ORIGINAL ARTICLE
Comparing the Effectiveness of Mindfulness-based Stress Reduction and Multidisciplinary Intervention Programs for Chronic Pain A Randomized Comparative Trial Samuel Yeung-Shan Wong, MD, MPH, CCFPC, FRACGP,* Frank Wan-Kin Chan, MB ChB, MPH,* Rebecca Lai-Ping Wong, BSc,* Ming-Chi Chu, MBBS, FFPMANZCA,w Yu-Yuk Kitty Lam, MSc, BSc,z Stewart W. Mercer, MB ChB, PhD,y and S. Helen Ma, PhDJ
Objectives: Research suggests that an 8-week Mindfulness-Based Stress Reduction (MBSR) program (a structured form of meditation) might be effective in the treatment of various health problems including chronic pain. Our objective was to compare the clinical effectiveness of the MBSR program with a multidisciplinary pain intervention (MPI) program in terms of pain intensity, pain-related distress, quality of life, and mood in patients with chronic pain. Methods: A randomized, comparative clinical trial was conducted, including 6-month posttreatment follow-up. Ninety-nine participants, aged 24 to 64 years, with pain for a minimum of 3 months, were recruited from community-based clinics, hospitals, and community service centers. Participants were randomly allocated to either the MBSR program (51 participants) or a MPI program (48 participants). The study used validated Chinese versions of selfreported questionnaires measuring pain, mood symptoms, and health-related quality of life. Results: Thirty-nine participants (77%) completed the MBSR program and 44 (90%) completed the MPI program. Patients in both the groups were comparable with regard to demographical characteristics, pain intensity, mood symptoms, and health-related quality-of-life measures before intervention. In both the groups, patients who completed the trial demonstrated statistically significant improvements in pain intensity and pain-related distress. However, no statistically significant differences were observed in overall results between the MBSR and MPI groups. Conclusions: This randomized, clinical trial showed that both MBSR and MPI programs reduced pain intensity and pain-related distress although no statistically significant differences were observed between the 2 groups and the improvements were small.
Received for publication December 9, 2009; revised January 31, 2011; accepted March 2, 2011. From the *School of Public Health and Primary Care; yJockey Club Centre for Osteoporosis Care and Control, The Chinese University of Hong Kong; zDepartment of Anaesthesia and Intensive Care, Prince of Wales Hospital, Hong Kong SAR; wGeneral Practice and Primary Care, Centre for Population and Health Sciences, University of Glasgow, UK; and JPrivate Clinical Psychologist and MBSR instructor, Hong Kong. Funded by The Health and Health Services Research Fund was established and granted by the Food and Health Bureau, Hong Kong SAR Government, Hong Kong, and a dissemination report (a requirement of all funded grants) was published in a supplement of the Hong Kong Medical Journal, October 2009. The authors declare no conflict of interest. The reported trial was registered with the Centre for Clinical Trials, the Chinese University of Hong Kong and was approved by the joint CUHK-NTE Clinical Ethics Committee. Reprints: Samuel Yeung-Shan Wong, MD, MPH, CCFPC, FRACGP, 4/F School of Public Health, Prince of Wales Hospital, Shatin, N.T., Hong Kong. (e-mail: yeungshanwong@cuhk.edu.hk). Copyright r 2011 by Lippincott Williams & Wilkins
724 | www.clinicalpain.com
Key Words: cognitive therapy, meditation, pain, comparative study, health education
(Clin J Pain 2011;27:724–734)
C
hronic pain is a common health problem and is a frequent cause of disability and suffering.1–4 It is also a condition associated with significant health-care costs.5,6 Although psychological interventions are established alternatives to traditional medical approaches,7,8 it is increasingly recognized that a significant proportion of individuals with chronic pain do not benefit from these traditional or psychological interventions.9 Research suggests that fear of pain and pain-related anxiety may account for significant disability, suffering, and health-care utilization in patients with chronic pain10–12 and that such patients may benefit from treatments that allow them to respond to their pain more neutrally, with less fear and avoidance.9 Reitsma and Meijler13 suggest that a cognitive shift, entailing psychological acceptance of the concept of “functioning regardless of pain,” was a positive prognostic indicator of treatment outcome. A number of approaches that incorporate both acceptance and mindfulness therapy have been developed.14,15 Hayes et al16 developed Acceptance and Commitment Therapy (ACT) during the 1990s. It includes acceptance and mindfulness strategies, and commitment and behavioral change strategies, to improve functioning in people with a wide range of behavioral problems. Recent evidence from clinical trials,17,18 review,14 and meta-analysis19 showed that ACT may be effective for a variety of psychologically-related problems including the treatment of chronic pain. In addition to ACT, a mindfulness-based stress reduction (MBSR) program has also been suggested as another alternative therapy in the management of pain.20 MBSR is a clinical program, originally developed to increase self-acceptance, coping, and reduction of suffering in patients with complex medical illnesses by providing systematic training in mindfulness meditation as a selfregulatory approach to stress reduction and emotion management.21 The MBSR program uses stress-reduction skills that include sitting and walking meditation, yoga, and a somatically-focused technique called the body scan.20 Participants are instructed on how to focus on their immediate experience with an attitude of openness, acceptance, and compassion.22 Studies over the past 25 years have demonstrated the efficacy of MBSR in the treatment of various health problems Clin J Pain
Volume 27, Number 8, October 2011
Clin J Pain
Volume 27, Number 8, October 2011
that include eating disorders,23 fibromyalgia,24 psoriasis25 and anxiety disorders,26,27 mood symptoms in cancer patients,28,29 and the prevention of relapse of depression.30,31 Early studies of the effectiveness of MBSR in the treatment of chronic pain21,31,32 used a preintervention and postintervention design without the use of randomization of interventions. Recently, a small number of randomized, controlled, and comparative trials have been conducted to evaluate the feasibility and effectiveness of MBSR on chronic pain,33–36 but there continues to be a lack of comparative studies with adequate sample size to conclusively show the effectiveness of MBSR in the treatment of pain. Using a validated Chinese versions of standard outcome measures, we set out to conduct a randomized, clinical trial to evaluate the effectiveness of the MBSR program as compared with a multidisciplinary pain intervention (MPI) strategy, both of which were designed to improve perception of pain, quality-of-life and mood symptoms, and disability in patients with chronic pain.
