Moderators and Nonspecific Predictors of Treatment Benefits in a Randomized Trial of Mindfulness-Based Stress Reduction vs. Cognitive-Behavioral Therapy vs. Usual Care for Chronic Low Back Pain

Jessica A Chen; Melissa L Anderson; Daniel C Cherkin; Benjamin H Balderson; Andrea J Cook; Karen J Sherman; Judith A Turner

doi:10.1016/j.jpain.2022.09.014

. Author manuscript; available in PMC: 2024 Feb 1.

Published in final edited form as: J Pain. 2022 Sep 28;24(2):282–303. doi: 10.1016/j.jpain.2022.09.014

Moderators and Nonspecific Predictors of Treatment Benefits in a Randomized Trial of Mindfulness-Based Stress Reduction vs. Cognitive-Behavioral Therapy vs. Usual Care for Chronic Low Back Pain

Jessica A Chen ^a,^b,^*, Melissa L Anderson ^c, Daniel C Cherkin ^c, Benjamin H Balderson ^c, Andrea J Cook ^c, Karen J Sherman ^c, Judith A Turner ^a,^d

PMCID: PMC9898119 NIHMSID: NIHMS1839124 PMID: 36180008

Abstract

Both mindfulness-based stress reduction (MBSR) and cognitive-behavioral therapy (CBT) are effective for chronic low back pain (CLBP), but little is known regarding who might benefit more from one than the other. Using data from a randomized trial comparing MBSR, CBT, and usual care (UC) for adults aged 20–70 years with CLBP (N = 297), we examined baseline characteristics that moderated treatment effects or were associated with improvement regardless of treatment. Outcomes included 8-week function (modified Roland Disability Questionnaire), pain bothersomeness (0–10 numerical rating scale), and depression (Patient Health Questionnaire-8). There were differences in the effects of CBT vs. MBSR on pain based on participant gender (P = 0.03) and baseline depressive symptoms (P = 0.01), but the only statistically significant moderator after Bonferroni correction was the nonjudging dimension of mindfulness. Scores on this measure moderated the effects of CBT vs. MBSR on both function (P = 0.001) and pain (P = 0.04). Pain control beliefs (P < 0.001) and lower anxiety (P < 0.001) predicted improvement regardless of treatment. Replication of these findings is needed to guide treatment decision-making for CLBP.

Trial Registration:

The trial and analysis plan were preregistered in ClinicalTrials.gov (Identifier: NCT01467843).

Perspective:

Although few potential moderators and nonspecific predictors of benefits from CBT or MBSR for CLBP were statistically significant after adjustment for multiple comparisons, these findings suggest potentially fruitful directions for confirmatory research while providing reassurance that patients could reasonably expect to benefit from either treatment.

Keywords: chronic back pain, cognitive-behavioral therapy, mindfulness-based stress reduction, treatment effect modifiers, predictors

Introduction

Cognitive-behavioral therapy (CBT) is effective and widely recommended for chronic pain.^24,93 CBT teaches patients skills for managing pain and for appraising pain less negatively, resulting in increased self-efficacy and decreased catastrophizing.^9,78,81 There is also growing evidence for the effectiveness of mindfulness-based interventions for chronic pain,^15,65,68,89 which are believed to decrease patient reactivity to pain through awareness and nonjudgmental acceptance of current moment experience.^{27,30,32,65,78} In the Mind-body Approaches to Pain (MAP) randomized controlled trial (RCT), we found both CBT and mindfulness-based stress reduction (MBSR) to have benefits for pain and function among individuals with chronic low back pain (CLBP).¹³ These two interventions had generally similar benefits, suggesting that both are reasonable options for adults with CLBP. This conclusion has been corroborated by a recent comparative trial.⁸

Although the group-level treatment effects may be similar, there is substantial individual variability in treatment response to CBT and MBSR.^19,51 Patient sociodemographic factors,^5,88 clinical characteristics (e.g., pain diagnosis),^5,81 and psychological or behavioral variables^{5,19,51,56,81} predict differential treatment response in some studies, but the evidence is mixed^5,6,55,81 and may be influenced by study sample characteristics, length of follow-up, and other study design features.^28,56 Understanding which treatments work for whom is critical for matching patients with the most effective treatment and efficiently using limited resources.⁷⁵ Currently, data are insufficient to guide the choice for a given patient of a specific psychosocial treatment among the multiple available; in particular, there is a paucity of RCTs that examined which patient characteristics predict differential benefit from CBT vs. MBSR for CLBP.^18,88 Knowledge concerning patient characteristics that are nonspecific predictors (i.e., predict outcomes irrespective of treatment) is also needed. This can help identify patients at elevated risk for poor outcomes who may require more intensive or targeted interventions, as well as patients who may do well regardless of treatment.

Our primary objective in this study was to use data from the MAP RCT to identify baseline participant characteristics associated with differential benefit from MBSR or CBT, vs. each other and vs. usual care (UC). We hypothesized a priori that, relative to UC, the effects of CBT and MBSR on function, pain bothersomeness, and depressive symptoms at 8 weeks would be greater among participants with clinically meaningful levels of depression and anxiety, and greater pain bothersomeness and functional limitations, at baseline. We further hypothesized that the treatment benefit of CBT vs. MBSR would be greater among individuals with these characteristics. We reasoned that both interventions would have greater benefits than usual care, but that participants with low baseline scores on these measures would have little room to improve and hence be less likely to show treatment benefits. We also reasoned that CBT might be more effective than MBSR for those with higher levels of depression, anxiety, pain, and functional limitations because it was designed to treat these problems (and our CBT intervention was specifically designed for chronic back pain), whereas MBSR was designed for stress management for individuals with a wide variety of illnesses. Substantial evidence supports the effectiveness of CBT for depression,^17,21 anxiety,⁸⁷ and pain and pain-related disability.⁹³ At study initiation, MBSR had been little-studied for these problems, especially among patients with chronic back pain.

A secondary objective was to identify baseline characteristics associated with improvement regardless of treatment. Based on previous research on predictors of chronic pain outcomes, we hypothesized a priori that expectations of improvement,^84,91 higher levels of education,^7,84 less widespread pain,^2,52,81,84 and lower baseline levels of pain and functional limitations^2,7,22,52,84 would be associated with better outcomes. Finally, we conducted exploratory analyses to examine other possible moderators and nonspecific predictors of treatment response.

Materials and Methods

Setting, Participants, and Procedures

Study participants were enrolled in the MAP RCT comparing group MBSR, group CBT, and UC for non-specific CLBP between September 2012 and April 2014. We previously reported details of the MAP trial methods,¹⁴ Consolidated Standards of Reporting Trials (CONSORT) flow diagram,¹³ outcomes,^12,13 and findings regarding relationships and changes among the study measures of putative therapeutic mechanisms.⁷⁸ In brief, participants were recruited from Group Health (now Kaiser Permanente Washington), an integrated healthcare system in Washington State, and from mailings to residents of communities served by Group Health. Eligibility criteria included age 20 to 70 years, back pain for ≥3 months, patient-rated pain bothersomeness during the past week ≥4 (0–10 scale), and patient-rated pain-related activity interference during the past week ≥3 (0–10 scale). Using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM)⁶⁴ diagnostic codes from electronic health records of visits in the previous year and telephone screening, we excluded patients with specific causes of low back pain. Other exclusion criteria were pregnancy, spine surgery in the previous 2 years, disability compensation or litigation, fibromyalgia or cancer diagnosis, other major medical condition, plans to see a medical specialist for back pain, inability to read or speak English, and participation in a “mind-body” treatment for back pain in the past year. The Group Health institutional review board approved the study, and all participants provided informed consent.

We randomized 342 participants to MBSR, CBT, or UC. We stratified randomization based on the baseline value of the primary outcome, a modified version of the Roland Disability Questionnaire (mRDQ), which we categorized as moderate (mRDQ score ≤12 on the 0–23 scale) vs. high (mRDQ scores ≥13) back pain-related functional limitations.⁶³ Trained survey staff, masked to treatment group, administered study measures to participants in computer-assisted telephone interviews. Participants were paid $20 for each interview completed. For the analyses reported here, we include the 297 participants who completed both the baseline and 8-week follow-up assessments.

Measures

Participants provided descriptive information at the screening and baseline interviews (including age, gender, education, and pain duration [<1 year vs. ≥1 year since experiencing 1 week without low back pain]) and completed measures at baseline (before randomization) and 4, 8, 26, and 52 weeks post-randomization. For the purposes of this study, we focused on change in outcome at 8 weeks (post-treatment for those randomized to MBSR or CBT).

Outcome Measures

Outcomes of interest for the current study were the two primary outcomes of the parent RCT (back pain-related functional limitations and pain bothersomeness) and one secondary outcome from the parent trial (depressive symptom severity).

Back Pain-Related Functional Limitations

Participants completed the modified Roland-Morris Disability Questionnaire (mRDQ).⁶³ The original RDQ, a widely-used measure of back pain-related functional limitations, asks whether 24 specific activities are limited today by back pain (yes or no).⁶⁶ We used a modified version that included 23 items⁶³ and asked about the previous week rather than today only (score range 0–23; higher scores indicate greater limitations).

Back Pain Bothersomeness

Participants rated how bothersome their back pain was during the previous week on a 0 to 10 numerical rating scale (0 = ‘not at all bothersome’ and 10 = ‘extremely bothersome’).

Depressive Symptoms

Participants completed the 8-item Patient Health Questionnaire (PHQ-8), a widely-used, validated measure of depressive symptom severity (score range 0–24; higher scores indicate greater depressive symptom severity).⁴⁹

Potential Moderators and Nonspecific Predictors

Moderators (treatment effect modifiers) are baseline characteristics of individuals that are associated with differential treatment response. Moderators influence the relationship between treatment and outcome (i.e., the effect of treatment on individuals depends on their value of the moderator, which precedes treatment and is not associated with treatment).⁴⁵ To examine treatment effect modification, we dichotomized continuous baseline measures using pre-specified cut points based on clinically meaningful criteria because cut points are advantageous for future clinical decision-making. When there were not established clinical cut points for a given instrument, we used logical cut points based on possible item responses and the distribution in our sample in order to avoid very small subgroups. All dichotomization decisions were made prior to conducting analyses.

For the moderation analyses, we categorized the baseline mRDQ as scores <14 vs. ≥14 based on the prior definition of scores of 14 or higher as a poor outcome.⁶⁶ Previous studies have suggested different cut points on 0–10 numerical rating scales for severe pain, with the minimal score typically being either 7 or 8.^36,40,80,94 We categorized baseline pain bothersomeness scores as <8 vs. 8–10. We categorized PHQ-8 scores in 2 pre-specified ways to reflect differences between minimal, mild, and moderate to severe depressive symptom severity based on standard interpretation: <5 vs. ≥5 (minimal vs. mild or more severe) and <10 vs. ≥10 (minimal or mild vs. moderate or more severe).⁴⁹ In models examining nonspecific prediction, we analyzed the continuous-scale measures on their original scales. In addition to demographic characteristics such as age and gender, which have been identified as potential moderators in some prior studies of CBT and mindfulness,^5,86 and baseline measures of study outcomes, we also tested nonspecific predictor/moderator effects for the baseline values of the measures listed below based on previous research suggesting their potential importance in chronic pain problems.^{5,23,28,31,52–54,84}

Expectations.

Participants were asked, “One year from now, do you expect your back pain to be: completely gone (1), much better (2), somewhat better (3), a little better (4), about the same (5), a little worse (6), or much worse (7)?”. Participants were then asked, “Both of the back pain programs we are studying address the mind as well as the body. How helpful do you think such a program would be for your back pain, using a 0 to 10 scale, where 0 is ‘not at all helpful’ and 10 is ‘extremely helpful’?”. For the moderator analyses, we categorized the back pain improvement expectations scores as 1–2 (pain completely gone or much better) vs. 3–7 (pain somewhat better to much worse) and the treatment helpfulness expectation rating as 0–7 (not helpful to neutral) vs. 8–10 (helpful).

Widespread Pain

Participants rated how much they had been bothered (not at all [0], a little [1], or a lot [2]) by pain in 9 different body sites during the past 4 weeks: abdominal pain; pain in arms, legs, or joints other than spine or back; headache; chest pain; neck pain; pelvic or groin pain (both genders) or painful prostatitis (males only); pain in face or jaws; head pain other than headaches; and ‘widespread pain, pain in most of your body, or fibromyalgia.’ The first four items were from the Patient Health Questionnaire 15 (PHQ-15),⁴⁶ and the other pain items were developed and used in previous research.^84,85 Research has consistently demonstrated more widespread pain to predict worse pain outcomes.^52,84 Scores could range from 0–18, with higher scores indicating greater widespread pain bothersomeness. Based on the distribution of scores in this sample, for the moderator analyses, we categorized scores as <4 vs. ≥4.

Anxiety

Participants completed the Generalized Anxiety Disorder (GAD)-2, a widely-used, validated 2-item measure of anxiety with high sensitivity and specificity for detecting anxiety disorders in primary care (score range 0–6; higher scores indicate greater anxiety).^47,48 Using the recommended cut point for clinically significant anxiety disorders,⁴⁷ we categorized scores as <3 vs. ≥3 for the moderator analyses.

Pain Beliefs

Participants completed the 2-item versions of 4 scales from the Survey of Pain Attitudes (SOPA): Disability (belief that one is disabled by pain), Harm (belief that pain is a sign of damage and that therefore activity should be avoided), Control (belief in one’s personal control over pain), and Emotions (belief that emotions affect pain).^34,39 These scales have good test-retest stability, validity, and internal consistency.^33,35,39,69 Each item is rated on a 0–4 scale, where 3 and 4 indicate ‘somewhat true for me’ or ‘very true for me,’ respectively. The total score for each scale is the mean of the 2 item responses, with higher scores indicating stronger beliefs in the construct. For the moderator analyses, we categorized scores on each scale as 3–4 vs. <3 (very or somewhat true vs. not).

Pain Catastrophizing

Participants completed the Pain Catastrophizing Scale (PCS), a 13-item measure assessing pain-related catastrophizing, including rumination, magnification, and helplessness.^61,71 Participants rated the degree to which they had certain thoughts and feelings when experiencing pain, using a scale from 0 = ‘not at all’ to 4 = ‘all the time.’ The total score is the sum of the item responses (possible range = 0–52; higher scores indicate greater endorsement of catastrophic thinking in response to pain). The PCS user manual⁷⁰ suggests a total score of 30+ is clinically relevant (75^th percentile); therefore, we categorized scores as <30 vs. ≥30 for the moderator analyses.

Pain Acceptance

The Chronic Pain Acceptance Questionnaire-8 (CPAQ-8), an 8-item version of the 20-item CPAQ, has been shown to be reliable and valid.^25,26 It has 2 scales: Activity Engagement (AE; engagement in life activities in a normal manner even while pain is being experienced) and Pain Willingness (PW; disengagement from attempts to control or avoid pain). Participants rated each of the 8 items on a 0–6 scale (0 = ‘never true,’ 1 = ‘very rarely true,’ 2 = ‘seldom true,’ 3 = ‘sometimes true,’ 4 = ‘often true,’ 5 = ‘almost always true,’ 6 = ‘always true’). For the current study, we examined only the total score (sum of item responses; possible range 0–48). Higher scores indicate greater activity engagement/pain willingness/pain acceptance. For the moderator analyses, we categorized the CPAQ-8 total score as <32 vs. ≥32 (an average of at least ‘often true’ for each item).

Pain Self-Efficacy

Based on prior research that identified self-efficacy as a moderator of benefit from CBT,⁵¹ we included the Pain Self-efficacy Questionnaire (PSEQ), which consists of 10 items assessing individuals’ confidence in their ability to cope with their pain and engage in activities despite their pain. The questionnaire has been demonstrated to be valid, reliable, and sensitive to change.⁶⁰ Each item is rated on a scale from 0 = ‘not at all confident’ to 6 = ‘completely confident’ and these ratings are summed to yield a total score (possible range 0–60; higher scores indicate greater self-efficacy). For the moderator analyses, we categorized scores as <50 vs. ≥50 (average score of 5 on each item).

Mindfulness

Mindfulness has been defined as the awareness that emerges through purposeful, non-judgmental attention to the present moment.⁴³ Although increased observing, awareness, nonjudging, and nonreactivity are thought to be important mechanisms of change in mindfulness-based interventions,³⁰ baseline mindfulness may also predict or moderate outcomes.^19,67 Participants completed 4 subscales of the Five Facet Mindfulness Questionnaire-Short Form (FFMQ-SF; one FFMQ-SF scale, Describing, was not administered due to the need to reduce participant assessment burden and its questionable relevance to this study): Observing (noticing internal and external experiences; 4 items); Acting with Awareness (attending to present moment activities, in contrast to behaving automatically while focusing attention elsewhere; 5 items); Nonreactivity (the tendency to allow thoughts and feelings to arise and leave without attachment or aversion; 5 items); and Nonjudging (high scores reflect a nonevaluative acceptance of one’s thoughts and feelings; 5-item scale, although inadvertently one question [‘I make judgments about whether my thoughts are good or bad’] was not asked).⁴ Using a 5-point Likert scale (1 = ‘never or very rarely true’ to 5 = ‘very often or always true’), participants rated what generally is true for them in terms of their tendency to be mindful in their daily lives. For each scale, the score was calculated as the mean of the answered items (possible range = 1–5; higher scores indicate higher levels of the construct). The FFMQ-SF has been demonstrated to be reliable, valid, and sensitive to change.⁴ For the moderator analyses, scores on each scale were categorized as <4 vs. ≥4 (average response of ‘often true’ or ‘very often or always true’).

Opioid Use

Participants were asked whether or not they had used prescription opioid medication for pain in the week before the baseline assessment.

Covariates

The covariates for these analyses were the same as those in our prior analyses of the interventions’ effects on the outcomes¹³: age, gender, education, and pain duration (<1 year vs. ≥1 year since experiencing 1 week without low back pain). We decided a priori to control for these variables given their potential to affect participant response to treatment and likelihood of obtaining follow-up information. We also examined each of these variables as a potential moderator or nonspecific predictor. In the moderation analyses, we categorized participant age as <60 vs. ≥60 years based on the distribution of age in the sample and the need to avoid very small subgroups. There are known differences between middle-aged and older adults in pain-related variables.^58,59 We categorized education as college graduate vs. less than college education, similar to prior studies⁵ and balancing the subgroup sizes. In the nonspecific predictor analyses, we modeled age in 5-year units.

