- Email: [email protected]
A prospective study of acceptance of pain and patient functioning with chronic pain
Pain 118 (2005) 164–169 www.elsevier.com/locate/pain
A prospective study of acceptance of pain and patient functioning with chronic pain Lance M. McCracken*, Christopher Eccleston Pain Management Unit, Royal National Hospital for Rheumatic Diseases, The University of Bath, Bath BA1 1RL, UK Received 15 September 2004; received in revised form 23 June 2005; accepted 8 August 2005
Abstract Acceptance of chronic pain is emerging as an important concept in understanding ways that chronic pain sufferers can remain engaged with valued aspects of life. Recent studies have relied heavily on cross-sectional investigations at a single time point. The present study sought to prospectively investigate relations between acceptance of chronic pain and patient functioning. A sample of adults referred for interdisciplinary treatment of severe and disabling chronic pain was assessed twice, an average of 3.9 months apart. Results showed that pain and acceptance were largely unrelated. Pain at Time 2 was weakly related to measures of functioning at Time 2. On the other hand, acceptance at Time 1 was consistently related to patient functioning at Time 2. Those patients who reported greater acceptance at Time 1 reported better emotional, social, and physical functioning, less medication consumption, and better work status at Time 2. These data suggest that willingness to have pain, and to engage in activity regardless of pain, can lead to healthy functioning for patients with chronic pain. Treatment outcome and process studies may demonstrate the potential for acceptance-based clinical methods for chronic pain management. q 2005 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. Keywords: Chronic pain; Behavioral approaches; Patient adjustment; Acceptance
1. Introduction Chronic pain provides the sufferer with many occasions in which their responses to pain can move them away from healthy life functioning. They may evaluate their pain in distressing ways and take these evaluations to be true. They also may regard pain and other feelings as firm reasons to disengage from important valued aspects of their life, and generally act to control or avoid painful experiences, despite the lack of positive effect these behaviors produce. This type of behavior pattern has been identified as the source of significant human suffering in the course of normal life experiences (Hayes et al., 1999). Based on this framework we have argued for adoption of a contextual acceptancebased approach to chronic pain, an approach that aims to
* Corresponding author. Tel.: C44 1225 473 403; fax: 44 1225 473 461. E-mail address: [email protected] (L.M. McCracken).
reduce inflexible behavior patterns that fail the pain sufferer (McCracken, 2005; McCracken et al., 2004a). Acceptance of chronic pain includes an active willingness to have pain present, along with associated thoughts and feelings, when to do so results in overall greater involvement in valued activities and reaching of personal goals (McCracken et al., 2004a,b). Acceptance includes responding to pain-related experiences without attempts at control or avoidance, particularly when these attempts have limited the patient’s quality of life, and engaging in activities regardless of these experiences. There are at least seven cross-sectional studies of acceptance and patient functioning in patients with chronic pain (Evers et al., 2001; McCracken, 1998; McCracken and Eccleston, 2003; McCracken et al., 1999, 2004b, 2005a; Viane et al., 2003). The consistent message from these studies, using three separate measures of acceptance in four different countries, is that greater acceptance of chronic pain is associated with better emotional, physical, and social functioning, less health care and medication use, and better
0304-3959/$20.00 q 2005 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.pain.2005.08.015
L.M. McCracken, C. Eccleston / Pain 118 (2005) 164–169
work status. A primary limitation of these studies is their reliance on measures of acceptance and patient functioning at the same point in time. The general conclusion that behavior consistent with acceptance of pain results in better functioning remains an inference, without the benefit of observation over time. The purpose of the present study was to examine the relations between acceptance of chronic pain and patient functioning prospectively. We had access to data from a sample of persons seeking treatment for chronic pain at two points in time. The first was at an initial assessment (Time 1) and the second was at the start of treatment, an average of 3.9 months later (Time 2). Our primary aim was to analyse the relations of acceptance at Time 1 with patient functioning at Time 2. Our measures of patient functioning included emotional, physical, and social functioning, medication use, and work status. We predicted that acceptance of chronic pain at Time 1 would significantly correlate with patient functioning at Time 2. We also planned to examine subscale scores from the acceptance questionnaire for secondary, largely exploratory, purposes.
165
appointment. On the second occasion patients completed forms in the clinic on their first day of treatment. One quarter of patients assessed for treatment were either not deemed appropriate or refused treatment. Typically patients were deemed inappropriate for treatment if they were seeking additional invasive medical procedures or evaluations, or if their level of daily functioning was intact enough to not require an intensive treatment approach. Those who were assessed but did not attend treatment (and therefore, did not participate in this study) did not differ from the study sample on any background variables or any of the measures used in this study with the exception of work status. Patients who did not go on to treatment were more likely to be working at the time of initial assessment (out of work due to pain: attenders 81.9%, non-attenders 62.1%). The measures used for the basis of this study included a measure of acceptance of chronic pain, the Chronic Pain Acceptance Questionnaire (CPAQ), and a range of measures of patient functioning, taken both at initial assessment (Time 1) and at the start of treatment (Time 2). The measures of patient functioning included a 0–10 rating of usual pain in the past week, ratings of average daily rest and uptime (average time spent standing or walking per day in the last week), a count of analgesic medications, a categorical measure of work status that was dichotomized into work status reduced by pain or work status not reduced by pain, the Beck Depression Inventory, the Pain Anxiety Symptoms Scale, and the Sickness Impact Profile.