MATERIALS AND METHODS Study Participants The recruitment and intervention phase of this study took place between January and March 2006, with followup to November of that year. Participants were recruited between January and February 2006 from the Hong Kong’s primary care, geriatric, and pain clinics situated in both the community, and the hospitals which most chronic pain patients had been found to attend.5 Posters were put up in these clinics with instructions for interested individuals to contact our research assistant for more information on eligibility and enrollment in the study. A trained research assistant initially determined eligibility according to a predefined set of inclusion and exclusion criteria, and provided detailed information on the study to potential participants. Those initially screened as eligible for participation attended an interview with a health-care professional (S.Y.W.) to confirm his or her eligibility. After the interview, the health-care professional (S.Y.W.) explained the study in further details to the eligible participants and allowed time for further explanation of the study when requested. Eligibility of an interested individual was determined by the following inclusion criteria: 1. Age between 18 and 65 years 2. The presence of chronic pain, which had persisted for at least 3 months at the moderate-to-severe level [ie, at least 4 of 10 on an 11-point Numerical Rating Scale (NRS)37 pain score] 3. Agreement by the participant not to receive other new treatments during the intervention, including the use of new medication, topical treatment, medication or other over-the-counter medication, or other nonpharmacological treatment 4. Ability to give a written consent. Volunteers were not admitted to the study if any of the following exclusion criteria were present: 1. Receiving concurrent treatment with therapies other than medications for pain or psychological symptoms 2. Having a known, concurrent doctor-diagnosed DSM-IV Axis I disorder 3. Having previously participated in an MBSR program r
2011 Lippincott Williams & Wilkins
MBSR and Interaction Program for Chronic Pain
4. Having been engaged, currently or previously, in the practice of meditation or relaxation techniques, including an MBSR program 5. Those who were illiterate, as they would not be able to complete the meditation diary. All participants gave a written consent and the study was performed according to the international Good Clinical Practice Guideline.38 All interventions were provided free of charge and there was no financial incentive for participating in the trials.
MBSR Group The MBSR intervention was led by S.H. Ma, a clinical psychologist with >5 years’ experience in teaching the MBSR program in various locations in Hong Kong. She has completed both the Teacher Development Intensive Training and an intensive professional training program in MBSR, which was conducted by the originators of the program.21 Our program followed the same format and content as those performed and evaluated previously in other MBSR trials and comprised 8 weekly group sessions, each of 2 1=2 hours, with a 7-hour “retreat” session. Instructive, inductive, and experiential modes of learning were used to carry out the intervention and to convey the information content.31 There were 3 primary elements: 1. Theoretical material related to mindfulness, relaxation, meditation, yoga, and the body-mind connection 2. Experiential practice of meditation and yoga, both during the group meetings and at home, and 3. Group activities that focused on removing impediments to effective practice, practical day-to-day applications of mindfulness, and supportive intervention between group members. Participants were given a booklet on the MBSR program and CDs that guided them through the mindfulness meditation exercises, which they were instructed to practice daily.
MPI Group The MPI group was led by an experienced nurse and served as an active control intervention. The MPI program included a set of educational instructions on management of chronic pain, based on a popular self-help book, “Managing Pain Before It Manages You.”39,40 The aims of the MPI group were to act as a control for therapists’ attention and contact time, and for any unmeasured effects of taking part in a group intervention. To avoid overlap between what is taught in the MPI and the MBSR, any information in the self-help book concerning mind-body connection and cognitive techniques introduced were not taught to the MPI group. Participants in this group met with a nurse coordinator with >five years of clinical psychiatric experience for 8, weekly, 21=2-hour group sessions. These sessions took the form of instructional lectures on basic understanding of chronic pain, factors that increase or decrease chronic pain, and effective ways for participants to signal their chronic pain to others. One session was conducted by a registered physiotherapist with >5 years of physiotherapy clinical experience in instruction on exercises for chronic pain. Another session was conducted by a registered dietician with >5 years of dietetic clinical experience to advise on healthy diet and weight control. www.clinicalpain.com |
725
Wong et al
These participants were given a CD of classical music and were instructed to listen to the CD daily. As for the MBSR participants, they also participated in a 7-hour group “retreat.” All therapists who were involved in the MPI group were instructed not to include any mindfulness meditation techniques in their sessions. A statistician, who was not involved in any part of the study, independently randomized participants by using a predetermined random table generated by Microsoft Excel 2002. These numbers were not decoded until the intervention group was assigned. The majority of the participants were female; due to the low number of male participants, randomization was stratified for sex. The allocation was concealed from the researchers who carried out the baseline assessment or recorded the data, and from the statistician who carried out the analysis. The allocation was unknown to the participants until the first appointment. We asked the participants, without referring to the pretreatment assessment, to assess posttreatment pain ratings by completing questionnaires that were sent to them after treatment. The research assistant responsible for the postal questionnaires was also unaware of the treatment allocation. Nonresponders were initially reminded by post followed, where necessary, by telephone calls from the research assistant.
Objectives and Outcome measures The purpose of the study was to compare the effectiveness of the MBSR program with the MPI program in terms of pain ratings (pain intensity and pain-related distress) for individuals with chronic pain. Outcome measures were collected at baseline, at 8 weeks (immediately after intervention), and 3 months and 6 months after intervention. Our primary outcomes were selfreported pain intensity and pain-related distress, measured by two 11-point NRS, with one measuring pain intensity and the other measuring pain-related distress.41 Participants reported the pain intensity and pain-related distress on 2 lines that were marked with whole numbers from 0 to 10, where 0 signifies “no pain/distress” and 10 “pain/distress as bad as it could be.” The scales have been demonstrated to be a reliable and sensitive measure of these perceptions.37 Our secondary outcome measures were mood status and symptoms, assessed using the validated Chinese version of Profile of Mood States (POMS).42,43 The POMS consisted of 65 adjective rating scales measuring depression, anger, fatigue, vigor, tension, confusion, and total mood disturbance. Higher scores in each POMS subscale indicate more depressive symptoms, angrier mood, greater mental or physical exhaustion, more vigor, more tension or stress, more confusion, and greater mood disturbance. Depressive symptoms were measured by the validated Chinese version of the Centre for Epidemiological StudiesDepression Scale (CES-D).44 A higher CES-D score means the patients is feeling more depressed. Symptoms of anxiety were measured using the validated Chinese-version State-Trait Anxiety Inventory (STAI).45 State-anxiety reflects a transitory emotional state, whereas trait-anxiety is a general tendency towards anxiety caused by perceived threats. Higher STAI scores indicate more intense stateanxiety and traitanxiety. Health-related quality of life was measured by the validated Chinese-version Medical Outcomes Study Short-
726 | www.clinicalpain.com
Clin J Pain
Volume 27, Number 8, October 2011
Form Health Survey (SF-12),46 which consists of physical and mental components, with higher scores indicating better health. Sick leave was factored into the study by asking participants to recall the number of sick leave days taken over the past 12 months at baseline before the intervention. In addition, they were asked to report their sick leave days during the 8-week intervention, immediately postintervention, and at 3 months and 6 months after intervention. The frequency and duration of practice of both the MBSR and MPI in each group was recorded on a Weekly Meditation Form that was collected each week during class and 6 months after intervention.
