Abstract

Objective. The combination of chronic musculoskeletal pain and depression is associated with worse clinical outcomes than either condition alone. In this study, we report the predictors of pain intensity and activity interference in primary care patients with co-morbid pain and depression.

Methods. This is a secondary data analysis of the 250 persons who participated in a randomized clinical trial designed to test the effectiveness of 12 weeks of optimized antidepressant therapy for both depression and pain. Using multivariate linear regression analysis, we assessed the predictive value of baseline self-efficacy, fear of movement, pain beliefs, and demographic and clinical factors on 3-month Graded Chronic Pain Scale pain intensity and activity interference outcomes.

Results. In the full model, significant sociodemographic predictors of less activity interference included being non-white (β−5.8, P = 0.04) and being employed (β−13.3, P < 0.0001). The latter was also predictive of less pain intensity (β−5.6, P = 0.01). As expected, the optimized antidepressant treatment arm was associated with improved outcomes (pain intensity: β−3.7, P = 0.0005 and activity interference: β−6.4, P = 0.01). Whereas stronger perceived pain control (β 3.6, P = 0.01) was associated with greater activity interference, higher degree of fear of movement (or fear avoidance) predicted greater pain intensity (β 0.46, P = 0.04) and activity interference (β 0.57, P = 0.05). Neither the location (low back vs hip/knee) nor duration of pain were predictive of pain intensity or interference outcomes.

Conclusion. The findings are consistent with a bio-psychosocial model, implicating the need to consider the impact of sociodemographic variables and pain-related beliefs and cognition on pain-related outcomes for patients with co-morbid musculoskeletal pain and depression.

Introduction

Musculoskeletal pain is the most common symptom reported in both the general population and in primary care [1–6]. In the United States, pain complaints account for over half of all outpatient visits for somatic symptoms, including an estimated 25 million visits alone each year for back pain and 12.7 million visits for knee or hip pain [7]. In the United States, pain costs over $100 billion each year in health care and lost productivity [8].

Chronic musculoskeletal pain (CMP) co-morbid with depression is particularly common [9–12]. Pain is a strong predictor of both the onset and persistence of depression [13–16], and depression is likewise a powerful predictor of pain, particularly persistent pain [17–19]. Concurrent pain and depression have a much greater impact than either disorder alone on multiple domains of functional status as well as health care utilization [18]. Consequently, a treatment model that incorporates assessment and treatment of both pain and depression seems necessary for more optimal outcomes [20].

Despite the availability of medications and other therapies for managing co-morbid pain and depression, undertreatment of these co-existing illnesses persists throughout the world [21–26]. Hence, it is important to identify risk factors associated with persistent problems after treatment. For researchers, this knowledge might be used to create more tailored and effective study interventions. For clinicians, it is essential to identify predictive factors for poor outcome and address these factors during treatment in order to choose the most effective treatment for individual patients.

Findings from several systematic reviews have concluded that baseline pain intensity, work-related parameters (i.e., receiving compensation and work disability) and coping style at baseline consistently predicted future pain-related outcomes [27–32]. Interestingly, no consistent evidence was found for the predictive value of other sociodemographic and psychological variables. However, in almost all of the studies reviewed, only 40–50% of the participants had coexisting depression [18]. The greater majority had CMP only. Because most studies suggest that patients with pain and depression are less responsive to treatment than patients with pain only [33–36], there is a need to determine predictors of pain-related outcomes among individuals with co-morbid chronic pain and depression.

As it can guide rational approaches to treatment, the bio-psychosocial model is a particularly useful model for understanding chronic painful conditions. The bio-psychosocial model suggests that the experience of pain involves a complex interaction of biological factors (genetics), psychological factors (mood, thoughts, and beliefs) and the social context (interpersonal relationship) [37]. Using the bio-psychosocial theoretical approach, we sought to identify predictors of pain intensity and activity interference in patients with CMP with co-morbid depression. This report contains a secondary data analysis of Stepped Care for Affective Disorders and Musculoskeletal Pain (SCAMP)—a randomized controlled trial ( RCT) comparing optimized antidepressant therapy vs usual care in improving the co-primary outcomes of depression and pain [38].

Methods

Details of the SCAMP trial design, treatment intervention, study population, and measures have been previously described [38]. Briefly, primary care patients with persistent musculoskeletal pain and co-morbid depression were enrolled in a two-step RCT, with the intervention group first receiving 12 weeks of optimized antidepressant therapy (step 1) followed by 12 weeks of a pain self-management program (step 2). The pain had to be: 1) located in the low back, hip or knee; 2) persistent for 3 months or longer despite conventional analgesic treatment, defined as prior use of at least two different analgesics; and 3) at least moderate in severity, defined as a Brief Pain Inventory score of 5 or greater [39,40]. The depression had to be of at least moderate severity, that is, a Patient Health Questionnaire-9 (PHQ-9) score ≥10 and endorsement of depressed mood and/or anhedonia. More than 90% of patients fulfilling this PHQ-9 criterion have major depression and/or dysthymia, and the remaining patients have clinically significant depression with substantial functional impairment [41,42]. Excluded were individuals with severe cognitive impairment, bipolar disorder, substance use disorder, schizophrenia, a pain-related disability claim currently under adjudication, plans to become pregnant in the next year, a life expectancy less than 12 months, or inability to speak English.

Participants in the active intervention group received six clinical contacts (baseline and 1, 3, 6, 9, and 12 weeks) with the study nurse care manager. Following a rational algorithmic approach to antidepressant selection and dosing, the study nurse care manager assessed antidepressant adherence, adverse effects, and depression response. Depression rather than pain response dictated antidepressant adjustments. Approximately two thirds of the study nurse care manager contacts with patients were by telephone. For those randomized to usual care, participants were informed that they had depressive symptoms and advised to seek advice from their primary care provider about treatment. There were no other attempts by study personnel to influence depression or pain management unless a psychiatric emergency (e.g., suicidal ideation) arose.

All baseline and 3-month outcome assessments were conducted by a research assistant blinded to treatment allocation and uninvolved in care management of subjects. This trial was approved by the Indiana University Institutional Review Board and was monitored by a local independent Data Safety Monitoring Committee.

