Increasing evidence suggests endurance (ER) behavior as an individual risk factor for pain maintenance and treatment failure in low back pain (LBP) patients.1–3 However, it remains unclear as to whether or not bodily overuse or underuse could be mediators responsible for pain chronicity.
The avoidance-endurance model (AEM) distinguishes among 3 different maladaptive pain response patterns and 1 adaptive coping style.1 Whereas patients with fear-avoidance (FAR) tend to catastrophize pain and are afraid of increasing pain by moving, patients with endurance (ER) behavior tend to ignore pain, suppress thoughts about pain, and persist in performing tasks regardless of the presence of pain or an increase in pain. The avoidance of specific movements and physical activity of FARs may lead to structural and functional adaptive changes related to pain chronification. In contrast, ER behavior is believed to lead to an overuse and overload of physical structures resulting in injuries to muscle fibers, connective tissues, or nerves. Whereas eustress-endurers (EER) are characterized by positive mood and distraction despite pain, distress-endurers (DE) tend to feel more negative, with tendencies toward pessimistic thoughts. In comparison, chronic pain patients using adaptive response (AR) patterns are able to cope with pain in a flexible, adaptive way, neither overusing nor underusing their bodies.
Findings from recent studies show that patients with either FAR or ER behavior report higher pain levels or more disability,1–3 lower working capacity,2 or an inferior quality of life3 6 months after primary care or surgeries compared with adaptive pain copers. Despite these observations, a lack of knowledge seems to exist about ER behavior and the relation to overuse and over-activity. Over-activity is defined as the excessive engagement of an individual with chronic pain in bodily activity that may result in significant increases in pain and periods of incapacity.4 Spending very long periods on sedentary activities has also been accepted as a form of under-activity by clinicians. A previous study also showed static strain positions to be related to back pain.5 As a consequence of overuse or underuse, differences in structural and functional adaptations would be expected between patients with FAR and ER behavior.
To date, back muscle function has been investigated only in fear-avoidant LBP patients, and the results of this research are partly conflicting. Whereas some studies found back muscle strength to be weaker6,7 and the electromyographic activity from trunk flexion-relaxation tests to be more impaired in FARs than in non-FARs,8 others did not observe any differences in either back muscle strength or trunk range of motion (ROM) measures.9 In addition, among the limited research that sought to compare fear-avoidant pain patients with endurers, 1 study found higher activity levels in self-reported and accelerometer-based assessments and more strained postures in ER compared with FAR.10 Another study observed differences between ER and AR in self-reported activity levels.11 Moreover, possible effects of exercise therapy, an important therapeutic intervention offered to most LBP patients,12 has been investigated only in subgroups comparing FAR with non-FAR, with results indicating that exercise therapy13,14 and movement exposures15–17 led to reduced pain and fear of movement in patients with FA behavior. AEM-subgroup-tailored interventions recommend confronting fear-avoidant patients with their fear of movement and to stay active, whereas endurers should avoid overexerting themselves and take breaks in time.18 Therefore, it is not obvious whether FAR and endurers would benefit to the same extent from one and the same standardized physical training intervention.
Thus, the aims of this study were to investigate (1) whether or not chronic LBP (cLBP) patients classified according to the AEM would differ in muscle function and activity measures, and (2) whether or not AEM subgroups would benefit differently from a 6-month rehabilitation training.
MATERIALS AND METHODS
Participants and Study Design
Between January 2012 and July 2015, a total of 216 patients who referred from various settings to the referral ambulatory rehabilitation center applied for participation in this cross-sectional, prospective observational study. Inclusion criteria were LBP lasting for >3 months19 at a minimum pain level of 30 on a visual analog scale (VAS, 0 to 100)20 and a minimum age of 18 years. Exclusion criteria were moderate pain levels (>30 on a VAS) in areas other than the lower back; peripheral neurological deficits; spinal fractures, infection, or cancer; recent surgeries involving the back region; previous experience with trunk muscle strength testing; and a body mass index exceeding 35 kg/m2. Before inclusion, all patients received oral and written information about the study and signed a consent form. The study protocol was acknowledged by the ethics committee of the city of Vienna.