2. Method Participants were 118 consecutive adults attending a pain management unit, in the southwest of the UK, for interdisciplinary treatment of chronic pain. Treatment included physical rehabilitation and psychological therapy methods delivered by a team of physiotherapists, occupational therapists, nurses, physicians, and clinical psychologists (see McCracken, 2005 for a description of this treatment). Sixty four percent were women and the mean age at time of first contact was 44.2 years (SDZ10.7). Most were married or co-habitating (72.1%; 27.9% single, divorced, or widowed). They had completed a mean of 12.2 years of education (SDZ2.5). The median chronicity of pain was 87.5 months (range 12–528 months). Most patients had a primary complaint of low back pain (49.6%); other locations included lower limbs (13.8%), upper limbs (12.2%), neck (11.4%), or other locations (13.0%). Most patients presented with generalized pain and had no definitive, consensually agreed diagnoses other than ‘chronic pain syndrome’. Only 9.3% of participants were working either part time or full time, away from home, at the time of the initial assessment (53.4% were not working due to pain; 23.7% had retired early due to pain; 8.5% were working as homemakers; 4.2% were not working for other reasons; 0.9% other). Data for this study were collected at two points in time during the process of standard clinical services, once at an initial assessment to judge whether patients were suitable for interdisciplinary treatment. The second occasion was on the first day of treatment an average of 3.9 months following the initial assessment (SDZ2.6; range 0.5–15.0 months). Ninety-five percent of patients waited nine or fewer months between Time 1 and Time 2 and 84% waited between 2 and 6 months. Patients were consecutive referrals and completed the measures used for this study as part of their routine clinical assessment. On the first assessment occasion patients completed the measures at home and the assessing clinician collected them on arrival for their initial
3. Measures The CPAQ (McCracken et al., 2004b) is a 20-item measure of acceptance of chronic pain. It was derived from a measure created by Geiser (1992). The CPAQ has two subscales derived from factor analysis. These are labeled Activity Engagement (11 items; e.g. ‘Despite the pain, I am now sticking to a certain course in my life’) and Pain Willingness (nine reverse-keyed items; e.g. ‘I would gladly sacrifice important things in my life to control this pain better’). They assess patients’ participation in activities regardless of pain and relative absence of attempts to control or avoid pain, respectively. The subscale and total scores from the CPAQ are internally consistent (aZ.78–.82; McCracken et al., 2004b) and demonstrate significant correlations with measures of avoidance, emotional distress, and patient functioning in cross-sectional analyses, supporting their validity as indices of acceptance of chronic pain (McCracken et al., 2004b). The other measures used for this study are commonly used measures in studies of patients with chronic pain. The Beck Depression Inventory (BDI; Beck et al., 1961) is a well-established, 21-item measure of symptoms of depression. The Pain Anxiety Symptoms Scale (PASS; McCracken et al., 1992) is a 40-item measure of anxious responses linked with chronic pain including avoidance, cognitive, fearful, and physiological responses. The Sickness Impact Profile (SIP; Bergner et al., 1981) is a 136-item measure of disability from a health problem. The SIP yields composite scores for physical and psychosocial disability.
166
L.M. McCracken, C. Eccleston / Pain 118 (2005) 164–169
Each of the measures used in this study is frequently used, and has repeatedly shown acceptable psychometric properties, in previous studies with chronic pain sufferers (e.g. McCracken, 1998; McCracken and Eccleston, 2003; McCracken et al., 2004b, 2005b).
4. Results Initial correlation analyses showed that the CPAQ scores were moderately intercorrelated between Time 1 and Time 2 at rZ.76, P!.001, rZ.59, P!.001, rZ.75, P!.001, for the Activity Engagement, Pain Willingness, and Total scores, respectively. There were statistically significant increases from Time 1 to Time 2 in Activity Engagement; MZ29.2 (SDZ10.7) to MZ30.7 (SDZ10.5), t (117)Z2.2, P!.05; and the Total score; MZ46.5 (SDZ15.8) to MZ 49.0 (SDZ15.8), t (117)Z2.4, P!.05; but no change in the Pain Willingness score; MZ17.1 (SDZ9.2) to MZ18.4 (SDZ8.9), t (117)Z1.7, ns. Although statistically significant, the changes in the CPAQ scores were small. Each of the measures of functioning was significantly correlated across time. The smallest correlation was attained by the measure of estimated hours sleeping or resting during the day related to pain, rZ.32, P!.001, however, the average correlation of these measures across time was rZ.64. Only the measure of daily uptime showed a statistically significant shift over time, from 3.2 to 3.8 h, t (109)Z2.9, P!.01; none of the other measures showed a change at P!.05. Thus, overall there was no substantial shift in measures of acceptance or patient functioning from Time 1 to Time 2. Finally, correlation analyses were conducted to examine relations between the length of time that passed between the two assessment occasions for each participant and their measures of acceptance at Time 1 and functioning at Time 2. None of these reached significance at PO.05. Table 1 shows correlations between the CPAQ scores at Time 1 and nine measures of patient pain and functioning at Time 2. Although, there were trends toward negative correlations between Activity Engagement and the CPAQ total with the rating of pain, these did not reach significance.
There were relatively strong negative correlations between the acceptance scores and the scores for depression and pain-related anxiety. The acceptance scores also were consistently negatively correlated with scores for physical and psychosocial disability and daily rest due to pain. Uptime and work status were positively correlated with Activity Engagement but uncorrelated with Pain Willingness. Number of analgesic medications being taken was negatively correlated with Pain Willingness but uncorrelated with Activity Engagement. The consistent picture emerging was that patients who reported greater acceptance at Time 1 reported better emotional, social, and physical functioning, less medication consumption, and better work status at Time 2. Next, we conducted a series of multiple regression analyses including depression, pain-related anxiety, physical and psychosocial disability, uptime, rest, medications, and work status from Time 2 as the criterion variables and the acceptance scores from Time 1 as the primary predictor variables. The pain score from Time 2 was entered into each equation first to examine its influence on functioning at that point. Age, gender, education, and duration of pain were also tested for entry in each equation to control their influences on the criterion variables. Table 2 includes results from the eight regression equations. In general, demographic variables played very little role in the prediction of functioning. Pain also did not play a large role in prediction of functioning. At entry it contributed significant but small increments of explained variance in depression, psychosocial disability, and medication use. In the final equations it retained a significant regression coefficient only in the case of medication use. Duration of pain demonstrated a small negative relationship with depression, and age demonstrated a small positive relationship with physical disability. The combined acceptance scores accounted for significant increments in explained variance at entry in each of the eight equations. The variance increments ranged from 6.3% for number of medications to 29.0% for pain-related anxiety. In general the acceptance scores accounted for more variance in the four equations where the measures of functioning were standardized questionnaires (Mean change
Table 1 Correlations between acceptance scores at Time 1 and measures of patient functioning at Time 2
Activity engagement Pain willingness CPAQ total
Pain
Depression
Pain-related anxiety
Physical disability
Psychosocial disability
Uptime
Daily rest due to pain
Number of pain medications
Work status
K.18
K.37***
K.41***
K.23**
K.32***
.30**
K.22*
K.14
.38***
K.073 K.16
K.42*** K.50***
K.47*** K.48***
K.34*** K.36***
K.33*** K.41***
.052 .23*
K.22* K.27**
K.27** K.25**
.12 .32***
*P!.05, **P!.01, ***P!.001. Pain was measured with a 0–10 rating scale for usual pain in the last week; depression was measured with the Beck Depression Inventory; pain-related anxiety was measured with the Pain Anxiety Symptoms Scale; physical and psychosocial disability was measured with the Sickness Impact Profile; uptime and rest were assessed by asking for daily averages to time standing and walking and time resting related to pain in the last week; number of medications was a sum or the separate classes of analgesics taken by the patient; and work status was a dichotomous variable (0Znot working due to pain; 1Zworking or not working for some other reason).