Statistical Analyses We tested the hypothesis that the MBSR program would reduce both pain intensity and pain-related distress and improve the quality of life in individuals with chronic pain. We also hypothesized that the improvement would be long term and could last for a period of 6 months after intervention. We used an independent sample t test to compare differences in continuous variables between the 2 groups and the w2 test to compare differences in categorical variables between the 2 groups. A Mann-Whitney U test was performed where data were not normally distributed. These analyses were performed using SPSS software, version 13.0 (SPSS Inc., Chicago, IL). To investigate differences in changes in outcome measures between the 2 groups and within each group, we used multilevel modeling analyses that took into account the dependency47 of repeated observations for each participant. Outcome variables that were of interest (pain intensity, pain-related distress, SF-12, POMS, STAI, CESD, and sick leave days due to pain) were separately modeled against time, along with the interactions between time and the groups. These longitudinal multilevel analyses were carried out using MLwiN computer software, version 2.10 beta10 (Centre for Multilevel Modelling, University of Bristol, UK). Analysis was performed following the intention-totreat principle. We analyzed patients with complete data at all time points and, for those with missing data, the last known value was carried forward to replace the missing values. All tests were 2-sided with a significant level of 5%. A separate analysis was conducted per protocol (without following the intention-to-treat principle) in only those who attended at least 5 sessions (each of 21=2 h) to see if results from this separate analysis differed from those analyzed following the intention-to-treat principle.
Sample Size Determination Sample size planning was determined, using pain intensity as the primary outcome measure. According to the study of LeFort et al,48 the mean difference in pain problem severity (in the range 0 to 100) was 11.69 [standard deviation (SD)=18.51] and 1.80 (SD=17.41) in the treatment and control groups respectively. To detect a difference of 9.89 with a pooled SD of 17.97 based on LeFort’s study, we estimated that approximately 50 participants in each group would be required for a 2sample t test with a 5% 2-sided significance level and 80% statistical power. r
2011 Lippincott Williams & Wilkins
Clin J Pain
Volume 27, Number 8, October 2011
MBSR and Interaction Program for Chronic Pain
RESULTS Demographical Characteristics Figure 1 shows a flow diagram of recruitment for this study. Recruitment was stopped once the planned 100 participants were enrolled. The 100 enrolled participants were randomly divided into 2 treatment groups: 51 in the MBSR group and 49 in the MPI group. A total of 80 participants completed all 4 questionnaires at baseline, immediately postintervention (evaluation occurred immediately after completion of the intervention) and at 3 months and 6 months after intervention. One participant in the MPI group dropped out after 1 session; she claimed that she had no time to continue the program. The drop-out participant did not return any of the 4 questionnaires, resulting in only 99 participants being included in our analysis.
The baseline characteristics of the 99 participants with chronic pain were summarized in Table 1. There were no statistically significant differences between the 2 groups in baseline characteristics that included demographics, the 11point NRS pain score, the 11-point pain-related distress score, quality-of-life, mood symptoms, and pain-related characteristics. The information on the use of pain analgesics is presented in Table 2.
Attrition and Practice of the Intervention A total of 83 participants completed the interventions. Completion of intervention was defined by having attended more than half of the sessions of either intervention: 39 of 51 (76%) in the MBSR group and 44 of 49 (90%) in the MPI group. Participants in the MPI group demonstrated higher adherence to the intervention when compared with those of the MBSR group (w2df=1 =4.21, P=0.040).
Individuals showed interest (called & left contacts) n = 291
Screened by trained research assistant on phone Interviewed with healthcare professional n = 112
Enrolled participants n = 100
Not enrolled to interview n=179 •Excluded by RA n=163 •Refused n=16
Not enrolled n=12 · Excluded by healthcare professional= 11 · Declined n= 1
Randomization
Allocated to MBSR n= 51
Allocation
Baseline questionnaire n=51 Not complete MBSR program n= 10
8-week MBSR Intervention
Allocated to MPI n= 49
Baseline questionnaire n=49
Intervention
8-week MPI Intervention
Post-intervention questionnaire returned 0 months n= 44 3 months n= 40 6 months n= 41 Completed all 4 questionnaires n=38
Follow-up
Post-intervention questionnaire returned 0 months n= 46 3 months n= 45 6 months n= 45 Completed all 4 questionnaires n = 42
Analyzed MBSR n= 51
Analysis
Analyzed MPI n= 48
Not complete MBSR program n= 5
FIGURE 1. Mindfulness-based stress reduction for chronic pain participant flow diagram. r
2011 Lippincott Williams & Wilkins
www.clinicalpain.com |
727
Clin J Pain
Wong et al
Volume 27, Number 8, October 2011
TABLE 1. Baseline Characteristics of 100 Participants With Chronic Pain
All participants, N=99
MBSR Group, N=51
MPI Group, N=48
P*
47.9 (7.84)
48.7 (7.84)
47.1 (7.82)
0.285 0.233
27 18 54
11 12 28
16 6 26
22 53 11 13
10 31 4 6
12 22 7 7
19 65 15
7 36 8
12 29 7
22 23 19 15
9 10 13 8
13 13 6 7
46 13 19 12
27 7 7 3
19 6 12 9
25 16 11
15 8 5
10 8 6
7 16 20 16 33
2 7 9 8 20
5 9 11 8 13
83 7.77 (2.87)
44 7.22 (3.10)
39 8.50 (2.30)
57 (68%) 27 (32%) 2.39 (1.47)
26 (65%) 14 (35%) 2.22 (1.27)
31 (70%) 13 (30%) 2.56 (1.67)
Demographics Age (y), mean (SD) Employment status Housewife and student Retired and unemployed Employed and self-employed Education level Primary education Secondary education up to F.5 Matriculation course to diplomas’ Tertiary education Marital status Single Married Otherw Household monthly incomez <$10,000 $10,001-$20,000 $20,001-$30,000 >$30,000 Religious belief No religious belief Catholic Christian Othery Pain related characteristics Sick leaves in the past 12 mo (d)J 0d 1-7 d >7 d Year(s) that the pain lasts for 0.5-1 y 1-3 y 3-5 y 6-10 y >10 y Attrition rate and other characteristics Completed more than 50% of the 10 sessionsz Attendance (sessions) (SD)z Participants who keep the 45-minute exercise (8 mo after intervention) Yes (%) No (%) Expected pain level after intervention#
0.472
0.359
0.290
0.125
0.674
0.522
0.760 0.022 0.593
0.414
The bold values in the table indicate that the corresponding p-value is smaller than or equal to the 5% statistical significance level. *P value by statistical tests (t test for continuous data; w2 test for categorical data) for group differences. wOther marital status include: separated, live alone, divorced, widowed. zSimilar results with monthly personal income for those employed or self employed. yOther religious belief include: Buddhism, Daoism, and Islam. JSick leaves reported by employed and self-employed participants were included. zTotal of 10 sessions consists of 8 weekly 2.5-hour sessions and retreat (2.5 h in the morning and afternoon respectively). #Expected pain level after intervention is an 11-point scale of postintervention pain level that participants expected at baseline. Questions were asked before the start of the first session after they were randomly assigned to 1 of the 2 groups. MBSR indicates Mindfulness-based Stress Reduction; MPI, Multidisciplinary Pain Intervention.