Outcome Measures

The Graded Chronic Pain Scale (GCPS) at 3 months assessed pain intensity and interference with usual daily activities [43–45]. The GCPS pain intensity was calculated by averaging 0–10 ratings of “current pain” and “average” and “worst pain” in the past month. The GCPS activity interference was calculated by averaging 0–10 ratings of pain interference with “daily activities,”“work/housework activities,” and “recreational/social activities” in the past month. The GCPS pain intensity and activity interference mean scores were each multiplied by 10 to yield a score range of 0–100, with higher scores indicating worse outcome. The GCPS pain intensity and activity interference scores have good internal consistency, test–retest reliability, and validity [44,45]. Compared with the usual care group, the active intervention group in the original SCAMP study had significantly lower 3-month GCPS pain intensity (absolute between-group difference of −7.19; standardized effect size of 0.43) and GCPS activity interference (absolute between-group difference of −8.73; standardized effect size of 0.35) scores [46].

Baseline Covariates

Depression

The SCL-20 is a modified subscale of the Hopkins Symptom Checklist (SCL-90) that has demonstrated sensitivity to detect differences in depression severity change between treatment groups [41,47,48]. The 20 items are scored from 0 to 4 and averaged to provide a measure of overall severity from 0 to 4, with higher scores representing more severe depression.

Self Efficacy

Self-efficacy to manage symptoms of pain was assessed with the 6-item Arthritis Self-Efficacy Subscale. This subscale has good internal consistency and associated with measures of health status [49,50]. For each item, patients reported their degree of certainty to manage symptoms at the present time on a scale ranging from 1 (very uncertain) to 10 (very certain). The total score was calculated as the mean of the summed responses and ranged from 1 to 10, with higher scores indicating greater self-efficacy.

Kinesiophobia

The modified 10-item Tampa Scale for Kinesiophobia was used as a measure of fear of movement/(re)injury. Internal consistency and concurrent criterion validity have been established in different pain conditions [51–54]. Each item (e.g., I'm afraid that I might injure myself if I exercise) is scored on a 4-point Likert scale with scores ranging from 1 “strongly disagree” to 4 “strongly agree.” The total score can range from 10 to 40, with higher scores indicating a higher degree of kinesiophobia (i.e., greater fear that movement will exacerbate pain).

Pain Beliefs and Attitudes

The eight-item Survey of Pain Attitudes (SOPA), derived from the original 57-item SOPA, assesses a patient's attitudes and beliefs about pain. This eight-item SOPA, with its four subscales, is reliable and strongly associated with their parent subscales and also with measures of pain, psychological dysfunction, and disability [55–57]. The four subscales are as follows: 1) emotion (i.e., “There is a connection between my emotions and my pain level” and “Stress in my life increases the pain I feel”); 2) solicitude (i.e., “When I am hurting, I deserve to be treated with care and concern” and “When I hurt, I want my family to treat me better”); 3) medical cures (i.e., “I trust that doctors can cure my pain” and “I do not expect a medical cure for my pain”); and 4) pain control (i.e., “I have learned to control my pain” and “There is little I can do to ease my pain”). Study participants were asked to indicate how much they agreed with each item, using a scale of 0 = “this is very untrue for me” to 4 = “this is very true for me.” Reverse scoring was used for one item in Medical Cures (“I do not expect a medical cure for my pain”) and one in Pain Control (“There is little I can do to ease my pain”). Scores on each subscale were calculated as the mean of the summed responses and thus can range from 0 to 4, with higher scores indicating greater agreement with the belief.

The 12-item Pain Stages of Change questionnaire (PSOC) was used to provide a “snapshot” of readiness to change at baseline [58]. The PSOC was designed to operationalize readiness to adopt a self-management approach to chronic pain. The reliability and criterion, discriminant and predictive validity of the PSCO has now been replicated in several studies [58–61]. PSOC is comprised of four distinct scales with three items each. Each item is scored on a 5-point Likert scale with scores ranging from 1 “strongly disagree” to 5 “strongly agree.” Scores on each scale were calculated as the mean of the summed responses and thus can range from 1 to 5. High scores on the precontemplation scale characterize a person with little perceived responsibility for pain control and little interest in implementing behavioral changes. High contemplation scores represent a consideration of behavioral changes, along with an increasing awareness of personal responsibility for controlling pain. The action scale assesses the degree to which someone is actively involved in learning self-management strategies, whereas the maintenance scale addresses the extent to which these changes have been incorporated into daily life, along with a strong sense of personal responsibility for pain control.

Pain Characteristics and Demographics

Pain location (back pain vs leg pain), pain duration (number of years), treatment arm (optimized antidepressant treatment vs usual care) and baseline GCPS pain intensity and activity interference were the four clinical factors considered. For demographic variables, we collected gender, age, marital status (married vs not married), education (≤high school graduate vs >high school), employment status (employed vs not employed) and race (non-whites vs whites).

Statistical Analysis

Mean values and standard deviation, frequencies and percentages were calculated for continuous and categorical variables. The two dependent variables were the GCPS pain intensity and GCPS activity interference at the 3-month outcome assessment. Controlling for baseline GCPS pain intensity or GCPS activity interference, univariate linear regression was conducted to estimate the effects of baseline covariates on each of the dependent variables. Variables that reached a P value ≤ 0.2 were entered into the multivariate linear regression model. Interaction terms between treatment arm and each baseline variable with P value ≤ 0.05 were assessed for statistical significance.

Twenty-four (9.6%) of the 250 enrolled subjects were lost to follow-up at the 3-month assessment. Compared with the 226 remaining subjects, the 24 participants who were lost to follow-up did not differ in terms of gender, age, marital status, employment status, ethnic affiliation, pain location, baseline GCPS pain intensity and activity interference, and treatment arms.

Results

Baseline Characteristics

Overall, the mean age of the 250 participants was 55.5 years; 52.8% were women; 39.6% were non-white and 60.4% white. Fifty-two per cent reported their highest level of education to be high school graduate or less. Work status was 25.6% employed and 74.4% unemployed or unable to work or retired. The site of pain was the back in 60.4% of subjects and the hip or knee in 39.6%. The median duration of pain at baseline was 9 years.

In terms of depression severity, the mean SCL-20 score was 1.89 representing moderately severe depressive symptoms. For pain severity, the mean GCPS pain intensity and activity interference scores were 72.7 and 69.0, respectively. Both of these scores represent moderately severe pain.