Interventions

We developed structured, detailed therapist/instructor manuals for each intervention. The two interventions were comparable in format (group), duration, frequency, and number of participants per group cohort. Each intervention consisted of 8 weekly 2-hour sessions, home activity assignments, materials to read between sessions, and CDs with relevant content for home practice (meditation, body scan, and yoga in MBSR; relaxation and imagery exercises in CBT). Consistent with the standard MBSR program, our MBSR intervention included an optional 6-hour retreat between the 6^th and 7^th sessions. We previously published detailed descriptions of both interventions,^14,78 session attendance rates (56.6% and 50.9% of randomized participants attended at least 6 sessions of CBT and MBSR, respectively), and retreat attendance rate (26% of participants randomized to MBSR),¹³ but describe the intervention components briefly here.

MBSR

The MBSR intervention was modeled closely after the original program developed by Kabat-Zinn⁴¹ and based on the 2009 MBSR instructor’s manual.³ Participants were given the book Full Catastrophe Living⁴² and handouts. The intervention included experiential training in mindfulness meditation and mindful yoga. All sessions included mindfulness exercises (e.g., body scan, sitting meditation) and mindful movement (most commonly, yoga).

CBT

The group CBT protocol included the techniques most commonly applied in CBT for CLBP^{10,24,62,73,83} and used in prior studies^{50,77,79,82,90}. The intervention included: (1) education about (a) chronic pain, (b) maladaptive thoughts (including catastrophizing) and maladaptive beliefs (e.g., inability to control pain, hurt equals harm) common among individuals with chronic pain, (c) the relationships between thoughts and emotional and physical reactions, (d) sleep hygiene, and (e) relapse prevention and maintenance of gains; and (2) instruction and practice in identifying and challenging unhelpful thoughts, generating alternative appraisals that are more accurate and helpful, setting and working towards behavioral goals, abdominal breathing and progressive muscle relaxation techniques, activity pacing, thought-stopping and distraction techniques, positive coping self-statements, and coping with pain flare-ups. None of these techniques were included in the MBSR intervention, and mindfulness, meditation, and yoga techniques were not included in the CBT intervention. CBT participants were also given a workbook developed for this study and the book The Pain Survival Guide,⁷⁶ and were asked to read specific chapters between sessions. During each session, participants completed a personal action plan for between-session home activities.

Usual Care

Patients assigned to UC received no MBSR training or CBT as part of the study and received whatever health care they would customarily receive during the study period.

Instructors/Therapists and Treatment Fidelity Monitoring

Details of instructor training and supervision and treatment fidelity monitoring were published previously.¹³ In brief, all 8 MBSR instructors received formal training in teaching MBSR from the Center for Mindfulness at the University of Massachusetts or equivalent training and had extensive previous experience teaching MBSR. The CBT intervention was conducted by 4 Ph.D.-level licensed psychologists with previous experience providing individual and group CBT to patients with chronic pain.

Statistical Analysis

To test whether baseline variables were moderators of treatment effects, we constructed separate linear regression models for each outcome (mRDQ, pain bothersomeness, and PHQ-8). The dependent variable was the change from baseline to 8 weeks in the outcome measure and the independent variable was treatment group. A separate model was fit to evaluate each potential moderator. To test whether intervention effects differed by subgroup, each model included an indicator for subgroup defined by the binary moderator variable and an interaction between subgroup and the 3-level treatment group. A finding of a statistically significant interaction effect indicates that the effect of the intervention differs by subgroup. We computed separate P-values to test for differences by subgroup for the following treatment effects: CBT vs. UC, MBSR vs. UC, and CBT vs. MBSR. Models adjusted for age, gender, education, pain duration (<1 year vs. ≥1 year since experiencing 1 week without low back pain), and the baseline value of the outcome measure. We used the Bonferroni method⁵⁷ within each pairwise treatment group comparison to control family-wise error rates due to testing for treatment interactions of 23 potential moderators. Specifically, we defined statistical significance for a given moderator as a P-value < 0.002 (0.05/23). Although the Bonferroni method is conservative compared to other approaches, in our data it yielded conclusions similar to those of other (sequential) methods.⁹² Given the exploratory nature of these analyses and the fact that the trial was not powered to detect treatment effect moderators, our primary focus is on effect size and results should be interpreted accordingly. We used a “complete case” sample for the analyses.

To examine nonspecific predictors of improvement in outcomes, we constructed linear regression models with the change in outcome as the dependent variable and the potential predictors (including the baseline value of the outcome measure) as independent variables. Each potential predictor was evaluated in a separate model, which was also adjusted for age, gender, education, pain duration, and the baseline value of the outcome measure. Using the Bonferroni method, we defined statistical significance for a given nonspecific predictor as a P-value < 0.002 (0.05/22 to adjust for family-wise error rate over 22 comparisons).

Results

Moderators of treatment effects

Table 1 shows the numbers and proportions of participants in each of the three study groups with each baseline characteristic we examined. Results for each outcome are summarized below.

Table 1:

Baseline characteristics of the study sample

	UC (n=106)	CBT (n=97)	MBSR (n=94)
Baseline characteristics	n (%)	n (%)	n (%)
Age ≥ 60 years old	24 (22.6)	28 (28.9)	29 (30.9)
Female	82 (77.4)	57 (58.8)	58 (61.7)
Bachelor’s degree or higher education	68 (64.2)	63 (65.0)	50 (53.2)
Moderate-severe functional limitations (mRDQ ≥14)	30 (28.3)	33 (34.0)	32 (34.0)
High pain bothersomeness (≥7)	19 (17.9)	14 (14.4)	14 (14.9)
Moderate-severe depressive symptoms (PHQ-8 ≥10)	14 (13.2)	18 (18.6)	17 (18.1)
Mild or greater depressive symptoms (PHQ-8 ≥5)	52 (49.1)	52 (53.6)	49 (52.1)
Low expected pain improvement (pain worse to somewhat better)	75 (70.8)	70 (72.9)	62 (66.7)
High expected program helpfulness (≥8)	62 (58.5)	46 (47.4)	53 (56.4)
Widespread pain (≥4)	71 (67.0)	57 (58.8)	55 (58.5)
Anxiety (GAD-2 ≥3)	16 (15.1)	17 (17.5)	18 (19.2)
Belief that one is disabled by pain (SOPA Disability ≥2)	39 (36.8)	42 (43.3)	43 (45.7)
Belief that pain is a sign of damage (SOPA Harm ≥2)	57 (53.8)	40 (41.2)	44 (46.8)
Belief in one’s control over pain (SOPA Control ≥3)	41 (38.7)	37 (38.1)	36 (38.3)
Belief that emotions affect pain (SOPA Emotion ≥3)	40 (37.7)	43 (44.3)	40 (42.6)
High pain catastrophizing (PCS ≥30)	93 (87.7)	86 (88.7)	79 (85.0)
High pain acceptance (CPAQ-8 ≥32)	51 (48.1)	47 (48.5)	48 (51.1)
High pain self-efficacy (PSEQ ≥50)	44 (41.5)	46 (47.4)	37 (39.4)
Mindfulness: Observing (FFMQ-SF OB ≥4)	54 (50.9)	44 (45.4)	36 (38.3)
Mindfulness: Acting with awareness (FFMQ-SF AA ≥4)	31 (29.3)	34 (35.1)	31 (33.0)
Mindfulness: Nonreactivity (FFMQ-SF NR ≥4)	22 (20.8)	30 (30.9)	31 (33.0)
Mindfulness: Nonjudging (FFMQ-SF NJ ≥4)	55 (51.9)	54 (55.7)	48 (51.1)
Opioid use, past week	12 (11.3)	10 (10.3)	8 (8.5)

Open in a new tab

AA, Acting with Awareness; CBT, cognitive-behavioral therapy; CPAQ-8, Chronic Pain Acceptance Questionnaire-8; FFMQ-SF, Five Facet Mindfulness Questionnaire-short form; GAD, Generalized Anxiety Disorder; MBSR, Mindfulness-Based Stress Reduction; NJ, Nonjudging; NR, Nonreactivity; OB, Observing; PCS, Pain Catastrophizing Scale; PHQ, Patient Health Questionnaire; PSEQ, Pain Self-Efficacy Questionnaire; mRDQ, modified Roland Disability Questionnaire; SOPA, Survey of Pain Attitudes; UC, Usual Care

Missing values: Usual Care (none); CBT (none); MBSR (Expected pain improvement (1), Pain catastrophizing (1))

Moderators of Intervention Effects on Back Pain-Related Functional Limitations

Table 2 shows the results of analyses examining the effects of treatment on back pain-related functional limitations (as measured by the mRDQ) for the pre-specified subgroups of interest. Among the baseline variables examined, the largest differences observed were for the nonjudging subscale of the FFMQ-SF mindfulness questionnaire and pain self-efficacy. Comparing the effects of CBT vs. MBSR on functional limitations, CBT was superior to MBSR among participants who reported being less nonjudging of their thoughts and feelings (adjusted mean change from baseline to 8 weeks on the 0–23 mRDQ −3.84 in CBT vs. −1.70 in MBSR; difference between treatments −2.14 [95% CI, −3.91, −0.37]), whereas MBSR was superior to CBT among those who reported being more nonjudgmental (adjusted mean change: −2.77 in CBT vs. −4.69 in MBSR; difference 1.92 [0.29, 3.56]) (interaction P = 0.001, statistically significant with Bonferroni correction). The effect of CBT vs. UC on function was greater among participants with higher pain self-efficacy (adjusted mean change: CBT −4.40 vs. UC −1.42; difference −2.98 [−4.76, −1.20]) than among those with lower self-efficacy (adjusted mean change: CBT −2.18 vs. UC −1.89; difference −0.29 [−1.87, 1.29]) (P = 0.03), but this difference was not statistically significant with Bonferroni correction. No other comparisons were statistically significant.

Table 2:

Functional limitations (mRDQ): Adjusted mean change and treatment effects at 8 weeks for subgroups defined by baseline characteristics.

	Adjusted mean change at 8 weeks			Treatment effect^*
	UC	CBT	MBSR	CBT – UC	MBSR - UC	CBT - MBSR
Baseline characteristic	Mean (95% CI)	Mean (95% CI)	Mean (95% CI)	P-value^† Mean (95% CI)	P-value^† Mean (95% CI)	P-value^† Mean (95% CI)
Overall	−1.67 (−2.49, −0.85)	−3.22 (−4.07, −2.37)	−3.24 (−4.10, −2.37)	−1.55 (−2.74, −0.36)	−1.56 (−2.77, −0.36)	0.01 (−1.20, 1.23)
Age				0.24	0.74	0.39
<60 yrs	−1.64 (−2.58, −0.71)	−3.62 (−4.63, −2.60)	−3.27 (−4.31, −2.23)	−1.97 (−3.36, −0.59)	−1.63 (−3.03, −0.22)	−0.35 (−1.80, 1.11)
≥60 yrs	−1.90 (−3.62, −0.18)	−2.24 (−3.85, −0.62)	−3.06 (−4.62, −1.50)	−0.34 (−2.70, 2.02)	−1.16 (−3.49, 1.16)	0.82 (−1.41, 3.06)
Gender				0.97	0.11	0.08
Male	−2.11 (−3.82, −0.41)	−3.57 (−4.89, −2.25)	−2.24 (−3.63, −0.84)	−1.45 (−3.60, 0.70)	−0.12 (−2.32, 2.07)	−1.33 (−3.25, 0.59)
Female	−1.56 (−2.49, −0.64)	−2.96 (−4.06, −1.85)	−3.84 (−4.94, −2.74)	−1.39 (−2.83, 0.05)	−2.28 (−3.72, −0.84)	0.89 (−0.67, 2.45)
Education				0.83	0.31	0.23
Less than bachelor’s degree	−1.25 (−2.62, 0.12)	−2.62 (−4.07, −1.18)	−3.52 (−4.80, −2.25)	−1.37 (−3.36, 0.61)	−2.28 (−4.14, −0.41)	0.90 (−1.01, 2.81)
Bachelor’s degree	−1.92 (−2.95, −0.89)	−3.57 (−4.62, −2.51)	−2.94 (−4.12, −1.75)	−1.65 (−3.12, −0.17)	−1.02 (−2.58, 0.55)	−0.63 (−2.21, 0.95)
Functional limitations				0.93	0.83	0.89
mRDQ <14	−0.67 (−1.69, 0.36)	−2.27 (−3.40, −1.15)	−2.51 (−3.63, −1.39)	−1.60 (−3.12, −0.09)	−1.84 (−3.37, −0.32)	0.24 (−1.35, 1.82)
mRDQ ≥14	−3.62 (−5.24, −1.99)	−5.10 (−6.65, −3.55)	−5.15 (−6.73, −3.57)	−1.49 (−3.71, 0.74)	−1.54 (−3.80, 0.72)	0.05 (−2.14, 2.25)
Pain bothersomeness				0.70	0.63	0.93
<7	−1.74 (−2.79, −0.68)	−3.42 (−4.46, −2.39)	−3.50 (−4.57, −2.43)	−1.69 (−3.16, −0.22)	−1.76 (−3.27, −0.26)	0.08 (−1.41, 1.56)
≥7	−1.56 (−2.93, −0.18)	−2.76 (−4.35, −1.17)	−2.71 (−4.21, −1.21)	−1.20 (−3.26, 0.86)	−1.15 (−3.16, 0.86)	−0.05 (−2.21, 2.11)
Depressive symptoms				0.34	0.67	0.59
PHQ-8 <10	−1.82 (−2.71, −0.93)	−3.14 (−4.08, −2.19)	−3.30 (−4.26, −2.33)	−1.32 (−2.62, −0.02)	−1.48 (−2.80, −0.16)	0.16 (−1.18, 1.51)
PHQ-8 ≥10	−0.71 (−2.98, 1.55)	−3.62 (−5.62, −1.63)	−2.92 (−5.01, −0.83)	−2.91 (−5.91, 0.10)	−2.20 (−5.23, 0.83)	−0.71 (−3.59, 2.18)
Depressive symptoms				0.44	0.83	0.33
PHQ-8 <5	−2.23 (−3.39, −1.08)	−3.33 (−4.59, −2.07)	−3.96 (−5.21, −2.71)	−1.10 (−2.80, 0.60)	−1.73 (−3.43, −0.03)	0.63 (−1.14, 2.40)
PHQ-8 ≥5	−1.09 (−2.26, 0.08)	−3.13 (−4.31, −1.96)	−2.56 (−3.76, −1.35)	−2.04 (−3.70, −0.38)	−1.46 (−3.14, 0.22)	−0.58 (−2.24, 1.09)
Expected pain improvement				0.42	0.23	0.71
1–2 (pain gone or much better)	−2.30 (−3.84, −0.77)	−3.09 (−4.75, −1.43)	−2.76 (−4.29, −1.23)	−0.79 (−3.04, 1.46)	−0.46 (−2.63, 1.72)	−0.33 (−2.57, 1.91)
3–7 (pain worse, unchanged, or somewhat better)	−1.43 (−2.41, −0.45)	−3.31 (−4.33, −2.29)	−3.48 (−4.55, −2.41)	−1.88 (−3.30, −0.46)	−2.05 (−3.51, −0.60)	0.17 (−1.30, 1.65)
Expected program helpfulness				0.34	0.71	0.59
<8	−2.11 (−3.38, −0.85)	−3.05 (−4.23, −1.87)	−3.41 (−4.73, −2.09)	−0.93, −2.66, 0.79)	−1.30 (−3.13, 0.54)	−0.36 (−1.40, 2.13)
≥8	−1.34 (−2.43, −0.25)	−3.43 (−4.67, −2.18)	−3.10 (−4.26, −1.95)	−2.09 (−3.75, −0.43)	−1.76 (−3.36, −0.17)	−0.32 (−2.03, 1.38)
Widespread pain				0.46	0.08	0.31
<4	−2.51 (−3.94, −1.08)	−3.41 (−4.76, −2.06)	−2.67 (−4.03, −1.31)	−0.90 (−2.86, 1.06)	−0.15 (−2.14, 1.83)	−0.75 (−2.63, 1.14)
≥4	−1.26 (−2.27, −0.25)	−3.10 (−4.21, −1.98)	−3.63 (−4.76, −2.49)	−1.84 (−3.34, −0.33)	−2.37 (−3.88, −0.85)	0.53 (−1.06, 2.11)
Anxiety				0.96	0.81	0.78
GAD-2 <3	−1.68 (−2.57, −0.78)	−3.23 (−4.18, −2.29)	−3.16 (−4.14, −2.19)	−1.56 (−2.86, −0.26)	−1.49 (−2.82, −0.15)	−0.07 (−1.42, 1.28)
GAD-2 ≥3	−1.66 (−3.80, 0.47)	−3.15 (−5.23, −1.07)	−3.54 (−5.55, −1.52)	−1.48 (−4.46, 1.50)	−1.87 (−4.77, 1.03)	0.39 (−2.53, 3.31)
SOPA Disability				0.22	0.54	0.55
<2	−1.51 (−2.55, −0.47)	−3.67 (−4.80, −2.53)	−3.37 (−4.55, −2.19)	−2.16 (−3.69, −0.63)	−1.86 (−3.43, −0.28)	−0.30 (−1.94, 1.33)
≥2	−1.97 (−3.32, −0.62)	−2.63 (−3.94, −1.32)	−3.07 (−4.35, −1.78)	−0.66 (−2.54, 1.22)	−1.10 (−2.96, 0.77)	0.44 (−1.38, 2.26)
SOPA Harm				0.17	0.72	0.09
<2	−1.76 (−3.00, −0.52)	−3.97 (−5.08, −2.86)	−3.08 (−4.27, −1.88)	−2.21 (−3.88, −0.55)	−1.32 (−3.06, 0.43)	−0.90 (−2.53, 0.74)
≥2	−1.63 (−2.74, −0.51)	−2.15 (−3.48, −0.82)	−3.38 (−4.64, −2.12)	−0.52 (−2.25, 1.20)	−1.75 (−3.44, −0.07)	1.23 (−0.60, 3.06)
SOPA Control				0.23	0.99	0.25
<3	−1.44 (−2.48, −0.40)	−2.41 (−3.49, −1.33)	−2.98 (−4.10, −1.86)	−0.97 (−2.47, 0.53)	−1.54 (−3.08, −0.01)	0.57 (−0.98, 2.13)
≥3	−2.07 (−3.40, −0.73)	−4.53 (−5.90, −3.16)	−3.62 (−5.01, −2.23)	−2.47 (−4.37, −0.56)	−1.55 (−3.46, 0.36)	−0.92 (−2.87, 1.04)
SOPA Emotion				0.16	0.30	0.74
<3	−2.06 (−3.10, −1.03)	−2.93 (−4.08, −1.78)	−3.14 (−4.30, −1.99)	−0.87 (−2.42, 0.68)	−1.08 (−2.64, 0.48)	0.21 (−1.41, 1.83)
≥3	−1.01 (−2.36, 0.34)	−3.59 (−4.88, −2.29)	−3.38 (−4.71, −2.04)	−2.58 (−4.45, −0.71)	−2.37 (−4.26, −0.47)	−0.21 (−2.07, 1.65)
Catastrophizing				0.97	0.72	0.70
PCS <30	0.32 (−2.08, 2.71)	−1.12 (−3.69, 1.45)	−1.82 (−4.17, 0.54)	−1.44 (−4.86, 1.99)	−2.13 (−5.37, 1.10)	0.70 (−2.69, 4.08)
PCS ≥30	−1.97 (−2.85, −1.09)	−3.49 (−4.39, −2.59)	−3.46 (−4.40, −2.52)	−1.52 (−2.78, −0.26)	−1.49 (−2.78, −0.20)	−0.03 (−1.33, 1.28)
Pain acceptance				0.17	0.20	0.94
CPAQ-8 <32	−1.82 (−2.97, −0.67)	−2.54 (−3.74, −1.34)	−2.55 (−3.81, −1.29)	−0.72 (−2.37, 0.92)	−0.74 (−2.42, 0.95)	0.01 (−1.69, 1.72)
CPAQ-8 ≥32	−1.56 (−2.79, −0.34)	−3.93 (−5.16, −2.70)	−3.85 (−5.07, −2.64)	−2.36 (−4.07, −0.65)	−2.29 (−3.99, −0.59)	−0.07 (−1.8, 1.65)
Pain self-efficacy				0.03	0.18	0.40
PSEQ <50	−1.89 (−2.95, −0.83)	−2.18 (−3.36, −0.99)	−2.74 (−3.86, −1.62)	−0.29 (−1.87, 1.29)	−0.86 (−2.39, 0.67)	0.57 (−1.03, 2.17)
PSEQ ≥50	−1.42 (−2.72, −0.11)	−4.40 (−5.65, −3.14)	−3.91 (−5.29, −2.54)	−2.98 (−4.76, −1.20)	−2.50 (−4.37, −0.62)	−0.48 (−2.33, 1.36)
FFMQ-SF OB				0.91	0.49	0.43
<4	−1.87 (−3.04, −0.71)	−3.35 (−4.50, −2.19)	−3.71 (−4.82, −2.61)	−1.47 (−3.12, 0.17)	−1.84 (−3.45, −0.23)	0.37 (−1.23, 1.96)
≥4	−1.47 (−2.62, −0.32)	−3.08 (−4.35, −1.82)	−2.45 (−3.87, −1.04)	−1.61 (−3.33, 0.10)	−0.98 (−2.81, 0.84)	−0.63 (−2.53, 1.27)
FFMQ-SF AA				0.35	0.65	0.64
<4	−1.91 (−2.89, −0.92)	−3.07 (−4.12, −2.01)	−3.30 (−4.36, −2.23)	−1.16 (−2.61, 0.28)	−1.39 (−2.84, 0.06)	0.23 (−1.28, 1.73)
≥4	−1.11 (−2.63, 0.40)	−3.49 (−4.94, −2.04)	−3.11 (−4.64, −1.58)	−2.38 (−4.47, −0.29)	−1.99 (−4.15, 0.16)	−0.38 (−2.47, 1.70)
FFMQ-SF NR				0.40	0.99	0.37
<4	−1.73 (−2.66, −0.81)	−2.94 (−3.97, −1.92)	−3.33 (−4.40, −2.27)	−1.21 (−2.60, 0.18)	−1.60 (−3.00, −0.19)	0.39 (−1.09, 1.87)
≥4	−1.45 (−3.25, 0.35)	−3.84 (−5.40, −2.28)	−3.02 (−4.57, −1.49)	−2.39 (−4.78, −0.01)	−1.58 (−3.96, 0.80)	−0.81 (−2.99, 1.36)
FFMQ-SF NJ				0.09	0.08	0.001
<4	−1.18 (−2.34, −0.01)	−3.84 (−5.11, −2.57)	−1.70 (−2.93, −0.48)	−2.67 (−4.38, −0.95)	−0.53 (−2.21, 1.16)	−2.14 (−3.91, −0.37)
≥4	−2.09 (−3.21, −0.98)	−2.77 (−3.90, −1.64)	−4.69 (−5.89, −3.50)	−0.68 (−2.26, 0.91)	−2.60 (−4.24, −0.95)	1.92 (0.29, 3.56)
Opioid use, past wk				0.97	0.25	0.25
No	−1.72 (−2.60, −0.84)	−3.26 (−4.16, −2.35)	−3.47 (−4.37, −2.56)	−1.54 (−2.8, −0.27)	−1.75 (−3.02, −0.47)	0.21 (−1.07, 1.49)
Yes	−1.31 (−3.77, 1.16)	−2.93 (−5.59, −0.27)	−0.68 (−3.68, 2.32)	−1.62 (−5.24, 1.99)	0.62 (−3.22, 4.47)	−2.25 (−6.25, 1.76)