TABLE 2. Self-report Use of Analgesics at Baseline Acetaminophen Rheumatic painkiller Opioid None
MBSR Group, N=51
MPI Group, N=48
P*
33 15 1 9
31 15 0 6
0.99 0.842 —w 0.475
*The between-group difference was tested using w2 test. wThere were only 1 participant-reported use of narcotic analgesics, so, no w2 test was performed in this case.
728 | www.clinicalpain.com
r
2011 Lippincott Williams & Wilkins
Clin J Pain
Volume 27, Number 8, October 2011
In addition, participants in the MPI group attended on average 1.28 (standard error=0.55) more sessions (t, 2.351; P=0.021) than the MBSR group. Six months after intervention, 84 questionnaires were returned: 26 of 40 (65%) of participants in MBSR and 31 of 44 (71%) in the MPI group claimed that they were still practicing meditation or prescribed exercise. Overall, MBSR participants practiced meditation 3.6 (SD=2.02) times per week, whereas participants in the MPI group practiced prescribed exercises 3.9 (SD=2.18) times per week. The difference in practicing time of the 2 groups was not statistically significant (t, 0.51, P=0.609). At baseline, participants expected that the pain intensity would be reduced by 4.14 units (SD=1.78 units) on average after intervention. Mean expectation of the 2 groups was –4.15 units (SD=1.71 units) and 4.14 units (SD=1.86 units) for MBSR and MPI participants respectively.
Pain and Medical History Approximately half (49 of 99) of the participants had chronic pain for >5 years and one third (33 of 99) of the participants for >10 years. The majority of the participants listed back pain (85 of 99) and joint pain (83 of 99) as the most irritating type of pain that they experienced. Participants’ reported use of analgesics is shown in Table 2. Acetaminophen was the most commonly used pain-relief medication, with 65% of the participants stating that they had used it. The 2 groups did not differ in the use of analgesics. Approximately 70% of participants in both groups had no chronic diseases and about 47% of participants in each group reported no medical diagnosis or condition that caused their pain.
Within-group Differences in Primary and Secondary Outcome Measures Table 3 shows the results of both within-group and between-group differences, following multilevel analysis. The pain intensity and pain-related distress of both MBSR and MPI groups improved significantly from baseline. For the MBSR group, pain intensity measured by 11-point NRS reduced by 0.57 units and pain-related distress reduced by 0.37 units. For the MPI group, pain intensity reduced by 0.61 units and pain-related distress reduced by 1.08 units. Figures 2A and B show the change in pain intensity and pain-related distress over the study period in both intervention groups respectively. Figure 2C shows the within-group POMS vigoractivity component score plotted against time. Immediately after intervention, the mean POMS vigor-activity scores of MBSR group increased by 0.88 units, whereas the scores of the MPI group decreased by 0.93 units. The mean SF-12 physical component (PCS12) immediately after the 8-week intervention of both MBSR and MPI did not significantly differ from baseline. However, there were statistically significant increases in the PCS12 within both the MBSR and MPI groups at 3-month (Wald statistic=4.62, P=0.032) and 6-month (Wald statistic=10.503, P=0.001) postintervention, when compared with the scores at baseline. Other analyses of within-group effects—including scores on POMS, depression measured using CES-D, anxiety measured using the STAI, and the number of sick leaves taken in the past 3-months—did not show significant improvements. r
2011 Lippincott Williams & Wilkins
MBSR and Interaction Program for Chronic Pain
“Between Group” Differences in Outcomes Measures by Multilevel Modeling Analysis Baseline pain intensity and pain-related distress did not differ between MBSR and MPI groups. The mean pain intensity scores were 6.55 (SD=1.50) and 6.76 (SD=1.26) for MBSR and MPI respectively, whereas the mean pain-related distress scores were 6.49 (SD=2.11) and 6.58 (SD= 2.23). Table 3 shows the “between-group,” and the “withingroup” differences in outcome measures immediately after intervention. There was no statistically significant difference in pain intensity between the 2 groups. In contrast, there was a statistically significant difference between the 2 groups in pain-related distress (Wald statistic=3.98; P=0.046) with participants in the MPI group having more reduction of pain-related distress (a difference of 0.71 units; standard error=0.356 in the MPI group). There were also statistically significant “betweengroup” differences in the POMS vigor-activity component score immediately postintervention (Wald statistic=4.05, P=0.044), although there was no statistically significant “between-group” differences in these outcome measures at 3 months and 6 months postintervention. No statistically significant difference was observed in the SF-12 mean outcome scores between the 2 groups at baseline, immediately postintervention, 3 months and 6 months postintervention. There was no statistically significant difference between the 2 groups in the State-Trait Anxiety Score at baseline, immediately postintervention, 3 and 6 months postintervention, but there was a trend for further decrease in anxiety in the MBSR group. There was no statistically significant difference between the 2 groups in the change of depressive symptoms measured by the CES-D, in the mood (POMS) score, and the number of days of sick leave between the 2 groups at baseline, immediately postintervention and at 3 months and 6 months postintervention. When the results were analyzed per protocol on those who only attended for >50% of the sessions in each group, the results were unchanged.