Univariate Predictors of 3-month Pain Intensity and Activity Interference

Table 1 shows the univariate associations between baseline variables and pain intensity and activity interference, controlling for their baseline value. Current employment and assignment to the intervention arm (i.e., optimized antidepressant therapy) predicted lower 3-month GCPS pain intensity and activity interference, whereas greater baseline depression severity, and kinesiophobia (i.e., more fearful of movement) predicted worse pain outcomes. Non-white participants had less activity interference at follow-up, whereas those with back pain (vs leg pain) had greater activity interference. Age, gender, marital status, education, pain duration, self-efficacy, SOPA (pain attitudes), and PSOC (readiness to change) measures were not related with either of the pain outcome measures.

Table 1

Predictors of pain intensity and activity interference for patients with chronic musculoskeletal pain and co-morbid depression on bivariate analysis

Baseline Variables 3-month GCPS Pain Intensity
 
3-month GCPS Activity Interference
 
Beta Coefficient P-value Beta Coefficient P-value 
Demographics     
  Age (in years) −0.06 (0.10) 0.5 −0.10 (0.13) 0.4 
  Female (vs male) −0.45 (2.37) 0.8 0.58 (2.99) 0.8 
Marital status     
  Married (vs unmarried) −0.18 (2.39) 0.9 0.56 (3.07) 0.8 
Education     
  ≤high school (vs >high school) −0.85 (2.48) 0.7 0.14 (3.14) 0.9 
  Non-whites (vs whites) −0.06 (2.38) 1.0 −5.75 (2.99) 0.06 
  Employed (vs not employed) −6.78 (2.28) 0.003 −13.15 (2.92) <0.0001 
Treatment arm     
  Stepped care (vs usual care) −8.34 (2.21) 0.0002 −8.31 (2.89) 0.004 
  Severity of depression 3.57 (1.85) 0.05 5.09 (2.34) 0.03 
Pain location     
  Back pain (vs leg pain) 1.51 (2.32) 0.5 6.28 (2.97) 0.03 
  Pain duration (in years) −0.01 (0.08) 0.9 0.08 (0.11) 0.4 
Baseline pain     
  GCPS pain intensity 0.58 (0.06) <0.0001 — — 
  GCPS activity interference — — 0.66 (0.05) <0.0001 
Pain-related cognitions     
  ASES     
  Self-efficacy −0.79 (0.57) 0.16 −0.56 (0.81) 0.4 
  TSK     
  Kinesiophobia 0.56 (0.21) 0.007 0.68 (0.29) 0.01 
  SOPA     
  Pain control −0.66 (1.15) 0.5 2.37 (1.45) 0.1 
  Emotion 0.45 (0.99) 0.6 2.09 (1.29) 0.1 
  Solicitude 0.42 (0.93) 0.6 0.64 (1.21) 0.5 
  Medical cure 0.18 (0.98) 0.8 −1.20 (1.24) 0.3 
  PSOC     
  Pre-contemplation 1.80 (1.34) 0.1 2.66 (1.72) 0.1 
  Contemplation −0.57 (1.54) 0.7 1.45 (1.99) 0.4 
  Action −1.08 (1.40) 0.4 1.93 (1.76) 0.3 
  Maintenance −0.52 (1.40) 0.7 2.08 (1.85) 0.3 
Baseline Variables 3-month GCPS Pain Intensity
 
3-month GCPS Activity Interference
 
Beta Coefficient P-value Beta Coefficient P-value 
Demographics     
  Age (in years) −0.06 (0.10) 0.5 −0.10 (0.13) 0.4 
  Female (vs male) −0.45 (2.37) 0.8 0.58 (2.99) 0.8 
Marital status     
  Married (vs unmarried) −0.18 (2.39) 0.9 0.56 (3.07) 0.8 
Education     
  ≤high school (vs >high school) −0.85 (2.48) 0.7 0.14 (3.14) 0.9 
  Non-whites (vs whites) −0.06 (2.38) 1.0 −5.75 (2.99) 0.06 
  Employed (vs not employed) −6.78 (2.28) 0.003 −13.15 (2.92) <0.0001 
Treatment arm     
  Stepped care (vs usual care) −8.34 (2.21) 0.0002 −8.31 (2.89) 0.004 
  Severity of depression 3.57 (1.85) 0.05 5.09 (2.34) 0.03 
Pain location     
  Back pain (vs leg pain) 1.51 (2.32) 0.5 6.28 (2.97) 0.03 
  Pain duration (in years) −0.01 (0.08) 0.9 0.08 (0.11) 0.4 
Baseline pain     
  GCPS pain intensity 0.58 (0.06) <0.0001 — — 
  GCPS activity interference — — 0.66 (0.05) <0.0001 
Pain-related cognitions     
  ASES     
  Self-efficacy −0.79 (0.57) 0.16 −0.56 (0.81) 0.4 
  TSK     
  Kinesiophobia 0.56 (0.21) 0.007 0.68 (0.29) 0.01 
  SOPA     
  Pain control −0.66 (1.15) 0.5 2.37 (1.45) 0.1 
  Emotion 0.45 (0.99) 0.6 2.09 (1.29) 0.1 
  Solicitude 0.42 (0.93) 0.6 0.64 (1.21) 0.5 
  Medical cure 0.18 (0.98) 0.8 −1.20 (1.24) 0.3 
  PSOC     
  Pre-contemplation 1.80 (1.34) 0.1 2.66 (1.72) 0.1 
  Contemplation −0.57 (1.54) 0.7 1.45 (1.99) 0.4 
  Action −1.08 (1.40) 0.4 1.93 (1.76) 0.3 
  Maintenance −0.52 (1.40) 0.7 2.08 (1.85) 0.3 

ASES = Arthritis Self-Efficacy Scale; TSK = Tampa Scale for Kinesiophobia; SOPA = Survey of Pain Attitudes; PSOC = Pain Stages of Change.

Multivariate Predictors of 3-month Pain Intensity and Activity Interference

To identify factors that were independent predictors of 3-month pain intensity and activity interference, variables with a P ≤ 0.2 in the univariate models were entered into a multivariate linear regression model. Table 2 shows that employment status, treatment arm, and kinesiophobia remained significant predictors of GCPS pain intensity and activity interference. Specifically, current employment and assignment to the intervention arm predicted lower 3-month GCPS pain intensity and activity interference, and greater fear that movement will exacerbate pain predicted worse pain outcomes. High baseline SOPA-pain control predicted higher (worse) GCPS activity interference at 3 months, whereas being non-white predicted lower activity interference. Severity of depression, location and duration of pain, self-efficacy, and other SOPA and PSOC measures were not significant predictors. Finally, there were no significant interactions between treatment arm and any of the significant predictors in the model.