Open in a new tab

AA, Acting with Awareness; CBT, cognitive-behavioral therapy; CI, confidence interval; CPAQ-8, Chronic Pain Acceptance Questionnaire-8; FFMQ-SF, Five Facet Mindfulness Questionnaire-short form; GAD, Generalized Anxiety Disorder; MBSR, Mindfulness-Based Stress Reduction; NJ, Nonjudging; NR, Nonreactivity; OB, Observing; PCS, Pain Catastrophizing Scale; PHQ, Patient Health Questionnaire; PSEQ, Pain Self-Efficacy Questionnaire; mRDQ, modified Roland Disability Questionnaire; SOPA, Survey of Pain Attitudes; UC, Usual Care

Models adjust for age, gender, education, pain duration, and baseline mRDQ score

Treatment effect defined as the difference between the two treatment groups in the 8-week change on the outcome measure within a given subgroup. Treatment effect estimates are coded such that negative treatment effects favor CBT and MBSR in their comparisons with UC, and CBT in the comparison of CBT and MBSR

^†

P-value for test of differences between subgroups in treatment effects at 8 weeks

Bold font indicates P < 0.05; only comparisons with P < 0.002 were statistically significant after Bonferroni adjustment for multiple comparisons.

Moderators of Intervention Effects on Pain Bothersomeness

Table 3 shows the results of analyses examining the effects of treatment on pain bothersomeness for the pre-specified subgroups of interest. The largest differences observed were for gender, depression, conviction that pain is a sign of physical harm, pain self-efficacy, and the nonjudging mindfulness subscale. Although none of these comparisons were statistically significant after Bonferroni correction, we describe below the effects for these subgroups given the potential fruitfulness of examining them in future confirmatory research.

Table 3:

Pain bothersomeness: Adjusted mean change and treatment effects at 8 weeks for subgroups defined by baseline characteristics.

	Adjusted mean change at 8 weeks			Treatment effect^*
	UC	CBT	MBSR	CBT-UC	MBSR-UC	CBT-MBSR
Baseline characteristic	Mean (95% CI)	Mean (95% CI)	Mean (95% CI)	P-value^† Mean (95% CI)	P-value^† Mean (95% CI)	P-value^† Mean (95% CI)
Overall	−0.57 (−0.92, −0.22)	−1.15 (−1.51, −0.80)	−1.26 (−1.63, −0.90)	−0.58 (−1.09, −0.08)	−0.69 (−1.20, −0.19)	0.11 (−0.40, 0.62)
Age at baseline				0.58	0.78	0.77
<60 yrs	−0.58 (−0.97, −0.19)	−1.25 (−1.67, −0.82)	−1.31 (−1.74, −0.87)	−0.67 (−1.25, −0.09)	−0.73 (−1.32, −0.13)	0.06 (−0.55, 0.67)
≥60 yrs	−0.58 (−1.30, 0.15)	−0.92 (−1.59, −0.24)	−1.14 (−1.81, −0.47)	−0.34 (−1.33, 0.65)	−0.57 (−1.55, 0.42)	0.22 (−0.73, 1.17)
Gender				0.71	0.02	0.03
Male	−1.15 (−1.86, −0.43)	−1.51 (−2.07, −0.96)	−0.93 (−1.52, −0.35)	−0.37 (−1.27, 0.53)	0.21 (−0.71, 1.13)	−0.58 (−1.38, 0.22)
Female	−0.36 (−0.75, 0.02)	−0.93 (−1.40, −0.47)	−1.50 (−1.96, −1.03)	−0.57 (−1.17, 0.03)	−1.13 (−1.74, −0.53)	0.56 (−0.09, 1.22)
Education				0.62	0.31	0.62
Less than bachelor’s degree	−0.14 (−0.72, 0.43)	−0.90 (−1.50, −0.29)	−1.15 (−1.69, −0.61)	−0.75 (−1.59, 0.08)	−1.01 (−1.79, −0.22)	0.25 (−0.55, 1.06)
Bachelor’s degree	−0.83 (−1.26, −0.40)	−1.32 (−1.76, −0.87)	−1.31 (−1.81, −0.81)	−0.49 (−1.11, 0.13)	−0.48 (−1.14, 0.18)	−0.01 (−0.68, 0.65)
Functional limitations				0.64	0.98	0.66
mRDQ <14	−0.64 (−1.06, −0.23)	−1.16 (−1.61, −0.70)	−1.34 (−1.79, −0.89)	−0.51 (−1.12, 0.10)	−0.7 (−1.31, −0.09)	0.19 (−0.45, 0.82)
mRDQ ≥14	−0.39 (−1.05, 0.28)	−1.16 (−1.78, −0.53)	−1.10 (−1.75, −0.45)	−0.77 (−1.66, 0.12)	−0.72 (−1.63, 0.20)	−0.06 (−0.94, 0.83)
Pain bothersomeness				0.85	0.91	0.94
<7	−0.02 (−0.46, 0.43)	−0.70 (−1.14, −0.25)	−0.73 (−1.19, −0.26)	−0.68 (−1.31, −0.05)	−0.71 (−1.36, −0.06)	0.03 (−0.61, 0.67)
≥7	−1.61 (−2.20, −1.03)	−2.19 (−2.86, −1.52)	−2.27 (−2.92, −1.61)	−0.57 (−1.46, 0.31)	−0.65 (−1.53, 0.22)	0.08 (−0.86, 1.01)
Depressive symptoms				0.18	0.52	0.48
PHQ-8 <10	−0.62 (−1.00, −0.25)	−1.05 (−1.45, −0.66)	−1.25 (−1.66, −0.85)	−0.43 (−0.98, 0.12)	−0.63 (−1.18, −0.08)	0.20 (−0.36, 0.76)
PHQ ≥10	−0.22 (−1.17, 0.73)	−1.60 (−2.43, −0.77)	−1.31 (−2.21, −0.42)	−1.38 (−2.65, −0.11)	−1.09 (−2.39, 0.20)	−0.29 (−1.52, 0.94)
Depressive symptoms				0.69	0.004	0.01
PHQ-8 <5	−0.31 (−0.79, 0.17)	−0.99 (−1.51, −0.46)	−1.76 (−2.28, −1.24)	−0.67 (−1.38, 0.03)	−1.45 (−2.16, −0.73)	0.77 (0.03, 1.51)
PHQ-8 ≥5	−0.83 (−1.32, −0.34)	−1.30 (−1.79, −0.82)	−0.81 (−1.31, −0.31)	−0.47 (−1.16, 0.22)	0.02 (−0.68, 0.72)	−0.49 (−1.19, 0.20)
Expected pain improvement				0.76	0.31	0.20
1–2 (pain gone or much better)	−1.18 (−1.81, −0.54)	−1.90 (−2.58, −1.21)	−1.47 (−2.10, −0.84)	−0.72 (−1.65, 0.21)	−0.29 (−1.19, 0.61)	−0.43 (−1.35, 0.50)
3–7 (pain worse, unchanged, or somewhat better)	−0.33 (−0.74, 0.07)	−0.88 (−1.30, −0.46)	−1.18 (−1.63, −0.74)	−0.55 (−1.14, 0.04)	−0.85 (−1.45, −0.25)	0.30 (−0.31, 0.91)
Expected program helpfulness				0.37	0.95	0.36
<8	−0.63 (−1.17, −0.10)	−0.99 (−1.49, −0.50)	−1.35 (−1.91, −0.78)	−0.36 (−1.09, 0.37)	−0.71 (−1.49, 0.06)	0.35 (−0.39, 1.10)
≥8	−0.52 (−0.98, −0.06)	−1.34 (−1.86, 0.82)	−1.20 (−1.69, −0.72)	−0.82 (−1.51, −0.12)	−0.68 (−1.35, −0.01)	−0.13 (−0.85, 0.58)
Widespread pain				0.80	0.60	0.44
<4	−0.64 (−1.25, −0.04)	−1.15 (−1.72, −0.58)	−1.50 (−2.07, −0.92)	−0.51 (−1.33, 0.32)	−0.85 (−1.69, −0.02)	0.35 (−0.45, 1.15)
≥4	−0.53 (−0.95, −0.10)	−1.16 (−1.63, −0.70)	−1.10 (−1.58, −0.62)	−0.64 (−1.27, −0.01)	−0.57 (−1.21, 0.07)	−0.06 (−0.74, 0.61)
Anxiety				0.54	0.99	0.54
GAD-2 <3	−0.54 (−0.91, −0.17)	−1.04 (−1.44, −0.65)	−1.22 (−1.63, −0.81)	−0.50 (−1.05, 0.04)	−0.68 (−1.24, −0.13)	0.18 (−0.38, 0.75)
GAD-2 ≥3	−0.76 (−1.65, 0.14)	−1.69 (−2.55, −0.82)	−1.44 (−2.31, −0.58)	−0.93 (−2.18, 0.32)	−0.69 (−1.92, 0.54)	−0.24 (−1.48, 1.00)
SOPA Disability				0.16	0.22	0.86
<2	−0.45 (−0.88, −0.01)	−1.32 (−1.80, −0.85)	−1.40 (−1.89, −0.90)	−0.88 (−1.52, −0.23)	−0.95 (−1.61, −0.29)	0.08 (−0.61, 0.76)
≥2	−0.79 (−1.36, −0.22)	−0.93 (−1.48, −0.39)	−1.10 (−1.64, −0.56)	−0.14 (−0.93, 0.65)	−0.31 (−1.10, 0.48)	0.17 (−0.6, 0.94)
SOPA Harm				0.02	0.30	0.18
<2	−0.39 (−0.90, 0.12)	−1.54 (−2.00, −1.07)	−1.36 (−1.86, −0.86)	−1.14 (−1.84, −0.45)	−0.97 (−1.69, −0.24)	−0.18 (−0.86, 0.50)
≥2	−0.72 (−1.19, −0.26)	−0.62 (−1.17, −0.06)	−1.15 (−1.69, −0.62)	0.10 (−0.62, 0.83)	−0.43 (−1.14, 0.28)	0.53 (−0.24, 1.30)
SOPA Control				0.28	0.84	0.40
<3	−0.46 (−0.90, −0.02)	−0.83 (−1.28, −0.38)	−1.11 (−1.58, −0.63)	−0.37 (−1.00, 0.26)	−0.65 (−1.3, 0.00)	0.28 (−0.38, 0.93)
≥3	−0.75 (−1.31, −0.20)	−1.68 (−2.26, −1.10)	−1.51 (−2.09, −0.92)	−0.93 (−1.72, −0.13)	−0.75 (−1.55, 0.04)	−0.17 (−0.99, 0.65)
SOPA Emotion				0.83	0.54	0.69
<3	−0.53 (−0.97, −0.10)	−1.16 (−1.65, −0.68)	−1.36 (−1.86, −0.87)	−0.63 (−1.28, 0.02)	−0.83 (−1.49, −0.17)	0.20 (−0.49, 0.89)
≥3	−0.62 (−1.19, −0.05)	−1.14 (−1.69, −0.60)	−1.13 (−1.69, −0.57)	−0.52 (−1.31, 0.26)	−0.51 (−1.31, 0.29)	−0.01 (−0.8, 0.77)
Pain catastrophizing				0.75	0.79	0.57
PCS <30	−0.64 (−1.65, 0.36)	−1.01 (−2.01, 0.08)	−1.50 (−2.45, −0.54)	−0.37 (−1.82, 1.09)	−0.85 (−2.23, 0.52)	0.49 (−0.95, 1.92)
PCS ≥30	−0.56 (−0.93, −0.18)	−1.17 (−1.55, −0.78)	−1.21 (−1.61, −0.81)	−0.61 (−1.15, −0.08)	−0.66 (−1.2, −0.11)	0.04 (−0.51, 0.60)
Pain acceptance				0.11	0.08	0.09
CPAQ-8 <32	−0.84 (−1.32, −0.37)	−1.03 (−1.53, −0.53)	−1.09 (−1.61, −0.56)	−0.19 (−0.88, 0.50)	−0.24 (−0.96, 0.47)	0.05 (−0.67, 0.78)
CPAQ-8 ≥32	−0.28 (−0.78, 0.22)	−1.28 (−1.80, −0.77)	−1.43 (−1.94, −0.92)	−1.00 (−1.72, −0.29)	−1.15 (−1.86, −0.44)	0.14 (−0.58, 0.87)
Pain self-efficacy				0.11	0.03	0.59
PSEQ <50	−0.77 (−1.22, −0.33)	−1.00 (−1.50, −0.51)	−1.02 (−1.49, −0.55)	−0.23 (−0.9, 0.43)	−0.25 (−0.90, 0.40)	0.02 (−0.66, 0.70)
PSEQ ≥50	−0.28 (−0.82, 0.26)	−1.33 (−1.85, −0.80)	−1.63 (−2.21, −1.05)	−1.05 (−1.8, −0.30)	−1.35 (−2.14, −0.56)	0.30 (−0.48, 1.08)
FFMQ-SF OB				0.51	0.28	0.10
<4	−0.65 (−1.14, −0.16)	−1.08 (−1.56, −0.60)	−1.54 (−2.01, −1.07)	−0.43 (−1.12, 0.26)	−0.89 (−1.57, −0.21)	0.46 (−0.21, 1.13)
≥4	−0.49 (−0.97, −0.01)	−1.25 (−1.78, −0.72)	−0.82 (−1.42, −0.23)	−0.76 (−1.48, −0.04)	−0.33 (−1.10, 0.43)	−0.43 (−1.22, 0.37)
FFMQ-SF AA				0.42	0.31	0.83
<4	−0.50 (−0.91, −0.09)	−1.22 (−1.67, −0.78)	−1.37 (−1.82, −0.92)	−0.72 (−1.33, −0.11)	−0.87 (−1.48, −0.26)	0.15 (−0.49, 0.78)
≥4	−0.75 (−1.39, −0.11)	−1.02 (−1.64, −0.41)	−1.05 (−1.70, −0.41)	−0.28 (−1.16, 0.61)	−0.30 (−1.21, 0.60)	0.03 (−0.85, 0.91)
FFMQ-SF NR				0.21	0.68	0.38
<4	−0.67 (−1.06, −0.28)	−1.08 (−1.51, −0.64)	−1.34 (−1.79, −0.89)	−0.40 (−0.98, 0.18)	−0.67 (−1.26, −0.08)	0.27 (−0.35, 0.89)
≥4	−0.19 (−0.94, 0.57)	−1.33 (−1.98, −0.67)	−1.10 (−1.75, −0.46)	−1.14 (−2.14, −0.14)	−0.92 (−1.92, 0.08)	−0.22 (−1.13, 0.69)
FFMQ-SF NJ				0.40	0.21	0.04
<4	−0.48 (−0.98, 0.01)	−1.31 (−1.85, −0.78)	−0.86 (−1.38, −0.33)	−0.83 (−1.56, −0.10)	−0.37 (−1.09, 0.35)	−0.46 (−1.22, 0.30)
≥4	−0.64 (−1.11, −0.16)	−1.04 (−1.52, −0.56)	−1.65 (−2.16, −1.14)	−0.40 (−1.08, 0.28)	−1.01 (−1.71, −0.31)	0.61 (−0.09, 1.31)
Opioid use, past wk				0.74	0.09	0.18
No	−0.48 (−0.85, −0.11)	−1.10 (−1.48, −0.72)	−1.33 (−1.71, −0.95)	−0.62 (−1.15, −0.09)	−0.85 (−1.39, −0.31)	0.23 (−0.31, 0.77)
Yes	−1.25 (−2.29, −0.21)	−1.59 (−2.71, −0.48)	−0.56 (−1.89, 0.77)	−0.34 (−1.86, 1.18)	0.69 (−0.99, 2.37)	−1.04 (−2.78, 0.71)