DISCUSSION This is one of the few published studies that used a randomized clinical trial design to study the effects of MBSR on chronic pain intensity with a comparative group that simulated a multidisciplinary psycho-education group. This is also a randomized clinical trial that studied the effects of MBSR on chronic pain in a non-White population. Previous studies that showed positive effects of MBSR on chronic pain were conducted mainly using a nonrandomized, preintervention and postintervention design 21,48–50 although recently, more randomized controlled or comparative studies have been conducted33,51–55 Recently, Morone et al33 conducted a randomized, controlled clinical trial on 37 community dwelling older adults aged 65 years and above, to compare the effects of MBSR with a waiting-list control group on chronic pain. They found that those who were in the MBSR group showed significant improvement in the Chronic Pain Acceptance Questionnaire Total Score and the Activities Engagement subscale and the SF-36 Physical Function Scale. As it was a feasibility trial, it was not robust enough to detect any statistical significant difference between the 2 groups in pain intensity and quality-of-life score www.clinicalpain.com |
729
Clin J Pain
Wong et al
Volume 27, Number 8, October 2011
TABLE 3. Between-group Differences in Main Outcome Measures of Pain Intensity and Pain-related Distress Analyzed by Multilevel Model Analysis With Respect to Time of 99 Participants With Chronic Pain
MBSR Group
Outcome Variables Pain intensity Baseline Postintervention 3-month postintervention 6-month postintervention Pain-related distress Baseline Postintervention 3-month postintervention 6-month postintervention SF-12 Physical component Baseline Postintervention 3-month postintervention 6-month postintervention SF-12 mental component Baseline Postintervention 3-month postintervention 6-month postintervention STAI: state anxiety Baseline Postintervention 3-month postintervention 6-month postintervention STAI: trait anxiety Baseline Postintervention 3-month postintervention 6-month postintervention POMS: tension anxiety Baseline Postintervention 3-month postintervention 6-month postintervention POMS: depression dejection Baseline Postintervention 3-month postintervention 6-month postintervention POMS: anger hostility Baseline Postintervention 3-month postintervention 6-month postintervention POMS: vigor activity Baseline Postintervention 3-month postintervention 6-month postintervention POMS: fatigue inertia Baseline Postintervention 3-month postintervention 6-month postintervention POMS: confusion bewilderments Baseline Postintervention 3-month postintervention 6-month postintervention
Mean* (SD)
Change From Baseline, Mean (SE)
MPI Group Mean* (SD)
Multilevel Analysis
Change From Baseline, Mean (SE)
N
Wald
Pw
6.55 5.98 5.84 5.53
(1.50) (1.63) (1.49) (1.94)
0.57 (0.16) 0.71 (0.22) 1.02 (0.23)
6.76 6.15 5.85 5.79
(1.26) (1.65) (1.90) (1.84)
0.61 (0.22) 0.91 (0.27) 0.97 (0.29)
98 98 98
0.022 0.416 0.027
0.882 0.517 0.869
6.49 6.12 5.70 5.34
(2.12) (1.94) (1.20) (2.19)
0.37 (0.23) 0.79 (0.24) 1.15 (0.30)
6.75 5.67 5.60 5.56
(1.81) (1.88) (1.93) (1.85)
1.08 (0.25) 1.15 (0.31) 1.19 (0.31)
98 98 98
3.977 0.974 0.013
0.046 0.324 0.910
35.27 34.67 36.98 37.70
(8.04) (8.56) (9.34) (10.19)
0.59 (0.71) 1.71 (0.86) 2.44 (0.90)
32.46 31.79 32.57 33.64
(6.88) (7.60) (7.83) (8.56)
0.68 (0.74) 0.10 (0.84) 1.18 (0.85)
88 88 88
0.005 2.341 1.250
0.941 0.126 0.264
40.63 43.02 43.91 42.87
(11.21) (9.84) (10.92) (11.27)
2.39 (1.01) 3.28 (1.58) 2.24 (1.57)
39.32 42.05 43.08 42.32
(9.16) (11.52) (11.55) (10.30)
2.73 (1.32) 3.76 (1.32) 3.00 (1.37)
88 88 88
0.030 0.002 0.026
0.862 0.960 0.872
48.21 46.66 45.72 45.06
(12.35) (12.55) (12.63) (13.16)
1.55 (1.11) 2.49 (1.31) 3.15 (1.59)
46.76 46.55 45.25 45.10
(9.73) (10.20) (10.96) (12.04)
0.21(1.02) 1.50 (1.37) 1.67 (1.37)
88 88 88
0.644 0.486 1.730
0.422 0.486 0.188
44.98 44.96 45.00 44.56
(9.50) (9.30) (9.78) (9.67)
0.22 (0.86) 0.02 (1.09) 0.42 (1.13)
46.76 46.55 45.26 45.10
(9.73) (10.20) (10.96) (12.04)
0.16 (0.73) 1.07 (0.86) 0.47 (1.06)
87 87 87
0.011 0.157 0.264
0.916 0.692 0.607
12.52 11.86 11.72 11.50
(8.51) (8.95) (9.43) (8.45)
0.66 (0.75) 0.80 (0.99) 1.02 (1.01)
11.83 12.61 12.30 12.26
(7.33) (6.97) (7.31) (8.73)
0.78 (0.76) 0.48 (0.84) 0.44 (6.20)
95 95 95
1.565 0.980 1.603
0.211 0.322 0.205
15.29 14.53 14.61 14.00
(13.68) (13.77) (14.46) (12.57)
0.76 (1.08) 0.67 (1.62) 1.21 (1.49)
15.29 16.16 15.47 15.96
(11.72) (13.00) (12.21) (14.11)
0.87 (1.08) 0.18 (1.11) 0.67 (1.15)
93 93 93
1.006 0.275 1.430
0.316 0.600 0.232
9.85 9.69 9.31 9.19
(8.22) (9.11) (8.74) (8.29)
0.17 (0.95) 0.54 (1.29) 0.67 (1.27)
9.60 10.65 10.02 10.50
(7.10) (8.07) (7.90) (8.60)
1.04 (0.91) 0.42 (0.97) 0.90 (0.94)
95 95 95
0.794 0.536 1.943
0.373 0.464 0.163
12.77 13.65 14.06 13.59
(4.91) (5.48) (5.41) (5.26)
0.88 (0.59) 1.29 (0.71) 5.08 (0.71)
13.39 12.45 13.67 14.20
(5.93) (4.87) (5.68) (6.07)
0.93 (0.74) 0.283 (0.57) 0.80 (0.65)
96 96 96
4.052 1.343 0.0001
0.044 0.247 0.990
12.55 12.29 11.49 12.55
(6.69) (7.17) (7.52) (6.69)
0.25 (0.61) 1.06 (0.89) 0.10 (0.65)
13.32 13.55 13.19 13.02
(6.72) (6.33) (5.77) (6.66)
0.23 (0.84) 0.13 (0.84) 0.30 (5.23)
97 97 97
0.236 0.661 0.056
0.627 0.416 0.813
9.20 9.43 9.63 9.27
(5.18) (5.70) (6.14) (5.34)
0.24 (0.53) 0.42 (0.66) 0.088 (0.56)
9.19 9.83 9.23 9.47
(4.40) (4.89) (4.03) (4.97)
0.64 (0.52) 0.43 (0.51) 0.28 (0.62)
97 97 97
0.314 0.280 0.094
0.576 0.597 0.760
(continued)
730 | www.clinicalpain.com
r
2011 Lippincott Williams & Wilkins
Clin J Pain
Volume 27, Number 8, October 2011
MBSR and Interaction Program for Chronic Pain
TABLE 3. (continued) MBSR Group
Outcome Variables Depression (CES-D) Baseline Postintervention 3-month postintervention 6-month postintervention Sick leaves Baseline Postintervention 3-month postintervention 6-month postintervention
Mean* (SD) 35.85 33.78 34.24 34.48
(8.91) (8.40) (9.94) (8.85)
12.86 9.61 10.81 9.96
(36.31) (34.35) (34.95) (34.48)
Change From Baseline, Mean (SE)
MPI Group Mean* (SD)
2.07 (0.68) 1.61 (0.87) 1.37 (1.10)
35.73 34.49 34.11 34.60
(6.55) (7.45) (7.00) (8.02)
3.25 (3.08) 2.30 (3.33) 2.89 (3.00)
19.21 6.48 7.23 6.93
(73.82) (24.24) (24.19) (24.86)
Multilevel Analysis
Change From Baseline, Mean (SE)
N
Wald
Pw
1.24 (0.62) 1.62 (0.86) 1.13 (0.91)
90 90 90
0.00005 0.198 0.072
0.994 0.656 0.789
49 49 49
1.680 3.721 1.849
0.195 0.538 0.174
12.73 (10.15) 11.98 (10.16) 12.89 (10.57)
The bold values in the table indicate that the corresponding p-value is smaller than or equal to the 5% statistical significance level. *The mean of changes from the baseline score of individual participants. wTo test differences between groups, multilevel model analysis of outcome variables against time along with the interactions between time and group was performed; the corresponding P value was based on the Wald statistic which follows a w2 distribution with 1 degree of freedom. CES-D indicates Centre for Epidemiological Studies-Depression Scale; MBSR, mindfulness-based stress reduction; MPI, multidisciplinary pain intervention; POMS, profile of mood states; SE, standard error.