L.M. McCracken, C. Eccleston / Pain 118 (2005) 164–169 Table 2 Results of regression analyses predicting patient functioning at Time 2 from acceptance scores (Activity engagement and pain willingness) at Time 1 Predictors
DR2
Depression 1. Pain .050* 2. Duration of pain .040 3. Activity engagement .20*** Pain willingness Pain-related anxiety 1. Pain .017 2. Activity engagement .29*** Pain willingness Physical disability 1. Pain .016 2. Age .044* 3. Activity engagement .13*** Pain willingness Psychosocial disability 1. Pain .043* 2. Activity engagement .15*** Pain willingness Uptime 1. Pain .00 2. Activity engagement .087** Pain willingness Daily rest due to pain 1. Pain .00 2. Activity engagement .074* Pain willingness Number of pain medications 1. Pain .047* 2. Activity engagement .063* Pain willingness Work status 1. Pain .003 2. Activity engagement .14*** Pain willingness
Beta
167
had a significant regression coefficient in the final equation. As shown in the correlation results, the acceptance scores consistently predicted functioning in a positive fashion.
Total R2
5. Discussion
.15 K.13 K.24** K.33***
.29
.068 K.30*** K.38***
.31
.068 .22 K.19* K.27**
.19
.15 K.24** K.25**
.19
.009 .30*** K.020
.088
K.035 K.17 K.17
.074
.18* K.053 K.23*
.11
.002 .38*** K.001
.15
*P!.05, **P!.01, ***P!.001. Pain was forced into each equation in step one. Age, gender, years of education, and duration of pain were tested for entry and retained in each equation if they were significant predictors of the criterion variable. The two components of acceptance, Activity engagement and pain willingness, were entered as a block in the final step of each equation.
R2Z.19) than in the four equations where they were single ratings or dichotomous variables (Mean change R2Z.091). If a Bonferroni correction were applied to the eight significance tests for the variance increments attributable to acceptance, a P-value less than .006 would be required for significance. At this level five out of eight results remain significant. The individual acceptance scores achieved significant regression coefficients in the final equations in 11–16 instances. Similar to the correlation analyses Activity Engagement alone was a significant and unique predictor of uptime and work status while Pain Willingness alone was a significant unique predictor of medication use. The total variance accounted for in the equation for daily rest was the smallest of the group at 7.4% and neither acceptance score
This study provided much needed data on the relations of acceptance of chronic pain with patient functioning over time. The aspects of patient functioning that were examined included emotional, physical, and social functioning, medication use, and work status. Pain level at Time 2 played a relatively small role in these aspects of functioning at Time 2. The role of acceptance, on the other hand, was independent of patients’ level of pain and much larger than that of pain. The current results are consistent with results demonstrated in cross-sectional studies (McCracken, 1998; McCracken and Eccleston, 2003; McCracken et al., 1999; 2004b; Viane et al., 2003), and extend these earlier results in several respects. First, four out of five of the previous, crosssectional studies of the CPAQ were conducted in a pain management center in the US. The present results come from a center in the UK, and, thus, show that the relations observed are not limited to the particular circumstances of the early studies. Second, the present results demonstrate relations between acceptance and patient functioning even when these variables are measured at two separate points in time, an average of nearly 4 months apart. This method helps to minimize a number of influences that could inflate the apparent relationship between acceptance and functioning. These potential influences include current pain or mood states or social influences that might lead patients to respond consistently across separate measures regardless of whether that consistency reflects reality. The current results strengthen the case for a directional relationship whereby acceptance leads to better functioning. There is at least some specificity in the relations of the separate acceptance scores with aspects of patients functioning. Activity Engagement showed unique relations with uptime and work status. Pain Willingness showed unique relations with number of pain medications used. These results are somewhat predictable. When patients consistently choose to engage in activities regardless of pain the expected result would be more overall daily activity and more engagement in specific realms of functioning such as work. Pain Willingness on the other hand particularly includes a relative absence of attempts at avoidance and control of pain. There are naturally a number of potential influences on the numbers of medication prescriptions patients will acquire for pain. Some of these will result from the types of specialists they have seen, the diagnoses given, and the prescribing practices of their GP. It is also expected, however, that the more patients are unwilling to have pain the more they will seek multiple consultations, request medications, and retain their prescriptions.
168
L.M. McCracken, C. Eccleston / Pain 118 (2005) 164–169
Our results are strengthened by inclusion of multiple methods for assessment of patient functioning: numerical ratings, free numerical estimates, standardized questionnaire scores, a researcher tallied score for number of analgesic medications, and a report of work status. This suggests some generality of the results and further suggests that they are not merely due to shared variance between similar methods of assessing patient behavior. Also, variance increments accounted for by the acceptance scores were relatively large given the time interval involved. This provides further support for the potential clinical utility of the acceptance measure. There are limitations of our methods. First, we studied a highly selected sample of patients, patients seeking specialty interdisciplinary treatment in the UK. Our results cannot be assumed to apply to all chronic pain sufferers. Second, our results cannot demonstrate causality. Although, they show that acceptance is related to functioning at a later point in time, they cannot rule out that better functioning leads to acceptance, or that a third factor leads to both. The stable levels of acceptance and functioning over time that subjects show does not provide an opportunity to demonstrate relations between changes in acceptance and changes in functioning. Research designs including interventions (e.g. McCracken et al., 2002) will be required to examine causal relations more directly. Third, we employed one short self-report measure of acceptance of chronic pain. Further, work is needed to develop methods to assess the psychological processes that fall under the heading of acceptance. Some of the processes of experience and behavior we are asking patients to report may happen without full awareness and may go unobserved without the type of training that is done in acceptance-based treatments. Finally, all of the patients in our sample have their unique histories and circumstances. The acceptance-related processes we examined are not critical issues for all patients to the same extent. There are likely other issues such as social factors, practical problems of life, and other health problems, that also contribute to varying degrees to patient functioning over time. Data continue to accumulate supporting the role of acceptance in patient functioning in the context of chronic pain (see McCracken, 2005; McCracken et al., 2004a for a review). These data show that willingness to have pain, and engage in activity regardless of pain, may lead to healthy life functioning for patients with chronic pain, in many circumstances, in contrast to struggling with pain and inflexibly applying attempts at control or avoidance of pain, which seem not to lead that way. Although, a range of other experiences such as unwanted moods and emotions, thoughts and memories, and other sensations, such as fatigue or stiffness, have not been the direct focus of our study, these too, most certainly, will provide occasions for the same processes of struggling, attempts at control, and avoidance, with the same effects in many situations. It seems worthwhile to broaden our investigations to examine
issues of ‘experiential avoidance’ in chronic pain more broadly, including avoidance of emotions and other cognitive content. These topics are likely to be crucially important for understanding the disability and suffering of individual chronic pain patients, and clinically important for planning appropriate treatment.