Open in a new tab

Models adjust for age, sex, education, pain duration, and baseline pain bothersomeness score

^†

P-value for test of differences between subgroups in treatment effects at 8 weeks

Bold font indicates P < 0.05; only comparisons with P < 0.002 were statistically significant after Bonferroni adjustment for multiple comparisons.

The effect of CBT vs. MBSR on pain bothersomeness was greater among men (adjusted mean change on the 0–10 rating: CBT −1.51 vs. MBSR −0.93; difference −0.58 [−1.38, 0.22]), whereas MBSR was superior to CBT among women (adjusted mean change: CBT −0.93 vs. MBSR −1.50; difference: 0.56 [−0.09, 1.22]) (P = 0.03). MBSR was superior to UC among women (adjusted mean change: MBSR −1.50 vs. UC −0.36; difference −1.13 [−1.74, −0.53]) but not men (adjusted mean change: MBSR −0.93 vs. UC −1.15; difference 0.21 [−0.71, 1.13]) (P = 0.02).

Baseline level of depressive symptoms was a potential moderator of the effects of CBT and MBSR on the 0–10 pain bothersomeness rating when using a PHQ-8 cut point of 5 (mild level of symptoms or greater) but not when using a cut point of 10 (moderate or greater level of symptoms). CBT had greater benefit than MBSR among those with mild or greater depressive symptom severity (adjusted mean change: CBT −1.30 vs. MBSR −0.81; difference: −0.49 [−1.19, 0.20]), but MBSR had greater benefit than CBT among those with minimal or no depressive symptoms (adjusted mean change: CBT −0.99 vs. MBSR −1.76; difference: 0.77 [0.03, 1.51]) (P = 0.01). MBSR was superior to UC among those with minimal depressive symptoms (adjusted mean change: MBSR −1.76 vs. UC −0.31; difference: −1.45 [−2.16, −0.73]) but not among those with mild or greater depressive symptoms (adjusted mean change: MBSR −0.81 vs. UC −0.83; difference: 0.02 [−0.68, 0.72]) (P = 0.004). The effect of CBT vs. UC on pain did not differ based on baseline depressive symptom severity.

Conviction that pain is a sign of bodily damage and the need to avoid activity, as measured by the SOPA Harm scale, was a potential moderator of the effects of CBT vs. UC on pain bothersomeness. CBT was superior to UC among participants with low conviction that pain is a sign of physical harm (adjusted mean change: CBT −1.54 vs. UC −0.39; difference: −1.14 [−1.84, −0.45]) but not among those with greater conviction that pain is a sign of harm (adjusted mean change: CBT −0.62 vs. UC −0.72; difference: 0.10 [−0.62, 0.83]) (P = 0.02). The effects of MBSR on pain did not differ by SOPA Harm scores.

MBSR was more beneficial than UC among participants with greater pain self-efficacy at baseline (adjusted mean change: MBSR −1.63 vs. UC −0.28; difference: −1.35 [−2.14, −0.56]) but not among participants with lower pain self-efficacy at baseline (adjusted mean change: MBSR −1.02 vs. UC −0.77; difference: −0.25 [−0.90, 0.40]) (P = 0.03). Pain self-efficacy did not modify the effects of CBT on pain bothersomeness.

Finally, the tendency to engage in a non-evaluative stance towards thoughts and feelings, as measured by the FFMQ-SF nonjudging subscale, was a potential moderator of the effects of CBT vs. MBSR on pain bothersomeness. The benefit of CBT was greater among participants who reported being less nonjudging of their thoughts and feelings (adjusted mean change: CBT −1.31 vs. MBSR −0.86; difference: −0.46 [−1.22, 0.30]), whereas the benefit of MBSR was greater among those who reported being more nonjudgmental (adjusted mean change: CBT −1.04 vs. MBSR −1.65; difference: 0.61 [−0.09, 1.31]) (P = 0.04).

Moderators of Intervention Effects on Depressive Symptoms

When examining depressive symptoms as the outcome, we observed the largest differences for the baseline depressive symptom subgroups (see Table 4), but no comparisons were statistically significant after Bonferroni correction. When using a PHQ-8 cut point of 10 (moderate or greater level of symptoms), CBT had greater benefit than UC among participants with moderate or severe baseline depressive symptoms (adjusted mean change on the 0–24 depressive symptom scale: CBT −5.96 vs. UC −1.78; difference: −4.18 [−6.40, −1.97]) but not among those with mild or minimal symptoms at baseline (adjusted mean change: CBT −1.45 vs. UC 0.20; difference: −1.65 [−2.62, −0.68]) (P = 0.04). The benefit of MBSR vs. UC for depressive symptoms was greater among participants with moderate or severe baseline depressive symptoms (adjusted mean change: MBSR – 5.83 vs. UC −1.78; difference: −4.04 [−6.34, −1.75]) than among those with mild or minimal symptoms at baseline (adjusted mean change: MBSR −0.70 vs. UC 0.20; difference: −0.90 [−1.87, 0.08]) (P = 0.01). When using a cut point of 5 (mild level of symptoms or greater) to create baseline subgroups, improvement in depression was greater with CBT than with UC among participants with mild or greater baseline depressive symptoms (adjusted mean change: CBT −3.95 vs. UC −0.86; difference: −3.09 [−4.34, −1.84]) than among those with minimal symptoms at baseline (adjusted mean change: CBT −0.43 vs. UC 0.74; difference: 1.17 [−2.47, 0.13]) (P = 0.04).

Table 4:

Depressive symptoms (PHQ-8): Adjusted mean change and treatment effects at 8 weeks for subgroups defined by baseline characteristics.

	Adjusted mean change at 8 weeks			Treatment effect^*
	UC	CBT	MBSR	CBT-UC	MBSR-UC	CBT-MBSR
Baseline characteristic	Mean (95% CI)	Mean (95% CI)	Mean (95% CI)	P-value^† Mean (95% CI)	P-value^† Mean (95% CI)	P-value^† Mean (95% CI)
Overall	−0.11 (−0.68, 0.45)	−2.23 (−2.82, −1.63)	−1.58 (−2.18, −0.98)	−2.11 (−2.94, −1.29)	−1.47 (−2.30, −0.63)	−0.65 (−1.49, 0.20)
Age at baseline				0.57	0.96	0.53
<60 yrs	0.13 (−0.53, 0.78)	−2.16 (−2.87, −1.45)	−1.35 (−2.08, −0.62)	−2.29 (−3.25, −1.32)	−1.48 (−2.46, −0.49)	−0.81 (−1.83, 0.21)
≥60 yrs	−0.69 (−1.88, 0.49)	−2.44 (−3.55, −1.34)	−2.22 (−3.32, −1.13)	−1.75 (−3.37, −0.13)	−1.53 (−3.15, 0.09)	−0.22 (−1.78, 1.34)
Gender				0.83	0.98	0.85
Male	−0.16 (−1.33, 1.02)	−2.40 (−3.32, −1.48)	−1.65 (−2.61, −0.68)	−2.24 (−3.73, −0.75)	−1.49 (−3.01, 0.03)	−0.75 (−2.08, 0.59)
Female	−0.08 (−0.72, 0.57)	−2.12 (−2.89, −1.36)	−1.55 (−2.33, −0.77)	−2.05 (−3.05, −1.04)	−1.47 (−2.48, −0.46)	−0.58 (−1.67, 0.52)
Education				0.16	0.85	0.12
Less than bachelor’s degree	0.32 (−0.61, 1.26)	−2.57 (−3.57, −1.57)	−1.08 (−1.97, −0.18)	−2.89 (−4.26, −1.52)	−1.40 (−2.70, −0.11)	−1.49 (−2.83, −0.15)
Bachelor’s degree	−0.38 (−1.09, 0.33)	−2.07 (−2.79, −1.34)	−1.94 (−2.76, −1.13)	−1.69 (−2.71, −0.66)	−1.56 (−2.64, −0.49)	−0.12 (−1.21, 0.97)
Functional limitations				0.11	0.23	0.73
mRDQ <14	−0.24 (−0.92, 0.43)	−1.89 (−2.63, −1.14)	−1.37 (−2.11, −0.64)	−1.64 (−2.64, −0.64)	−1.13 (−2.13, −0.13)	−0.51 (−1.55, 0.52)
mRDQ ≥14	0.21 (−0.85, 1.28)	−2.86 (−3.88, −1.84)	−2.03 (−3.11, −0.94)	−3.07 (−4.51, −1.62)	−2.24 (−3.74, −0.74)	−0.83 (−2.29, 0.63)
Pain bothersomeness				0.60	0.28	0.60
<7	−0.39 (−1.1, 0.33)	−2.30 (−3.01, −1.60)	−1.51 (−2.25, −0.76)	−1.92 (−2.94, −0.90)	−1.12 (−2.16, −0.08)	−0.8 (−1.82, 0.23)
≥7	0.36 (−0.58, 1.30)	−2.03 (−3.12, −0.94)	−1.72 (−2.76, −0.69)	−2.39 (−3.82, −0.96)	−2.08 (−3.48, −0.69)	−0.31 (−1.81, 1.19)
Depressive symptoms				0.04	0.01	0.62
PHQ-8 <10	0.20 (−0.46, 0.86)	−1.45 (−2.16, −0.74)	−0.70 (−1.42, 0.01)	−1.65 (−2.62, −0.68)	−0.90 (−1.87, 0.08)	−0.75 (−1.75, 0.25).
PHQ-8 ≥10	−1.78 (−3.44, −0.12)	−5.96 (−7.42, −4.50)	−5.83 (−7.45, −4.21)	−4.18 (−6.40, −1.97)	−4.04 (−6.34, −1.75)	−0.14 (−2.33, 2.05)
Depressive symptoms				0.04	0.12	0.60
PHQ-8 <5	0.74 (−0.14, 1.63)	−0.43 (−1.38, 0.52)	−0.02 (−0.96, 0.92)	−1.17 (−2.47, 0.13)	−0.77 (−2.01, 0.53)	−0.41 (−1.75, 0.93)
PHQ-8 ≥5	−0.86 (−1.74, 0.03)	−3.95 (−4.83, −3.06)	−3.05 (−3.97, −2.12)	−3.09 (−4.34, −1.84)	−2.19 (−3.47, −0.91)	−0.90 (−2.18, 0.38)
Expected pain improvement				0.83	0.90	0.73
1–2 (pain gone or much better)	−0.09 (−1.15, 0.97)	−2.35 (−3.49, −1.20)	−1.46 (−2.55, −0.37)	−2.26 (−3.80, −0.71)	−1.37 (−2.89, 0.15)	−0.88 (−2.46, 0.69)
3–7 (pain worse, unchanged, or somewhat better)	−0.12 (−0.80, 0.56)	−2.18 (−2.89, −1.47)	−1.62 (−2.36, −0.88)	−2.05 (−3.04, −1.06)	−1.49 (−2.50, −0.49)	−0.56 (−1.58, 0.46)
Expected program helpfulness				0.32	0.75	0.20
<8	0.04 (−0.83, 0.91)	−2.50 (−3.32, −1.68)	−1.27 (−2.20, −0.34)	−2.54 (−3.73, −1.34)	−1.31 (−2.58, −0.04)	−1.23 (−2.46, 0.01)
≥8	−0.23 (−0.99, 0.52)	−1.93 (−2.78, −1.08)	−1.81 (−2.61, −1.01)	−1.69 (−2.84, −0.55)	−1.58 (−2.68, −0.47)	−0.12 (−1.29, 1.05)
Widespread pain				0.58	0.95	0.62
<4	−0.44 (−1.45, 0.57)	−2.23 (−3.18, −1.29)	−1.84 (−2.79, −0.88)	−1.79 (−3.16, −0.42)	−1.40 (−2.78, −0.01)	−0.40 (−1.71, 0.92)
≥4	0.05 (−0.65, 0.75)	−2.23 (−3.00, −1.45)	−1.40 (−2.19, −0.60)	−2.28 (−3.32, −1.24)	−1.45 (−2.50, −0.40)	−0.83 (−1.93, 0.27)
Anxiety				0.13	0.98	0.13
GAD-2 <3	−0.35 (−0.97, 0.26)	−2.23 (−2.88, −1.59)	−1.90 (−2.57, −1.23)	−1.88 (−2.77, −0.99)	−1.55 (−2.46, −0.64)	−0.33 (−1.26, 0.59)
GAD-2 ≥3	1.35 (−0.16, 2.86)	−2.23 (−3.77, −0.7)	−0.16 (−1.65, 1.32)	−3.58 (−5.63, −1.54)	−1.51 (−3.53, 0.50)	−2.07 (−4.13, −0.01)
SOPA Disability				0.73	0.18	0.10
<2	−0.21 (−0.92, 0.51)	−2.46 (−3.24, −1.68)	−1.18 (−1.99, −0.37)	−2.25 (−3.31, −1.19)	−0.97 (−2.05, 0.11)	−1.28 (−2.41, −0.16)
≥2	0.04 (−0.89, 0.96)	−1.92 (−2.82, −1.02)	−2.08 (−2.99, −1.17)	−1.96 (−3.26, −0.66)	−2.12 (−3.43, −0.82)	0.16 (−1.11, 1.44)
SOPA Harm				0.27	0.89	0.36
<2	−0.36 (−1.21, 0.50)	−2.02 (−2.8, −1.25)	−1.73 (−2.55, −0.90)	−1.67 (−2.82, −0.51)	−1.37 (−2.57, −0.17)	−0.30 (−1.43, 0.83)
≥2	0.08 (−0.68, 0.85)	−2.51 (−3.44, −1.59)	−1.40 (−2.30, −0.50)	−2.60 (−3.80, −1.39)	−1.49 (−2.67, −0.31)	−1.11 (−2.40, 0.18)
SOPA Control				0.75	0.78	0.56
<3	0.31 (−0.40, 1.03)	−1.89 (−2.63, −1.15)	−1.03 (−1.80, −0.26)	−2.20 (−3.23, −1.17)	−1.34 (−2.40, −0.28)	−0.86 (−1.94, 0.21)
≥3	−0.83 (−1.74, 0.07)	−2.77 (−3.72, −1.82)	−2.42 (−3.38, −1.45)	−1.93 (−3.25, −0.62)	−1.58 (−2.89, −0.28)	−0.35 (−1.70, 1.00)
SOPA Emotion				0.12	0.64	0.30
<3	−0.39 (−1.11, 0.33)	−1.96 (−2.77, −1.15)	−1.72 (−2.51, −0.92)	−1.57 (−2.65, −0.49)	−1.33 (−2.40, −0.26)	−0.24 (−1.36, 0.88)
≥3	0.34 (−0.59, 1.27)	−2.55 (−3.44, −1.66)	−1.40 (−2.35, −0.44)	−2.89 (−4.17, −1.61)	−1.74 (−3.05, −0.42)	−1.15 (−2.45, 0.14)
Pain self-efficacy				0.55	0.67	0.88
PSEQ <50	0.25 (−0.48, 0.98)	−2.01 (−2.83, −1.19)	−1.34 (−2.12, −0.57)	−2.26 (−3.36, −1.17)	−1.60 (−2.66, −0.53)	−0.67 (−1.79, 0.45)
PSEQ ≥50	−0.69 (−1.60, 0.22)	−2.45 (−3.32, −1.58)	−1.92 (−2.87, −0.96)	−1.76 (−3.02, −0.51)	−1.23 (−2.54, 0.09)	−0.54 (−1.83, 0.76)
Pain catastrophizing				0.81	0.77	0.61
PCS <30	0.50 (−1.11, 2.11)	−1.33 (−3.09, 0.42)	−1.26 (−2.87, 0.34)	−1.83 (−4.20, 0.53)	−1.77 (−4.04, 0.51)	−0.07 (−2.44, 2.30)
PCS ≥30	−0.21 (−0.82, 0.40)	−2.35 (−2.98, −1.72)	−1.62 (−2.28, −0.96)	−2.14 (−3.02, −1.26)	−1.41 (−2.31, −0.51)	−0.73 (−1.64, 0.18)
Pain acceptance				0.89	0.55	0.65
CPAQ-8 <32	−0.17 (−0.96, 0.62)	−2.22 (−3.05, −1.38)	−1.37 (−2.25, −0.49)	−2.05 (−3.20, −0.90)	−1.2 (−2.39, −0.02)	−0.85 (−2.04, 0.35)
CPAQ-8 ≥32	−0.06 (−0.89, 0.77)	−2.23 (−3.09, −1.37)	−1.77 (−2.62, −0.93)	−2.17 (−3.36, −0.98)	−1.71 (−2.89, −0.54)	−0.46 (−1.66, 0.75)
FFMQ-SF OB				0.67	0.13	0.28
<4	−0.02 (−0.81, 0.78)	−2.30 (−3.10, −1.50)	−2.00 (−2.77, −1.23)	−2.28 (−3.41, −1.15)	−1.99 (−3.10, −0.88)	−0.29 (−1.40, 0.81)
≥4	−0.21 (−1.01, 0.59)	−2.14 (−3.02, −1.26)	−0.90 (−1.88, 0.08)	−1.93 (−3.12, −0.73)	−0.69 (−1.95, 0.58)	−1.24 (−2.55, 0.07)
FFMQ-SF AA				0.93	0.60	0.54
<4	−0.04 (−0.72, 0.64)	−2.18 (−2.92, −1.43)	−1.34 (−2.09, −0.60)	−2.14 (−3.15, −1.13)	−1.31 (−2.31, −0.30)	−0.83 (−1.88, 0.22)
≥4	−0.27 (−1.33, 0.78)	−2.33 (−3.33, −1.32)	−2.05 (−3.11, −1.00)	2.06 (−3.50, −0.61)	−1.78 (−3.27, −0.29)	−0.28 (−1.71, 1.15)
FFMQ-SF NR				0.75	0.17	0.26
<4	−0.23 (−0.87, 0.41)	−2.32 (−3.03, −1.61)	−1.33 (−2.08, −0.59)	−2.09 (−3.05, −1.13)	−1.10 (−2.08, −0.13)	−0.99 (−2.02, 0.04)
≥4	0.38 (−0.87, 1.64)	−2.02 (−3.11, −0.93)	−2.07 (−3.12, −1.02)	−2.40 (−4.07, −0.73)	−2.45 (−4.10, −0.80)	0.05 (−1.44, 1.55)
FFMQ-SF NJ				0.26	0.93	0.24
<4	0.46 (−0.36, 1.28)	−2.13 (−3.05, −1.22)	−0.96 (−1.85, −0.07)	−2.60 (−3.80, −1.39)	−1.43 (−2.61, −0.24)	−1.17 (−2.43, 0.08)
≥4	−0.65 (−1.45, 0.15)	−2.30 (−3.09, −1.51)	−2.15 (−3.00, −1.30)	−1.65 (−2.76, −0.54)	−1.50 (−2.65, −0.35)	−0.15 (−1.29, 0.98)
Opioid use, past wk				0.80	0.45	0.34
No	−0.14 (−0.75, 0.47)	−2.21 (−2.84, −1.58)	−1.69 (−2.33, −1.06)	−2.07 (−2.95, −1.19)	−1.55 (−2.44, −0.67)	−0.52 (−1.41, 0.38)
Yes	0.07 (−1.65, 1.79)	−2.34 (−4.19, −0.50)	−0.41 (−2.46, 1.63)	−2.41 (−4.90, 0.08)	−0.48 (−3.12, 2.16)	−1.93 (−4.68, 0.82)