measured by SF-36 (Global, Physical, and Mental Health Composite), although a trend favoring MBSR was observed. We did not measure pain acceptance, and therefore this aspect of possible effects of MBSR could not be compared. However, we used 3 scales that measured patients’ mood symptoms, including POMS, STAI, and CES-D, which were not used in Morone’s study and we showed that there was no statistically significant difference between the MBSR and active control groups in these mood symptoms using a larger sample size. For pain intensity and pain-related distress, although each group showed significant improvement, no statistically significant difference was seen between the 2 groups. Moreover, the decrease in pain intensity and pain-related distress was small and inconsistent suggesting that the effects from both interventions were rather weak. Another recent pilot study was conducted by Kingston et al36 on 42 university students, to evaluate the effects of mindfulness training on pain tolerance and psychological well-being and acquisition of mindfulness skills. They showed that pain tolerance significantly increased in those assigned to the mindfulness group, as compared with those assigned to the control group. However, their study was small and was conducted on individuals with no history or symptoms of chronic pain, making it hard to generalize to patients with chronic pain. In the Kingston et al36 study, twice as many participants in the MBSR group dropped out, as compared with participants in the control group. One of the reasons given for the drop out was that “they had difficulty in understanding the course materials and physical constraint in the posture.” As there were explicit verbal and written instructions that they did not need to practice any postures that they were not comfortable with, the reason for dropping out was probably their difficulty in understanding the course materials. This raised the important issue of patient-therapy match,9 which has been shown to be an important factor in the improvement of psychological symptoms in studies on mindfulness-based cognitive therapy.30,31 Some patients with chronic pain may simply find that the practice of paying attention to their body in an accepting and compassionate r
2011 Lippincott Williams & Wilkins
way, does not make any sense to them. Thus, it is possible that a subset of chronic pain patients who are more accepting of MBSR may still benefit from it. Indeed, researchers9,51 have suggested that one possible way to improve treatment efficacy in chronic pain is to better match treatments to patient characteristics, treatment components, and patienttreatment interactions similar to those in psychological research. This is an important issue as a recent study demonstrated that MBSR may be particularly effective in this patient population with chronic pain.52 Although there was no statistically significant difference in the duration of chronic pain history in both groups at baseline, participants in the MBSR group had a longer duration of chronic pain when compared with that of the MPI group. Thus, inherent differences in patient characteristics could still be present between the 2 groups that might have affected the results and that were not accounted for in this study due to the relatively small sample size in each group. We excluded patients with a concurrent reported diagnosis of Axis I psychiatric disorder, with the aim to decrease heterogeneity of participants. This limits the general applicability of our results to chronic pain patients with depression (which is quite common in this category of patient). Our study demonstrated a clear superiority over previous studies in investigating the effects of MBSR on chronic pain. Ours was the only study that used a randomized trial design that could account for the benefits or improvement of pain due to attention received from therapist and group attention. In addition, as we showed that the expectation of treatment effectiveness was similar between the intervention and control group at baseline, we have controlled for differences in health outcomes, attributed to differences in expectation of treatment effectiveness, which could be a potential confounding factors in trials with a waiting-list control. In contrast, the multidisciplinary education program used in this study was constructed, based on the contents of a popular self-help book with its purpose based on controlling, rather than accepting, pain. Improvements observed in pain intensity in both www.clinicalpain.com |
731
Clin J Pain
Wong et al
B Change in pain intensity of participants in MBSR and MPI groups over the study period
7.0
Pain Intensity
*
+
+
* *
6.0
Treatment group MBSR group MPI group
7.0
+ 6.5
Change in pain-related distress of participants in MBSR and MPI groups over the study period 7.5
Treatment group MBSR group MPI group
Pain-related Distress
A
Volume 27, Number 8, October 2011
5.5
O 6.5 *
+
+
+ *
6.0
5.5
5.0 5.0
C
3m in on te th rv en po tio stn 6m in on te th rv en po tio stn
n
po
st
-in
te
rv
en
tio
Ba se lin e
Change in Profile of Mood States Vigor-activity subscale of participants in MBSR and MPI groups over the study period 16.00
CPOMS: Vigor-activity subscale
6m in on te th rv en po tio stn
3m in on te th rv en po tio stn
po st -in te rv en tio n
Ba se lin e
4.5
Treatment group MBSR group o MPI group
*
*
15.00
+
14.00
13.00
12.00
3m in on te th rv en po tio stn 6m in on te th rv en po tio stn
po st -in te rv en tio n
Ba se lin e
11.00
FIGURE 2. Change in (A) pain intensity, (B) pain-related distress, and (C) Profile of Mood States Vigor-activity subscale, of participants in Mindfulness-Based Stress Reduction (MBSR) and Multidisplinary Pain Intervention (MPI) group over the study period. The asterisk (*) indicates significant difference in the specific outcome measure compared with the baseline measure in MBSR; the plus sign (+) indicates significant difference in the specific outcome measure compared with the baseline measure in MPI; the circle (o) indicates significant group difference at the time point. Error bars show 95% confidence interval of means.
mindfulness and MPI group showed that these nonmindfulness techniques may be beneďŹ cial to patients as well. We used measurements that included a number of outcome measures that were important for patients with chronic pain that included both depressive and anxiety symptoms and quality-of-life measures, which have all been
732 | www.clinicalpain.com
validated locally. Moreover, all these outcome measures were recommended by the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials group53 as valid outcome measures for studies on chronic pain. Our randomization procedures were rigorous, which ensured that bias was minimized. Furthermore, our study r
2011 Lippincott Williams & Wilkins
Clin J Pain
Volume 27, Number 8, October 2011
participant’s retention was high and we have analyzed our results using both per protocol and intention-to-treat analysis, which yielded similar results.