6. Clinical implications As data accrue the case becomes clearer for the translation of acceptance-based treatment methods into treatment programs for chronic pain sufferers. There are already encouraging data for approaches that incorporate acceptance for a range of behavior problems such as depression (Teasdale et al., 2000; Zettle and Raines, 1989), borderline personality and self-harm (Linehan, 1993), cancer-related distress (Speca et al., 2000) work stress (Bond and Bunce, 2000), and schizophrenia (Bach and Hayes, 2002). Approaches such as mindfulness-based treatments (e.g. Kabat-Zinn, 1990) and Acceptance and Commitment Therapy (ACT, Hayes et al., 1999) show considerable promise for the further development of chronic pain treatment. Preliminary results from a small randomized trial of ACT for patients at risk for painrelated disability (Dahl et al., 2004) and from a nonrandomized within- subjects comparison, including a sample of patients with complex longstanding pain problems (McCracken et al., 2005b), both demonstrated generally beneficial results from contextual acceptancebased approaches. ‘Acceptance’ may be a limited word for all that it is supposed to imply. As it is currently conceptualized and measured it includes patients (a) seeing thoughts about pain as just thoughts about pain, that may be unnecessarily adding to their suffering, (b) being present with potentially disturbing thoughts and feelings without defense or struggling, and (c) choosing actions that move them toward things they most value in life, in the presence of these same thoughts and feelings. These processes appear to play a significant role in chronic pain and may be appropriate targets for continuing therapy development.
References Bach P, Hayes SC. The use of acceptance and commitment therapy to prevent the rehospitalization of psychotic patients: a randomized controlled trial. J Clin Consult Psychol 2002;70:129–1139. Bond FW, Bunce D. Mediators of change in emotion-focused and problemfocused worksite stress management intervention. J Occup Health Psychol 2000;5:156–63. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry 1961;4:561–71. Bergner M, Bobbitt RA, Carter WB, Gilson BS. The sickness impact profile: development and final revision of a health status measure. Med Care 1981;29:787–805.
L.M. McCracken, C. Eccleston / Pain 118 (2005) 164–169 Dahl J, Wilson KG, Nilsson A. Acceptance and commitment therapy and the treatment of persons at risk for long-term disability resulting from stress and pain symptoms: a preliminary randomized trial. Behav Ther 2004;35:785–802. Evers AWM, Kraaimaat FW, van Lankveld W, Jongen PJH, Jacobs JWG, Bijlsma JWJ. Beyond unfavorable thinking: the illness cognition questionnaire for chronic diseases. J Consult Clin Psychol 2001;69: 1026–36. Geiser DS. A comparison of acceptance-focused and control-focused psychological treatments in a chronic pain treatment center. Unpublished doctoral dissertation. University of Nevada, Reno; 1992. Hayes SC, Strosahl KD, Wilson KG. Acceptance and commitment therapy: an experiential approach to behavior change. New York: The Guilford Press; 1999. Kabat-Zinn J. Full catastrophe living: using the wisdom of your body and mind to face stress, pain, and illness. New York: Dell Publishing; 1990. Linehan MM. Cognitive-behavioral treatment of borderline personality disorder. New York: Guilford; 1993. McCracken LM. Learning to live with the pain: acceptance of pain predicts adjustment in persons with chronic pain. Pain 1998;74:21–7. McCracken LM. Contextual cognitive-behavioral therapy for chronic pain. Seattle, WA: IASP press; 2005. McCracken LM, Eccleston C. Coping or acceptance: what to do about chronic pain? Pain 2003;105:197–204. McCracken LM, Zayfert C, Gross RT. The pain anxiety symptom scale: development and validation of a scale to measure fear of pain. Pain 1992;50:67–73. McCracken LM, Spertus IL, Janeck AS, Sinclair D, Wetzel FT. Behavioral dimensions of adjustment in persons with chronic pain: pain-related anxiety and acceptance. Pain 1999;80:283–9.
169
McCracken LM, Gross RT, Eccleston C. Multimethod assessment of treatment process in chronic low back pain: comparison of reported pain-related anxiety with directly measured physical capacity. Behav Res Ther 2002;40:585–94. McCracken LM, Carson JW, Eccleston C, Keefe FJ. Topical review: acceptance and change in the context of chronic pain. Pain 2004a;109: 4–7. McCracken LM, Vowles KE, Eccleston C. Acceptance of chronic pain: component analysis and a revised assessment method. Pain 2004b;107: 159–66. McCracken LM, Eccleston C, Bell L. Clinical assessment of behavioral coping responses: results from a brief inventory. Eur J Pain 2005a;9: 69–78. McCracken LM, Vowles KE, Eccleston C. Acceptance-based treatment for persons with complex, long standing chronic pain: a preliminary analysis of treatment outcome in comparison to a waiting phase. Behav Res Ther 2005b;43:1335–46. Speca M, Carlson LE, Goodey E, Angen M. A randomized, wait-list controlled clinical trial: the effect of a mindfulness meditation-based stress reduction program on mood and symptoms of stress in cancer outpatients. Psychosom Med 2000;62:613–22. Teasdale JD, Segal ZV, Williams JMG, Ridgeway VA, Soulsby JM, Lau MA. Prevention of relapse/recurrence in major depression by mindfulness-based cognitive therapy. J Consult Clin Psychol 2000;68: 615–23. Viane I, Crombez G, Eccleston C, Poppe C, Devulder J, Van Houdenhove B, De Corte W. Acceptance of pain is an independent predictor of mental well-being in patients with chronic pain: empirical evidence and reappraisal. Pain 2003;106:65–72. Zettle RD, Rains JC. Group cognitive and contextual therapies in treatment of depression. J Clin Psychol 1989;45:436–45.