Table 2

Multivariate predictors of pain intensity and activity interference in patients with chronic musculoskeletal pain and co-morbid depression

Baseline Variables 3-month GCPS Pain Intensity
 
3-month GCPS Activity Interference
 
Beta Coefficient P-value B Coefficient P-value 
Demographics     
  Non-whites (vs whites) — — −5.83 (2.85) 0.04 
  Employed (vs not employed) −5.61 (2.24) 0.01 −13.38 (2.85) <0.0001 
Treatment arm     
  Stepped care (vs usual care) −3.73 (2.19) 0.0005 −6.40 (2.70) 0.01 
Severity of depression 2.92 (1.85) 0.1 3.77 (2.39) 0.1 
Pain location     
  Back pain (vs leg pain) — — 4.69 (2.76) 0.09 
Baseline pain     
  GCPS pain intensity 0.48 (0.07) <0.0001 — — 
  GCPS activity interference — — 0.51 (0.07) <0.0001 
Pain-related cognitions     
  ASES     
  Self-efficacy −0.12 (0.58) 0.8 — — 
  TSK     
  Kinesiophobia 0.46 (0.23) 0.04 0.57 (0.29) 0.05 
  SOPA     
  Pain control — — 3.64 (1.48) 0.01 
  Emotion — — 1.32 (1.28) 0.3 
  PSOC     
  Pre-contemplation 0.10 (1.42) 0.9 2.85 (1.76) 0.1 
Baseline Variables 3-month GCPS Pain Intensity
 
3-month GCPS Activity Interference
 
Beta Coefficient P-value B Coefficient P-value 
Demographics     
  Non-whites (vs whites) — — −5.83 (2.85) 0.04 
  Employed (vs not employed) −5.61 (2.24) 0.01 −13.38 (2.85) <0.0001 
Treatment arm     
  Stepped care (vs usual care) −3.73 (2.19) 0.0005 −6.40 (2.70) 0.01 
Severity of depression 2.92 (1.85) 0.1 3.77 (2.39) 0.1 
Pain location     
  Back pain (vs leg pain) — — 4.69 (2.76) 0.09 
Baseline pain     
  GCPS pain intensity 0.48 (0.07) <0.0001 — — 
  GCPS activity interference — — 0.51 (0.07) <0.0001 
Pain-related cognitions     
  ASES     
  Self-efficacy −0.12 (0.58) 0.8 — — 
  TSK     
  Kinesiophobia 0.46 (0.23) 0.04 0.57 (0.29) 0.05 
  SOPA     
  Pain control — — 3.64 (1.48) 0.01 
  Emotion — — 1.32 (1.28) 0.3 
  PSOC     
  Pre-contemplation 0.10 (1.42) 0.9 2.85 (1.76) 0.1 

ASES = Arthritis Self-Efficacy Scale; TSK = Tampa Scale for Kinesiophobia; SOPA = Survey of Pain Attitudes; PSOC = Pain Stages of Change.

Discussion

We have previously reported the effectiveness of 12-weeks of optimized antidepressant therapy in improving both depression and pain among patients with moderately severe CMP and co-morbid depression [46]. Our multivariate modeling of 3-month pain outcomes reveals several important findings. First, optimized antidepressant therapy proved beneficial for chronic musculoskeletal pain co-morbid with depression, controlling for age, gender, severity of baseline depression, duration or location of the pain, self-efficacy, and pain stage of change. Second, several nonmodifiable and modifiable factors predicted short-term outcomes. Specifically, participants who were unemployed as well as those with greater baseline pain and higher kinesiophobia (i.e., fear of movement) had worse pain intensity and activity interference at 3 months. Also, white race and those with a stronger belief that pain is under their own control had worse activity interference at 3 months.

The significant benefits of the intervention that were demonstrated in our models is likely a combination of both a specific effect (i.e., optimized antidepressant therapy) as well as a nonspecific “attention” effect of care manager contacts. This is true of all effectiveness trials where a care management intervention is compared with a treatment-as-usual control group. However, both the specific and nonspecific effects of the intervention are accounted for by entering treatment group (intervention vs usual care) in our multivariable model. Thus, controlling for this “bundled” intervention effect allows us to draw valid conclusions about the independent effects of other predictors.

The predictive value of baseline pain intensity or interference shown in this study has been reported by other authors [27,31,62,63]. Not only is higher pain intensity related with worse outcome [27,63], but higher interference of pain with activities is also associated with reduced treatment success [27,62,64]. One implication is that patients with depression co-morbid with moderate to severe pain may require co-management with antidepressants and optimized analgesics or other nonpharmacological treatments directed at pain. Regarding work-related parameters, prior studies have also demonstrated that employment status or the ability to work, predicted better outcome [32,63,65]. Our finding regarding the beneficial relationship between employment and better pain outcomes is not likely due to preselecting patients who were less disabled because: 1) our entry criteria did not include a specific GCPS pain interference cutoff score; 2) a greater majority (74.4%) of our subjects were unemployed or unable to work; and 3) the mean GCPS activity interference score was 69, which indicates that the cohort had a moderate to severe level of physical impairment at study entry. Certainly, our findings combined with previous research suggest that maximizing employability and work rehabilitation should be a priority.

Our finding of the association between race/ethnicity and pain-related activity interference is consistent with some previous studies but differs from others [66–71]. Disparate results related to race/ethnicity may partly depend on differences in study settings, sample characteristics, and different pain measures assessed across studies. For example, Edwards et al. previously reported no significant ethnic group differences on the Multidimensional Pain Inventory measures of pain-related disability, while administration of the Oswestry Disability Questionnaire produced significantly greater ratings of disability among African Americans [66]. However, a review of the literature suggests that African Americans with chronic pain report more pain severity and disability due to pain than non-Hispanic Whites with chronic pain [72].