Open in a new tab

Models adjust for age, sex, education, pain duration, and baseline PHQ-8 score

^†

P-value for test of differences between subgroups in treatment effects at 8 weeks

Bold font indicates large effect sizes of interest; only those with P < 0.002 were statistically significant after Bonferroni adjustment for multiple comparisons.

Post-Hoc Analysis for Moderators

Although the different treatment effects based on baseline pain self-efficacy were not statistically significant after Bonferroni correction, given the sizes of the treatment effects and the fact that differences were found for both functional limitations (CBT vs. UC) and pain bothersomeness (MBSR vs. UC), we conducted a post-hoc analysis to examine whether those with high vs. low baseline pain self-efficacy differed in CBT and MBSR class attendance. We speculated that those with higher pain self-efficacy at baseline might have been more likely to attend more treatment sessions (due to belief that they could successfully apply the techniques taught to manage their pain) and thus benefit more from treatment. We previously reported that among the participants randomized to CBT or MBSR, those who completed at least 6 sessions had significantly higher pain self-efficacy scores at baseline.⁷⁸ Our additional analyses revealed that, among those with high self-efficacy, 63% attended 6 or more classes vs. 47% of those with lower scores (chi-square = 6.10, P = 0.01).

Nonspecific Predictors of Improvement

Table 5 shows the results of regression models estimating the associations between baseline predictors and baseline to 8-week change in outcomes, adjusting for treatment allocation and the prespecified covariates. Stronger beliefs in one’s personal control over pain (SOPA Control; P = 0.002) and greater endorsement of catastrophic thinking in response to pain (PCS; P = 0.005) were associated with greater improvement in functional limitations for participants regardless of assigned treatment group. Stronger belief in one’s personal control over pain (P = 0.03) was also associated with greater improvement in pain bothersomeness, as was college education (P = 0.049). Nonspecific baseline predictors of greater improvement in depressive symptoms were older age (P = 0.007), less widespread pain (P = 0.009), lower anxiety (GAD-2; P < 0.001), greater belief in one’s control over pain (SOPA Control; P < 0.001), and greater endorsement of mindfulness on the FFMQ-SF Acting with Awareness (P = 0.008) and Nonjudging (P = 0.008) scales. After Bonferroni correction for multiple comparisons, only anxiety and belief in one’s control over pain were statistically significant predictors of improvement in depression. No predictors of function or pain were statistically significant after correction.

Table 5:

Nonspecific predictors of baseline to 8-week improvement on measures of functional limitations (mRDQ), pain bothersomeness, and depressive symptoms (PHQ-8).

	mRDQ		Pain Bothersomeness		PHQ-8
Baseline characteristic	ß (95% CI)^*	P-value	ß (95% CI)^*	P-value	ß (95% CI)^*	P-value
Age^†	0.03 (−0.01, 0.07)	0.21	0.01 (−0.01, 0.03)	0.33	−0.04 (−0.07, −0.01)	0.007
Gender (female vs male)	−0.17 (−1.23, 0.88)	0.75	0.24 (−0.20, 0.69)	0.29	0.16 (−0.57, 0.88)	0.67
Education (bachelor’s degree vs less education)	−0.34 (−1.37, 0.69)	0.52	−0.43 (−0.85, −0.00)	0.049	−0.38 (−1.08, 0.32)	0.28
Functional limitations (mRDQ)	−0.41 (−0.52, −0.30)	<0.001	0.01 (−0.04, 0.06)	0.60	−0.01 (−0.09, 0.07)	0.79
Pain bothersomeness	0.17 (−0.16, 0.51)	0.31	−0.54 (−0.68, −0.41)	<0.001	0.14 (−0.07, 0.36)	0.20
Depressive symptoms (PHQ-8)	0.11 (−0.02, 0.24)	0.09	0.00 (−0.05, 0.05)	0.94	−0.47 (−0.56, −0.39)	<0.01
Expected pain improvement	−0.13 (−0.47, 0.21)	0.47	0.10 (−0.04, 0.25)	0.16	−0.01 (−0.25, 0.22)	0.91
Expected program helpfulness	−0.01 (−0.27, 0.26)	0.96	−0.00 (−0.12, 0.11)	0.94	−0.10 (−0.29, 0.09)	0.29
Widespread pain	0.06 (−0.15, 0.28)	0.57	0.03 (−0.06, 0.12)	0.54	0.20 (0.05, 0.35)	0.009
Anxiety (GAD-2)	0.05 (−0.30, 0.39)	0.79	−0.01 (−0.16, 0.13)	0.88	0.54 (0.25, 0.84)	<0.001
SOPA Disability	0.03 (−0.47, 0.52)	0.92	0.04 (−0.16, 0.24)	0.71	0.10 (−0.24, 0.43)	0.56
SOPA Harm	0.02 (−0.54, 0.58)	0.95	0.08 (−0.16, 0.31)	0.53	0.15 (−0.23, 0.53)	0.45
SOPA Control	−0.76 (−1.25, −0.27)	0.002	−0.24 (−0.45, −0.03)	0.03	−0.67 (−1.00, −0.34)	<0.001
SOPA Emotion	0.03 (−0.40, 0.47)	0.88	0.05 (−0.13, 0.22)	0.60	0.06 (−0.26, 0.37)	0.71
Pain catastrophizing (PCS)	−0.12 (−0.20, −0.03)	0.005	−0.00 (−0.03, 0.03)	0.85	−0.00 (−0.05, 0.04)	0.92
Pain acceptance (CPAQ-8)	−0.06 (−0.15, 0.03)	0.17	−0.00 (−0.03, 0.03)	0.99	0.01 (−0.05, 0.07)	0.79
Pain self-efficacy (PSEQ)	−0.04 (−0.10, 0.03)	0.27	0.01 (−0.02, 0.03)	0.49	−0.02 (−0.07, 0.02)	0.26
FFMQ-SF OB	0.38 (−0.22, 0.99)	0.21	0.21 (−0.05, 0.46)	0.12	0.20 (−0.21, 0.62)	0.34
FFMQ-SF AA	−0.08 (−0.79, 0.63)	0.83	0.04 (−0.26, 0.34)	0.80	−0.70 (−1.22, −0.18)	0.008
FFMQ-SF NR	−0.02 (−0.65, 0.62)	0.96	0.13 (−0.14, 0.40)	0.35	−0.27 (−0.72, 0.17)	0.23
FFMQ-SF NJ	−0.48 (−1.03, 0.07)	0.09	−0.05 (−0.29, 0.18)	0.65	−0.58 (−1.00, −0.15)	0.008
Opioid use, past wk (vs no use)	1.03 (−0.62, 2.68)	0.22	−0.28 (−0.98, 0.42)	0.43	0.40 (−0.75, 1.55)	0.50

Open in a new tab

Adjusted regression coefficients and 95% CIs were estimated using regression models that included the predictor of interest and adjusting for an indicator of treatment allocation and age, gender, education, pain duration, and baseline value of the outcome measure. The dependent variable was the change from baseline on the outcome measure. Regression coefficients reflect the average effect across all study arms. Negative regression coefficients indicate that higher values of the baseline characteristic were associated with greater improvement on the outcome measure; positive regression coefficients indicate that lower values of the baseline characteristic were associated with greater improvement on the outcome measures. For binary variables, the coefficient is the difference in adjusted mean change on the outcome measure between the two groups (e.g., for females, the decrease in mRDQ scores was 0.17 points more than it was for males; i.e., greater improvement in functional limitations). For continuous variables, the coefficient is the difference in adjusted mean change on the outcome measure associated with a 1-unit change in the predictor.

^†

Coefficient for age represents the change in outcome associated with a 5-year difference in age Bold font indicates P < 0.05; only associations with P < 0.002 were statistically significant after Bonferroni adjustment for multiple comparisons.

Summary

Table 6 summarizes the baseline patient characteristics examined, and those hypothesized and observed (before and after Bonferroni correction) to be moderators and nonspecific predictors across the three outcomes examined. It can be seen that few characteristics were statistically significant after adjustment for the familywise error rate and there were inconsistencies in findings across outcomes examined.

Table 6:

Baseline characteristics examined: hypothesized (●) moderators and nonspecific predictors of 8-week improvement in functional limitations (mRDQ), pain bothersomeness, and depressive symptoms (PHQ-8); effects with P < 0.05 but not statistically significant after Bonferroni correction (X) and statistically significant after Bonferroni correction (B).

	Moderators									Nonspecific Predictors
	CBT – UC			MBSR - UC			CBT - MBSR
Baseline characteristic	mRDQ	Pain Bothersomeness	PHQ-8	mRDQ	Pain Bothersomeness	PHQ-8	mRDQ	Pain Bothersomeness	PHQ-8	mRDQ	Pain Bothersomeness	PHQ-8
Age												X
Gender (female vs male)					X			X
Education (bachelor’s degree vs less education)										●	● X	●
Depressive symptoms, moderate-severe (PHQ-8 ≥10)	●	●	● X	●	●	● X	●	●	●
Depressive symptoms, mild or more severe (PHQ-8 ≥5)	●	●	●X	●	● X	●	●	● X	●
Pain bothersomeness	●	●	●	●	●	●	●	●	●	●	●	●
Functional limitations (mRDQ)	●	●	●	●	●	●	●	●	●	●	●	●
Widespread pain										●	●	● X
Expected pain improvement										●	●	●
Expected program helpfulness
Anxiety (GAD-2)	●	●	●	●	●	●	●	●	●			B
SOPA Disability
SOPA Harm		X
SOPA Control										X	X	B
SOPA Emotion
Pain self-efficacy (PSEQ)	X				X
Pain catastrophizing (PCS)										X
Pain acceptance (CPAQ-8)
FFMQ-SF OB
FFMQ-SF AA												X
FFMQ-SF NR
FFMQ-SF NJ							B	X				X
Opioid use, past wk (vs no use)

Open in a new tab

● Hypothesized moderator or nonspecific predictor

X Effect with P < 0.05

B Effect that is statistically significant after Bonferroni correction.

Discussion

To our knowledge, this is the first report from a large RCT of moderators and nonspecific predictors of treatment response to CBT vs. MBSR vs. UC for CLBP. After adjusting for multiple comparisons, only one baseline participant characteristic was associated with a statistically significant difference in response to CBT vs. MBSR. Below we discuss this finding, as well as results that, although not statistically significant after adjustment, may merit further study given their size and the possibility that Bonferroni correction may have resulted in defining as non-significant some effects that were “real.”

The only potential moderator that was statistically significant after correction was the nonjudging dimension of mindfulness. Although differences in effects of CBT vs. MBSR based on nonjudging were observed for both function and pain, the difference was statistically significant after correction only for function. Among participants who reported being less nonjudging of their thoughts and feelings, benefits were greater from CBT; among those who reported being more nonjudging, benefits were greater from MBSR. If confirmed in further studies, this finding suggests the potential clinical utility of this measure for guiding patients to MBSR vs. CBT.

In a trial of treatments for CLBP similar but not identical to those in our trial,¹⁹ participants reporting high levels of another dimension of mindfulness, mindful nonreactivity, showed greater improvement in physical function with mindfulness-based cognitive therapy than with mindfulness meditation. Consistent with our results, a moderating effect was not found for the FFMQ-SF Observing scale.¹⁹ Nonjudging was not examined. Although we did not find a moderating effect for nonreactivity, taken as a whole, these results support the potential importance of mindfulness dimensions in selecting patients for mind-body interventions and in future research regarding moderators and treatment mechanisms.

As hypothesized, participants with greater depressive symptoms at baseline showed greater improvement in depressive symptoms with CBT and MBSR vs. UC, and with CBT vs. MBSR, although no comparisons were statistically significant after correction. These results suggest that the finding in the MAP RCT of CBT having greater benefit than MBSR for depression post-treatment¹³ may have been due largely to treatment differences among those who were more depressed. A meta-analysis of mind-body treatments for individuals with chronic pain found that mindfulness-based interventions are less effective for depression as compared with acceptance and commitment therapy, a type of CBT.⁸⁹ Another meta-analysis found that MBSR was not effective for mental health among patients with back pain.¹ Further research is needed to confirm the possibility that more depressed individuals with chronic pain may show greater improvement in depression with CBT for pain than with MBSR.

Our hypotheses that the effects of CBT and MBSR vs. UC, and of CBT vs. MBSR, would be greater among participants with higher baseline levels of anxiety, pain, and disability were not confirmed. At the time we planned the study, little research had evaluated the effectiveness of MBSR for these problems. Subsequent research has provided more support for the effectiveness of mindfulness interventions as a group for depression and anxiety,²⁹ but not for benefits of MBSR for mental health or long-term benefits for pain and function among patients with back pain.¹

Two exploratory psychological variables were identified as potential moderators: pain self-efficacy and belief that pain is a sign of physical harm. Prior studies have found both to predict pain outcomes.^20,37 Participants with higher pain self-efficacy benefited more from CBT vs. UC for functional limitations and from MBSR vs. UC for pain bothersomeness, although neither finding remained statistically significant after correction. A study of CBT for temporomandibular disorder pain also found greater benefits for those with higher pain self-efficacy.⁵¹ If individuals believe they can self-manage their pain, they may be more likely to engage in, and hence benefit from, treatment.⁷² This hypothesis is supported by our finding that participants with higher pain self-efficacy were significantly more likely to attend 6 or more classes of either intervention.