A prospective study of acceptance of pain and patient functioning with chronic pain Lance M. McCracken*, Christopher Eccleston Pain Management Unit, Royal National Hospital for Rheumatic Diseases, The University of Bath, Bath BA1 1RL, UK Received 15 September 2004; received in revised form 23 June 2005; accepted 8 August 2005
Abstract Acceptance of chronic pain is emerging as an important concept in understanding ways that chronic pain sufferers can remain engaged with valued aspects of life. Recent studies have relied heavily on cross-sectional investigations at a single time point. The present study sought to prospectively investigate relations between acceptance of chronic pain and patient functioning. A sample of adults referred for interdisciplinary treatment of severe and disabling chronic pain was assessed twice, an average of 3.9 months apart. Results showed that pain and acceptance were largely unrelated. Pain at Time 2 was weakly related to measures of functioning at Time 2. On the other hand, acceptance at Time 1 was consistently related to patient functioning at Time 2. Those patients who reported greater acceptance at Time 1 reported better emotional, social, and physical functioning, less medication consumption, and better work status at Time 2. These data suggest that willingness to have pain, and to engage in activity regardless of pain, can lead to healthy functioning for patients with chronic pain. Treatment outcome and process studies may demonstrate the potential for acceptance-based clinical methods for chronic pain management. q 2005 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. Keywords: Chronic pain; Behavioral approaches; Patient adjustment; Acceptance
1. Introduction Chronic pain provides the sufferer with many occasions in which their responses to pain can move them away from healthy life functioning. They may evaluate their pain in distressing ways and take these evaluations to be true. They also may regard pain and other feelings as firm reasons to disengage from important valued aspects of their life, and generally act to control or avoid painful experiences, despite the lack of positive effect these behaviors produce. This type of behavior pattern has been identified as the source of significant human suffering in the course of normal life experiences (Hayes et al., 1999). Based on this framework we have argued for adoption of a contextual acceptancebased approach to chronic pain, an approach that aims to
* Corresponding author. Tel.: C44 1225 473 403; fax: 44 1225 473 461. E-mail address: [email protected] (L.M. McCracken).
reduce inflexible behavior patterns that fail the pain sufferer (McCracken, 2005; McCracken et al., 2004a). Acceptance of chronic pain includes an active willingness to have pain present, along with associated thoughts and feelings, when to do so results in overall greater involvement in valued activities and reaching of personal goals (McCracken et al., 2004a,b). Acceptance includes responding to pain-related experiences without attempts at control or avoidance, particularly when these attempts have limited the patient’s quality of life, and engaging in activities regardless of these experiences. There are at least seven cross-sectional studies of acceptance and patient functioning in patients with chronic pain (Evers et al., 2001; McCracken, 1998; McCracken and Eccleston, 2003; McCracken et al., 1999, 2004b, 2005a; Viane et al., 2003). The consistent message from these studies, using three separate measures of acceptance in four different countries, is that greater acceptance of chronic pain is associated with better emotional, physical, and social functioning, less health care and medication use, and better
0304-3959/$20.00 q 2005 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.pain.2005.08.015
L.M. McCracken, C. Eccleston / Pain 118 (2005) 164–169
work status. A primary limitation of these studies is their reliance on measures of acceptance and patient functioning at the same point in time. The general conclusion that behavior consistent with acceptance of pain results in better functioning remains an inference, without the benefit of observation over time. The purpose of the present study was to examine the relations between acceptance of chronic pain and patient functioning prospectively. We had access to data from a sample of persons seeking treatment for chronic pain at two points in time. The first was at an initial assessment (Time 1) and the second was at the start of treatment, an average of 3.9 months later (Time 2). Our primary aim was to analyse the relations of acceptance at Time 1 with patient functioning at Time 2. Our measures of patient functioning included emotional, physical, and social functioning, medication use, and work status. We predicted that acceptance of chronic pain at Time 1 would significantly correlate with patient functioning at Time 2. We also planned to examine subscale scores from the acceptance questionnaire for secondary, largely exploratory, purposes.
165
appointment. On the second occasion patients completed forms in the clinic on their first day of treatment. One quarter of patients assessed for treatment were either not deemed appropriate or refused treatment. Typically patients were deemed inappropriate for treatment if they were seeking additional invasive medical procedures or evaluations, or if their level of daily functioning was intact enough to not require an intensive treatment approach. Those who were assessed but did not attend treatment (and therefore, did not participate in this study) did not differ from the study sample on any background variables or any of the measures used in this study with the exception of work status. Patients who did not go on to treatment were more likely to be working at the time of initial assessment (out of work due to pain: attenders 81.9%, non-attenders 62.1%). The measures used for the basis of this study included a measure of acceptance of chronic pain, the Chronic Pain Acceptance Questionnaire (CPAQ), and a range of measures of patient functioning, taken both at initial assessment (Time 1) and at the start of treatment (Time 2). The measures of patient functioning included a 0–10 rating of usual pain in the past week, ratings of average daily rest and uptime (average time spent standing or walking per day in the last week), a count of analgesic medications, a categorical measure of work status that was dichotomized into work status reduced by pain or work status not reduced by pain, the Beck Depression Inventory, the Pain Anxiety Symptoms Scale, and the Sickness Impact Profile.