The result that baseline depression severity was not predictive of pain-related outcomes was surprising given the finding that depression in patients with pain is associated with more pain complaints and greater impairment [18]. Also, observational studies have suggested that the presence of depression adversely affects pain outcomes [18,73]. One possible explanation for this finding is that the predictive value of baseline depression severity was reduced or washed out by including in the model treatment arm which itself was highly predictive. Of note, baseline depression severity was predictive at the univariate level. Our study suggests that optimized antidepressant therapy may be effective in improving pain outcomes in patients with a spectrum of depression severity.

Psychosocial factors play a major role in the maintenance and progression of chronic disability in musculoskeletal disorders. One variable that has attracted increasing theoretical and empirical interest is pain-related fear [74,75]. Fear of movement and injury (or kinesiophobia) has been shown to be a robust predictor of disability in patients with acute and chronic low back pain [51–53,76], fibromyalgia [52], neck pain [77], and osteoarthritis [54]. In our study, we confirmed the association of fear of movement and activity interference—a surrogate marker of disability. Additionally, our analysis showed that kinesiophobia is a predictor of pain intensity. Thus, fear of movement may be an important target for psychologically-based interventions.

Previous research has often shown that the stronger the belief in one's personal control over pain (SOPA-pain control), the better the outcome [78–80]. Surprisingly, we found the opposite, that is, the stronger the belief in control over pain at baseline, the greater the pain activity interference was at follow-up. We initially thought that subjects got confused with the reverse worded item, “There is little I can do to ease my pain” and the response options of “very untrue” to “very true.” However, our correlation analysis table (data not shown) demonstrated that the relationships between SOPA-pain control with depression, pain intensity, activity interference and other measures of disability were in the expected direction confirming previous reports. One potential explanation for our unexpected finding is that all of the patients in our trial had co-morbid depression, which may have influenced the degree or directionality of pain beliefs as a predictor. Second, because the SOPA questionnaire was interviewer-administered, the participants may have provided answers that would be viewed as more socially acceptable, which in this case, a more adaptive style of coping with pain. If such was the case, high SOPA-pain control at baseline would be associated with greater activity interference at follow-up. A third possibility may be regression to the mean. Specifically, subjects scoring high on SOPA-pain control at baseline may be more likely to regress to the mean at follow-up. And if lower SOPA-pain control at 3 months is associated with greater activity interference at this time point, high SOPA-pain control at study entry might predict greater activity interference at follow-up. Unfortunately, we did not collect SOPA-pain control at the 3-month time point; hence, we cannot verify the validity of this hypothesis. Finally, recent data have suggested the apparently “unhelpful” role of pain control in relation to future physical functioning. In a prospective investigation of acceptance and control-oriented coping among chronic pain sufferers, McCraken et al. have reported that attempting to control pain was associated with relatively worse functioning over time [81]. Certainly, using pain medication to control pain can be quite adaptive response on some occasions. On other occasions it may serve as part of a pattern of avoidance of pain, a pattern that could constrict daily functioning. In other words, attempts to control pain, which on the surface might look like adaptive ways of coping, appear less useful than responses that engender more acceptance of pain. At least 15 laboratory and clinical studies make the growing case for the role of acceptance in improved functioning of people with chronic pain [82].

The pain stages of change construct is a psychologic construct that maps the process of behavior change [60]. In our study, the pain stages of change did not predict 3-month outcomes. Although the pain stages of change model is intuitively plausible and has the potential to increase the effectiveness of the practitioners' counseling, a recent review of eight studies has concluded that the pain stages of change model is in its infancy and its predictive validity should be proven first before it is applied clinically [83].

Several potential limitations of the study warrant consideration. First, the use of self-reported data may introduce potential biases. However, self-reported pain-related fear has not only been associated with self-reported outcomes but also to physical performance [84–87]. For example, when faced with performing physical tasks such as lifting an arm weight or engaging in trunk extension and flexion exercises, patients scoring high on pain-related fear perform these tasks significantly slower [84]. Second, 3 months is a relatively short follow-up period for painful syndromes that often last many years. However, most of our findings are consistent with prior research in chronic pain with longer follow-up periods.

Conclusion

To our knowledge, this is the first study to identify prognostic factors of pain intensity and activity interference for patients with both CMP and depression. In addition to employment status and race serving as predictors of pain-related outcomes, a clinically relevant finding of our study is that kinesiophobia predicts greater pain intensity and activity interference. One wonders whether the combined use of medications to reduce pain, and psychological-based therapy to reduce kinesiophobia would enhance treatment outcomes. Given that mono-therapies (pharmacologic or psychologically based therapies) produce only modest reductions in pain, it may be time to consider more careful study of treatment components that are designed to complement each other such as exercise, cognitive behavioral therapy, pain self-management programs, and other types of nonpharmacological interventions [88]. Given the strength of “employment” status we may also want to consider interventions in the workplace or incorporate a vocational rehabilitation component.

References

1
Kroenke
K
.
Patients presenting with somatic complaints: Epidemiology, psychiatric comorbidity and management
.
Int J Methods Psychiatr Res
 
2003
;
12
(
1
):
34
43
.
2
Kroenke
K
Mangelsdorff
AD
.
Common symptoms in ambulatory care: Incidence, evaluation therapy, and outcome
.
Am J Med
 
1989
;
86
(
3
):
262
6
.
3
Kroenke
K
Arrington
ME
Mangelsdorff
AD
.
The prevalence of symptoms in medical outpatients and the adequacy of therapy
.
Arch Intern Med
 
1990
;
150
(
8
):
1685
9
.
4
Kroenke
K
Price
RK
.
Symptoms in the community. Prevalence, classification, and psychiatric comorbidity
.
Arch Intern Med
 
1993
;
153
(
21
):
2474
80
.
5
Verhaak
PF
Kerssens
JJ
Dekker
J
Sorbi
MJ
Bensing
JM
.
Prevalence of chronic benign pain disorder among adults: A review of the literature
.
Pain
 
1998
;
77
(
3
):
231
9
.
6
Elliott
AM
Smith
BH
Penny
KI
Smith
WC
Chambers
WA
.
The epidemiology of chronic pain in the community
.
Lancet
 
1999
;
354
(
9186
):
1248
52
.
7
Schappert
SM
.
National Ambulatory Medical Care Survey: 1989 summary
.
Vital Health Stat 13
 
