCS have been published

1167

Pain Catastrophizing in CFS [30–33]: the Cronbach alpha coefficient varied across studies but was always ≥0.85, and the test– retest reliability coefficient was 0.92.

Exercise Testing The patients performed a bicycle ergometric test against a graded increase in workload until exhaustion was reached [15]. There was continuous monitoring of electrocardiographic and ventilatory parameters. (For a detailed description of the exercise test as applied in the present study, see [5]). The following parameters were measured and/or extrapolated: heart rate at rest (HRREST), peak heart rate (HRPEAK), exercise duration, peak work capacity attained, peak oxygen uptake (VO2PEAK; i.e., the highest amount of oxygen consumed by the tissues during the exercise test), VO2PEAK per kilogram of body weight, functional aerobic impairment, peak respiratory exchange ratio (REREAK), and the percentage of target heart rate achieved. The age-predicted HRPEAK was calculated as 220 minus the subject’s age in years [33]. The functional aerobic impairment is the percentage difference between observed VO2PEAK and the VO2PEAK predicted for a healthy person of the same age, gender, and habitual activity status [34]. The VO2PEAK for any age can be predicted using the regression equations from Bruce et al. [34]; for the purpose of this study, we used the equations for sedentary individuals. Exercise capacity testing is widely used for the assessment of patients with CFS, and it appears to be both reproducible and valid [15,28,35,36]. Statistical Analysis A one-sample Kolmogorov-Smirnov test was used to examine whether the variables entering a Pearson correlation analysis were normally distributed. If a variable was not normally distributed, then the nonparametric Spearman correlation analysis was used. In order to examine the associations between pain catastrophizing, depression, self-reported disability (i.e., the CFS-APQ and SF-36 scores), self-reported employment rate, and exercise capacity, nonparametric Spearman correlation analyses were used. The associations between pain severity, self-reported disability, and exercise performance were examined using the parametric Pearson correlation coefficient. For interpreting correlation coefficients, they were squared to obtain the coefficient of determination [37]. In order to protect against potential type I errors, a Bonferroni correction was used to adapt the significance level. In case of a statistically significant

association (P value < 0.05), a power analysis was performed. A power of 80% was considered fair. A hierarchical multiple regression analysis was conducted to examine the relative contribution of depression and pain catastrophizing to the prediction of bodily pain. Likewise, a hierarchical multiple regression analysis was performed to examine the relative contribution of depression and helplessness to the prediction of bodily pain. Retained variables were checked for multicollinearity (the variance inflating factor, or VIF, should be below 2). The power analysis was performed using SigmaStat 3.1 (Systat Software, Inc., San Jose, CA). Except for the power analysis, data were analyzed using SPSS 11.0© for Windows (SPSS Inc., Chicago, IL). Results

The mean age of the subjects was 39 ± 8 years (±SD; range 22–54), and the illness duration ranged between 6 and 360 months (median: 48 months; interquartile range: 63 months). When comparing the current and premorbid percentage employment rate, a mean difference of 65% was found (SD = 40). All descriptive statistics of the data obtained with the self-reported measures are presented in Table 1. Table 2 displays the descriptives of the exercise capacity parameters. The percentage functional aerobic impairment can be used Table 1 Descriptive statistics of the data obtained with the self-reported measures (N = 36) Variable PCS total score PCS rumination PCS magnification PCS helplessness Beck Depression Inventory

Median (IQR) 22.5 (19.0) 9.5 (6.0) 4.0 (5.8) 11.5 (9.0) 18.5 (11.8)

Range (6–52) (0–16) (0–12) (3–24) (4–32)

Variable CFS-APQ1 CFS-APQ2 SF-36 bodily pain SF-36 physical functioning SF-36 role limitations due to physical functioning SF-36 role limitations due to emotional problems SF-36 social functioning SF-36 mental health SF-36 vitality SF-36 general health perception Premorbid employment rate (%) Current employment rate (%) Difference between premorbid and current employment rate (%)

Mean ± SD 9.1 ± 2.7 2.9 ± 0.65 29.5 ± 22.2 36.7 ± 20.0 11.1 ± 27.7

Range (3.8–14.7) (1.4–3.9) (0–74) (5–80) (0–100)

51.9 ± 48.8

(0–100)

36.5 ± 24.7 54.6 ± 21.4 27.9 ± 17.8 16.4 ± 9.5 93.2 ± 16.4 28.5 ± 40.3 64.7 ± 40.0

(0–100) (16–100) (5–70) (0–35) (50–100) (0–100) (0–100)

CFS-APQ = Chronic Fatigue Syndrome Activities and Participation Questionnaire; IQR = interquartile range; PCS = Pain Catastrophizing Scale; SF-36 = Medical Outcomes Short-Form 36 Health Status Survey.

1168

Nijs et al.