2. Method Participants were 118 consecutive adults attending a pain management unit, in the southwest of the UK, for interdisciplinary treatment of chronic pain. Treatment included physical rehabilitation and psychological therapy methods delivered by a team of physiotherapists, occupational therapists, nurses, physicians, and clinical psychologists (see McCracken, 2005 for a description of this treatment). Sixty four percent were women and the mean age at time of first contact was 44.2 years (SDZ10.7). Most were married or co-habitating (72.1%; 27.9% single, divorced, or widowed). They had completed a mean of 12.2 years of education (SDZ2.5). The median chronicity of pain was 87.5 months (range 12–528 months). Most patients had a primary complaint of low back pain (49.6%); other locations included lower limbs (13.8%), upper limbs (12.2%), neck (11.4%), or other locations (13.0%). Most patients presented with generalized pain and had no definitive, consensually agreed diagnoses other than ‘chronic pain syndrome’. Only 9.3% of participants were working either part time or full time, away from home, at the time of the initial assessment (53.4% were not working due to pain; 23.7% had retired early due to pain; 8.5% were working as homemakers; 4.2% were not working for other reasons; 0.9% other). Data for this study were collected at two points in time during the process of standard clinical services, once at an initial assessment to judge whether patients were suitable for interdisciplinary treatment. The second occasion was on the first day of treatment an average of 3.9 months following the initial assessment (SDZ2.6; range 0.5–15.0 months). Ninety-five percent of patients waited nine or fewer months between Time 1 and Time 2 and 84% waited between 2 and 6 months. Patients were consecutive referrals and completed the measures used for this study as part of their routine clinical assessment. On the first assessment occasion patients completed the measures at home and the assessing clinician collected them on arrival for their initial
3. Measures The CPAQ (McCracken et al., 2004b) is a 20-item measure of acceptance of chronic pain. It was derived from a measure created by Geiser (1992). The CPAQ has two subscales derived from factor analysis. These are labeled Activity Engagement (11 items; e.g. ‘Despite the pain, I am now sticking to a certain course in my life’) and Pain Willingness (nine reverse-keyed items; e.g. ‘I would gladly sacrifice important things in my life to control this pain better’). They assess patients’ participation in activities regardless of pain and relative absence of attempts to control or avoid pain, respectively. The subscale and total scores from the CPAQ are internally consistent (aZ.78–.82; McCracken et al., 2004b) and demonstrate significant correlations with measures of avoidance, emotional distress, and patient functioning in cross-sectional analyses, supporting their validity as indices of acceptance of chronic pain (McCracken et al., 2004b). The other measures used for this study are commonly used measures in studies of patients with chronic pain. The Beck Depression Inventory (BDI; Beck et al., 1961) is a well-established, 21-item measure of symptoms of depression. The Pain Anxiety Symptoms Scale (PASS; McCracken et al., 1992) is a 40-item measure of anxious responses linked with chronic pain including avoidance, cognitive, fearful, and physiological responses. The Sickness Impact Profile (SIP; Bergner et al., 1981) is a 136-item measure of disability from a health problem. The SIP yields composite scores for physical and psychosocial disability.
166
L.M. McCracken, C. Eccleston / Pain 118 (2005) 164–169
Each of the measures used in this study is frequently used, and has repeatedly shown acceptable psychometric properties, in previous studies with chronic pain sufferers (e.g. McCracken, 1998; McCracken and Eccleston, 2003; McCracken et al., 2004b, 2005b).
4. Results Initial correlation analyses showed that the CPAQ scores were moderately intercorrelated between Time 1 and Time 2 at rZ.76, P!.001, rZ.59, P!.001, rZ.75, P!.001, for the Activity Engagement, Pain Willingness, and Total scores, respectively. There were statistically significant increases from Time 1 to Time 2 in Activity Engagement; MZ29.2 (SDZ10.7) to MZ30.7 (SDZ10.5), t (117)Z2.2, P!.05; and the Total score; MZ46.5 (SDZ15.8) to MZ 49.0 (SDZ15.8), t (117)Z2.4, P!.05; but no change in the Pain Willingness score; MZ17.1 (SDZ9.2) to MZ18.4 (SDZ8.9), t (117)Z1.7, ns. Although statistically significant, the changes in the CPAQ scores were small. Each of the measures of functioning was significantly correlated across time. The smallest correlation was attained by the measure of estimated hours sleeping or resting during the day related to pain, rZ.32, P!.001, however, the average correlation of these measures across time was rZ.64. Only the measure of daily uptime showed a statistically significant shift over time, from 3.2 to 3.8 h, t (109)Z2.9, P!.01; none of the other measures showed a change at P!.05. Thus, overall there was no substantial shift in measures of acceptance or patient functioning from Time 1 to Time 2. Finally, correlation analyses were conducted to examine relations between the length of time that passed between the two assessment occasions for each participant and their measures of acceptance at Time 1 and functioning at Time 2. None of these reached significance at PO.05. Table 1 shows correlations between the CPAQ scores at Time 1 and nine measures of patient pain and functioning at Time 2. Although, there were trends toward negative correlations between Activity Engagement and the CPAQ total with the rating of pain, these did not reach significance.
There were relatively strong negative correlations between the acceptance scores and the scores for depression and pain-related anxiety. The acceptance scores also were consistently negatively correlated with scores for physical and psychosocial disability and daily rest due to pain. Uptime and work status were positively correlated with Activity Engagement but uncorrelated with Pain Willingness. Number of analgesic medications being taken was negatively correlated with Pain Willingness but uncorrelated with Activity Engagement. The consistent picture emerging was that patients who reported greater acceptance at Time 1 reported better emotional, social, and physical functioning, less medication consumption, and better work status at Time 2. Next, we conducted a series of multiple regression analyses including depression, pain-related anxiety, physical and psychosocial disability, uptime, rest, medications, and work status from Time 2 as the criterion variables and the acceptance scores from Time 1 as the primary predictor variables. The pain score from Time 2 was entered into each equation first to examine its influence on functioning at that point. Age, gender, education, and duration of pain were also tested for entry in each equation to control their influences on the criterion variables. Table 2 includes results from the eight regression equations. In general, demographic variables played very little role in the prediction of functioning. Pain also did not play a large role in prediction of functioning. At entry it contributed significant but small increments of explained variance in depression, psychosocial disability, and medication use. In the final equations it retained a significant regression coefficient only in the case of medication use. Duration of pain demonstrated a small negative relationship with depression, and age demonstrated a small positive relationship with physical disability. The combined acceptance scores accounted for significant increments in explained variance at entry in each of the eight equations. The variance increments ranged from 6.3% for number of medications to 29.0% for pain-related anxiety. In general the acceptance scores accounted for more variance in the four equations where the measures of functioning were standardized questionnaires (Mean change
Table 1 Correlations between acceptance scores at Time 1 and measures of patient functioning at Time 2
Activity engagement Pain willingness CPAQ total
Pain
Depression
Pain-related anxiety
Physical disability
Psychosocial disability
Uptime
Daily rest due to pain
Number of pain medications
Work status
K.18
K.37***
K.41***
K.23**
K.32***
.30**
K.22*
K.14
.38***
K.073 K.16
K.42*** K.50***
K.47*** K.48***
K.34*** K.36***
K.33*** K.41***
.052 .23*
K.22* K.27**
K.27** K.25**
.12 .32***
*P!.05, **P!.01, ***P!.001. Pain was measured with a 0–10 rating scale for usual pain in the last week; depression was measured with the Beck Depression Inventory; pain-related anxiety was measured with the Pain Anxiety Symptoms Scale; physical and psychosocial disability was measured with the Sickness Impact Profile; uptime and rest were assessed by asking for daily averages to time standing and walking and time resting related to pain in the last week; number of medications was a sum or the separate classes of analgesics taken by the patient; and work status was a dichotomous variable (0Znot working due to pain; 1Zworking or not working for some other reason).