1992
;
110
:
1
80
.
8
NIH.PA 03-152. Biobehavioral Pain Research. 7-11-2003. Ref Type: Pamphlet.
9
Magni
G
Caldieron
C
Rigatti-Luchini
S
Merskey
H
.
Chronic musculoskeletal pain and depressive symptoms in the general population. An analysis of the 1st National Health and Nutrition Examination Survey data
.
Pain
 
1990
;
43
(
3
):
299
307
.
10
Magni
G
Marchetti
M
Moreschi
C
Merskey
H
Luchini
SR
.
Chronic musculoskeletal pain and depressive symptoms in the National Health and Nutrition Examination. I. Epidemiologic follow-up study
.
Pain
 
1993
;
53
(
2
):
163
8
.
11
Carroll
LJ
Cassidy
JD
Cote
P
.
The Saskatchewan health and back pain survey: The prevalence and factors associated with depressive symptomatology in Saskatchewan adults
.
Can J Public Health
 
2000
;
91
(
6
):
459
64
.
12
Ang
DC
Kroenke
K
.
Depression in osteoarthritis
. In:
Brandt
KB
Doherty
MD
, eds.
Osteoarthritis
 ,
2nd edition
.
Oxford
:
Oxford University Press
;
2003
:
333
8
.
13
Livingston
G
Watkin
V
Milne
B
Manela
MV
Katona
C
.
Who becomes depressed? The Islington community study of older people
.
J Affect Disord
 
2000
;
58
(
2
):
125
33
.
14
Ohayon
MM
Schatzberg
AF
.
Using chronic pain to predict depressive morbidity in the general population
.
Arch Gen Psychiatry
 
2003
;
60
(
1
):
39
47
.
15
Patten
SB
.
Long-term medical conditions and major depression in a Canadian population study at waves 1 and 2
.
J Affect Disord
 
2001
;
63
(
1–3
):
35
41
.
16
Rethelyi
JM
Berghammer
R
Kopp
MS
.
Comorbidity of pain-associated disability and depressive symptoms in connection with sociodemographic variables: Results from a cross-sectional epidemiological survey in Hungary
.
Pain
 
2001
;
93
(
2
):
115
21
.
17
Gureje
O
Von
KM
Simon
GE
Gater
R
.
Persistent pain and well-being: A World Health Organization study in primary care
.
JAMA
 
1998
;
280
(
2
):
147
51
.
18
Bair
MJ
Robinson
RL
Katon
W
Kroenke
K
.
Depression and pain comorbidity: A literature review
.
Arch Intern Med
 
2003
;
163
(
20
):
2433
45
.
19
Pincus
T
Burton
AK
Vogel
S
Field
AP
.
A systematic review of psychological factors as predictors of chronicity/disability in prospective cohorts of low back pain
.
Spine
 
2002
;
27
(
5
):
E109
20
.
20
Psychological Approaches to Pain Management
.
A Practitioner's Handbook
 ,
2nd edition
.
New York
:
The Guilford Press
;
2002
.
21
Robinson
CL
.
Relieving pain in the elderly
.
Health Prog
 
2007
;
88
(
1
):
48
53
,
70
.
22
Kerns
RD
Haythornthwaite
JA
.
Depression among chronic pain patients: Cognitive-behavioral analysis and effect on rehabilitation outcome
.
J Consult Clin Psychol
 
1988
;
56
(
6
):
870
6
.
23
Turk
DC
.
Clinical effectiveness and cost-effectiveness of treatments for patients with chronic pain
.
Clin J Pain
 
2002
;
18
(
6
):
355
65
.
24
Curatolo
M
Bogduk
N
.
Pharmacologic pain treatment of musculoskeletal disorders: Current perspectives and future prospects
.
Clin J Pain
 
2001
;
17
(
1
):
25
32
.
25
North
RB
Ewend
MG
Lawton
MT
Kidd
DH
Piantadosi
S
.
Failed back surgery syndrome: 5-year follow-up after spinal cord stimulator implantation
.
Neurosurgery
 
1991
;
28
(
5
):
692
9
.
26
Turner
JA
Loeser
JD
Bell
KG
.
Spinal cord stimulation for chronic low back pain: A systematic literature synthesis
.
Neurosurgery
 
1995
;
37
(
6
):
1088
95
.
27
Van Der
HM
Vollenbroek-Hutten
MM
Ijzerman
MJ
.
A systematic review of sociodemographic, physical, and psychological predictors of multidisciplinary rehabilitation-or, back school treatment outcome in patients with chronic low back pain
.
Spine
 
2005
;
30
(
7
):
813
25
.
28
Carragee
EJ
Alamin
TF
Miller
JL
Carragee
JM
.
Discographic MRI and psychosocial determinants of low back pain disability and remission: A prospective study in subjects with benign persistent back pain
.
Spine J
 
2005
;
5
(
1
):
24
35
.
29
Michaelson
P
Sjolander
P
Johansson
H
.
Factors predicting pain reduction in chronic back and neck pain after multimodal treatment
.
Clin J Pain
 
2004
;
20
(
6
):
447
54
.
30
Petersen
T
Larsen
K
Jacobsen
S
.
One-year follow-up comparison of the effectiveness of McKenzie treatment and strengthening training for patients with chronic low back pain: Outcome and prognostic factors
.
Spine
 
2007
;
32
(
26
):
2948
56
.
31
Bendix
AF
Bendix
T
Haestrup
C
.
Can it be predicted which patients with chronic low back pain should be offered tertiary rehabilitation in a functional restoration program? A search for demographic, socioeconomic, and physical predictors
.
Spine
 
1998
;
23
(
16
):
1775
83
.
32
McCracken
LM
Turk
DC
.
Behavioral and cognitive-behavioral treatment for chronic pain: Outcome, predictors of outcome, and treatment process
.
Spine
 
2002
;
27
(
22
):
2564
73
.
33
Dworkin
RH
Gitlin
MJ
.
Clinical aspects of depression in chronic pain patients
.
Clin J Pain
 
1991
;
7
(
2
):
79
94
.
34
Dworkin
RH
Richlin
DM
Handlin
DS
Brand
L
.
Predicting treatment response in depressed and non-depressed chronic pain patients
.
Pain
 
1986
;
24
(
3
):
343
53
.
35
Cherkin
DC
Deyo
RA
Street
JH
Barlow
W
.
Predicting poor outcomes for back pain seen in primary care using patients' own criteria
.
Spine
 