MBSR and Interaction Program for Chronic Pain
for Clinical Trials and Epidemiological Research, The Chinese University of Hong Kong, Hong Kong, for his professional statistical consultation.
Limitations Our study had several limitations. First, we only followed our participants for a maximum of 6 months postintervention, and thus were unable to demonstrate any potential between group differences in outcome measures beyond 6 months. Second, for the MBSR group, only a proportion of participants practiced daily mindfulness meditation for the duration recommended by the therapist. As a result, our results might have been affected by the extent of participants’ adherence to recommended meditation practice. In addition, due to the number of drop outs, our sample size might not have been large enough to show significant differences in outcome measures between the 2 groups. Third, we only used self-reported sick leave as a proxy measure for the effects of intervention on disability. As a result, it was subject to recall bias. No objective measures for disability associated with chronic pain were used in our study, as the authors could not find a validated Chinese-version questionnaire on disability associated with chronic pain. Thus, changes in disability in participants, which was an important outcome measure for chronic pain intervention, might not be reflected by our study results. Furthermore, no measurement of pain acceptance was made, which might have been an important outcome measure. Fourth, the effectiveness of any remedial therapy is highly dependent on the therapist’s technique and experience. The importance of therapist effect is a powerful and important influence in psychological treatments,54 and it is difficult to differentiate therapist from treatment variance in our study.55 Although both our therapists were experienced health-care workers, they both had limited experience in treating chronic pain. To reduce the effects of therapist variation, we only included 1 main therapist in each group with a well-structured, standardized protocol for intervention.
CONCLUSIONS This randomized comparative clinical trial showed that both MBSR and a multidisciplinary intervention group improved perceptions of pain intensity and painrelated distress, although there was no statistically significant difference in these attributes between the 2 groups at 6 months after intervention. ACKNOWLEDGMENTS The authors thank Winny Lau, RD (dietician), Dominic Lo, MSc (physiotherapist) and Anthony Kwok (M Health Admin & Info Sys, Centre Manager and Physiotherapist) from The Jockey Club Center for Osteoporosis Care and Control, The Chinese University of Hong Kong, Hong Kong, for contributing to the Multidisciplinary Pain interaction Program. The authors also thank Karen Lee, MA, MSocSc (Research Assistant) and Claudia Wong, BPsySc (Research Assistant) from The Department of Community and Family Medicine, The Chinese University of Hong Kong, Hong Kong, for their logistical support on the interventions for chronic pain sufferers. The authors are also grateful to David Wilmshurst, PhD, BA, The Research Administration Office, CUHK’s Academic Editor, for editing a draft of this study. Last but not least, the authors thank Eric Wong, MSc, Centre r
2011 Lippincott Williams & Wilkins
REFERENCES 1. Bair MJ, Robinson RL, Katon W, et al. Depression and pain comorbidity: a literature review. Arch Intern Med. 2003;163: 2433–2445. 2. Benca RM, Ancoli-Israel S, Moldofsky H. Special considerations in insomnia diagnosis and management: depressed, elderly, and chronic pain populations. J Clin Psychiatry. 2004; 8:26–35. 3. Cooper JK, Kohlmann T. Factors associated with health status of older Americans. Age Ageing. 2001;30:495–501. 4. Loeser JD, Melzack R. Pain: an overview. Lancet. 1999;353: 1607–1609. 5. Ng KF, Tsui SL, Chan WS. Prevalence of common chronic pain in Hong Kong adults. Clin J Pain. 2002;18:275–281. 6. Lachaine J, Beauchemin C, Landry PA. Clinical and economic characteristics of patients with fibromyalgia syndrome. Clin J Pain. 2010;26:284–290. 7. Keefe FJ, Beaupre PM, Gil KM, et al. Group therapy for patients with chronic pain. In: Turk DC, Gatchel RJ, ed. Psychological Approaches to Pain Management: a Practitioner’s Handbook. New York: Guilford Press; 2002:234–255. 8. Turk DC. Cognitive-behavioral approach to the treatment of chronic pain patients. Reg Anesth Pain Med. 2003;28: 573–579. 9. McCracken LM, Turk DC. Behavioral and cognitive-behavioral treatment for chronic pain. Spine. 2002;27:2564–2573. 10. Asmundson GJG, Norton GR, Allergings MD. Fear and avoidance in dysfunctional chronic back pain patients. Pain. 1997;69:231–236. 11. Asmundson GJG, Norton PJ, Norton GR. Beyond pain: the role of fear and avoidance in chronicity. Clin Psychol Rev. 1999;19:97–119. 12. Crombez G, Vlaeyen JWS, Heuts PHTG, et al. Pain-related fear is more disabling than pain itself: evidence on the role of pain-related fear in chronic back pain disability. Pain. 1999;80:329–339. 13. Reitsma B, Meijler WJ. Pain and patienthood. Clin J Pain. 1997;13:9–21. 14. Pull CB. Current empirical status of acceptance and commitment therapy. Curr Opin Psychiatry. 2008;22:55–60. 15. McCracken LM, Spertus IL, Janeck AS, et al. Behavioral dimensions of adjustment in persons with chronic pain: painrelated anxiety and acceptance. Pain. 1999;80:283–289. 16. Hayes SC, Strosahl KD, Wilson KG. Acceptance and Commitment Therapy: an Experiential Approach to Behaviour Change. New York: Guilford Press; 1999. 17. Bransteetter-Rost A, Cushing C, Douleh T. Personal values and pain tolerance: does a values intervention add to acceptance? J Pain. 2009;10:887–892. 18. Wicksell RK, Mellin L, Lekander M, et al. Evaluating the effectiveness of exposure and acceptance strategies to improve functioning and quality of life in long standing pediatric pain: a randomized controlled trial. Pain. 2008;141:248–257. 19. Ost LG. Efficacy of the third wave of behavioural therapies: a systematic review and meta-analysis. Behav Res Ther. 2008;46: 296–321. 20. Kabat-Zinn J. An outpatient programme in behavioural medicine for chronic pain patients based on the practice of mindfulness meditation: theoretical considerations and preliminary results. Gen Hosp Psychiatry. 1982;4:33–42. 21. Kabat-Zinn J. Full Catastrophe Living: Using the Wisdom of Your Body and Mind to Face Stress, Pain, and Illness. New York: Delacourt; 1990. 22. Shapiro SL, Carlson LE, Astin JA, et al. Mechanisms of mindfulness. J Clin Psychol. 2006;62:373–386.