L.M. McCracken, C. Eccleston / Pain 118 (2005) 164–169 Table 2 Results of regression analyses predicting patient functioning at Time 2 from acceptance scores (Activity engagement and pain willingness) at Time 1 Predictors
DR2
Depression 1. Pain .050* 2. Duration of pain .040 3. Activity engagement .20*** Pain willingness Pain-related anxiety 1. Pain .017 2. Activity engagement .29*** Pain willingness Physical disability 1. Pain .016 2. Age .044* 3. Activity engagement .13*** Pain willingness Psychosocial disability 1. Pain .043* 2. Activity engagement .15*** Pain willingness Uptime 1. Pain .00 2. Activity engagement .087** Pain willingness Daily rest due to pain 1. Pain .00 2. Activity engagement .074* Pain willingness Number of pain medications 1. Pain .047* 2. Activity engagement .063* Pain willingness Work status 1. Pain .003 2. Activity engagement .14*** Pain willingness
Beta
167
had a significant regression coefficient in the final equation. As shown in the correlation results, the acceptance scores consistently predicted functioning in a positive fashion.
Total R2
5. Discussion
.15 K.13 K.24** K.33***
.29
.068 K.30*** K.38***
.31
.068 .22 K.19* K.27**
.19
.15 K.24** K.25**
.19
.009 .30*** K.020
.088
K.035 K.17 K.17
.074
.18* K.053 K.23*
.11
.002 .38*** K.001
.15
*P!.05, **P!.01, ***P!.001. Pain was forced into each equation in step one. Age, gender, years of education, and duration of pain were tested for entry and retained in each equation if they were significant predictors of the criterion variable. The two components of acceptance, Activity engagement and pain willingness, were entered as a block in the final step of each equation.
R2Z.19) than in the four equations where they were single ratings or dichotomous variables (Mean change R2Z.091). If a Bonferroni correction were applied to the eight significance tests for the variance increments attributable to acceptance, a P-value less than .006 would be required for significance. At this level five out of eight results remain significant. The individual acceptance scores achieved significant regression coefficients in the final equations in 11–16 instances. Similar to the correlation analyses Activity Engagement alone was a significant and unique predictor of uptime and work status while Pain Willingness alone was a significant unique predictor of medication use. The total variance accounted for in the equation for daily rest was the smallest of the group at 7.4% and neither acceptance score
This study provided much needed data on the relations of acceptance of chronic pain with patient functioning over time. The aspects of patient functioning that were examined included emotional, physical, and social functioning, medication use, and work status. Pain level at Time 2 played a relatively small role in these aspects of functioning at Time 2. The role of acceptance, on the other hand, was independent of patients’ level of pain and much larger than that of pain. The current results are consistent with results demonstrated in cross-sectional studies (McCracken, 1998; McCracken and Eccleston, 2003; McCracken et al., 1999; 2004b; Viane et al., 2003), and extend these earlier results in several respects. First, four out of five of the previous, crosssectional studies of the CPAQ were conducted in a pain management center in the US. The present results come from a center in the UK, and, thus, show that the relations observed are not limited to the particular circumstances of the early studies. Second, the present results demonstrate relations between acceptance and patient functioning even when these variables are measured at two separate points in time, an average of nearly 4 months apart. This method helps to minimize a number of influences that could inflate the apparent relationship between acceptance and functioning. These potential influences include current pain or mood states or social influences that might lead patients to respond consistently across separate measures regardless of whether that consistency reflects reality. The current results strengthen the case for a directional relationship whereby acceptance leads to better functioning. There is at least some specificity in the relations of the separate acceptance scores with aspects of patients functioning. Activity Engagement showed unique relations with uptime and work status. Pain Willingness showed unique relations with number of pain medications used. These results are somewhat predictable. When patients consistently choose to engage in activities regardless of pain the expected result would be more overall daily activity and more engagement in specific realms of functioning such as work. Pain Willingness on the other hand particularly includes a relative absence of attempts at avoidance and control of pain. There are naturally a number of potential influences on the numbers of medication prescriptions patients will acquire for pain. Some of these will result from the types of specialists they have seen, the diagnoses given, and the prescribing practices of their GP. It is also expected, however, that the more patients are unwilling to have pain the more they will seek multiple consultations, request medications, and retain their prescriptions.
168
L.M. McCracken, C. Eccleston / Pain 118 (2005) 164–169
Our results are strengthened by inclusion of multiple methods for assessment of patient functioning: numerical ratings, free numerical estimates, standardized questionnaire scores, a researcher tallied score for number of analgesic medications, and a report of work status. This suggests some generality of the results and further suggests that they are not merely due to shared variance between similar methods of assessing patient behavior. Also, variance increments accounted for by the acceptance scores were relatively large given the time interval involved. This provides further support for the potential clinical utility of the acceptance measure. There are limitations of our methods. First, we studied a highly selected sample of patients, patients seeking specialty interdisciplinary treatment in the UK. Our results cannot be assumed to apply to all chronic pain sufferers. Second, our results cannot demonstrate causality. Although, they show that acceptance is related to functioning at a later point in time, they cannot rule out that better functioning leads to acceptance, or that a third factor leads to both. The stable levels of acceptance and functioning over time that subjects show does not provide an opportunity to demonstrate relations between changes in acceptance and changes in functioning. Research designs including interventions (e.g. McCracken et al., 2002) will be required to examine causal relations more directly. Third, we employed one short self-report measure of acceptance of chronic pain. Further, work is needed to develop methods to assess the psychological processes that fall under the heading of acceptance. Some of the processes of experience and behavior we are asking patients to report may happen without full awareness and may go unobserved without the type of training that is done in acceptance-based treatments. Finally, all of the patients in our sample have their unique histories and circumstances. The acceptance-related processes we examined are not critical issues for all patients to the same extent. There are likely other issues such as social factors, practical problems of life, and other health problems, that also contribute to varying degrees to patient functioning over time. Data continue to accumulate supporting the role of acceptance in patient functioning in the context of chronic pain (see McCracken, 2005; McCracken et al., 2004a for a review). These data show that willingness to have pain, and engage in activity regardless of pain, may lead to healthy life functioning for patients with chronic pain, in many circumstances, in contrast to struggling with pain and inflexibly applying attempts at control or avoidance of pain, which seem not to lead that way. Although, a range of other experiences such as unwanted moods and emotions, thoughts and memories, and other sensations, such as fatigue or stiffness, have not been the direct focus of our study, these too, most certainly, will provide occasions for the same processes of struggling, attempts at control, and avoidance, with the same effects in many situations. It seems worthwhile to broaden our investigations to examine
issues of ‘experiential avoidance’ in chronic pain more broadly, including avoidance of emotions and other cognitive content. These topics are likely to be crucially important for understanding the disability and suffering of individual chronic pain patients, and clinically important for planning appropriate treatment.