After screening for inclusion and exclusion criteria, a total of 178 cLBP patients were eligible for inclusion in the study. The study design was cross-sectional and longitudinal. Before (T1) and at the end of a 6-month training intervention (T2), patients answered psychological and pain-related questionnaires and completed physical testing. Six months later, a brief follow-up examination (T3) was conducted by mail, which comprised of pain intensity ratings and an assessment of back-related quality of life.
Psychological and Back-related Measures
Demographic and Pain History Variables. Patients’ sex, age, educational level, marital status, and pain history were all assessed with a general demographic and medical history checklist.
Pain Intensity. Pain intensity was rated on VAS ranging from 0 (no pain) to 100 (highest pain).20 The VAS has been shown to be a valid and reliable measurement.21
Avoidance and Eendurance Pain Response Patterns. Avoidance-endurance behavior was measured with the Avoidance-Endurance Questionnaire (AEQ),22 which is comprised of 49 items and 9 subscales. Patients rated the items on a 7-point rating scale (0=never, 6=always). The subscales of FA behavior contain the Anxiety/Depression scale, the Catastrophizing scale, the Help/Hopelessness scale, the Avoidance of Social Activities scale, and the Avoidance of Physical Activities scale. Subscales measuring ER behavior include the Pain Persistence scale, the Humor/Distraction scale, the Thought Suppression Scale (TSS), and the Positive Mood scale. The Behavioral Endurance scale (BES) is a sum scale calculated by the Humor/Distraction scale and the Pain Persistence scale. All of the AEQ subscales have been proven to have high levels of validity and internal consistency.22
Disability. The Roland Morris Disability Questionnaire (RMDQ)23 checks 24 items that patients either agree (1) or disagree (0) with. The sum score ranges from 0 to 24, with higher scores indicating higher disability levels. The RMDQ has proven to be both valid and reliable.23,24
The Pain Disability Index (PDI)25 assesses experienced pain-related disability in 7 domains of life measured on an 11-point rating scale (0=no disability, 10=total disability). The sum score ranges between 0 and 70, with higher scores indicating higher disability. The PDI has been shown to be valid and reliable.25,26
Quality of Life and Working Capacity. The 36-Item Short-Form Health Survey (SF-36) is a valid and reliable measure that assesses back-related quality of life.27 It is comprised of 36 items, which can be summarized in 8 subscales and 2 sum scales ranging from 0 (maximum disability) to 100 (no disability). The Physical Health sum scale contains the subscales of Physical Functioning, Physical Role Functioning, Bodily Pain, and General Health. The Mental Health sum scale includes the subscales measuring Vitality, Social Functioning, Emotional Role Functioning, and the Mental Health Inventory (MHI). The MHI demonstrated excellent validity and reliability and good to excellent correlations with the Beck Depression Inventory.28 The subscales of Physical and Emotional Role Functioning assess working capacity.27
Physical Measures and Activity Levels
Maximum Back Extension Strength and Range of Motion Measures. Maximum isometric back extension strength was tested using a special test device (F110 DAVID, Helsinki, Finland) (Fig. 1). This device consists of a hip fixation mechanism to guarantee valid measurements. Patients were seated with their knees flexed and their trunks flexed forward at 30 degrees relative to the vertical. Trunk extensor torque was displayed in real time on a monitor (EVE). After adequate submaximal dynamic warm-up, patients generated 2 maximum isometric back extensions, while examiners provided standardized verbal encouragement. If maximum back extensions varied by >10%, a third trial was given. The best value of 2 consistent attempts was recorded. Patients’ maximum trunk ROM was assessed on the David F110 device in the sagittal plane from the maximum flexed to the maximum extended positions, and was measured in degrees.
Flexion-Relaxation Test. Dynamic trunk flexion-extension records the electromyographic back muscle activity during a maximum trunk anteflexion from the upright standing position. This method has been found to be an accurate diagnostic test that identifies impaired neuromuscular back extensor function (Fig. 2).29,30
Double parallel-bar electrode sensors (Trigno; DelSys Inc., Boston, MA) were attached to the skin bilaterally over the multifidi muscles at L5.31,32 After a few familiarization trials, patients performed 2 full flexions and reextensions at a given velocity.29,30 They slowly flexed their trunk forward until reaching a half-way position, then slowly further flexed their trunk to the point of maximum flexion. Thereafter, they reextended their trunk back to the half-way and to the upright positions. The procedure was repeated twice without pausing, and both trials were recorded. The variables of the flexion-relaxation test comprised of measures in upright position, full flexion and the ratio between the half-flexed position and the full flexion (half flexion-relaxation ratio).