www.clinicalpain.com |
733
Wong et al
23. Kristeller JL, Hallett CB. An exploratory study of a meditation-based intervention for binge eating disorder. J Health Psychol. 1999;4:357–363. 24. Kaplan KH, Goldenberg DL, Galvin-Nadeau M. The impact of a meditation-based stress reduction programme on fibromyalgia. Gen Hosp Psychiatry. 1993;15:284–289. 25. Kabat-Zinn J, Wheeler E, Light T, et al. Influence of a mindfulness meditation-based stress reduction intervention on rates of skin clearing in patients with moderate to severe Psoriasis Undergoing Phototherapy (UVB) and Photochemotherapy (PUVA). Psychosom Med. 1998;60:635. 26. Kabat-Zinn J, Massion AO, Kristeller J, et al. Effectiveness of a meditation-based stress reduction programme in the treatment of anxiety disorders. Am J Psychiatry. 1992;149:936–943. 27. Miller JJ, Fletcher K, Kabat-Zinn J. Three-year follow-up and clinical implications of a mindfulness meditation-based stress reduction intervention in the treatment of anxiety disorders. Gen Hosp Psychiatry. 1995;17:192–200. 28. Speca M, Carlson LE, Goodey E, et al. A Randomized, waitlist controlled clinical trial: the effect of a mindfulness meditation-based stress reduction programme on mood and symptoms of stress in cancer outpatients. Psychosom Med. 2000;62:613–622. 29. Carlson LE, Ursuliak Z, Goodey E, et al. The effects of a mindfulness meditation based stress reduction programme on mood and symptoms of stress in cancer outpatients: 6-Month follow-up. Support Care Cancer. 2001;9:112–123. 30. Teasdale JD, Segal ZV, Williams JMG, et al. Prevention of relapse/recurrence in major depression by mindfulness-based cognitive therapy. J Consult Clin Psychol. 2000;68: 615–623. 31. Ma SH, Teasdale JD. Mindfulness-based cognitive therapy for depression: replication and exploration of differential relapse prevention effects. J Consult Clin Psychol. 2004;72:31–40. 32. Sephton SE, Salmon P, Weissbecker I, et al. Mindfulness meditation alleviates depressive symptoms in women with fibromyalgia: results of a randomized clinical trial. Arthritis Rheum. 2007;57:77–85. 33. Morone NE, Greco CM, Weiner DK. Mindfulness meditation for the treatment of chronic low back pain in older adults: a randomized controlled pilot study. Pain. 2008;134:310–319. 34. Plews-Ogan M, Owens JE, Goodman M, et al. A pilot study evaluating mindfulness based stress reduction and massage for the management of chronic pain. J Gen Intern Med. 2005;20: 1136–1138. 35. Astin JA. Stress reduction through mindfulness meditation: effects on psychological symptomatology, sense of control and spiritual experiences. Psychother Psychosom. 1997;66: 97–106. 36. Kingston J, Chadwick P, Meron D, et al. A pilot randomized control trial investigating the effect of mindfulness practice on pain tolerance, psychological well-being and physiological activity. J Psychosom Res. 2007;62:297–300.46. 37. Jensen MP, Karoly P, O’Riordan EF, et al. The subjective experience of acute pain: an assessment of the study of 10 indices. Clin J Pain. 1989;5:153–159.
734 | www.clinicalpain.com
Clin J Pain
Volume 27, Number 8, October 2011
38. http://www.cct.cuhk.edu.hk/eng/services/info.html accessed on 29 June 2007. 39. Caudill M, Schnable R, Zuttenneister P, et al. Decreased clinic use by chronic pain patients: responses to behavioural medicine intervention. Clin J Pain. 1991;7:305–310. 40. Caudill MA. Managing Pain Before it Manages You (revised edition). New York: Guilford Press; 2002. [Chapters 6-9]. 41. Von Korff M, Jensen MP, Karoly P. Assessing global pain severity by self-report in clinical and health services research. Spine. 2000;25:310–351. 42. McNair DM, Lorr M, Droppleman LF. Profile of Mood States (POMS], Educational and Industrial Testing Service. Calif: San Diego; 1971. 43. Molassiotis A, Holroyd E, Taylor-Piliae R. Psychometric Evaluation of the Chinese Translation of the Profile of Mood States. HK Nurs J. 1999;35:7–12. 44. Cheung CK, Bagley C. Validating an American scale in Hong Kong: the Centre of Epidemiological Studies Depression Scale (CES-D]. J Psychol. 1998;132:169–186. 45. Shek DTL. The Chinese version of the State-Trait Anxiety Inventory: its relationship to different measures of psychological well-being. J Clin Psychol. 1993;49:349–358. 46. Lam CLK, Tse EYY, Gandek B. Is The Standard SF-12 Health Survey valid and equivalent for a Chinese population? Qual Life Res. 2005;14:539–547. 47. Twish J. Applied Multilevel Analysis: a Practical Guide. Cambridge, UK; New York: Cambridge University Press; 2006. 48. LeFort SM, Gray-Donald K, Rowat KM, et al. Randomized controlled trial of a community-based psychoeducation programme for the self-management of chronic pain. Pain. 1998;74:297–306. 49. Kabat-Zinn J, Lipworth L, Burney R, et al. Four-year followup of a meditation based programme for the self-regulation of chronic pain: treatment outcomes and compliance. Clin J Pain. 1986;2:159–173. 50. Creamer P, Singh BB, Hochberg MC, et al. Sustained improvement produced by non-pharmacologic intervention in Fibromyalgia: results of a pilot study. Arthritis Care Res. 2000;13:198–204.45. 51. Vlaeyen JWS, Morley S. Cognitive-behavioral treatments for chronic pain: what works for whom? Clin J Pain. 2005; 21:1–8. 52. Zautra AJ, Davis MC, Reich JW, et al. Comparison of cognitive behavioral and mindfulness meditation interventions on adaptation to rheumatoid arthritis for patients with and without history of recurrent depression. J Consult Clin Psychol. 2008;76:408–421. 53. Dworkin RH, Turk DC, Haythornthwaite J, et al. Core outcome measures for chronic pain clinical trials: IMMPACT recommendations. Pain. 2005;113:9–19. 54. Wampold BE, Brown CS. Estimating variability in outcomes attributable to therapists: a naturalistic study of outcomes in managed care. J Consult Clin Psychol. 2005;73: 914–923. 55. Morley S. RCTs of psychological treatments for chronic pain: progress and challenges. Pain. 2006;121:171–172.
r
2011 Lippincott Williams & Wilkins