6. Clinical implications As data accrue the case becomes clearer for the translation of acceptance-based treatment methods into treatment programs for chronic pain sufferers. There are already encouraging data for approaches that incorporate acceptance for a range of behavior problems such as depression (Teasdale et al., 2000; Zettle and Raines, 1989), borderline personality and self-harm (Linehan, 1993), cancer-related distress (Speca et al., 2000) work stress (Bond and Bunce, 2000), and schizophrenia (Bach and Hayes, 2002). Approaches such as mindfulness-based treatments (e.g. Kabat-Zinn, 1990) and Acceptance and Commitment Therapy (ACT, Hayes et al., 1999) show considerable promise for the further development of chronic pain treatment. Preliminary results from a small randomized trial of ACT for patients at risk for painrelated disability (Dahl et al., 2004) and from a nonrandomized within- subjects comparison, including a sample of patients with complex longstanding pain problems (McCracken et al., 2005b), both demonstrated generally beneficial results from contextual acceptancebased approaches. ‘Acceptance’ may be a limited word for all that it is supposed to imply. As it is currently conceptualized and measured it includes patients (a) seeing thoughts about pain as just thoughts about pain, that may be unnecessarily adding to their suffering, (b) being present with potentially disturbing thoughts and feelings without defense or struggling, and (c) choosing actions that move them toward things they most value in life, in the presence of these same thoughts and feelings. These processes appear to play a significant role in chronic pain and may be appropriate targets for continuing therapy development.
References Bach P, Hayes SC. The use of acceptance and commitment therapy to prevent the rehospitalization of psychotic patients: a randomized controlled trial. J Clin Consult Psychol 2002;70:129–1139. Bond FW, Bunce D. Mediators of change in emotion-focused and problemfocused worksite stress management intervention. J Occup Health Psychol 2000;5:156–63. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry 1961;4:561–71. Bergner M, Bobbitt RA, Carter WB, Gilson BS. The sickness impact profile: development and final revision of a health status measure. Med Care 1981;29:787–805.
L.M. McCracken, C. Eccleston / Pain 118 (2005) 164–169 Dahl J, Wilson KG, Nilsson A. Acceptance and commitment therapy and the treatment of persons at risk for long-term disability resulting from stress and pain symptoms: a preliminary randomized trial. Behav Ther 2004;35:785–802. Evers AWM, Kraaimaat FW, van Lankveld W, Jongen PJH, Jacobs JWG, Bijlsma JWJ. Beyond unfavorable thinking: the illness cognition questionnaire for chronic diseases. J Consult Clin Psychol 2001;69: 1026–36. Geiser DS. A comparison of acceptance-focused and control-focused psychological treatments in a chronic pain treatment center. Unpublished doctoral dissertation. University of Nevada, Reno; 1992. Hayes SC, Strosahl KD, Wilson KG. Acceptance and commitment therapy: an experiential approach to behavior change. New York: The Guilford Press; 1999. Kabat-Zinn J. Full catastrophe living: using the wisdom of your body and mind to face stress, pain, and illness. New York: Dell Publishing; 1990. Linehan MM. Cognitive-behavioral treatment of borderline personality disorder. New York: Guilford; 1993. McCracken LM. Learning to live with the pain: acceptance of pain predicts adjustment in persons with chronic pain. Pain 1998;74:21–7. McCracken LM. Contextual cognitive-behavioral therapy for chronic pain. Seattle, WA: IASP press; 2005. McCracken LM, Eccleston C. Coping or acceptance: what to do about chronic pain? Pain 2003;105:197–204. McCracken LM, Zayfert C, Gross RT. The pain anxiety symptom scale: development and validation of a scale to measure fear of pain. Pain 1992;50:67–73. McCracken LM, Spertus IL, Janeck AS, Sinclair D, Wetzel FT. Behavioral dimensions of adjustment in persons with chronic pain: pain-related anxiety and acceptance. Pain 1999;80:283–9.
169
McCracken LM, Gross RT, Eccleston C. Multimethod assessment of treatment process in chronic low back pain: comparison of reported pain-related anxiety with directly measured physical capacity. Behav Res Ther 2002;40:585–94. McCracken LM, Carson JW, Eccleston C, Keefe FJ. Topical review: acceptance and change in the context of chronic pain. Pain 2004a;109: 4–7. McCracken LM, Vowles KE, Eccleston C. Acceptance of chronic pain: component analysis and a revised assessment method. Pain 2004b;107: 159–66. McCracken LM, Eccleston C, Bell L. Clinical assessment of behavioral coping responses: results from a brief inventory. Eur J Pain 2005a;9: 69–78. McCracken LM, Vowles KE, Eccleston C. Acceptance-based treatment for persons with complex, long standing chronic pain: a preliminary analysis of treatment outcome in comparison to a waiting phase. Behav Res Ther 2005b;43:1335–46. Speca M, Carlson LE, Goodey E, Angen M. A randomized, wait-list controlled clinical trial: the effect of a mindfulness meditation-based stress reduction program on mood and symptoms of stress in cancer outpatients. Psychosom Med 2000;62:613–22. Teasdale JD, Segal ZV, Williams JMG, Ridgeway VA, Soulsby JM, Lau MA. Prevention of relapse/recurrence in major depression by mindfulness-based cognitive therapy. J Consult Clin Psychol 2000;68: 615–23. Viane I, Crombez G, Eccleston C, Poppe C, Devulder J, Van Houdenhove B, De Corte W. Acceptance of pain is an independent predictor of mental well-being in patients with chronic pain: empirical evidence and reappraisal. Pain 2003;106:65–72. Zettle RD, Rains JC. Group cognitive and contextual therapies in treatment of depression. J Clin Psychol 1989;45:436–45.