Participants with a low baseline belief that pain is a sign of physical harm improved more in pain bothersomeness from CBT vs. UC, although the effect was not statistically significant after correction. If this finding is confirmed in future studies, we speculate that patients with low harm beliefs, like those with greater pain self-efficacy, may be more receptive to engaging in CBT and practicing the skills taught, and hence more likely to benefit. Those with greater harm beliefs may be more focused on correcting the perceived biological causes of pain. Harm beliefs did not moderate any MBSR effects or any effects on functional limitations.

An exploratory demographic variable was identified as a potential moderator, although it was not statistically significant after correction: CBT improved pain bothersomeness more than did MBSR among men, whereas MBSR was more effective for pain than was CBT among women. We are unaware of any prior research regarding gender differences in effects of mind-body treatments on pain. If this finding is confirmed, it has potential to be a fruitful area for future research. Treatment outcome did not vary substantially by other demographic characteristics examined, including age, which was found in a prior study to moderate effects of CBT for CLBP on function (greater response among younger patients).⁸⁶

A secondary objective of this study was to identify participant characteristics associated with improvement regardless of treatment. Such knowledge may help inform clinical decision-making and covariate selection in clinical trials. Consistent with our hypothesis and prior studies,^16,84 college-educated participants showed greater improvement in pain bothersomeness than did those with less education, although this was not statistically significant after correction. Given the potential for inequity in pain outcomes based on education level, it may be important to incorporate literacy-adapted CBT, which has been shown to be effective for low literacy populations.^74,88

Among our other hypothesized nonspecific predictors (widespread pain, pain bothersomeness, functional limitations, expectations of improvement), only less widespread pain predicted improvement, and this association was not statistically significant after correction. Contrary to our hypotheses and past research,^{52,56,81,84,91} we did not find that higher expectations of improvement or lower levels of pain or disability at baseline predicted better outcomes at 8 weeks.

In exploratory analyses, lower anxiety and greater perceived control over pain were significantly associated with better depression outcomes irrespective of treatment. Multiple studies have reported similar findings for perceived control;^11,20,38 as a group, these findings support the importance of perceived control and the potential benefit of interventions designed to increase patients’ confidence in their ability to control their pain.

A number of findings in our moderator and predictor analyses were statistically significant prior to, but not after, correcting for familywise Type I error rate. Previous studies provide conflicting evidence for pain self-efficacy, pain catastrophizing, depressive symptoms, and treatment expectations as moderators and predictors in behavioral trials for pain.^{5,19,51,56,81,86} Differences in findings may reflect differences in sampling, interventions, measures, and statistical power to detect effects. It is possible that some of our findings were originally statistically significant due to chance, but we note that the RCT was not powered to detect moderator effects and we applied a conservative correction. Further research is needed to replicate our findings, which may be useful in generating hypotheses for confirmatory studies.

We acknowledge other study limitations. Approximately half of the participants randomized to MBSR or CBT attended at least 6 of the 8 sessions. Results could differ in studies with different numbers of sessions or treatment adherence rates. Study participants volunteered for a study of mind-body treatments for pain, had generally high expectations of benefitting, and had fairly low levels of psychosocial distress. The extent to which these findings would generalize to more highly distressed and disabled patients or those less receptive to psychosocial treatments for pain is unclear. Finally, CBT was delivered in a group format in our study and results could differ for CBT delivered individually.

Despite these limitations, our study, by leveraging data from a large RCT, offers unique insights into potential moderators and nonspecific predictors of improvement with two evidence-based treatments for chronic pain, CBT and MBSR. Although only one potential moderator was statistically significant after applying a conservative correction for familywise error rate, other subgroup differences could indicate potentially clinically meaningful differences. Our findings point to specific patient characteristics that show promise for further investigation. If confirmed in future studies, such findings could have utility for patient-treatment matching, which could lead to improved allocation of limited resources and more cost-effective care, as well as to enhanced understanding of treatments mechanisms and potential ways to increase efficacy.

In conclusion, when considering our findings in the context of other studies, consistent, strong associations have not been found between any patient baseline characteristic and differential response to psychosocial interventions for chronic pain.^{5,19,81,86,88} Our findings are reassuring that patients with CLBP, regardless of these baseline characteristics examined to date, could reasonably expect to benefit from either CBT or MBSR.⁴⁴ At least at this time, it appears reasonable for patient preference and treatment availability to guide treatment selection. Yet further research has the potential to shed light on variability in treatment response, optimal patient-treatment matching, and therapeutic mechanisms.

Highlights.

Few patient characteristics predicted benefits of CBT vs MBSR
Nonjudging of one’s thoughts/feelings predicted greater benefit from MBSR vs CBT
Less nonjudging of thoughts/feelings predicted greater benefit from CBT vs. MBSR
Pain control beliefs and lower anxiety predicted improvement regardless of treatment

Research Funding

This research was supported by the National Center for Complementary & Integrative Health of the National Institutes of Health under Award Number R01AT006226. Jessica A. Chen is supported by a VA Health Services Research & Development Career Development Award (IK2HX002866).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Declaration of Competing Interest

Judith A. Turner receives royalties from PAR, Inc. on sales of the Chronic Pain Coping Inventory (CPCI) and CPCI/Survey of Pain Attitudes (SOPA) score report software. She also has served as a consultant on research and clinical applications related to CBT and CBT-based treatments. Benjamin H. Balderson has served as a consultant on research evaluating CBT and MBSR treatments. The other authors report no conflicts of interest.

References

1.Anheyer D, Haller H, Barth J, Lauche R, Dobos G, Cramer H: Mindfulness-based stress reduction for treating low back pain: A systematic review and meta-analysis. Ann Intern Med 166:799, 2017. [DOI] [PubMed] [Google Scholar]
2.Artus M, Campbell P, Mallen CD, Dunn KM, van der Windt DAW: Generic prognostic factors for musculoskeletal pain in primary care: a systematic review. BMJ Open 7:e012901, 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Blacker M, Meleo-Meyer F, Kabat-Zinn J, Santorelli S: Stress reduction clinic mindfulness-based stress reduction (MBSR) curriculum guide. Worcester, MA: Center for Mindfulness in Medicine, Health Care and Society, University of Massachusetts Medical School; 2009. [Google Scholar]
4.Bohlmeijer E, ten Klooster PM, Fledderus M, Veehof M, Baer R: Psychometric properties of the Five Facet Mindfulness Questionnaire in depressed adults and development of a short form. Assessment 18:308–20, 2011. [DOI] [PubMed] [Google Scholar]
5.Broderick JE, Keefe FJ, Schneider S, Junghaenel DU, Bruckenthal P, Schwartz JE, Kaell AT, Caldwell DS, McKee D, Gould E: Cognitive behavioral therapy for chronic pain is effective, but for whom?: PAIN 157:2115–23, 2016. [DOI] [PubMed] [Google Scholar]
6.Brown MM, Arigo D, Wolever RQ, Smoski MJ, Hall MH, Brantley JG, Greeson JM: Do gender, anxiety, or sleep quality predict mindfulness-based stress reduction outcomes? J Health Psychol 26:2656–62, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Burgess R, Mansell G, Bishop A, Lewis M, Hill J: Predictors of functional outcome in musculoskeletal healthcare: An umbrella review. Eur J Pain 24:51–70, 2020. [DOI] [PubMed] [Google Scholar]
8.Burns JW, Jensen MP, Thorn B, Lillis TA, Carmody J, Newman AK, Keefe F: Cognitive therapy, mindfulness-based stress reduction, and behavior therapy for the treatment of chronic pain: randomized controlled trial. PAIN 163:376–89, 2022. [DOI] [PubMed] [Google Scholar]
9.Burns JW, Van Dyke BP, Newman AK, Morais CA, Thorn BE: Cognitive behavioral therapy (CBT) and pain education for people with chronic pain: Tests of treatment mechanisms. J Consult Clin Psychol 88:1008–18, 2020. [DOI] [PubMed] [Google Scholar]
10.Caudill M: Managing pain before it manages you. Rev. ed. New York, NY: Guilford Press; 2002. [Google Scholar]
11.Chen Y, Campbell P, Strauss VY, Foster NE, Jordan KP, Dunn KM: Trajectories and predictors of the long-term course of low back pain: cohort study with 5-year follow-up. PAIN 159:252–60, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Cherkin DC, Anderson ML, Sherman KJ, Balderson BH, Cook AJ, Hansen KE, Turner JA: Two-year follow-up of a randomized clinical trial of mindfulness-based stress reduction vs cognitive behavioral therapy or usual care for chronic low back pain. JAMA 317:642, 2017. [DOI] [PubMed] [Google Scholar]
13.Cherkin DC, Sherman KJ, Balderson BH, Cook AJ, Anderson ML, Hawkes RJ, Hansen KE, Turner JA: Effect of mindfulness-based stress reduction vs cognitive behavioral therapy or usual care on back pain and functional limitations in adults with chronic low back pain: a randomized clinical trial. JAMA 315:1240, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Cherkin DC, Sherman KJ, Balderson BH, Turner JA, Cook AJ, Stoelb B, Herman PM, Deyo RA, Hawkes RJ: Comparison of complementary and alternative medicine with conventional mind–body therapies for chronic back pain: protocol for the Mind–body Approaches to Pain (MAP) randomized controlled trial. Trials 15:211, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Chiesa A, Serretti A: Mindfulness-based interventions for chronic pain: a systematic review of the evidence. J Altern Complement Med 17:83–93, 2011. [DOI] [PubMed] [Google Scholar]
16.Costa L d. CM, Maher CG, McAuley JH, Hancock MJ, Herbert RD, Refshauge KM, Henschke N: Prognosis for patients with chronic low back pain: inception cohort study. BMJ 339:b3829–b3829, 2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Cuijpers P, Noma H, Karyotaki E, Cipriani A, Furukawa TA: Effectiveness and acceptability of cognitive behavior therapy delivery formats in adults with depression: A network meta-analysis. JAMA Psychiatry 76:700, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Day MA, Ehde DM, Jensen MP: Psychosocial pain management moderation: the limit, activate, and enhance model. J Pain 16:947–60, 2015. [DOI] [PubMed] [Google Scholar]
19.Day MA, Thorn BE, Ehde DM, Burns JW, Barnier A, Mattingley JB, Matthews N, Jensen MP: Moderators of mindfulness meditation, cognitive therapy, and mindfulness-based cognitive therapy for chronic low back pain: a test of the limit, activate, and enhance model. J Pain 21:161–9, 2020. [DOI] [PubMed] [Google Scholar]
20.de Rooij A, van der Leeden M, Roorda LD, Dekker J: Predictors of outcome of multidisciplinary treatment in chronic widespread pain: an observational study. BMC Musculoskelet Disord 13:133, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Driessen E, Hollon SD: Cognitive behavioral therapy for mood disorders: Efficacy, moderators and mediators. Psychiatr Clin North Am 33:537–55, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Dunn KM, Jordan KP, Croft PR: Contributions of prognostic factors for poor outcome in primary care low back pain patients. Eur J Pain 15:313–9, 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Ecija C, Catala P, Lopez-Gomez I, Bedmar D, Peñacoba C: What does the psychological flexibility model contribute to the relationship between depression and disability in chronic pain? The role of cognitive fusion and pain acceptance. Clin Nurs Res :1–13, 2021. [DOI] [PubMed] [Google Scholar]
24.Ehde DM, Dillworth TM, Turner JA: Cognitive-behavioral therapy for individuals with chronic pain: Efficacy, innovations, and directions for research. Am Psychol 69:153–66, 2014. [DOI] [PubMed] [Google Scholar]
25.Fish RA, Hogan MJ, Morrison TG, Stewart I, McGuire BE: Willing and able: a closer look at pain willingness and activity engagement on the Chronic Pain Acceptance Questionnaire (CPAQ-8). J Pain 14:233–45, 2013. [DOI] [PubMed] [Google Scholar]
26.Fish RA, McGuire B, Hogan M, Morrison TG, Stewart I: Validation of the Chronic Pain Acceptance Questionnaire (CPAQ) in an internet sample and development and preliminary validation of the CPAQ-8: Pain 149:435–43, 2010. [DOI] [PubMed] [Google Scholar]
27.Garland EL, Gaylord SA, Palsson O, Faurot K, Douglas Mann J, Whitehead WE: Therapeutic mechanisms of a mindfulness-based treatment for IBS: effects on visceral sensitivity, catastrophizing, and affective processing of pain sensations. J Behav Med 35:591–602, 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Gilpin HR, Keyes A, Stahl DR, Greig R, McCracken LM: Predictors of treatment outcome in contextual cognitive and behavioral therapies for chronic pain: A systematic review. J Pain 18:1153–64, 2017. [DOI] [PubMed] [Google Scholar]
29.Goldberg SB, Tucker RP, Greene PA, Davidson RJ, Wampold BE, Kearney DJ, Simpson TL: Mindfulness-based interventions for psychiatric disorders: A systematic review and meta-analysis. Clin Psychol Rev 59:52–60, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Gu J, Strauss C, Bond R, Cavanagh K: How do mindfulness-based cognitive therapy and mindfulness-based stress reduction improve mental health and wellbeing? A systematic review and meta-analysis of mediation studies. Clin Psychol Rev 37:1–12, 2015. [DOI] [PubMed] [Google Scholar]
31.Hayden JA, Wilson MN, Riley RD, Iles R, Pincus T, Ogilvie R: Individual recovery expectations and prognosis of outcomes in non-specific low back pain: prognostic factor review. Cochrane Database Syst Rev Issue 11:Art. No.: CD011284, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Jensen MP: Psychosocial approaches to pain management: An organizational framework. PAIN 152:717–25, 2011. [DOI] [PubMed] [Google Scholar]
33.Jensen MP, Karoly P: Pain-specific beliefs, perceived symptom severity, and adjustment to chronic pain. Clin J Pain United States; 8:123–30, 1992. [DOI] [PubMed] [Google Scholar]
34.Jensen MP, Karoly P: Survey of pain attitudes: professional manual. Lutz, FL: Psychological Assessment Resources; 2008. [Google Scholar]
35.Jensen MP, Romano JM, Turner JA, Good AB, Wald LH: Patient beliefs predict patient functioning: further support for a cognitive-behavioural model of chronic pain. PAIN 81:95–104, 1999. [DOI] [PubMed] [Google Scholar]
36.Jensen MP, Smith DG, Ehde DM, Robinson LR: Pain site and the effects of amputation pain: further clarification of the meaning of mild, moderate, and severe pain. PAIN 3:317–22, 2001. [DOI] [PubMed] [Google Scholar]
37.Jensen MP, Turner JA, Romano JM: Changes after multidisciplinary pain treatment in patient pain beliefs and coping are associated with concurrent changes in patient functioning. PAIN 131:38–47, 2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Jensen MP, Turner JA, Romano JM: Changes in beliefs, catastrophizing, and coping are associated with improvement in multidisciplinary pain treatment. J Consult Clin Psychol 69:655–62, 2001. [DOI] [PubMed] [Google Scholar]
39.Jensen MP, Turner JA, Romano JM, Lawler BK: Relationship of pain-specific beliefs to chronic pain adjustment. PAIN 57:301–9, 1994. [DOI] [PubMed] [Google Scholar]
40.Jones KR, Vojir CP, Hutt E, Fink R: Determining mild, moderate, and severe pain equivalency across pain-intensity tools in nursing home residents. J Rehabil Res Dev 44:305, 2007. [DOI] [PubMed] [Google Scholar]
41.Kabat-Zinn J: An outpatient program in behavioral medicine for chronic pain patients based on the practice of mindfulness meditation: Theoretical considerations and preliminary results. Gen Hosp Psychiatry 4:33–47, 1982. [DOI] [PubMed] [Google Scholar]
42.Kabat-Zinn J: Full catastrophe living: Using the wisdom of your body and mind to face stress, pain, and illness. New York, NY: Random House; 2005. [Google Scholar]
43.Kabat-Zinn J: Mindfulness-based interventions in context: past, present, and future. Clin Psychol Sci Pract 10:144–56, 2003. [Google Scholar]
44.Khoo E-L, Small R, Cheng W, Hatchard T, Glynn B, Rice DB, Skidmore B, Kenny S, Hutton B, Poulin PA: Comparative evaluation of group-based mindfulness-based stress reduction and cognitive behavioural therapy for the treatment and management of chronic pain: A systematic review and network meta-analysis. Evid Based Ment Health 22:26–35, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Kraemer H, Wilson G, Fairburn CG, Agras W: Mediators and moderators of treatment effects in randomized clinical trials. Arch Gen Psychiatry 59:877–83, 2002. [DOI] [PubMed] [Google Scholar]
46.Kroenke K, Spitzer RL, Williams JBW: The PHQ-15: validity of a new measure for evaluating the severity of somatic symptoms. Psychosom Med 64:258–66, 2002. [DOI] [PubMed] [Google Scholar]
47.Kroenke K, Spitzer RL, Williams JBW, Löwe B: The Patient Health Questionnaire somatic, anxiety, and depressive symptom scales: a systematic review. Gen Hosp Psychiatry 32:345–59, 2010. [DOI] [PubMed] [Google Scholar]
48.Kroenke K, Spitzer RL, Williams JBW, Monahan PO, Löwe B: Anxiety disorders in primary care: prevalence, impairment, comorbidity, and detection. Ann Intern Med 146:317, 2007. [DOI] [PubMed] [Google Scholar]
49.Kroenke K, Strine TW, Spitzer RL, Williams JBW, Berry JT, Mokdad AH: The PHQ-8 as a measure of current depression in the general population. J Affect Disord 114:163–73, 2009. [DOI] [PubMed] [Google Scholar]
50.Lamb SE, Hansen Z, Lall R, Castelnuovo E, Withers EJ, Nichols V, Potter R, Underwood MR: Group cognitive behavioural treatment for low-back pain in primary care: a randomised controlled trial and cost-effectiveness analysis. The Lancet 375:916–23, 2010. [DOI] [PubMed] [Google Scholar]
51.Litt MD, Shafer DM, Kreutzer DL: Brief cognitive-behavioral treatment for TMD pain: Long-term outcomes and moderators of treatment. PAIN 151:110–6, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Mallen CD, Peat G, Thomas E, Dunn KM, Croft PR: Prognostic factors for musculoskeletal pain in primary care: a systematic review. Br J Gen Pract 57:655, 2007. [PMC free article] [PubMed] [Google Scholar]
53.Martinez-Calderon J, Jensen MP, Morales-Asencio JM, Luque-Suarez A: Pain catastrophizing and function in individuals with chronic musculoskeletal pain: A systematic review and meta-analysis. Clin J Pain 35:279–93, 2019. [DOI] [PubMed] [Google Scholar]
54.Martinez-Calderon J, Zamora-Campos C, Navarro-Ledesma S, Luque-Suarez A: The role of self-efficacy on the prognosis of chronic musculoskeletal pain: A systematic review. J Pain 19:10–34, 2018. [DOI] [PubMed] [Google Scholar]
55.Mascaro JS, Singh V, Wehrmeyer K, Scott B, Juan J, McKenzie-Brown AM, Lane OP, Haack C: Randomized, wait-list–controlled pilot study of app-delivered mindfulness for patients reporting chronic pain. PAIN Rep 6:1–e924, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
56.McCracken LM, Turk DC: Behavioral and cognitive–behavioral treatment for chronic pain: outcome, predictors of outcome, and treatment process. Spine 27:2564–73, 2002. [DOI] [PubMed] [Google Scholar]
57.Jr Miller. RG: Simultaneous statistical inference [Internet]. Second Edition. New York, NY: Springer-Verlag; 1981. Available from: 10.1007/978-1-4613-8122-8 [DOI] [Google Scholar]
58.Molton IR, Terrill AL: Overview of persistent pain in older adults. Am Psychol 69:197–207, 2014. [DOI] [PubMed] [Google Scholar]
59.Murray CB, Patel KV, Twiddy H, Sturgeon JA, Palermo TM: Age differences in cognitive–affective processes in adults with chronic pain. Eur J Pain 25:1041–52, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Nicholas MK: The pain self-efficacy questionnaire: Taking pain into account. Eur J Pain 11:153–63, 2007. [DOI] [PubMed] [Google Scholar]
61.Osman A, Barrios FX, Kopper BA, Hauptmann W, Jones J, O’Neill E: Factor structure, reliability, and validity of the Pain Catastrophizing Scale. J Behav Med 20:589–605, 1997. [DOI] [PubMed] [Google Scholar]
62.Otis JD: Managing chronic pain: a cognitive-behavioral therapy approach. Therapist guide. Oxford, England: Oxford University Press; 2007. [Google Scholar]
63.Patrick DL, Deyo RA, Atlas SJ, Singer DE, Chapin A, Keller RB: Assessing health-related quality of life in patients with sciatica. Spine 20:1899–909, 1995. [DOI] [PubMed] [Google Scholar]
64.Public Health Service and Health Care Financing Administration: International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). Washington, DC: Public Health Service; 2011. [Google Scholar]
65.Reiner K, Tibi L, Lipsitz JD: Do mindfulness-based interventions reduce pain intensity? A critical review of the literature. Pain Med 14:230–42, 2013. [DOI] [PubMed] [Google Scholar]
66.Roland M, Morris R: A study of the natural history of back pain. Part I: development of a reliable and sensitive measure of disability in low-back pain. Spine 8:141–4, 1983. [DOI] [PubMed] [Google Scholar]
67.Shapiro SL, Brown KW, Thoresen C, Plante TG: The moderation of Mindfulness-based stress reduction effects by trait mindfulness: Results from a randomized controlled trial. J Clin Psychol John Wiley & Sons, Ltd; 67:267–77, 2011. [DOI] [PubMed] [Google Scholar]
68.Skelly AC, Chou R, Dettori JR, Turner JA, Friedly JL, Rundell SD, Fu R, Brodt ED, Wasson N, Kantner S, Ferguson AJR: Noninvasive nonpharmacological treatment for chronic pain: a systematic review update [Internet]. Rockville, MD, USA: Agency for Healthcare Research and Quality (AHRQ); 2020. Apr. Report No.: No. 227. Available from: https://effectivehealthcare.ahrq.gov/products/noninvasive-nonpharm-pain-update/research [PubMed] [Google Scholar]
69.Strong J, Ashton R, Chant D: The measurement of attitudes towards and beliefs about pain. PAIN 48:227–36, 1992. [DOI] [PubMed] [Google Scholar]
70.Sullivan MJ: The Pain Catastrophizing Scale user manual [Internet]. 2009. Available from: http://www.aci.health.nsw.gov.au/__data/assets/pdf_file/0004/257422/Pain_Catastrophizing_Scale_Manual.pdf
71.Sullivan MJL, Bishop SR, Pivik J: The Pain Catastrophizing Scale: development and validation. Psychol Assess 7:524–32, 1995. [Google Scholar]
72.Thompson EL, Broadbent J, Bertino MD, Staiger PK: Do pain-related beliefs influence adherence to multidisciplinary rehabilitation?: A systematic review. Clin J Pain 32:164–78, 2016. [DOI] [PubMed] [Google Scholar]
73.Thorn BE: Cognitive therapy for chronic pain: a step-by-step guide. New York, NY: Guilford Press; 2004. [Google Scholar]
74.Thorn BE, Day MA, Burns J, Kuhajda MC, Gaskins SW, Sweeney K, McConley R, Ward CL, Cabbil C: Randomized trial of group cognitive behavioral therapy compared with a pain education control for low-literacy rural people with chronic pain: PAIN 152:2710–20, 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
75.Turk DC, Wilson HD, Cahana A: Treatment of chronic non-cancer pain. The Lancet 377:2226–35, 2011. [DOI] [PubMed] [Google Scholar]
76.Turk DC, Winter F: The pain survival guide: How to reclaim your life. Washington, DC: American Psychological Association; 2006. [Google Scholar]
77.Turner JA: Comparison of group progressive-relaxation training and cognitive-behavioral group therapy for chronic low back pain. J Consult Clin Psychol 50:757–65, 1982. [DOI] [PubMed] [Google Scholar]
78.Turner JA, Anderson ML, Balderson BH, Cook AJ, Sherman KJ, Cherkin DC: Mindfulness-based stress reduction and cognitive behavioral therapy for chronic low back pain: similar effects on mindfulness, catastrophizing, self-efficacy, and acceptance in a randomized controlled trial. PAIN 157:2434–44, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
79.Turner JA, Clancy S: Comparison of operant behavioral and cognitive-behavioral group treatment for chronic low back pain. J Consult Clin Psychol 56:261–6, 1988. [DOI] [PubMed] [Google Scholar]
80.Turner JA, Franklin G, Heagerty PJ, Wu R, Egan K, Fulton-Kehoe D, Gluck JV, Wickizer TM: The association between pain and disability. PAIN 112:307–14, 2004. [DOI] [PubMed] [Google Scholar]
81.Turner JA, Holtzman S, Mancl L: Mediators, moderators, and predictors of therapeutic change in cognitive–behavioral therapy for chronic pain: PAIN 127:276–86, 2007. [DOI] [PubMed] [Google Scholar]
82.Turner JA, Mancl L, Aaron LA: Short- and long-term efficacy of brief cognitive-behavioral therapy for patients with chronic temporomandibular disorder pain: A randomized, controlled trial. PAIN 121:181–94, 2006. [DOI] [PubMed] [Google Scholar]
83.Turner JA, Romano JM: Cognitive-behavioral therapy for chronic pain. In: Loeser JD, editor. Bonicas Manag Pain Philadelphia, PA: Lippincott Williams & Wilkins; page 1751–82001. [Google Scholar]
84.Turner JA, Shortreed SM, Saunders KW, LeResche L, Berlin JA, Korff MV: Optimizing prediction of back pain outcomes. PAIN 154:1391–401, 2013. [DOI] [PubMed] [Google Scholar]
85.Turner JA, Shortreed SM, Saunders KW, LeResche L, Thielke S, Von Korff M: Does association of opioid use with pain and function differ by fibromyalgia or widespread pain status?: PAIN 157:2208–16, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
86.Underwood M, Mistry D, Lall R, Lamb S: Predicting response to a cognitive-behavioral approach to treating low back pain: Secondary analysis of the BeST data set. Arthritis Care Res 63:1271–9, 2011. [DOI] [PubMed] [Google Scholar]
87.van Dis EAM, van Veen SC, Hagenaars MA, Batelaan NM, Bockting CLH, van den Heuvel RM, Cuijpers P, Engelhard IM: Long-term outcomes of cognitive behavioral therapy for anxiety-related disorders: A systematic review and meta-analysis. JAMA Psychiatry 77:265, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
88.Van Dyke BP, Newman AK, Moraís CA, Burns JW, Eyer JC, Thorn BE: Heterogeneity of treatment effects in a randomized trial of literacy-adapted group cognitive-ehavioral therapy, pain psychoeducation, and usual medical care for multiply disadvantaged patients with chronic pain. J Pain 20:1236–48, 2019. [DOI] [PubMed] [Google Scholar]
89.Veehof MM, Trompetter HR, Bohlmeijer ET, Schreurs KMG: Acceptance- and mindfulness-based interventions for the treatment of chronic pain: a meta-analytic review. Cogn Behav Ther 45:5–31, 2016. [DOI] [PubMed] [Google Scholar]
90.Vitiello MV, McCurry SM, Shortreed SM, Balderson BH, Baker LD, Keefe FJ, Rybarczyk BD, Von Korff M: Cognitive-behavioral treatment for comorbid insomnia and osteoarthritis pain in primary care: the lifestyles randomized controlled trial. J Am Geriatr Soc 61:947–56, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
91.Vlaeyen JWS, Morley S: Cognitive-behavioral treatments for chronic pain: What works for whom? Clin J Pain 21:1–8, 2005. [DOI] [PubMed] [Google Scholar]
92.Wang R, Ware JH: Detecting moderator effects using subgroup analyses. Prev Sci 14:111–20, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
93.Williams AC de C, Fisher E, Hearn L, Eccleston C: Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database Syst Rev [Internet], 2020. Available from: http://doi.wiley.com/10.1002/14651858.CD007407.pub4 [DOI] [PMC free article] [PubMed] [Google Scholar]
94.Zelman DC, Hoffman DL, Seifeldin R, Dukes EM: Development of a metric for a day of manageable pain control: derivation of pain severity cut-points for low back pain and osteoarthritis. PAIN 106:35–42, 2003. [DOI] [PubMed] [Google Scholar]