Table 2 Descriptive statistics of the exercise capacity parameters (N = 36) Exercise capacity parameter

Mean ± SD

Exercise duration (minutes) HRPEAK (bpm) Workload (W) Workload per body weight (W/kg) VO2PEAK (L/min) VO2PEAK /body weight (ml/kg per min) % functional aerobic impairment Peak respiratory exchange ratio % target heart rate achieved

−8.2 ± 3.2 144.5 ± 25 −80.8 ± 13.4 1.4 ± 0.6 1,030 ± 339 16.9 ± 5.8

(2.7–16) (93–197) (20–160) (0.24–2.8) (528–1,915) (7.5–30.1)

40.0 ± 22.0 1.1 ± 0.09 79.8 ± 13.5

(−15.7–72.2) (0.94–1.3) (52.8–108.4)

Range

HRREST = resting heart rate; SD = standard deviation; VO2PEAK = peak oxygen uptake.

for the interpretation of the exercise testing data. A mean functional aerobic impairment of 40% was observed, indicating that on average, the CFS patients studied obtained a VO2PEAK that was 40%

lower compared with the VO2PEAK predicted for a healthy person of the same age, gender, and habitual activity status. The outcome of the correlation analyses is displayed in Tables 3 and 4. Pain catastrophizing was related to bodily pain (the SF-36 “bodily pain” subscale score; r = −0.70; P = 0.012), depression (r = 0.55; P = 0.012), activity limitations/participation restrictions (r = 0.68; P < 0.012), and various aspects of quality of life (physical functioning, role limitations due to emotional problems, mental health, vitality, and general health perception). When analysing the subscale scores of the PCS in relation to bodily pain (not presented in the tables), it was found that helplessness (r = −0.74; P = 0.003) and rumination (r = −0.69; P = 0.003), rather than magnification (r = −0.32; P = 0.058), were associated with bodily pain. Depression, as assessed using the BDI, was associated with bodily

Table 3 Correlations of pain catastrophizing, depression, and pain severity by activity limitations/participation restrictions, quality of life, and employment status (N = 36) Spearman’s rho Pain Catastrophizing Scale (power‡)

Variable CFS-APQ1 CFS-APQ2 SF-36 bodily pain SF-36 physical functioning SF-36 role limitations due to physical functioning SF-36 role limitations due to emotional problems SF-36 social functioning SF-36 mental health SF-36 vitality SF-36 general health perception Current employment rate (%) Difference between premorbid and current employment rate (%)

0.68, 0.59, −0.70, −0.51, −0.23, −0.51, −0.27, −0.63, −0.51, −0.64, 0.03, −0.09,

P = 0.012* P = 0.012* P = 0.012* P = 0.024* P = 0.19 P = 0.012* P = 0.11 P = 0.012* P = 0.024* P = 0.012* P = 0.86 P = 0.59

(0.99) (92) (0.99) (0.84) (0.76) (0.96) (0.84) (0.97)

Spearman’s rho Beck Depression Inventory (power‡) 0.54, 0.51, −0.56, −0.64, −0.41, −0.55, −0.39, −0.77, −0.70, −0.48, −0.11, 0.06,

P = 0.012* P = 0.024* P = 0.012* P = 0.012* P = 0.24* P = 0.012* P = 0.22* P = 0.012* P = 0.012* P = 0.036* P = 0.52 P = 0.75

(0.83) (0.84) (0.87) (0.97) (0.85) (1.00) (0.99) (0.82)

Pearson’s r SF-36 bodily pain (power‡) −0.62, P = 0.012* (0.95) −0.61, P = 0.012* (0.94) — 0.64, P = 0.012* (0.97) 0.31, P = 0.07† 0.63, P = 0.012*† (0.96) 0.52, P = 0.012* (0.79) 0.41, P = 0.24* 0.63, P = 0.012* (0.96) 0.49, P = 0.036* (0.84) 0.22, P = 0.21† −0.11, P = 0.52†

* Bonferroni-corrected P value. † Nonparametric Spearman’s rho because of the outcome of the Kolmogorov-Smirnov test. ‡ The power is presented only in case of statistically significant correlations. CFS-APQ = Chronic Fatigue Syndrome Activities and Participation Questionnaire; SF-36 = Medical Outcomes Short-Form 36 Health Status Survey.

Table 4

Pain, pain catastrophizing, and depression vs exercise capacity (N = 36)

Exercise capacity parameter

Spearman’s rho Pain Catastrophizing Scale (power†)

Spearman’s rho Beck Depression Inventory

Pearson’s r SF-36 bodily pain (power†)

Exercise duration (minutes) HRPEAK (bpm) Workload (W) Workload per body weight (W/kg) VO2PEAK (L/min) VO2PEAK/body weight (ml/kg per min) % functional aerobic impairment Peak respiratory exchange ratio % target heart rate achieved

−0.54, −0.42, −0.42, −0.51, −0.43, −0.61, 0.51, −0.23, −0.39,

−0.23, −0.03, −0.14, −0.23, −0.10, −0.21, 0.17, −0.31, 0.01,

0.54, 0.43, 0.49, 0.46, 0.49, 0.50, −0.43, 0.49, 0.41,

P = 0.012* P = 0.14* P = 0.13* P = 0.012* P = 0.09* P = 0.012* P = 0.024* P = 0.84* P = 0.20*

(0.83)

(0.76) (0.94) (0.84)

* Bonferroni-corrected P values. † The power is presented only in case of statistically significant correlations. HRPEAK = peak heart rate; VO2PEAK = peak oxygen uptake.