1996
;
21
(
24
):
2900
7
.
36
Painter
JR
Seres
JL
Newman
RI
.
Assessing benefits of the pain center: Why some patients regress
.
Pain
 
1980
;
8
(
1
):
101
13
.
37
Gallagher
RM
.
Biopsychosocial pain medicine and mind-brain-body science
.
Phys Med Rehabil Clin North Am
 
2004
;
15
(
4
):
855
82
,
vii
.
38
Kroenke
K
Bair
M
Damush
T
et al
Care for Affective Disorders and Musculoskeletal Pain (SCAMP) study: Design and practical implications of an intervention for comorbid pain and depression
.
Gen Hosp Psychiatry
 
2007
;
29
(
6
):
506
17
.
39
Williams
LS
Jones
WJ
Shen
J
et al
Prevalence and impact of depression and pain in neurology outpatients
.
J Neurol Neurosurg Psychiatry
 
2003
;
74
(
11
):
1587
9
.
40
Cleeland
CS
Gonin
R
Hatfield
AK
et al
Pain and its treatment in outpatients with metastatic cancer
.
N Engl J Med
 
1994
;
330
(
9
):
592
6
.
41
Kroenke
K
West
SL
Swindle
R
et al
Similar effectiveness of paroxetine, fluoxetine, and sertraline in primary care: A randomized trial
.
JAMA
 
2001
;
286
(
23
):
2947
55
.
42
Kroenke
K
Spitzer
RL
Williams
JB
.
The PHQ-9: Validity of a brief depression severity measure
.
J Gen Intern Med
 
2001
;
16
(
9
):
606
13
.
43
Von
KM
Ormel
J
Keefe
FJ
Dworkin
SF
.
Grading the severity of chronic pain
.
Pain
 
1992
;
50
(
2
):
133
49
.
44
Dworkin
SF
Von Korff
M
Whitney
CW
et al
Measurement of characteristic pain intensity in field research
.
Pain
  Suppl.
1990
;
5
:
S290
.
45
Von Korff
M
.
Epidemiological and survey methods: Assessment of chronic pain
. In:
Turk
D
Melzack
R
, eds.
Handbook of Pain Assessment
 .
New York
:
The Guilford Press
;
2001
:
603
18
.
46
Bair
MJ
Damush
TM
Wu
J
Hoke
S
Sutherland
JM
Kroenke
K
.
Optimizing antidepressant therapy in primary care patients with musculoskeletal pain and depression: A randomized controlled trial
.
J Gen Intern Med
 
2008
;
23
(
suppl 2
):
353
4
.
47
Katon
W
Von
KM
Lin
E
et al
Stepped collaborative care for primary care patients with persistent symptoms of depression: A randomized trial
.
Arch Gen Psychiatry
 
1999
;
56
(
12
):
1109
15
.
48
Unutzer
J
Katon
W
Callahan
CM
et al
Collaborative care management of late-life depression in the primary care setting: A randomized controlled trial
.
JAMA
 
2002
;
288
(
22
):
2836
45
.
49
Damush
TM
Weinberger
M
Perkins
SM
et al
Randomized trial of a self-management program for primary care patients with acute low back pain: Short-term effects
.
Arthritis Rheum
 
2003
;
49
(
2
):
179
86
.
50
Lorig
K
Chastain
RL
Ung
E
Shoor
S
Holman
HR
.
Development and evaluation of a scale to measure perceived self-efficacy in people with arthritis
.
Arthritis Rheum
 
1989
;
32
(
1
):
37
44
.
51
Swinkels-Meewisse
IE
Roelofs
J
Verbeek
AL
Oostendorp
RA
Vlaeyen
JW
.
Fear of movement/(re)injury, disability and participation in acute low back pain
.
Pain
 
2003
;
105
(
1–2
):
371
9
.
52
Roelofs
J
Goubert
L
Peters
ML
Vlaeyen
JW
Crombez
G
.
The Tampa scale for Kinesiophobia: Further examination of psychometric properties in patients with chronic low back pain and fibromyalgia
.
Eur J Pain
 
2004
;
8
(
5
):
495
502
.
53
Goubert
L
Crombez
G
Van
DS
et al
Confirmatory factor analysis of the Tampa Scale for Kinesiophobia: Invariant two-factor model across low back pain patients and fibromyalgia patients
.
Clin J Pain
 
2004
;
20
(
2
):
103
10
.
54
Heuts
PH
Vlaeyen
JW
Roelofs
J
et al
Pain-related fear and daily functioning in patients with osteoarthritis
.
Pain
 
2004
;
110
(
1–2
):
228
35
.
55
Tait
RC
Chibnall
JT
.
Development of a brief version of the Survey of Pain Attitudes
.
Pain
 
1997
;
70
(
2–3
):
229
35
.
56
Tan
G
Nguyen
Q
Cardin
SA
Jensen
MP
.
Validating the use of two-item measures of pain beliefs and coping strategies for a veteran population
.
J Pain
 
2006
;
7
(
4
):
252
60
.
57
Jensen
MP
Keefe
FJ
Lefebvre
JC
Romano
JM
Turner
JA
.
One- and two-item measures of pain beliefs and coping strategies
.
Pain
 
2003
;
104
(
3
):
453
69
.
58
Kerns
RD
Rosenberg
R
.
Predicting responses to self-management treatments for chronic pain: Application of the pain stages of change model
.
Pain
 
2000
;
84
(
1
):
49
55
.
59
Dijkstra
A
Vlaeyen
JW
Rijnen
H
Nielson
W
.
Readiness to adopt the self-management approach to cope with chronic pain in fibromyalgic patients
.
Pain
 
2001
;
90
(
1–2
):
37
45
.
60
Jensen
MP
Nielson
WR
Romano
JM
Hill
ML
Turner
JA
.
Further evaluation of the pain stages of change questionnaire: Is the transtheoretical model of change useful for patients with chronic pain?
Pain
 
2000
;
86
(
3
):
255
64
.
61
Biller
N
Arnstein
P
Caudill
MA
Federman
CW
Guberman
C
.
Predicting completion of a cognitive-behavioral pain management program by initial measures of a chronic pain patient's readiness for change
.
Clin J Pain
 