[R1] 1.Anheyer D, Haller H, Barth J, Lauche R, Dobos G, Cramer H: Mindfulness-based stress reduction for treating low back pain: A systematic review and meta-analysis. Ann Intern Med 166:799, 2017. [DOI] [PubMed] [Google Scholar]

[R2] 2.Artus M, Campbell P, Mallen CD, Dunn KM, van der Windt DAW: Generic prognostic factors for musculoskeletal pain in primary care: a systematic review. BMJ Open 7:e012901, 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R3] 3.Blacker M, Meleo-Meyer F, Kabat-Zinn J, Santorelli S: Stress reduction clinic mindfulness-based stress reduction (MBSR) curriculum guide. Worcester, MA: Center for Mindfulness in Medicine, Health Care and Society, University of Massachusetts Medical School; 2009. [Google Scholar]

[R4] 4.Bohlmeijer E, ten Klooster PM, Fledderus M, Veehof M, Baer R: Psychometric properties of the Five Facet Mindfulness Questionnaire in depressed adults and development of a short form. Assessment 18:308–20, 2011. [DOI] [PubMed] [Google Scholar]

[R5] 5.Broderick JE, Keefe FJ, Schneider S, Junghaenel DU, Bruckenthal P, Schwartz JE, Kaell AT, Caldwell DS, McKee D, Gould E: Cognitive behavioral therapy for chronic pain is effective, but for whom?: PAIN 157:2115–23, 2016. [DOI] [PubMed] [Google Scholar]

[R6] 6.Brown MM, Arigo D, Wolever RQ, Smoski MJ, Hall MH, Brantley JG, Greeson JM: Do gender, anxiety, or sleep quality predict mindfulness-based stress reduction outcomes? J Health Psychol 26:2656–62, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R7] 7.Burgess R, Mansell G, Bishop A, Lewis M, Hill J: Predictors of functional outcome in musculoskeletal healthcare: An umbrella review. Eur J Pain 24:51–70, 2020. [DOI] [PubMed] [Google Scholar]

[R8] 8.Burns JW, Jensen MP, Thorn B, Lillis TA, Carmody J, Newman AK, Keefe F: Cognitive therapy, mindfulness-based stress reduction, and behavior therapy for the treatment of chronic pain: randomized controlled trial. PAIN 163:376–89, 2022. [DOI] [PubMed] [Google Scholar]

[R9] 9.Burns JW, Van Dyke BP, Newman AK, Morais CA, Thorn BE: Cognitive behavioral therapy (CBT) and pain education for people with chronic pain: Tests of treatment mechanisms. J Consult Clin Psychol 88:1008–18, 2020. [DOI] [PubMed] [Google Scholar]

[R10] 10.Caudill M: Managing pain before it manages you. Rev. ed. New York, NY: Guilford Press; 2002. [Google Scholar]

[R11] 11.Chen Y, Campbell P, Strauss VY, Foster NE, Jordan KP, Dunn KM: Trajectories and predictors of the long-term course of low back pain: cohort study with 5-year follow-up. PAIN 159:252–60, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R12] 12.Cherkin DC, Anderson ML, Sherman KJ, Balderson BH, Cook AJ, Hansen KE, Turner JA: Two-year follow-up of a randomized clinical trial of mindfulness-based stress reduction vs cognitive behavioral therapy or usual care for chronic low back pain. JAMA 317:642, 2017. [DOI] [PubMed] [Google Scholar]

[R13] 13.Cherkin DC, Sherman KJ, Balderson BH, Cook AJ, Anderson ML, Hawkes RJ, Hansen KE, Turner JA: Effect of mindfulness-based stress reduction vs cognitive behavioral therapy or usual care on back pain and functional limitations in adults with chronic low back pain: a randomized clinical trial. JAMA 315:1240, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R14] 14.Cherkin DC, Sherman KJ, Balderson BH, Turner JA, Cook AJ, Stoelb B, Herman PM, Deyo RA, Hawkes RJ: Comparison of complementary and alternative medicine with conventional mind–body therapies for chronic back pain: protocol for the Mind–body Approaches to Pain (MAP) randomized controlled trial. Trials 15:211, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R15] 15.Chiesa A, Serretti A: Mindfulness-based interventions for chronic pain: a systematic review of the evidence. J Altern Complement Med 17:83–93, 2011. [DOI] [PubMed] [Google Scholar]

[R16] 16.Costa L d. CM, Maher CG, McAuley JH, Hancock MJ, Herbert RD, Refshauge KM, Henschke N: Prognosis for patients with chronic low back pain: inception cohort study. BMJ 339:b3829–b3829, 2009. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R17] 17.Cuijpers P, Noma H, Karyotaki E, Cipriani A, Furukawa TA: Effectiveness and acceptability of cognitive behavior therapy delivery formats in adults with depression: A network meta-analysis. JAMA Psychiatry 76:700, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R18] 18.Day MA, Ehde DM, Jensen MP: Psychosocial pain management moderation: the limit, activate, and enhance model. J Pain 16:947–60, 2015. [DOI] [PubMed] [Google Scholar]

[R19] 19.Day MA, Thorn BE, Ehde DM, Burns JW, Barnier A, Mattingley JB, Matthews N, Jensen MP: Moderators of mindfulness meditation, cognitive therapy, and mindfulness-based cognitive therapy for chronic low back pain: a test of the limit, activate, and enhance model. J Pain 21:161–9, 2020. [DOI] [PubMed] [Google Scholar]

[R20] 20.de Rooij A, van der Leeden M, Roorda LD, Dekker J: Predictors of outcome of multidisciplinary treatment in chronic widespread pain: an observational study. BMC Musculoskelet Disord 13:133, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R21] 21.Driessen E, Hollon SD: Cognitive behavioral therapy for mood disorders: Efficacy, moderators and mediators. Psychiatr Clin North Am 33:537–55, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R22] 22.Dunn KM, Jordan KP, Croft PR: Contributions of prognostic factors for poor outcome in primary care low back pain patients. Eur J Pain 15:313–9, 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R23] 23.Ecija C, Catala P, Lopez-Gomez I, Bedmar D, Peñacoba C: What does the psychological flexibility model contribute to the relationship between depression and disability in chronic pain? The role of cognitive fusion and pain acceptance. Clin Nurs Res :1–13, 2021. [DOI] [PubMed] [Google Scholar]

[R24] 24.Ehde DM, Dillworth TM, Turner JA: Cognitive-behavioral therapy for individuals with chronic pain: Efficacy, innovations, and directions for research. Am Psychol 69:153–66, 2014. [DOI] [PubMed] [Google Scholar]

[R25] 25.Fish RA, Hogan MJ, Morrison TG, Stewart I, McGuire BE: Willing and able: a closer look at pain willingness and activity engagement on the Chronic Pain Acceptance Questionnaire (CPAQ-8). J Pain 14:233–45, 2013. [DOI] [PubMed] [Google Scholar]

[R26] 26.Fish RA, McGuire B, Hogan M, Morrison TG, Stewart I: Validation of the Chronic Pain Acceptance Questionnaire (CPAQ) in an internet sample and development and preliminary validation of the CPAQ-8: Pain 149:435–43, 2010. [DOI] [PubMed] [Google Scholar]

[R27] 27.Garland EL, Gaylord SA, Palsson O, Faurot K, Douglas Mann J, Whitehead WE: Therapeutic mechanisms of a mindfulness-based treatment for IBS: effects on visceral sensitivity, catastrophizing, and affective processing of pain sensations. J Behav Med 35:591–602, 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R28] 28.Gilpin HR, Keyes A, Stahl DR, Greig R, McCracken LM: Predictors of treatment outcome in contextual cognitive and behavioral therapies for chronic pain: A systematic review. J Pain 18:1153–64, 2017. [DOI] [PubMed] [Google Scholar]