P = 0.17 P = 0.87 P = 0.40 P = 0.17 P = 0.56 P = 0.23 P = 0.32 P = 0.07 P = 0.98

P = 0.012* P = 0.096* P = 0.036* P = 0.06* P = 0.036* P = 0.024* P = 0.12* P = 0.024* P = 0.17*

(0.83) (0.84) (0.84) (0.82) (0.80)

1169

Pain Catastrophizing in CFS Activity Limitations / Participation Restrictions

r = .54

Exercise Capacity

r = .68

Depression

r = .55

r = -.55

r = [-.51; -.61]

Pain Catastrophizing

r = -.70

PAIN Figure 1 Female CFS patients with chronic widespread pain: associations between bodily pain, pain catastrophizing, depression, exercise performance, and self-reported disability. The associations presented are Bonferronicorrected bivariate relationships among variables. CFS = chronic fatigue syndrome.

pain (r = −0.56; P < 0.012), activity limitations/ participation restrictions (r = 0.54; P = 0.012), and all but two SF-36 subscale scores. Neither pain catastrophizing nor depression was related to employment status. Except for SF-36 role limitations due to emotional problems vs pain catastrophizing and SF-36 bodily pain vs SF-36 social functioning (power = 0.76 and 0.79, respectively), all statistically significant associations displayed a power > 0.80. Various exercise capacity parameters (exercise duration, body weight-adjusted peak workload, body weight-adjusted peak oxygen uptake, functional aerobic impairment) were significantly associated with both pain catastrophizing (r varied between −0.51 and −0.61) and bodily pain (r varied between 0.49 and 0.54). Except for the correlation between body weight-adjusted workload and pain catastrophizing (power = 0.76), all statistically significant associations displayed a power ≥ 0.80. Depression was not related to any of the exercise capacity parameters. A summary of the most important findings of the correlation analysis is displayed in Figure 1. The primary findings of the correlation analysis were strengthened by the outcome of the hierarchical multiple regression analysis. A hierarchical multiple regression analysis was performed to examine the relative contribution of pain catastrophizing and depression to bodily pain. Pain catastrophizing entered the model in step 1 (R2 = 0.42), succeeded by depression in step 2 (R2 = 0.48; Table 5). No significant additional percentage of

the variance in SF-36 bodily pain scores could be predicted by the addition of the BDI total scores to the model (P = 0.068). Examination of the beta weights revealed that only pain catastrophizing contributed significant unique variance to the prediction of bodily pain. Thus, depression does not mediate the effects of catastrophizing on bodily pain. Conversely, entering depression in step 1 (R2 = 0.31; F = 15.6; VIF = 1.0; P < 0.001), succeeded by pain catastrophizing in step 2 (R2 = 0.48; R2change = 0.17; Fchange = 10.5; VIF = 1.5; P = 0.003), generated two statistically significant models (P < 0.001). Thus, pain catastrophizing accounted for 17% of the variance in SF-36 bodily pain scores beyond the variance accounted for by BDI total scores. As the helplessness subscale score displayed the strongest correlation with bodily pain, a hierarchical multiple regression analysis was performed to examine the relative contribution of helplessness and depression to bodily pain. The helplessness subscale scores entered the model in step 1 (R2 = 0.51; Fchange = 35.4; P < 0.001), succeeded by depression in step 2 (R2 = 0.74; R2change = 0.04; Fchange = 3.2; P = 0.08). No significant additional percentage of the variance in SF36 bodily pain scores could be predicted by the addition depression to the model. Discussion

Pain catastrophizing accounts for 41% of the variance in bodily pain in female CFS patients who experience chronic widespread musculoskeletal pain. Pain catastrophizing was found to be a predictor of bodily pain, independent from depression as assessed using the BDI. Among the three subscale scores of the PCS, helplessness and rumination rather than magnification were strongly related to bodily pain. Helplessness was found to predict bodily pain, independent from depression. To our knowledge, this is the first study exploring at least part of the nature of chronic pain in Table 5 Hierarchical regression of catastrophizing and depression on bodily pain (dependent variable = SF-36 bodily pain) Variable

Beta

Model 1 Pain Catastrophizing Scale

−0.65

Model 2 Pain Catastrophizing Scale Beck Depression Inventory

−0.49 −0.29

R2

F

VIF P 1.0