2000
;
16
(
4
):
352
9
.
62
Feuerstein
M
Beattie
P
.
Biobehavioral factors affecting pain and disability in low back pain: Mechanisms and assessment
.
Phys Ther
 
1995
;
75
(
4
):
267
80
.
63
Hasenbring
M
.
Predictors of efficacy in treatment of chronic low back pain
.
Curr Opin Anaesthesiol
 
1998
;
11
(
5
):
553
8
.
64
Talo
S
Puukka
P
Rytokoski
U
Ronnemaa
T
Kallio
V
.
Can treatment outcome of chronic low back pain be predicted? Psychological disease consequences clarifying the issue
.
Clin J Pain
 
1994
;
10
(
2
):
107
21
.
65
Walsh
NE
Dumitru
D
.
The influence of compensation on recovery from low back pain
.
Occup Med
 
1988
;
3
(
1
):
109
21
.
66
Edwards
RR
Doleys
DM
Fillingim
RB
Lowery
D
.
Ethnic differences in pain tolerance: Clinical implications in a chronic pain population
.
Psychosom Med
 
2001
;
63
(
2
):
316
23
.
67
Green
CR
Baker
TA
Sato
Y
Washington
TL
Smith
EM
.
Race and chronic pain: A comparative study of young black and white Americans presenting for management
.
J Pain
 
2003
;
4
(
4
):
176
83
.
68
Green
CR
Baker
TA
Smith
EM
Sato
Y
.
The effect of race in older adults presenting for chronic pain management: A comparative study of black and white Americans
.
J Pain
 
2003
;
4
(
2
):
82
90
.
69
McCracken
LM
Matthews
AK
Tang
TS
Cuba
SL
.
A comparison of blacks and whites seeking treatment for chronic pain
.
Clin J Pain
 
2001
;
17
(
3
):
249
55
.
70
Novy
DM
Nelson
DV
Hetzel
RD
Squitieri
P
Kennington
M
.
Coping with chronic pain: Sources of intrinsic and contextual variability
.
J Behav Med
 
1998
;
21
(
1
):
19
34
.
71
Riley
JL
III
Wade
JB
Myers
CD
Sheffield
D
Papas
RK
Price
DD
.
Racial/ethnic differences in the experience of chronic pain
.
Pain
 
2002
;
100
(
3
):
291
8
.
72
Green
CR
Anderson
KO
Baker
TA
et al
The unequal burden of pain: Confronting racial and ethnic disparities in pain
.
Pain Med
 
2003
;
4
(
3
):
277
94
.
73
Williams
LS
Jones
WJ
Shen
J
Robinson
RL
Kroenke
K
.
Outcomes of newly referred neurology outpatients with depression and pain
.
Neurology
 
2004
;
63
(
4
):
674
7
.
74
Leeuw
M
Goossens
ME
Linton
SJ
et al
The fear-avoidance model of musculoskeletal pain: Current state of scientific evidence
.
J Behav Med
 
2007
;
30
(
1
):
77
94
.
75
Vlaeyen
JW
Crombez
G
.
Fear of movement/(re)injury, avoidance and pain disability in chronic low back pain patients
.
Man Ther
 
1999
;
4
(
4
):
187
95
.
76
Crombez
G
Vlaeyen
JW
Heuts
PH
Lysens
R
.
Pain-related fear is more disabling than pain itself: Evidence on the role of pain-related fear in chronic back pain disability
.
Pain
 
1999
;
80
(
1–2
):
329
39
.
77
Nederhand
MJ
Ijzerman
MJ
Hermens
HJ
Turk
DC
Zilvold
G
.
Predictive value of fear avoidance in developing chronic neck pain disability: Consequences for clinical decision making
.
Arch Phys Med Rehabil
 
2004
;
85
(
3
):
496
501
.
78
Jensen
MP
Turner
JA
Romano
JM
Lawler
BK
.
Relationship of pain-specific beliefs to chronic pain adjustment
.
Pain
 
1994
;
57
(
3
):
301
9
.
79
Turner
JA
Jensen
MP
Romano
JM
.
Do beliefs, coping, and catastrophizing independently predict functioning in patients with chronic pain?
Pain
 
2000
;
85
(
1–2
):
115
25
.
80
Nielson
WR
Jensen
MP
.
Relationship between changes in coping and treatment outcome in patients with Fibromyalgia Syndrome
.
Pain
 
2004
;
109
(
3
):
233
41
.
81
McCracken
LM
Vowles
KE
Gauntlett-Gilbert
J
.
A prospective investigation of acceptance and control-oriented coping with chronic pain
.
J Behav Med
 
2007
;
30
(
4
):
339
49
.
82
McCracken
LM
Vowles
KE
.
Acceptance of chronic pain
.
Curr Pain Headache Rep
 
2006
;
10
(
2
):
90
4
.
83
Dijkstra
A
.
The validity of the stages of change model in the adoption of the self-management approach in chronic pain
.
Clin J Pain
 
2005
;
21
(
1
):
27
37
.
84
Vlaeyen
JW
Kole-Snijders
AM
Boeren
RG
Van
EH
.
Fear of movement/(re)injury in chronic low back pain and its relation to behavioral performance
.
Pain
 
1995
;
62
(
3
):
363
72
.
85
Pfingsten
M
Leibing
E
Harter
W
et al
Fear-avoidance behavior and anticipation of pain in patients with chronic low back pain: A randomized controlled study
.
Pain Med
 
2001
;
2
(
4
):
259
66
.
86
Thomas
JS
France
CR
Lavender
SA
Johnson
MR
.
Effects of fear of movement on spine velocity and acceleration after recovery from low back pain
.
Spine
 
2008
;
33
(
5
):
564
70
.
87
Thomas
JS
France
CR
.
Pain-related fear is associated with avoidance of spinal motion during recovery from low back pain
.
Spine
 
2007
;
32
(
16
):
E460
6
.
88
Kroenke
K
Krebs
EE
Bair
MJ
.
Treatment of chronic pain syndromes
. In:
Schatzberg
AB
Nemeroff
CB
, eds.
Textbook of Psychopharmacology
 ,
4th edition
.
Washington, DC
:
American Psychiatric Publishing
;
2009
:
1441
74
.
Funding/Support: This study was supported by a grant from the National Institute of Mental Health to Dr. Kroenke (MH-071268).