Anticipatory and Post Eexpositional Emotional States. Immediately before and after the physical tests, patients were asked to rate their anticipatory and post expositional emotional states regarding the physical measures on Borg Category Ratio scales ranging from 0 (nothing at all) to 10 (extremely strong).33 The emotions were derived from the affective response patterns of the AEQ,22 assessing positive as well as negative emotions. All scales showed satisfactory internal consistency estimates (Cronbach α coefficient ranging from 0.85 to 0.91).
Activity Levels. The International Physical Activity Questionnaire (IPAQ) long-form interview version comprising 27 items was used to record average physical activity levels34,35 in different domains of life (domestic/gardening, work, transport, recreation/sports), weekly average moderate and vigorous activity levels, as well as daily time spent sitting or walking. The interview consists of a total Physical Activity sum scale to yield weekly average activity levels converted in metabolic equivalent of tasks with 1 metabolic equivalent of tasks equating the energy consumption in an inactive state (=3.5 mL O2/kg/min).36
Patients performed a progressive resistance training to improve muscle strength and endurance according to the evidence-based recommendations provided by the American College of Sports Medicine.37 The training equipment used was comprised of 10 DAVID and Technogym training machines for the back muscles, abdominal muscles, and adjacent muscles (hip adductors, abductors). The training was based on the initial maximum isometric muscle strength and mobility measures, and the resistance was gradually increased every 4 weeks. At each of the training sessions, patients performed a 10-minute warm-up. Thereafter, they trained in a standardized way completing 2 sets of 15 repetitions per device within their predefined ROM. All the training procedures, the number of repetitions, movement velocity, and the range of movement were displayed in real time on monitors attached to the devices providing visual feedback when performing the exercises. Each training session lasted approximately 1 hour. Patients were instructed to train twice a week. Training frequency and intensity levels were automatically recorded and stored by the DAVID devices.
Cross-Sectional Analysis at Baseline. For the cross-sectional analysis examining differences between the AEM subgroups in physical trunk muscle function, the following variables were defined as outcomes: back muscle extension strength, trunk ROM from full flexion to maximum extension, the electromyographic activity recorded during erect standing and full trunk flexion, the respective flexion-relaxation ratio,29,30 and the anticipatory and post exposition emotional states related to these measures. Furthermore, the activity levels as assessed by the IPAQ34 were investigated.
Longitudinal Outcomes. The longitudinal outcome variables were chosen according to previous studies investigating prospective changes of AEM subgroups after intervention.1–3 Pain intensity (VAS),20 quality of life (SF-36), and working capacity (SF-36)27 served as primary outcome measures and were investigated after intervention and at follow-up. Disability levels (RMDQ23; PDI25), back extensor strength and the FA subscales “pain catastrophizing,” “avoidance of physical activities,” “help and hopelessness,” as well as the endurance-subscale of “pain persistence behavior” derived from the AEQ22 were defined as secondary outcome parameters and were investigated directly after intervention.
According to the AEM, cLBP patients were classified into the subgroups by a cluster analysis using the TSS and BES subscales of the AEQ22 and the MHI.27 The raw scores were transformed to z-scores to provide standardized scores for subsequent cluster analysis. On the basis of an input distance matrix (Euclidean distances), a single-linkage hierarchical clustering was applied to detect and subsequently eliminate outliers. After removing the outliers, a Ward hierarchical clustering was performed, which minimizes the within-cluster sum-of-squares. To determine the number of clusters, statistical (NbClust package,38 cluster R2) as well as theoretical/substantive criteria1,10,11 were taken into account.
The measures of physical function were compared between subgroups using analysis of covariances, controlling for age, sex, and body mass index. In case of a significant group effect, Tukey honestly significant difference post hoc tests were computed to evaluate differences between the AEM subgroups. Longitudinal differences in treatment effect were evaluated using mixed-effects models with groups (AR, FAR, DER, EER) serving as the between-subjects factor and time (baseline, after intervention, 6-month follow-up) as the within-subjects factor. In case of significant group effects, simple effect analyses were applied to test for differences among AEM subgroups at each point in time. For a more detailed exploration of changes in pain levels, simple effects analyses were computed.
All statistical analyses were performed in the R environment for statistical Computing.39
A total of 178 patients were eligible for the study and completed the baseline measurements. Of these 178 patients, 29 dropped out during the half-year training period, which equated to a dropout rate of 16.29%. The reasons for dropout from training were (1) disorders of the musculoskeletal system [shoulder injuries (n=3), disk herniation (n=2), pain in the iliac sacral joint (n=1), increased pain in the lower back (n=1)] that were unrelated to training therapy, (2) diseases such as malignant tumor (n=2), hepatic inflammation (n=1), and (3) not further specified health reasons (n=2). Further reasons for early cessation of training included reported lack of time due to employment (n=7), dissatisfaction with the training (n=1), pregnancy (n=1), and being overweight (n=1). No information was available for 7 patients. A descriptive analysis of the clustering variables showed that the dropouts were not overly represented in 1 of the 4 specific AEM subgroups. Another 12 patients had to be excluded from further statistical analyses: 2 patients had missing data within the AEQ, and the single linkage clustering method proposed an exclusion of 10 extreme outliers. Finally, a total of 137 patients were stratified into the AEM subgroups. Baseline characteristics of the whole group are presented in Table 1. The follow-up examination was conducted through mail. About 97 of the 137 patients answered the postal follow-up, which yielded a return rate of 68.92%. AEM subgroups did not differ significantly in their dropout rate.
Results of the Cluster Analysis
A total of 23 indices for determining the number of clusters as implemented in the NbClust package38 were used. Of these, 15 indices suggested a 3-cluster solution, whereas the cluster R-squared (ie, the ratio of between-cluster sum-of-squares and total sum-of-squares) revealed that the 3-cluster solution explained only 50.47%, the 4-cluster solution 58.90%, or the 5-cluster solution 64.49% of the variance, respectively. By taking all these statistical measures and the theoretical considerations into account, a 4-cluster solution was chosen. According to the AEM, low back pain patients were subclassified into (1) FAR with low TSS, BES, and low MHI scores, (2) DER with high values at TSS and BES and low MHI scores, (3) EER with high scores at TSS, BES, and MHI, or (4) AR with low TSS, BES, and high MHI. The means and SDs of the clustering variables are provided in Table 2. Thus, a total of 24% of the patients were classified as FA, 34% as DE, 17% as EER, and 25% as AR.
Differences of AEM-based Subgroups in Physical Measures, Activity Levels, and Emotional States at Baseline
At baseline, all the outcome variables related to back extensor strength, trunk ROM, and the flexion-extension test were similar between groups (Table 3). Moreover, AEM subgroups did not differ within any of the IPAQ subscales or the sum scale from each other.
Shortly before and immediately after the physical testing, patients classified as FAR reported significantly more negative and less positive emotions toward the testing compared with the other subgroups (Table 4). Likewise, after completion of the tests, patients classified as DE reported more negative emotions regarding the test procedure than the AR group.
Outcomes and Longitudinal Changes After the 6 Months’ Rehabilitation Training
The amount of training performed was similar in all AEM subgroups. On average, patients trained once a week within the 6-month training period (mean=27.74, SD=16.31). Training intensity levels did not differ between subgroups.
Primary Outcomes After Intervention and at the 6-Month Follow-up
Means and SEs of the 3 test time points and the results of the mixed-effects models are provided in Table 5.
Pain Intensity. A significant time effect was found for pain intensity. Although the overall interaction effect was nonsignificant, simple effects analysis revealed that the AR, EER, and the FAR showed a significant reduction of pain from baseline to the end of intervention. However, at the 6-month follow-up, the pain levels of the FAR had increased significantly again. The DER did not demonstrate any significant changes in their pain levels at the end of training or at the 6-month follow-up.
Quality of Life and Working Capacity. All subgroups revealed significant improvements in physical health with training (time effect). The FAR subgroup, which had demonstrated the lowest mental health levels at baseline, achieved the best improvements in this category (group×time effect). Notably, after the end of the 6-month training intervention, the FAR still showed significantly lower mental health levels than the other subgroups, and at the 6-month follow-up, an ongoing tendency toward lower mental health compared with AR.
AEM subgroups differed significantly in their working capacity (physical role functioning and emotional role functioning) at each test time point (group effect). At baseline, the FAR and DER exhibited lower physical role functioning compared with AR, and the FAR had lower values compared with EER, whereas at the end of the intervention, the DER continued to have lower values in physical role functioning compared with AR. Regarding emotional role functioning, before the intervention, the FAR showed a significantly worse emotional role functioning compared with all the other subgroups, and after intervention, the FAR still reported significantly lower values than EER and AR. At the end of intervention, the DER also differed significantly from AR. At follow-up, no more differences in subgroup working capacity were observed.
Means and SEs of the secondary outcome variables and the results of the mixed-effects models are outlined in Table 6.
Disability. Throughout the training, all subgroups showed significant reductions in their disability levels (RMDQ, PDI) (time effect). At baseline, the FAR and DER had significantly higher values in RMDQ and PDI compared with AR, and the FAR had significantly higher PDI levels compared with EER (group effect), whereas subgroups did not differ significantly at the end of intervention.
Back Extension Strength. All AEM subgroups had a similar significant improvement in back extensor strength at the end of intervention (time effect).
Avoidance-Endurance Pain Response Patterns. Feelings of help and hopelessness as well as catastrophizing thoughts were found to be significantly decreased in all the AEM subgroups at the end of the training intervention (time effect). However, simple effects analysis for significant group differences revealed that after the intervention, the FAR still demonstrated higher pain catastrophizing compared with AR (group effect). At the end of intervention, the DER and FAR also reported significantly more feelings of help and hopelessness compared with AR (group effect). Both at baseline and at the end of intervention, the DER showed significantly higher levels in physical avoidance compared with EER and AR (group effect). The pain persistence behavior did not change over time. Before as well as at the end of the intervention, the EER showed significantly higher levels of pain persistence, whereas the pain persistence of the AR was significantly lower compared with that of other subgroups (group effect).
This study found for the first time that (1) measurements of physical tests and activity levels revealed similar results in all four AEM subgroups of cLBP patients and (2) that the AEM subgroups benefitted differently in their outcomes after a 6-month training intervention, that is, the FAR and the DER showed the worst therapy outcomes.
The comparisons among physical variables did not reveal any differences in bodily overuse or underuse between FAR or patients with endurance response patterns, although patients with predominant FAR behavior stated more negative emotions toward the physical measures. These results are in line with those of a previous study by Demoulin et al,9 which found comparable results in back muscle strength and ROM measures of FAR and non-FAR. Interestingly, the results of the flexion-extension test seem to contrast the findings from another study.8 The authors of this previous research found higher lumbar extensor electromyography muscle activity at the maximum flexion position in fear-avoidant cLBP patients, leading to higher flexion-relaxation ratios in these patients. However, cLBP patients in this study also showed a significant restriction in their lumbothoracic ROM, which could cause more activated lumbar extensor muscles to overcome gravity in a more erect maximum trunk flexion position. Thus, it remains unclear as to whether the observations from this study were related to either the FAR behavior or the restricted trunk flexibility, or both.
We compared the AEM subgroups in their self-reported activity levels in recreation and leisure, domestic and gardening, work-related, and transport-related tasks, as well as their total physical activity, according to the recommendations of Abenhaim et al40 to investigate subgroups’ activity levels in different domains of life. Most surprisingly, and in contrast to the results of previous studies,10,11 the AEM subgroups did not reveal any significant differences in the activity levels in any of the domains of life or in their moderate, vigorous, or total activity levels. However, there is evidence that FAR avoid only specific movements, for example turning or bending, but not movements or activity in general,41 which might explain the different results. In addition, no disuse and deconditioning syndrome of FARs is sufficiently evident to date.42 Furthermore, Plaas et al10 found a higher amount of accelerometer-based measures of static postures (sitting, standing) in endurers compared with FAR, which was related to task persistence behavior in daily duties. In contrast, our study did not reveal any differences in self-reported daily sitting time between subgroups. However, overt activity levels could not be compared one-to-one with self-reported measures.
Training therapy is a worldwide common intervention for LBP patients with proven effectiveness.43 Although all subgroups benefitted from the rehabilitation training program, the FAR and DER had the least positive outcomes, and reported lower quality of life and inferior working capacity, higher FAR attitudes (catastrophizing thoughts, feelings of help and hopelessness), and FAR behavior (avoidance of physical activities) after intervention compared with AR or EER. Scholich et al3 also investigated patients with cLBP and similarly concluded that FAR and DER have the worst treatment outcome, whereas the EER came close to the results of the AR. In a study of back pain patients engaged in competitive sports,44 authors also found a higher risk of ongoing disability and pain chronicity in the FAR and DER subgroups, which reported more pessimistic thoughts and negative emotions. Accordingly, Pincus et al45 summarized in a review that negative thoughts and depressed mood might better explain pain chronicity and worsening health conditions than the mechanisms of FAR and ER alone.
To conclude, we did not find differences in physical performance and activity levels between the AEM subgroups that would stand for bodily overuse or underuse and could explain a higher risk of pain chronicity. Rather, AEM subgroups were differentiated through psychological differences and treatment outcomes regarding pain, quality of life, and working capacity. We found that the classification into the AEM-based subgroups had a prognostic value for the rehabilitation outcome, and that it is therefore important to provide subgroup-based, individualist therapy adapted to patients’ needs. There are different concepts to classify patients into subgroups: for example, Turk and Rudy classified patients on the basis of psychosocial variables using the West Haven-Yale Multidimensional Pain Inventory46, the STarT-back screening tool47 classifies patients in low-risk, middle-risk, or high-risk groups, and Huijnen et al,48 McCracken and Samuel,49 and Andrews et al4 built subgroups based on activity-related behavior. As the FAR and DER had a less positive outcome after rehabilitation, they might benefit from subgroup-tailored psychological interventions. Mindfulness-based interventions could help these patients to become aware of their own pain-coping style, and acceptance-based interventions could help them to change pessimistic thought contents and depressed mood.
Some limitations have to be considered in our study. The voluntary aspect of the training therapy could have resulted in selection bias, as patients with higher FAR behavior may have been less likely to apply. Furthermore, patients reported relatively low to moderate pain and disability levels at baseline, and thus patients with higher scores could reveal different results. Besides, FAR beliefs might have been decreased in physical measures through the motivation through the examiners, and the received information about the upcoming procedures. Thus, it might not be possible to generalize the results of the supervised physical measures to all cLBP patients.
Overall, the results of this study underline the importance of AEM-subgroup-specific interventions for patients with cLBP.
The authors thank Richard Habenicht and Sara Riegler BSc for their help in acquisition of data, and Savo Ristic and Margit Ritter for their technical assistance and valuable support. They also thank Birgit Paul for her essential contributions during the planning of the study and Haley Milko for providing English language corrections.
1. Hasenbring M, Hallner D, Klasen B, et al. Pain-related avoidance versus endurance in primary care patients with subacute back pain: psychological characteristics and outcome at a 6-months follow-up. Pain. 2012;153:211–217.
2. Grebner M, Breme K, Rothoerl R, et al. Coping und Genesungsverlauf nach lumbaler Bandscheibenoperation. [Coping and convalescence course after lumbar disk operations]. Schmerz. 1999;13:19–30.
3. Scholich S, Hallner D, Wittenberg R, et al. The relationship between pain, disability, quality of life and cognitive-behavioural factors in chronic back pain. Disabil Rehabil. 2012;34:1993–2000.
4. Andrews EA, Strong J, Meredith PJ. Overactivity in chronic pain: is it a valid construct? Pain. 2015;156:1991–2000.
5. Wong KC, Lee RY, Yeung SS. The association between back pain and trunk posture of workers in a special school for the severe handicaps. BMC Musculoskelet Disord. 2009;10:43.
6. Al-Obaidi S, Nelson R, Al-Awadhi S, et al. The role of anticipation and fear of pain in the persistence of avoidance behavior in patients with chronic low back pain
. Spine. 2000;25:1126–1131.
7. Thomas JS, France CR, Sha D, et al. The influence of pain-related fear on peak muscle activity and force generation during maximal isometric trunk exertions. Spine. 2008;33:342–348.
8. Neblett R, Brede E, Mayer T, et al. What is the best surface EMG measure of lumbar flexion-relaxation for distinguishing chronic low back pain
patients from pain-free controls? Clin J Pain. 2013;29:334–340.
9. Demoulin C, Huijnen IP, Somville PR, et al. Relationship between different measures of pain-related fear and physical capacity of the spine in patients with chronic low back pain
. Spine J. 2013;13:1039–1047.
10. Plaas H, Sudhaus S, Willburger R, et al. Physical activity and low back pain
: the role of subgroups based on the avoidance-endurance model
. Disabil Rehabil. 2013;36:749–755.
11. Hasenbring M, Plaas H, Fischbein B, et al. The relationship between activity and pain in patients 6 months after lumbar disc surgery: do pain-related coping modes act as moderator variables? Eur J Pain. 2006;10:701–709.
12. Middelkoop M, Rubinstein SM, Kuijpers T, et al. A systematic review on the effectiveness of physical and rehabilitation interventions for chronic non-specific low back pain
. Eur Spine J. 2011;20:19–39.
13. Mannion AF, Dvorak J, Taimela S. Lumbale Rückenschmerzen: Vergleich von drei aktiven Therapieverfahren. [Chronic low back pain
: a comparison of three active therapies]. Manuelle Medizin. 2001;39:170–176.
14. Moffett JA, Carr J, Howarth E. High fear-avoiders of physical activity benefit from an exercise program for patients with back pain. Spine. 2004;29:1167–1172.
15. Boersma K, Linton S, Overmeer T, et al. Lowering fear-avoidance and enhancing function through exposure in vivo a multiple baseline study across six patients with back pain. Pain. 2004;108:8–16.
16. Vlaeyen JW, De Jong J, Geilen M, et al. Graded exposure in vivo in the treatment of pain-related fear: a replicated single-case experimental design in four patients with chronic low back pain
. Behav Res Ther. 2001;39:151–166.
17. Vlaeyen JW, De Jong J, Geilen M, et al. The treatment of fear of movement/(re)injury in chronic low back pain
: further evidence on the effectiveness of exposure in vivo. Clin J Pain. 2002;18:251–261.
18. Hasenbring MI, Klasen BW, Rusu ACHasenbring MI, Rusu AC, Turk DC. Risk factor-based cognitive-behavioural therapy for acute and subacute back pain. From Acute to Chronic Back Pain. New York, NY: Oxford University Press; 2012:513–528.
19. Merskey H, Bogduk N. Classification of Chronic Pain. Seattle, WA: IASP Press; 1994.
20. Knop C, Oeser M, Bastian L, et al. Development and validation of the Visual Analogue Scale (VAS) Spine Score. Unfallchirurg. 2001;104:488–497.
21. Sindhu BS, Shechtman O, Tuckey L. Validity, reliability, and responsiveness of a digital version of the visual analog scale. J Hand Ther. 2011;24:356–363.
22. Hasenbring M, Hallner D, Rusu A. Fear-avoidance- and endurance-related responses to pain: development and validation of the Avoidance-Endurance Questionnaire (AEQ). Eur J Pain. 2009;13:620–628.
23. Wiesinger GF, Nuhr M, Qittan M, et al. Cross-cultural adaptation of the Roland-Morris questionnaire for German-speaking patients with low back pain
. Spine. 1999;24:1099–1103.
24. Roland M, Fairbank J. The Roland-Morris Disability Questionnaire and the Oswestry Disability Questionnaire. Spine. 2000;25:3115–3124.
25. Dillmann U, Nilges P, Saile H, et al. Behinderungseinschätzung bei chronischen Schmerzpatienten. [Assessing disability in chronic pain patients]. Schmerz. 1994;8:100–110.
26. Chibnall JT, Tait RC. The Pain Disability Index: factor structure and normative data. Arch Phys Med Rehabil. 1994;75:1082–1086.
27. Bullinger M. German translation and psychometric testing of the SF-36 Health Survey: preliminary results from the IQOLA Project. International Quality of Life Assessment. Soc Sci Med. 1995;41:1359–1366.
28. Beukel van den TO, Siegert CE, Dijk van S, et al. Comparison of the SF-36 Five-item Mental Health Inventory and Beck Depression Inventory for the screening of depressive symptoms in chronic dialysis patients. Nephrol Dial Transplant. 2012;27:4453–44537.
29. Watson PJ, Booker CK, Main CJ, et al. Surface electromyography in the identification of chronic low back pain
patients: the development of the flexion relaxation ratio. Clin Biomech. 1997;12:165–171.
30. Kienbacher T, Fehrmann E, Habenicht R, et al. Age and gender related neuromuscular pattern during trunk flexion-extension in chronic low back pain
patients. J Neuroeng Rehabil. 2016;19:13–16.
31. Hermens HJ, Freriks B, Merletti R, et al. European recommendations for surface electromyography. Roessingh Res Dev. 1999;8:13–54.
32. Larivière C, Arsenault AB, Gravel D, et al. Evaluation of measurement strategies to increase the reliability of EMG indices to assess back muscle fatigue and recovery. J Electromyogr Kinesiol. 2002;12:91–102.
33. Borg G. Borg’s Perceived Exertion and Pain Scales. Champaign, IL: Human Kinetics; 1998.
34. Hagströmer M, Bergman P, De Bourdeaudhuij I, et al. Concurrent validity of a modified version of the International Physical Activity Questionnaire (IPAQ-A) in European adolescents: The HELENA Study. Int J Obes. 2008;32:42–48.
35. Craig CL, Marshall AL, Sjöström M, et al. International Physical Activity Questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. 2003;35:1381–1395.
36. Hagströmer M, Oja P, Sjöström M. The International Physical Activity Questionnaire (IPAQ): a study of concurrent and construct validity. Public Health Nutr. 2005;9:755–762.
38. Charrad M, Ghazzali N, Boiteau V, et al. NbClust: an R package for determining the relevant number of clusters in a data set. J Stat Soft. 2014;61:1–36.
39. R Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing 2016, Vienna, Austria. 2016. Available at: http://www.R-project.org/
Accessed February 2016.
40. Abenhaim L, Rossignol M, Valat JP, et al. The role of activity in the therapeutic management of back pain. Report of the International Paris Task Force on Back Pain. Spine. 2000;25:1–33.
41. Pincus T, Smeets RJ, Simmonds MJ, et al. The fear avoidance model disentangled: improving the clinical utility of the fear avoidance model. Clin J Pain. 2010;26:739–746.
42. Verbunt JA, Seelen HA, Vlaeyen JW, et al. Disuse and deconditioning in chronic low back pain
: concepts and hypotheses on contributing mechanisms. Eur J Pain. 2003;7:9–21.
43. Scharrer M, Ebenbichler G, Pieber K, et al. A systematic review on the effectiveness of medial training therapy for subacute and chronic low back pain
. Eur J Phys Rehabil Med. 2012;48:361–370.
44. Gajsar H, Titze C, Levenig C, et al. Predicting pain and disability outcomes in high-performance athletes using the endurance-avoidance approach. Poster, presented at the 9th Congress of the European Pain Federation, September 2-5, 2016, Vienna, Austria.
45. Pincus T, Vogel S, Burton AK, et al. Fear avoidance and prognosis of back pain: a systematic review and synthesis of current evidence. Arthritis Rheum. 2006;54:3999–4010.
46. Turk DC, Rudy TE. Toward an empirically derived taxonomy of chronic pain patients: integration of psychological assessment data. J Consult Clin Psychol. 1988;56:233–238.
47. Hill JC, Whitehurst DG, Lewis M, et al. Comparison of stratified primary care management for low back pain
with current best practice (STarT Back): a randomised controlled trial. Lancet. 2011;378:1560–1571.
48. Huijnen IP, Verbunt JA, Peters ML, et al. Differences in activity-related behavior among patient with chronic low back pain
. Eur J Pain. 2011;15:748–755.
49. McCracken LM, Samuel VM. The role of avoidance, pacing, and other activity patterns in chronic pain. Pain. 2007;130:119–125.