[R29] 29.Goldberg SB, Tucker RP, Greene PA, Davidson RJ, Wampold BE, Kearney DJ, Simpson TL: Mindfulness-based interventions for psychiatric disorders: A systematic review and meta-analysis. Clin Psychol Rev 59:52–60, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R30] 30.Gu J, Strauss C, Bond R, Cavanagh K: How do mindfulness-based cognitive therapy and mindfulness-based stress reduction improve mental health and wellbeing? A systematic review and meta-analysis of mediation studies. Clin Psychol Rev 37:1–12, 2015. [DOI] [PubMed] [Google Scholar]

[R31] 31.Hayden JA, Wilson MN, Riley RD, Iles R, Pincus T, Ogilvie R: Individual recovery expectations and prognosis of outcomes in non-specific low back pain: prognostic factor review. Cochrane Database Syst Rev Issue 11:Art. No.: CD011284, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R32] 32.Jensen MP: Psychosocial approaches to pain management: An organizational framework. PAIN 152:717–25, 2011. [DOI] [PubMed] [Google Scholar]

[R33] 33.Jensen MP, Karoly P: Pain-specific beliefs, perceived symptom severity, and adjustment to chronic pain. Clin J Pain United States; 8:123–30, 1992. [DOI] [PubMed] [Google Scholar]

[R34] 34.Jensen MP, Karoly P: Survey of pain attitudes: professional manual. Lutz, FL: Psychological Assessment Resources; 2008. [Google Scholar]

[R35] 35.Jensen MP, Romano JM, Turner JA, Good AB, Wald LH: Patient beliefs predict patient functioning: further support for a cognitive-behavioural model of chronic pain. PAIN 81:95–104, 1999. [DOI] [PubMed] [Google Scholar]

[R36] 36.Jensen MP, Smith DG, Ehde DM, Robinson LR: Pain site and the effects of amputation pain: further clarification of the meaning of mild, moderate, and severe pain. PAIN 3:317–22, 2001. [DOI] [PubMed] [Google Scholar]

[R37] 37.Jensen MP, Turner JA, Romano JM: Changes after multidisciplinary pain treatment in patient pain beliefs and coping are associated with concurrent changes in patient functioning. PAIN 131:38–47, 2007. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R38] 38.Jensen MP, Turner JA, Romano JM: Changes in beliefs, catastrophizing, and coping are associated with improvement in multidisciplinary pain treatment. J Consult Clin Psychol 69:655–62, 2001. [DOI] [PubMed] [Google Scholar]

[R39] 39.Jensen MP, Turner JA, Romano JM, Lawler BK: Relationship of pain-specific beliefs to chronic pain adjustment. PAIN 57:301–9, 1994. [DOI] [PubMed] [Google Scholar]

[R40] 40.Jones KR, Vojir CP, Hutt E, Fink R: Determining mild, moderate, and severe pain equivalency across pain-intensity tools in nursing home residents. J Rehabil Res Dev 44:305, 2007. [DOI] [PubMed] [Google Scholar]

[R41] 41.Kabat-Zinn J: An outpatient program in behavioral medicine for chronic pain patients based on the practice of mindfulness meditation: Theoretical considerations and preliminary results. Gen Hosp Psychiatry 4:33–47, 1982. [DOI] [PubMed] [Google Scholar]

[R42] 42.Kabat-Zinn J: Full catastrophe living: Using the wisdom of your body and mind to face stress, pain, and illness. New York, NY: Random House; 2005. [Google Scholar]

[R43] 43.Kabat-Zinn J: Mindfulness-based interventions in context: past, present, and future. Clin Psychol Sci Pract 10:144–56, 2003. [Google Scholar]

[R44] 44.Khoo E-L, Small R, Cheng W, Hatchard T, Glynn B, Rice DB, Skidmore B, Kenny S, Hutton B, Poulin PA: Comparative evaluation of group-based mindfulness-based stress reduction and cognitive behavioural therapy for the treatment and management of chronic pain: A systematic review and network meta-analysis. Evid Based Ment Health 22:26–35, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R45] 45.Kraemer H, Wilson G, Fairburn CG, Agras W: Mediators and moderators of treatment effects in randomized clinical trials. Arch Gen Psychiatry 59:877–83, 2002. [DOI] [PubMed] [Google Scholar]

[R46] 46.Kroenke K, Spitzer RL, Williams JBW: The PHQ-15: validity of a new measure for evaluating the severity of somatic symptoms. Psychosom Med 64:258–66, 2002. [DOI] [PubMed] [Google Scholar]

[R47] 47.Kroenke K, Spitzer RL, Williams JBW, Löwe B: The Patient Health Questionnaire somatic, anxiety, and depressive symptom scales: a systematic review. Gen Hosp Psychiatry 32:345–59, 2010. [DOI] [PubMed] [Google Scholar]

[R48] 48.Kroenke K, Spitzer RL, Williams JBW, Monahan PO, Löwe B: Anxiety disorders in primary care: prevalence, impairment, comorbidity, and detection. Ann Intern Med 146:317, 2007. [DOI] [PubMed] [Google Scholar]

[R49] 49.Kroenke K, Strine TW, Spitzer RL, Williams JBW, Berry JT, Mokdad AH: The PHQ-8 as a measure of current depression in the general population. J Affect Disord 114:163–73, 2009. [DOI] [PubMed] [Google Scholar]

[R50] 50.Lamb SE, Hansen Z, Lall R, Castelnuovo E, Withers EJ, Nichols V, Potter R, Underwood MR: Group cognitive behavioural treatment for low-back pain in primary care: a randomised controlled trial and cost-effectiveness analysis. The Lancet 375:916–23, 2010. [DOI] [PubMed] [Google Scholar]

[R51] 51.Litt MD, Shafer DM, Kreutzer DL: Brief cognitive-behavioral treatment for TMD pain: Long-term outcomes and moderators of treatment. PAIN 151:110–6, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R52] 52.Mallen CD, Peat G, Thomas E, Dunn KM, Croft PR: Prognostic factors for musculoskeletal pain in primary care: a systematic review. Br J Gen Pract 57:655, 2007. [PMC free article] [PubMed] [Google Scholar]

[R53] 53.Martinez-Calderon J, Jensen MP, Morales-Asencio JM, Luque-Suarez A: Pain catastrophizing and function in individuals with chronic musculoskeletal pain: A systematic review and meta-analysis. Clin J Pain 35:279–93, 2019. [DOI] [PubMed] [Google Scholar]

[R54] 54.Martinez-Calderon J, Zamora-Campos C, Navarro-Ledesma S, Luque-Suarez A: The role of self-efficacy on the prognosis of chronic musculoskeletal pain: A systematic review. J Pain 19:10–34, 2018. [DOI] [PubMed] [Google Scholar]

[R55] 55.Mascaro JS, Singh V, Wehrmeyer K, Scott B, Juan J, McKenzie-Brown AM, Lane OP, Haack C: Randomized, wait-list–controlled pilot study of app-delivered mindfulness for patients reporting chronic pain. PAIN Rep 6:1–e924, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R56] 56.McCracken LM, Turk DC: Behavioral and cognitive–behavioral treatment for chronic pain: outcome, predictors of outcome, and treatment process. Spine 27:2564–73, 2002. [DOI] [PubMed] [Google Scholar]

[R57] 57.Jr Miller. RG: Simultaneous statistical inference [Internet]. Second Edition. New York, NY: Springer-Verlag; 1981. Available from: 10.1007/978-1-4613-8122-8 [DOI] [Google Scholar]

[R58] 58.Molton IR, Terrill AL: Overview of persistent pain in older adults. Am Psychol 69:197–207, 2014. [DOI] [PubMed] [Google Scholar]

[R59] 59.Murray CB, Patel KV, Twiddy H, Sturgeon JA, Palermo TM: Age differences in cognitive–affective processes in adults with chronic pain. Eur J Pain 25:1041–52, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R60] 60.Nicholas MK: The pain self-efficacy questionnaire: Taking pain into account. Eur J Pain 11:153–63, 2007. [DOI] [PubMed] [Google Scholar]

[R61] 61.Osman A, Barrios FX, Kopper BA, Hauptmann W, Jones J, O’Neill E: Factor structure, reliability, and validity of the Pain Catastrophizing Scale. J Behav Med 20:589–605, 1997. [DOI] [PubMed] [Google Scholar]

[R62] 62.Otis JD: Managing chronic pain: a cognitive-behavioral therapy approach. Therapist guide. Oxford, England: Oxford University Press; 2007. [Google Scholar]

[R63] 63.Patrick DL, Deyo RA, Atlas SJ, Singer DE, Chapin A, Keller RB: Assessing health-related quality of life in patients with sciatica. Spine 20:1899–909, 1995. [DOI] [PubMed] [Google Scholar]

[R64] 64.Public Health Service and Health Care Financing Administration: International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). Washington, DC: Public Health Service; 2011. [Google Scholar]

[R65] 65.Reiner K, Tibi L, Lipsitz JD: Do mindfulness-based interventions reduce pain intensity? A critical review of the literature. Pain Med 14:230–42, 2013. [DOI] [PubMed] [Google Scholar]

[R66] 66.Roland M, Morris R: A study of the natural history of back pain. Part I: development of a reliable and sensitive measure of disability in low-back pain. Spine 8:141–4, 1983. [DOI] [PubMed] [Google Scholar]

[R67] 67.Shapiro SL, Brown KW, Thoresen C, Plante TG: The moderation of Mindfulness-based stress reduction effects by trait mindfulness: Results from a randomized controlled trial. J Clin Psychol John Wiley & Sons, Ltd; 67:267–77, 2011. [DOI] [PubMed] [Google Scholar]

[R68] 68.Skelly AC, Chou R, Dettori JR, Turner JA, Friedly JL, Rundell SD, Fu R, Brodt ED, Wasson N, Kantner S, Ferguson AJR: Noninvasive nonpharmacological treatment for chronic pain: a systematic review update [Internet]. Rockville, MD, USA: Agency for Healthcare Research and Quality (AHRQ); 2020. Apr. Report No.: No. 227. Available from: https://effectivehealthcare.ahrq.gov/products/noninvasive-nonpharm-pain-update/research [PubMed] [Google Scholar]

[R69] 69.Strong J, Ashton R, Chant D: The measurement of attitudes towards and beliefs about pain. PAIN 48:227–36, 1992. [DOI] [PubMed] [Google Scholar]

[R70] 70.Sullivan MJ: The Pain Catastrophizing Scale user manual [Internet]. 2009. Available from: http://www.aci.health.nsw.gov.au/__data/assets/pdf_file/0004/257422/Pain_Catastrophizing_Scale_Manual.pdf

[R71] 71.Sullivan MJL, Bishop SR, Pivik J: The Pain Catastrophizing Scale: development and validation. Psychol Assess 7:524–32, 1995. [Google Scholar]

[R72] 72.Thompson EL, Broadbent J, Bertino MD, Staiger PK: Do pain-related beliefs influence adherence to multidisciplinary rehabilitation?: A systematic review. Clin J Pain 32:164–78, 2016. [DOI] [PubMed] [Google Scholar]

[R73] 73.Thorn BE: Cognitive therapy for chronic pain: a step-by-step guide. New York, NY: Guilford Press; 2004. [Google Scholar]

[R74] 74.Thorn BE, Day MA, Burns J, Kuhajda MC, Gaskins SW, Sweeney K, McConley R, Ward CL, Cabbil C: Randomized trial of group cognitive behavioral therapy compared with a pain education control for low-literacy rural people with chronic pain: PAIN 152:2710–20, 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R75] 75.Turk DC, Wilson HD, Cahana A: Treatment of chronic non-cancer pain. The Lancet 377:2226–35, 2011. [DOI] [PubMed] [Google Scholar]

[R76] 76.Turk DC, Winter F: The pain survival guide: How to reclaim your life. Washington, DC: American Psychological Association; 2006. [Google Scholar]

[R77] 77.Turner JA: Comparison of group progressive-relaxation training and cognitive-behavioral group therapy for chronic low back pain. J Consult Clin Psychol 50:757–65, 1982. [DOI] [PubMed] [Google Scholar]

[R78] 78.Turner JA, Anderson ML, Balderson BH, Cook AJ, Sherman KJ, Cherkin DC: Mindfulness-based stress reduction and cognitive behavioral therapy for chronic low back pain: similar effects on mindfulness, catastrophizing, self-efficacy, and acceptance in a randomized controlled trial. PAIN 157:2434–44, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R79] 79.Turner JA, Clancy S: Comparison of operant behavioral and cognitive-behavioral group treatment for chronic low back pain. J Consult Clin Psychol 56:261–6, 1988. [DOI] [PubMed] [Google Scholar]

[R80] 80.Turner JA, Franklin G, Heagerty PJ, Wu R, Egan K, Fulton-Kehoe D, Gluck JV, Wickizer TM: The association between pain and disability. PAIN 112:307–14, 2004. [DOI] [PubMed] [Google Scholar]

[R81] 81.Turner JA, Holtzman S, Mancl L: Mediators, moderators, and predictors of therapeutic change in cognitive–behavioral therapy for chronic pain: PAIN 127:276–86, 2007. [DOI] [PubMed] [Google Scholar]

[R82] 82.Turner JA, Mancl L, Aaron LA: Short- and long-term efficacy of brief cognitive-behavioral therapy for patients with chronic temporomandibular disorder pain: A randomized, controlled trial. PAIN 121:181–94, 2006. [DOI] [PubMed] [Google Scholar]

[R83] 83.Turner JA, Romano JM: Cognitive-behavioral therapy for chronic pain. In: Loeser JD, editor. Bonicas Manag Pain Philadelphia, PA: Lippincott Williams & Wilkins; page 1751–82001. [Google Scholar]

[R84] 84.Turner JA, Shortreed SM, Saunders KW, LeResche L, Berlin JA, Korff MV: Optimizing prediction of back pain outcomes. PAIN 154:1391–401, 2013. [DOI] [PubMed] [Google Scholar]

[R85] 85.Turner JA, Shortreed SM, Saunders KW, LeResche L, Thielke S, Von Korff M: Does association of opioid use with pain and function differ by fibromyalgia or widespread pain status?: PAIN 157:2208–16, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R86] 86.Underwood M, Mistry D, Lall R, Lamb S: Predicting response to a cognitive-behavioral approach to treating low back pain: Secondary analysis of the BeST data set. Arthritis Care Res 63:1271–9, 2011. [DOI] [PubMed] [Google Scholar]

[R87] 87.van Dis EAM, van Veen SC, Hagenaars MA, Batelaan NM, Bockting CLH, van den Heuvel RM, Cuijpers P, Engelhard IM: Long-term outcomes of cognitive behavioral therapy for anxiety-related disorders: A systematic review and meta-analysis. JAMA Psychiatry 77:265, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R88] 88.Van Dyke BP, Newman AK, Moraís CA, Burns JW, Eyer JC, Thorn BE: Heterogeneity of treatment effects in a randomized trial of literacy-adapted group cognitive-ehavioral therapy, pain psychoeducation, and usual medical care for multiply disadvantaged patients with chronic pain. J Pain 20:1236–48, 2019. [DOI] [PubMed] [Google Scholar]

[R89] 89.Veehof MM, Trompetter HR, Bohlmeijer ET, Schreurs KMG: Acceptance- and mindfulness-based interventions for the treatment of chronic pain: a meta-analytic review. Cogn Behav Ther 45:5–31, 2016. [DOI] [PubMed] [Google Scholar]

[R90] 90.Vitiello MV, McCurry SM, Shortreed SM, Balderson BH, Baker LD, Keefe FJ, Rybarczyk BD, Von Korff M: Cognitive-behavioral treatment for comorbid insomnia and osteoarthritis pain in primary care: the lifestyles randomized controlled trial. J Am Geriatr Soc 61:947–56, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R91] 91.Vlaeyen JWS, Morley S: Cognitive-behavioral treatments for chronic pain: What works for whom? Clin J Pain 21:1–8, 2005. [DOI] [PubMed] [Google Scholar]

[R92] 92.Wang R, Ware JH: Detecting moderator effects using subgroup analyses. Prev Sci 14:111–20, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R93] 93.Williams AC de C, Fisher E, Hearn L, Eccleston C: Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database Syst Rev [Internet], 2020. Available from: http://doi.wiley.com/10.1002/14651858.CD007407.pub4 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R94] 94.Zelman DC, Hoffman DL, Seifeldin R, Dukes EM: Development of a metric for a day of manageable pain control: derivation of pain severity cut-points for low back pain and osteoarthritis. PAIN 106:35–42, 2003. [DOI] [PubMed] [Google Scholar]

PERMALINK

Moderators and Nonspecific Predictors of Treatment Benefits in a Randomized Trial of Mindfulness-Based Stress Reduction vs. Cognitive-Behavioral Therapy vs. Usual Care for Chronic Low Back Pain

Jessica A Chen

Melissa L Anderson

Daniel C Cherkin

Benjamin H Balderson

Andrea J Cook

Karen J Sherman

Judith A Turner

Abstract

Trial Registration:

Perspective:

Introduction

Materials and Methods

Setting, Participants, and Procedures

Measures

Outcome Measures

Back Pain-Related Functional Limitations

Back Pain Bothersomeness

Depressive Symptoms

Potential Moderators and Nonspecific Predictors

Expectations.

Widespread Pain

Anxiety

Pain Beliefs

Pain Catastrophizing

Pain Acceptance

Pain Self-Efficacy

Mindfulness

Opioid Use

Covariates

Interventions

MBSR

CBT

Usual Care

Instructors/Therapists and Treatment Fidelity Monitoring

Statistical Analysis

Results

Moderators of treatment effects

Table 1:

Moderators of Intervention Effects on Back Pain-Related Functional Limitations

Table 2:

Moderators of Intervention Effects on Pain Bothersomeness

Table 3:

Moderators of Intervention Effects on Depressive Symptoms

Table 4:

Post-Hoc Analysis for Moderators

Nonspecific Predictors of Improvement

Table 5:

Summary

Table 6:

Discussion

Highlights.

Research Funding

Footnotes

References

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases