Archives of Physical Medicine and Rehabilitation
Volume 90, Issue 5 , Pages 768-777, May 2009

Clinical Outcomes for Patients Classified by Fear-Avoidance Beliefs and Centralization Phenomenon

  • Mark W. Werneke, PT, MS

      Affiliations

    • Spine Rehabilitation at CentraState Medical Center, Freehold, NJ
    • Corresponding Author InformationReprint requests to Mark W. Werneke, PT, MS, Dip. MDT at CentraState Medical Center, 901 West Main St, Freehold, NJ 07728
    • ,
  • Dennis L. Hart, PT, PhD

      Affiliations

    • Focus On Therapeutic Outcomes Inc, White Stone, VA
    • ,
  • Steven Z. George, PT, PhD

      Affiliations

    • Brooks Center for Rehabilitation Studies, University of Florida, Gainsville, FL
    • ,
  • Paul W. Stratford, PT, MS

      Affiliations

    • School of Rehabilitation Science and Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, ON, Canada
    • ,
  • James W. Matheson, PT, DPT

      Affiliations

    • Therapy Partners Inc, Burnsville, MN
    • ,
  • Adrian Reyes, PT

      Affiliations

    • Spine Rehabilitation at CentraState Medical Center, Freehold, NJ

Article Outline

Abstract 

Werneke MW, Hart DL, George SZ, Stratford PW, Matheson JW, Reyes A. Clinical outcomes for patients classified by fear-avoidance beliefs and centralization phenomenon.

Objectives

To (1) determine the prevalence of pain pattern classification subgroups (centralization, noncentralization, and not classified) observed during the initial evaluation of patients experiencing high versus low Fear-Avoidance Beliefs Questionnaire for physical activity (FABQ-PA) scores, (2) examine the association between discharge pain intensity and functional status (FS) outcomes based on FABQ-PA and pain pattern subgroups, and (3) compare minimal clinically important improvement for FS and pain intensity for FABQ-PA and pain pattern classification subgroups.

Design

Observational cohort design.

Setting

Suburban hospital-based outpatient rehabilitation clinic.

Participants

Consecutive patients with low back syndromes (N=238, mean ± SD, 59.1±17.0y; minimum=20, maximum=91).

Interventions

Interventions were designed to match patient classification by fear-avoidance level and pain pattern.

Main Outcome Measures

Two outcome measures were assessed: patient self-reported FS and pain intensity. FS was assessed by using computerized adaptive testing methods. Maximal pain intensity was assessed by using an 11-point numeric pain scale: 0 (no pain) to 10 (worst imaginable pain).

Results

There were no differences (χ22=3.7, P=.16) in proportion of patients classified by pain pattern experiencing high or low fear-avoidance beliefs. After controlling for the effect of available risk-adjustment variables, only dual-level classification subgroups, symptom acuity, payer type, and intake FS or pain intensity were associated with discharge FS or pain outcomes. The highest proportion of patients achieving minimal clinically important improvement in pain and FS were reported by the following patient subgroup: centralization and low fear.

Conclusions

Pain pattern and FABQ-PA characteristics impacted rehabilitation outcomes. We recommend that both factors be considered when managing patients with low back pain in an effort to optimize rehabilitation outcomes.

Key Words: Low back pain, Outcome assessment (health care), Rehabilitation

List of Abbreviations: CI, confidence interval, FS, functional status, FABQ, Fear Avoidance Belief Questionnaire, FABQ-PA, Fear Avoidance Belief Questionnaire for Physical Activity, LBP, low back pain

 

THE USE OF A BIOPSYCHOSOCIAL model1, 2 and multilevel patient classification methods3 that concurrently consider psychologic, social, and physical domains to evaluate and manage patients with nonspecific acute or chronic LBP syndromes has gained widespread acceptance. For example, such an approach is now recommended by clinical practice guidelines and by the World Health Organization's International Classification of Functioning, Disability and Health.4, 5 Consistent with these recommendations, there is evidence of the effectiveness for clinical assessment and management by physical therapists for pain-related fear6, 7 and the centralization of symptoms8, 9 experienced by patients with LBP.

Pain-related fear is part of the Fear-Avoidance Model.10 The Fear-Avoidance Model has been used to explain, in part, poor treatment outcomes, the transition between acute and chronic stages of pain, and the sustainability of chronic disability.10, 11, 12 Patients with LBP syndromes and elevated fear-avoidance traits typically avoid spinal movements and activities of daily living in an effort to avoid pain or perceived harm. Over time, the avoidance of movement may sustain and/or contribute to chronic pain and disability troubles.11 Recent studies have shown that the effectiveness of treatments in reducing disability for patients with lumbar impairment was associated with decreases in fear-avoidance beliefs. For example, exercise programs to improve fitness and strength,6, 13, 14, 15 educational programs to stay active and avoid unnecessary medical procedures and treatments,16, 17, 18 and cognitive behavioral techniques to alter the patient's exaggerated threat value of pain and to improve physical activity tolerance19, 20, 21, 22, 23 were all successful in reducing fear-avoidance beliefs. Taken collectively, these studies suggest that clinicians should emphasize and use treatment strategies specifically designed to reduce elevated pain-related fears to optimize patient outcomes.

The centralization phenomenon is a common physical examination sign to guide specific exercise prescription.24, 25 The centralization phenomenon is characterized by spinal-referred symptoms that are progressively abolished in a distal-to-proximal direction in response to a standard clinical assessment method in which the clinician documents the patient's pain and symptom response(s) during repeated and end-range lumbar movement tests.25 The classification and management of patients with lumbar impairments based on centralization appear important because classification results have been used to successfully direct treatment, and the presence of centralizing symptoms consistently predicts good short- and long-term outcomes.8, 9, 26, 27, 28, 29, 30 A standardized pain pattern assessment and measurement technique to identify centralization have been recommended28 and supported by strong reliability and validity data.26, 27, 31, 32

It seems theoretically feasible that combining both factors (ie, fear avoidance and pain pattern) in a dual-level classification method and tailoring management techniques to each factor might benefit patients with lumbar spine syndromes. However, to date, research27, 30 investigating the combined influence of pain pattern and elevated fear-avoidance beliefs on rehabilitation outcomes using multivariate regression models has been scant. George et al30 reported that both intake centralization and elevated fear-avoidance beliefs of work activities were important for explaining long-term disability outcomes from LBP. In contrast, Werneke and Hart27 reported that when intake pain pattern classification and elevated fear avoidance were in a multivariate model, only the noncentralization of symptoms predicted disability at 1 year after discharge from rehabilitation. The ability to generalize these data are limited by small sample sizes,30 selected patient characteristics including only young patients with acute onset of LBP27, 30 or only patients whose symptoms centralized,30 and clinical interventions not matched to the level of fear avoidance at intake.27, 30 Therefore, further direction is needed for therapists interested in managing patients based on the level of fear avoidance and centralization. For example, if a patient is experiencing elevated fear-avoidance beliefs of physical activity at intake and he/she is managed based solely on the Fear-Avoidance Model, the patient may be missing the beneficial effects of also modifying management based on pain pattern response. The converse may be true as well, with patients managed solely by pain pattern response missing the benefits of management by the Fear-Avoidance Model.

We believe that the relative contribution or interaction between these factors during treatment is complex, and observational data are lacking from the literature. Therefore, we implemented an exploratory study to improve our understanding of patient outcomes when patients are classified at intake using a dual-level classification scheme when both pain pattern classification and fear-avoidance belief levels are considered by treating clinicians. Clinicians assessed and classified patients by pain pattern and FABQ-PA and administered specific treatments that have supporting evidence individually but have not been studied well when performed concurrently. The specific aims were to (1) determine the percentage of patients classified at intake by the centralization and noncentralization of symptoms who experience either high or low levels of fear-avoidance beliefs of physical activity, (2) examine the association between discharge pain intensity and FS and patients subgrouped by pain pattern and levels of fear at intake, and (3) compare minimal clinically important improvement for FS and pain intensity for patients subgrouped by pain pattern and fear. We hypothesized that the percentage of patients classified at intake by centralization would be the same for patients reporting either low or high fear-avoidance beliefs. In addition, we hypothesized that symptom centralization would be associated with better outcomes regardless of the level of fear avoidance and noncentralization and high fear would be associated with worse outcomes. Study results may allow physical therapists to better weigh the relative importance of 2 common initial examination findings for patients with LBP.

Back to Article Outline

Methods 

Subjects 

This study involved a secondary analysis of 238 patients with nonspecific lumbar spine syndromes from a previously described sample of patients.32 The original study32 was a prospective observational cohort design collecting data from 517 consecutive patients with mechanical neck or LBP who were treated at 1 suburban hospital-based outpatient rehabilitation clinic between June 2003 and December 2004. The clinic was participating with Focus On Therapeutic Outcomes Inc, a medical rehabilitation data-management company.33, 34 Focus On Therapeutic Outcomes Inc's Institutional Review Board for the Protection of Human Subjects approved the project.

Procedures 

Patient classification at intake 

Procedures for patients' physical examination and data collection have been previously described.32 Briefly, all patients before physical examination completed a battery of questionnaires gathering functional and demographic information, pain, and psychosocial factors (ie, fear-avoidance belief questionnaire,35 depression, somatization subscales36, 37) by using Patient Inquiry software.a Patients were evaluated by 1 of 2 physical therapists who were trained and credentialed at examining and classifying patients using mechanical diagnosis and therapy methods.25

For this study, patients were classified using a dual-level classification method combining both pain pattern and FABQ-PA classification subgroups as follows: centralization and fear low, centralization and fear high, noncentralization and fear low, noncentralization and fear high, not classified by pain pattern and fear low, and nonclassified by pain pattern and fear high.

Patients were also classified at intake using the FABQ, which has been recommended as a screening tool for patients with lumbar spine syndromes who are fearful of physical or work activities.35 The FABQ-PA has been reported to exhibit an acceptable level of test-retest reliability35, 38 and has been used to assess fear-avoidance beliefs in patients with lumbar spine syndromes.15, 39, 40 For this study, we included only the FABQ-PA survey because many patients in our sample were older than 65 years and retired, making the FABQ work subscale not applicable. We classified patients according to their level of fear of physical activities based on a median split previously recommended in the literature18 (ie, high fear was operationally defined as FABQ-PA scores of 15 points or more, and low fear was operationally defined as FABQ-PA scores of 0 to 14).

In addition to FABQ-PA classification, patients were also classified at intake using pain pattern criteria for categorizing patients by centralization, noncentralization, and not classified by pain pattern.32 Although centralization is a commonly known clinical phenomenon, operational definitions and assessment methods to identify centralization vary widely.28, 31 Using nonstandardized assessment methods and subjective techniques to identify centralization may explain variations in centralization's prevalence rates and validity.41 We specifically chose a standardized assessment and measurement technique using pain pattern criteria to identify centralization because of strong reliability and validity data supporting this pain pattern classification method.26, 27, 31, 32

Pain pattern classification methods have been described in detail elsewhere.32 Pain pattern is a method of classifying patients with back pain according to a change in anatomic location of pain reported in response to clinician-directed repeated and end-range lumbar movements. All changes in pain location were measured using body diagrams and a numeric body diagram overlay template tool, the reliability of which is very good.28 Because pain responses were numerically defined, if the patient's pain location score immediately before mechanical diagnosis and therapy assessment was 0, the patient was unable to be classified using this method.32 For example, if patients, especially those patients 65 years of age or older with persistent yet intermittent pains during the subjective history, report no pain immediately before repeated lumbar movement testing, they were classified as “not classified” using this method. Patients who could not be classified were retained in the analyses to serve as a comparison group for those patients who could be classified using pain pattern methods; the implications and limitations of not classified pain pattern subgroup will be discussed.

Intervention 

Patients were evaluated using a standard mechanical diagnosis and therapy assessment method at intake and reassessed during subsequent visits.25 Subsequent visits were scheduled at the therapist's discretion. Both therapists were experienced with intervention techniques described by McKenzie25 and with a specific cognitive-behavioral intervention “graded exposure in vivo” described by Vlaeyen et al.20, 42 Overall treatment strategies were guided by pain pattern and levels and FABQ-PA observed during intake and over the treatment episode. For instance, the patient subgroups centralization and fear low and centralization and fear high were prescribed specific exercises.25 For patients classified with elevated fear at intake, therapy treatment was augmented by integrating carefully selected educational information with an active management strategy based on “graded exposure in vivo” methods.21, 42 The graded exposure in vivo program was specifically designed by using cognitive-behavioral techniques to change the patient's perception concerning hurt versus harm, to reduce the patient's pain-related fears, and to regain the patient's confidence in returning to physical activities, which were commonly part of the prescribed specific exercises. Treatments for patients with noncentralizing symptoms or patients who could not be classified by pain pattern criteria consisted of reassurance, education to stay active, functional activities and exercises, mobilization/manipulation, and avoidance of passive modalities as recommended by recent clinical practice guidelines.5, 43 Because of this project's observational cohort research design, we did not standardize care beyond these guidelines.

Outcomes 

Two primary outcome measures, patient self-reported FS and pain intensity, have been described in a prior study32 and a detailed technical report.44 Briefly, FS was assessed by using computerized adaptive testing methods.44, 45, 46 The FS measures generated from the computerized adaptive testing range from 0 (low) to 100 (high functioning) on a linear metric. The items have adequate internal consistency reliability (r=.86) and person separation (2.49).46, 47 FS measures have good discriminant validity for condition severity, symptom acuity, age, and surgical history, and FS measures estimated by using the computerized adaptive testing are sensitive to change (lumbar effect size=1.05).44 The maximal pain intensity experienced during the preceding 24 hours was assessed by using an 11-point numeric pain scale, 0 (no pain) to 10 (worst imaginable pain).48, 49 The reliability and validity for this 11-point pain scale have been reported to be good and recommended for clinical assessment of subjective pain intensity.48, 49

Data Analysis 

To better understand possible patient selection bias, we assessed differences between patients who had missing pain pattern and fear classification data compared with those who had classification data, and we assessed differences between patients who dropped out of treatment compared with those who completed treatment by using chi-square tests of independence (dichotomous and categoric) or 2-sample t tests (continuous data) using available independent variables.

For purpose 1, we calculated the percentage of patients classified for each pain pattern subgroup (centralization, noncentralization, not classified by pain pattern) and level of fear-avoidance beliefs (high, low fear) and tested differences in the percentage of patients per subgroup by using chi-square tests for independence with standardized deviates.

For purpose 2, we used ordinary least squares multiple regressions to examine associations between intake patient classification and characteristics and discharge functional status and pain, which allowed us to control for risk-adjustment variables. One regression was run for each dependent variable, discharge FS and pain. The primary factor of interest was patient dual-level classification (table 1). To control for the influence of important risk-adjustment factors, we included the following independent variables in the regression models: intake FS or pain, acuity, age, sex, surgical history, payer, pain location, and therapist. The importance of each level of each variable was tested using a t statistic. The value for a type 1 error was set to .05 for all analyses.

Table 1. Patient Characteristics by Dual-Level Classification (N=238)
CharacteristicCen Fear LowCen Fear HighNoncen Fear LowNoncen Fear HighNot Classified Fear LowNot Classified Fear High
Sex
Male305830392748
Female704270617352
Pain location
Above657956577548
Below352144432552
Age
18–443050163288
45–64604238481112
65–7410823113616
≥75002394564
Surgery
None9010080817776
1 or more10020192324
Acuity
Acute455823302612
Subacute20418191616
Chronic353859525872
Payer⁎
Litigation081404
Medicaid5232701
Medicare6940258174
HMO3218181489
PPO471427201113
WC5146700
Other51462300
Therapist
First857165675976
Second152935334124

NOTE. Values are column percents per dual-level classification group for categoric data. For continuous data, values are least squares means (SEs).

Abbreviations: Cen, centralization; HMO, health maintenance organization; noncen, noncentralization; PPO, preferred provider organization; WC, workers' compensation.

⁎Chi-square statistics, P<.01.

For purpose 3, we compared the proportions of patients classified by both pain pattern and levels of fear who achieved at least a minimal clinically important improvement as determined by using receiver operating characteristics for change in FS and pain intensity.

The data analyses to calculate minimal clinically important improvement50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60 for FS and pain change were described at length in our original article.32 Minimal clinically important improvement represents the smallest observed change in patients estimated to have important change.50 The external measure or criterion standard to assess minimal clinically important improvement for FS and pain intensity change was a 15-point global rating of change scale.55, 58 For minimal clinically important improvement assessment, patients were dichotomized by their global rating of change on a 15 point scale (−7 to 7) as patients who did not improve (ie, global rating of change scores <3) versus patients who improved (ie, global rating of change scores ≥3).57 In addition, we used nonparametric receiver operating characteristic analyses to quantify the accuracy52 of the change measures to discriminate between patients whose FS and pain had improved in an important way compared with patients whose FS and pain had not improved.61 Minimal clinically important improvement cut points were identified by selecting the cut point with the largest average specificity/sensitivity. Receiver operator curve results were described using area under the receiver operator curves, SEs, and 95% confidence intervals.

Back to Article Outline

Results 

Subjects 

Over the data-collection time, 517 consecutive patients with either lumbar spine or cervical syndromes were solicited to participate in the original study (fig 1). Of these patients, 99 could not complete intake surveys or did not receive a standard physical examination. Of the remaining 418 patients, 316 had lumbar syndromes, and, of these, 48 dropped out before treatment discharge. Patients who dropped out had lower intake functional status measures (P=.005), higher pain at intake (P<.001), and were younger (P<.001), and proportionately fewer patients who dropped out could not be classified by pain pattern classification (P=.002). Of the remaining 268 patients with lumbar syndromes who started and completed treatment, 238 had complete pain pattern and FABQ-PA data (table 2). There were no differences between patients who were not classified at intake by both pain pattern and FABQ-PA (n=30) compared with patients who were classified (n=238) for age, intake FS, fear, or pain (P>.05).

  • View full-size image.
  • Fig 1. 

    A flow diagram showing patients with low back pain available for analyses. Abbreviation: MDT, Mechanical Diagnosis and Therapy Assessment method is described by McKenzie and May.25

Table 2. Patient Characteristics at Intake (N=238)
CharacteristicValue
Sex
Men (%)36.1
Age (y)
Mean ± SD59.1±17.5
Range20–91
Age by group
18–44y21.8
45–64y34.9
65–74y19.4
≥75y23.9
Most distal location of pain complaints
Above elbow or knee62.2
Below elbow or knee37.8
Fear-avoidance physical
Mean ± SD12.7±6.7
Range0–24
Functional status
Mean ± SD45.9±11.1
Range0–74
Pain
Mean± SD5.7±3.1
Range0–10
Symptom acuity
Acute (0–21d)29.0
Subacute (22–90d)16.4
Chronic (>90d)54.2
Missing0.4
Payer source
Litigation2.1
Medicaid4.2
Medicare Part B36.1
Health maintenance organization13.4
Preferred provider organization18.9
Workers' compensation4.6
Other7.7
Missing13.0
Surgical history
None82.4
One or more17.6
Pain pattern classification
Could not be classified30.0
Symptoms centralized18.5
Symptoms did not centralize52.5
Fear-avoidance beliefs physical activity
Low fear56.7
High fear43.3

NOTE. Values are percentages unless otherwise identified.

Patient Characteristics at Initial Evaluation 

Patient characteristics by pain pattern and levels of FABQ-PA are displayed in table 1. There were no differences in the proportion of patients between fear and pain pattern classification groups for sex, pain location, surgical history, or therapist. There were differences between treatment groups for age, symptom acuity, and payer. There were proportionally more patients aged 18 to 44 years old whose symptoms centralized and had high fear, proportionally more patients aged 75 years old or older whose symptoms could not be classified and had high fear, and proportionally more patients aged 65 years old or older whose symptoms could not be classified and had low fear (χ152=88.9, P<.001). There were proportionally more patients whose symptoms centralized and had high fear whose symptoms were acute (χ102=19.7, P=.032). There were proportionally more patients whose symptoms centralized and had high fear who received litigation or Medicaid benefits, more patients whose symptoms centralized and had low fear who received preferred provider organization benefits, and more patients whose symptoms could not be classified and had high fear or low fear who received Medicare Part B benefits (χ302=100.1, P<.001).

Purpose 1: Percentage of Patients Classified by Pain Pattern and Levels of Fear 

The overall percentages of patients for pain patterns at intake were 18%, 52%, and 30% for centralization, noncentralization, and not classified by pain pattern, respectively, and the overall percentages of patients for levels of fear at intake were 56% and 44% for low and high fear, respectively. The percentage of patients for pain patterns by level of fear are displayed in table 3. There were no differences (χ22=3.7, P=.16) in the percentage of patients classified by pain pattern reporting high or low fear.

Table 3. Percentage of Patients Classified by Pain Pattern by Levels of Fear
Pain PatternsFear High (%)Fear Low (%)
Centralization2315
Noncentralization5352
Not classified (reference)2433

Purpose 2: Associations Between Discharge Pain and Functional Status Outcomes and Intake Patient Dual-Level Classification 

The results from multivariate modeling for dependent variables FS and pain intensity are displayed in table 4. Only 4 independent variables at intake including intake FS or pain, symptom acuity, payer type, and patient dual-level classification were associated with or predicted discharge FS or pain outcomes after controlling the effect of all assessed risk-adjustment variables in the models (see table 4). Results regarding intake patient dual-level classification can be interpreted as follows: the noncentralization and fear high subgroup (independent variable) averaged 14 FS points lower and 2 pain intensity points higher at discharge compared with the patient subgroup centralization and fear low (reference variable) after controlling for all other independent variables in the model. In addition, the noncentralization and fear low subgroup of patients averaged 2 pain intensity points higher at discharge compared with the patient subgroup centralization and fear low after controlling for all other independent variables in the model.

Table 4. Associations Between Pain and Functional Status Outcomes and Intake Patient Dual-Level Classification
SubgroupsVariableReference VariableCoefficient95% CI
FS⁎Intake FSNA0.750.54to0.96
Chronic symptomsAcute symptoms−10.4−15.5to−5.2
Payer PPOPayer litigation15.50.84to30.1
Non-CEN, fear highCEN, fear low−14.2−22.8to−5.58
Pain†Intake painNA0.460.32to0.62
Subacute symptomsAcute symptoms1.340.19to2.49
Chronic symptomsAcute symptoms1.490.62to2.37
Payer PPOPayer litigation−3.24−5.72to−0.76
Payer MedicaidPayer litigation−3.76−6.55to−0.97
Payer Medicare Part BPayer litigation−3.20−6.20to−0.19
Payer HMOPayer litigation−2.62−5.10to−0.14
Payer WCPayer litigation−3.34−6.10to−0.58
Payer otherPayer litigation−3.66−6.27to−1.05
Non-CEN, fear lowCEN, fear low1.700.31to3.09
Non-CEN, fear highCEN, fear low2.270.82to3.72

Abbreviations: CEN, patients whose symptoms centralized; HMO, health maintenance organization; NA, not applicable; Non-CEN, patients whose symptoms did not centralize; PPO, preferred provider organization; WC, Workers' compensation.

⁎Discharge functional status regression model (n=207, R2=.42).

†Dscharge pain regression model (n=207, R2=.38).

Purpose 3: Patient Dual-Level Classification and Proportion of Patients Achieving Minimal Clinically Important Improvement 

Receiver operating curve and minimal clinically important improvement results are displayed in table 5. The proportion of patients achieving minimal clinically important improvement classified by pain pattern and levels of FABQ-PA are displayed in table 6. Standardized deviates for FS change suggested more than the predicted proportion of patients whose symptoms did not centralize and had high fear did not obtain minimal clinically important improvement, and fewer than predicted patients whose symptoms centralized and had low fear did not obtain minimal clinically important improvement. Fewer than predicted patients whose symptoms centralized and had low fear did not obtain minimal clinically important improvement by pain improvement. In each of the 4 analyses, the percent of patients with an improvement of functional status or pain change greater than or equal to minimal clinically important improvement was highest for centralization compared with noncentralization, which was supported by standardized deviate assessment. Standardized deviates did not support a difference for the percent of patients with an improvement of functional status or pain change greater than or equal to the minimal clinically important improvement within each pain pattern classification subgroup stratified by the level of fear.

Table 5. Sensitivity to Change Using Receiver Operating Characteristic Analyses
SubgroupsImproved (Change ≥3)No Change (Change <3)ROC Cut PointAUCSE95% CI% ≥MCII
FS116289.894.031.833–.95665
Pain113272.786.052.684–.88763

Abbreviations: AUC, area under the ROC curve; MCII, minimal clinically important improvement; ROC, receiver operating characteristic analysis.

Table 6. Proportion of Patients by Dual-Level Classification Achieving Minimal Clinically Important Improvement in Functional Status or Pain
FS ChangePain Change
Patient SubgroupsN=235⁎N=235⁎
Cen and fear low95.095.0
Cen and fear high79.287.0
Noncen and fear low74.366.2
Noncen and fear high51.957.4
Not classified and fear low64.352.4
Not classified and fear high60.048.0

NOTE. Values are percentages of patients whose changes scores were equal to or greater than minimal clinically important improvement.

Abbreviations: Cen, centralization; Noncen, noncentralization.

⁎P<.01 for chi-square tests. Three patients did not have global rating of change data.

Back to Article Outline

Discussion 

Our purpose was to improve our understanding of outcomes when patients were examined by using a dual-level classification (ie, pain pattern classification and fear-avoidance beliefs of physical activities), and patient management was modified to account for each classification subgroup. Overall, the results supported our hypotheses.

First, patients in this study were able to centralize their symptoms at intake during repeated end-range spinal movement tests despite experiencing elevated fear-avoidance beliefs of physical activity, supporting our first hypothesis. Based on the Fear-Avoidance Model, one may expect that patients with elevated fear would limit or not be able to perform such trunk movements, and centralization would potentially not be observed. For instance, recent studies62, 63, 64 have reported that patients with either acute, subacute, or chronic LBP who experience elevated pain-related fear show submaximal lifting capability and avoid or alter movements of the lumbar spine during physical tasks compared with similar patients with low levels of fear. Lumbar flexion range of motion has also been consistently shown to be inversely related to pain-related fear for patients with either acute or chronic LBP.65, 66, 67 Considering that clinical observation of centralization was enhanced when patients performed repeated trunk flexion and extension movements versus single trunk movements tests,68, 69 it was of interest that in our sample centralization was observed in patients who performed repeated lumbar movements despite experiencing elevated fear-avoidance beliefs. We believe that our data contrast with previous studies62, 63, 64, 70 showing an association between elevated pain-related fear and impaired physical performance or restricted lumbar spine movements because these studies did not examine or consider centralization when interpreting results.

Second, results from multivariate models support our second hypothesis indicating that symptom centralization would be associated with better outcomes regardless of the level of fear avoidance at intake and noncentralization and high fear would be associated with worse outcomes. For patients whose symptoms centralize, fear-avoidance beliefs of physical activity do not appear to influence patient outcomes, and augmenting specific exercise prescription based on centralization with fear-avoidance treatment targeting patients with elevated fear-avoidance beliefs of physical activity may not be necessary. Our hypothesis is consistent with data from a previous study30 investigating both centralization and fear-avoidance beliefs. George et al30, 40 did not recommend that intake FABQ-PA data be considered by clinicians managing patients with LBP because FABQ-PA was not significant in their proposed prognostic models. However, George et al30 limited their analyses to centralization and did not consider other commonly observed pain patterns from repeated movement testing such as noncentralization or patients whose symptoms could not be classified.31, 32 When using multivariate techniques, current results indicate the presence of noncentralizing symptoms, and elevated fear at intake was associated with a 14-unit reduction in discharge functional status (95% CI, −22.85 to −5.58) and a 2-unit increase in discharge pain (95% CI, 0.82–3.72) compared with patient subgroup centralization and fear low. Findings suggest that the assessment and management of both pain pattern and FABQ-PA classification at intake are important particularly for the patient subgroup experiencing elevated fear-avoidance beliefs of physical activity whose symptoms do not centralize during physical examination.

We are cautious with this interpretation of the results because we targeted patients with elevated fear as recommended by present professional knowledge with a graded exposure in vivo intervention,7, 20, 21, 42 which may have modified or masked the association between FABQ-PA and pain pattern levels and outcomes. It is also possible that elevated fears may be adequately addressed by treatment methods described by McKenzie and May,25 which emphasize a patient-centered approach and patient educational principles of empowerment and self-care, which may reduce elevated fear and the prognostic importance of intake FABQ-PA levels, all of which emphasize the complexity of this type of research and merit further study.

Third, our second hypothesis was confirmed by our minimal clinically important improvement analyses for pain and FS change. Specifically, FS change for patients whose symptoms did not centralize and reported elevated fear did not achieve minimal clinically important improvement for FS (52%) as frequently as patients whose symptoms centralized and reported low fear at intake (95%). Interestingly, although we observed a trend favoring a higher proportion of patients whose symptoms centralized at intake with low fear (95%) compared with centralizers at intake with high fear (79%) who achieved minimal clinically important improvement for FS, this trend was not supported statistically by standardized deviate analyses. Although the trend may be clinically important, judgment awaits future studies performed with larger samples. Pain change for patients whose symptoms did not centralize and reported elevated fear did not achieve minimal clinically important improvement for pain (57%) as frequently as the patient subgroup centralization and fear low (95%). The influence of fear avoidance for patients whose change in pain was clinically important appeared similar to its effect on FS. Our finding that fear avoidance influenced both pain and FS changes for certain subgroups of patients contrasts with earlier studies reporting that FABQ-PA levels were related more to disability compared with pain intensity.22, 64 We believe the difference is best explained by the influence of centralization-based treatment for reducing pain intensity. Again, this hypothesis awaits future studies.

Study Limitations 

There are several limitations that should be considered when interpreting our study's results. We classified patients into high or low FABQ-PA subgroups based on a median previously recommended in the literature.18 However, the validation of the FABQ-PA cut point used in this study to identify patients at risk is required. Although some investigators recommend using continuous data to analyze FABQ-PA,30 our goal was to classify patients by high versus low fear at intake to assist treatment direction. Our present research (unpublished observations) examining single items of FABQ scales of physical activity support an FABQ-PA cut point of 15 or higher for identifying patients with high fear.

Although recent randomized controlled trials support prescribing specific exercises and interventions for patients with LBP showing centralization8, 9 or experiencing high fear,6, 7 treatments for patients with noncentralizing symptoms or patients who could not be classified by pain pattern criteria as outlined in this study require validation. However, treatments rendered to these patient subgroups followed recommendations from current clinical practice guidelines.5, 43 In addition, 30% of our sample could not be classified by pain pattern. The inability to classify all patients following 1 specific classification technique is not a unique limitation to pain pattern classification methods. For example, in a recent study,24 75% of patients with nonspecific LBP referred to physical therapy were unable to be classified by using a different decision-making algorithm for subgrouping patients. The majority of patients not classified in our sample were older than 65 years of age (see table 2); therefore, the clinical use of pain pattern classification for this patient population appears limited. Developing specific treatment-based classification systems to improve outcomes for geriatric patients is recommended.

Finally, clinicians tailored treatment based on baseline FABQ-PA scores and pain patterns generated by the classification methods. Therefore, the assessment of outcomes and the interpretation of the results are confounded by the specific treatments that were rendered, which differed for each patient classification. Furthermore, results are affected by patient characteristics, which impact the heterogeneity of the patients within each treatment group. We attempted to control for the effect of patient differences by including several important and available independent variables in our regression models. Thus, the results of this study cannot be generalized to different cohorts with different independent variables and those who did not receive this type of classification and treatment.

Despite these limitations, clinically logical trends and associations seem to be present in the current data, which may benefit physical therapists interested in patient dual-level classification and treatment by centralization and fear-avoidance beliefs. Clinicians during routine clinical practice evaluate, classify, and treat a heterogeneous group of patients with diverse presentations of LBP, many of which are different than those studied in randomized controlled trials.71, 72 Therefore, there is a need to understand both the importance of using and difficulty investigating complex biopsychosocial treatment-based classification systems3 in actual practice settings using data from observational designs.73 Multilevel or dual-level classification processes as examined in this study that are used to direct the management of diverse patients add complexity to the assessment of treatment outcomes. Further research is warranted for not only examining the relationship between centralization and fear-avoidance beliefs when collected concurrently for patients with LBP syndromes in other treatment cohorts but also to concurrently investigate other important demographic, biomedical, psychologic, and social constructs that could affect patient outcomes.73

Future randomized controlled trials are necessary to advance the understanding of the effect of a dual-level classification process when it seems clinically logical, and our data suggest that patients could be classified 2 ways and treated accordingly for better outcomes. Our findings suggest that pain pattern is more important than FABQ classifications for patients whose symptoms centralize, yet both appear important for patients whose symptoms do not centralize and in whom fear is high. If future analyses support the trends and associations observed in these data, the management of patients with lumbar spine syndromes could benefit.

Back to Article Outline

Conclusions 

This study investigated clinical outcomes of patients managed with treatments tailored to address both a psychosocial and a physical domain commonly evaluated by clinicians (ie, fear-avoidance beliefs of physical activity and pain pattern based on symptom change after physical examination). Results suggest that when centralization is observed, augmenting treatment for patients with LBP with a fear-avoidance program may not be necessary. However, targeting patients experiencing high fear and noncentralization of symptoms for formal cognitive behavioral techniques appears important. Future studies are recommended to understand interactions between multiple classification and management strategies and their effect on outcomes.

Suppliers

Back to Article Outline

References 

  1. Waddell G. The back pain revolution. Edinburgh: Churchill Livingstone; 1998;
  2. Weiner BK. Spine update: the biopsychosocial model and spine care. Spine. 2008;33:219–223
  3. McCarthy C, Arnall F, Strimpakos N, Freemont A, Oldham J. The biopsychosocial classification of non-specific low back pain: a systematic review. Phys Ther Rev. 2004;9:17–30
  4. World Health Organization. International Classification of Functioning, Disability and Health. Geneva: World Health Organization; 2001;
  5. Rossignol M, Arsenault B, Dionne C. Clinic on low-back pain in interdisciplinary practice (CLIP) guidelines. Montreal: Agence de la sante et des services sociaux de Montreal; 2006;
  6. George SZ, Fritz JM, Bialosky JE, Donald DA. The effect of a fear-avoidance-based physical therapy intervention for patients with acute low back pain: results of a randomized clinical trial. Spine. 2003;28:2551–2560
  7. Leeuw M, Goossens ME, van Breukelen GJ, et al. Exposure in vivo versus operant graded activity in chronic low back pain patients: results of a randomized controlled trial. Pain. 2008;138:192–207
  8. Long A, Donelson R, Fung T. Does it matter which exercise? (A randomized control trial of exercise for low back pain). Spine. 2004;29:2593–2602
  9. Browder DA, Childs JD, Cleland JA, Fritz JM. Effectiveness of an extension-oriented treatment approach in a subgroup of subjects with low back pain: a randomized clinical trial. Phys Ther. 2007;87:1608–1618
  10. Lethem J, Slade PD, Troup JD, Bentley G. Outline of a Fear-Avoidance Model of exaggerated pain perception–I. Behav Res Ther. 1983;21:401–408
  11. Leeuw M, Goossens ME, Linton SJ, Crombez G, Boersma K, Vlaeyen JW. The fear-avoidance model of musculoskeletal pain: current state of scientific evidence. J Behav Med. 2007;30:77–94
  12. Vlaeyen JW, Linton SJ. Fear-avoidance and its consequences in chronic musculoskeletal pain: a state of the art. Pain. 2000;85:317–332
  13. Klaber 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
  14. Mannion AF, Junge A, Taimela S, Muntener M, Lorenzo K, Dvorak J. Active therapy for chronic low back pain: part 3 (Factors influencing self-rated disability and its change following therapy). Spine. 2001;26:920–929
  15. Marshall P, Murphy B. Self-report measures best explain changes in disability compared with physical measures after exercise rehabilitation for chronic low back pain. Spine. 2008;33:326–338
  16. Buchbinder R, Jolley D. Effects of a media campaign on back beliefs is sustained 3 years after its cessation. Spine. 2005;30:1323–1330
  17. Buchbinder R, Jolley D, Wyatt M. 2001 Volvo Award Winner in Clinical Studies: effects of a media campaign on back pain beliefs and its potential influence on management of low back pain in general practice. Spine. 2001;26:2535–2542
  18. Burton AK, Waddell G, Tillotson KM, Summerton N. Information and advice to patients with back pain can have a positive effect (A randomized controlled trial of a novel educational booklet in primary care). Spine. 1999;24:2484–2491
  19. Brox JI, Sorensen R, Friis A, et al. Randomized clinical trial of lumbar instrumented fusion and cognitive intervention and exercises in patients with chronic low back pain and disc degeneration. Spine. 2003;28:1913–1921
  20. Vlaeyen JW, de Jong J, Geilen M, Heuts PH, van Breukelen G. 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
  21. Vlaeyen JW, de Jong J, Geilen M, Heuts PH, van Breukelen G. 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
  22. Woby SR, Watson PJ, Roach NK, Urmston M. Are changes in fear-avoidance beliefs, catastrophizing, and appraisals of control, predictive of changes in chronic low back pain and disability?. Eur J Pain. 2004;8:201–210
  23. Woby SR, Watson PJ, Roach NK, Urmston M. Adjustment to chronic low back pain–the relative influence of fear-avoidance beliefs, catastrophizing, and appraisals of control. Behav Res Ther. 2004;42:761–774
  24. Brennan GP, Fritz JM, Hunter SJ, Thackeray A, Delitto A, Erhard RE. Identifying subgroups of patients with acute/subacute “nonspecific” low back pain: results of a randomized clinical trial. Spine. 2006;31:623–631
  25. McKenzie R, May S. The lumbar spine: mechanical diagnosis and therapy. 2nd ed. Waikanae: Spinal Publication; 2003;
  26. Werneke M, Hart DL. Centralization phenomenon as a prognostic factor for chronic low back pain and disability. Spine. 2001;26:758–764
  27. Werneke MW, Hart DL. Categorizing patients with occupational low back pain by use of the Quebec Task Force Classification system versus pain pattern classification procedures: discriminant and predictive validity. Phys Ther. 2004;84:243–254
  28. Aina A, May S, Clare H. The centralization phenomenon of spinal symptoms–a systematic review [erratum appears in J Orthop Sports Phys Ther 2007;37:769]. Man Ther. 2004;9:134–143
  29. Fritz JM, Cleland JA, Childs JD. Subgrouping patients with low back pain: evolution of a classification approach to physical therapy. J Orthop Sports Phys Ther. 2007;37:290–302
  30. George SZ, Bialosky JE, Donald DA. The centralization phenomenon and fear-avoidance beliefs as prognostic factors for acute low back pain: a preliminary investigation involving patients classified for specific exercise. J Orthop Sports Phys Ther. 2005;35:580–588
  31. Werneke M, Hart DL, Cook D. A descriptive study of the centralization phenomenon (A prospective analysis). Spine. 1999;24:676–683
  32. Werneke MW, Hart DL, Resnik L, Stratford PW, Reyes A. Centralization: prevalence and effect on treatment outcomes using a standardized operational definition and measurement method. J Orthop Sports Phys Ther. 2008;38:116–125
  33. Swinkels IC, van den Ende CH, de Bakker D, et al. Clinical databases in physical therapy. Physiother Theory Pract. 2007;23:153–167
  34. Deutscher D, Hart DL, Dickstein R, Horn SD, Gutvirtz M. Implementing an integrated electronic outcomes and electronic health record process to create a foundation for clinical practice improvement. Phys Ther. 2008;88:270–285
  35. Waddell G, Newton M, Henderson I, Somerville D, Main CJ. A Fear-Avoidance Beliefs Questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability. Pain. 1993;52:157–168
  36. Dionne CE. Psychological distress confirmed as predictor of long-term back-related functional limitations in primary care settings. J Clin Epidemiol. 2005;58:714–718
  37. Dionne CE, Koepsell TD, Von Korff M, Deyo RA, Barlow WE, Checkoway H. Predicting long-term functional limitations among back pain patients in primary care settings. J Clin Epidemiol. 1997;50:31–43
  38. Williamson E. Fear Avoidance Beliefs Questionnaire (FABQ). Aust J Physiother. 2006;52:149
  39. George SZ. Fear: a factor to consider in musculoskeletal rehabilitation. J Orthop Sports Phys Ther. 2006;36:264–266
  40. George SZ, Fritz JM, Childs JD. Investigation of elevated fear-avoidance beliefs for patients with low back pain: a secondary analysis involving patients enrolled in physical therapy clinical trials. J Orthop Sports Phys Ther. 2008;38:50–58
  41. Schmidt I, Rechter L, Hansen VK, Andreasen J, Overvad K. Prognosis of subacute low back pain patients according to pain response. Eur Spine J. 2008;17:57–63
  42. de Jong JR, Vlaeyen JW, Onghena P, Goossens ME, Geilen M, Mulder H. Fear of movement/(re)injury in chronic low back pain: education or exposure in vivo as mediator to fear reduction?. Clin J Pain. 2005;21:9–17
  43. Chou R, Huffman LH. Nonpharmacologic therapies for acute and chronic low back pain: a review of the evidence for an American Pain Society/American College of Physicians clinical practice guideline [erratum appears in Ann Intern Med 2008;148:247-8]. Ann Intern Med. 2007;147:492–504
  44. Hart D, Connolly J. Pay-for-performance for physical therapy and occupational therapy: Medicare Part B Services. Grant #: 18-P-93066/9-01: Health & Human Services/Centers for Medicare & Medicaid Services. 2006;
  45. In:  Wainer H editors. Computerized adaptive testing (A primer). 2nd ed. Mahway: Lawrence Erlbaum Associates; 2000;
  46. Wright B. Rating scale analysis. Chicago: Mesa Pr; 1982;
  47. Mallinson T, Stelmack J, Velozo C. A comparison of the separation ratio and coefficient alpha in the creation of minimum item sets. Med Care. 2004;42(1 Suppl):I17–I24
  48. Childs JD, Piva SR, Fritz JM. Responsiveness of the numeric pain rating scale in patients with low back pain. Spine. 2005;30:1331–1334
  49. Jensen MP, Turner JA, Romano JM, Fisher LD. Comparative reliability and validity of chronic pain intensity measures. Pain. 1999;83:157–162
  50. Beaton DE, Bombardier C, Katz JN, Wright JG. A taxonomy for responsiveness. J Clin Epidemiol. 2001;54:1204–1217
  51. Crosby RD, Kolotkin RL, Williams GR. Defining clinically meaningful change in health-related quality of life. J Clin Epidemiol. 2003;56:395–407
  52. Deyo RA, Centor RM. Assessing the responsiveness of functional scales to clinical change: an analogy to diagnostic test performance. J Chronic Dis. 1986;39:897–906
  53. Hart DL, Wang YC, Stratford PW, Mioduski JE. Computerized adaptive test for patients with knee impairments produced valid and responsive measures of function. J Clin Epidemiol. 2008;61:1113–1124
  54. Hsieh YW, Wang CH, Wu SC, Chen PC, Sheu CF, Hsieh CL. Establishing the minimal clinically important difference of the Barthel Index in stroke patients. Neurorehabil Neural Repair. 2007;21:233–238
  55. Jaeschke R, Singer J, Guyatt GH. Measurement of health status (Ascertaining the minimal clinically important difference). Control Clin Trials. 1989;10:407–415
  56. Lingard EA, Riddle DL. Impact of psychological distress on pain and function following knee arthroplasty. J Bone Joint Surg Am. 2007;89:1161–1169
  57. Stratford PW, Binkley FM, Riddle DL. Health status measures: strategies and analytic methods for assessing change scores. Phys Ther. 1996;76:1109–1123
  58. Stratford PW, Binkley J, Solomon P, Gill C, Finch E. Assessing change over time in patients with low back pain. Phys Ther. 1994;74:528–533
  59. Tubach F, Ravaud P, Baron G, et al. Evaluation of clinically relevant changes in patient reported outcomes in knee and hip osteoarthritis: the minimal clinically important improvement. Ann Rheum Dis. 2005;64:29–33
  60. Tubach F, Ravaud P, Beaton D, et al. Minimal clinically important improvement and patient acceptable symptom state for subjective outcome measures in rheumatic disorders. J Rheumatol. 2007;34:1188–1193
  61. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology. 1982;143:29–36
  62. Thomas JS, France CR. Pain-related fear is associated with avoidance of spinal motion during recovery from low back pain. Spine. 2007;32:E460–E466
  63. Swinkels-Meewisse IE, Roelofs J, Oostendorp RA, Verbeek AL, Vlaeyen JW. Acute low back pain: pain-related fear and pain catastrophizing influence physical performance and perceived disability. Pain. 2006;120:36–43
  64. Crombez G, Vlaeyen JW, Heuts PH, Lysens R. Pain-related fear is more disabling than pain itself: evidence on the role of pain-related fear in chronic back pain disability. Pain. 1999;80:329–339
  65. Geisser ME, Haig AJ, Wallbom AS, Wiggert EA. Pain-related fear, lumbar flexion, and dynamic EMG among persons with chronic musculoskeletal low back pain. Clin J Pain. 2004;20:61–69
  66. George SZ, Fritz JM, McNeil DW. Fear-avoidance beliefs as measured by the fear-avoidance beliefs questionnaire: change in fear-avoidance beliefs questionnaire is predictive of change in self-report of disability and pain intensity for patients with acute low back pain. Clin J Pain. 2006;22:197–203
  67. Thomas JS, France CR. The relationship between pain-related fear and lumbar flexion during natural recovery from low back pain. Eur Spine J. 2008;17:97–103
  68. Donelson R, Grant W, Kamps C, Medcalf R. Pain response to sagittal end-range spinal motion (A prospective, randomized, multicentered trial). Spine. 1991;16(6 Suppl):S206–S212
  69. Wetzel FT, Donelson R. The role of repeated end-range/pain response assessment in the management of symptomatic lumbar discs. Spine J. 2003;3:146–154
  70. Vlaeyen JW, Kole-Snijders AM, Boeren RG, van Eek H. Fear of movement/(re)injury in chronic low back pain and its relation to behavioral performance. Pain. 1995;62:363–372
  71. Horn SD. Performance measures and clinical outcomes. JAMA. 2006;296:2731–2732
  72. Horn SD, DeJong G, Ryser DK, Veazie PJ, Teraoka J. Another look at observational studies in rehabilitation research: going beyond the holy grail of the randomized controlled trial. Arch Phys Med Rehabil. 2005;86(12 Suppl 2):S8–S15
  73. Resnik L, Liu D, Hart DL, Mor V. Benchmarking physical therapy clinic performances: statistical methods to enhance internal validity when using observational data. Phys Ther. 2008;88:1078–1087
  • a Focus On Therapeutic Outcomes Inc, 6832 S Northshore Dr, Knoxville,TN 37919.

 A commercial party having a direct financial interest in the results of the research supporting this article has conferred or will confer a financial benefit on the author or one or more of the authors. Dr Hart is an investor in and employee of Focus On Therapeutic Outcomes Inc. Analyses of data and reporting of results of those analyses such as those reported in this manuscript are part of his routine, daily activities.

PII: S0003-9993(09)00084-7

doi:10.1016/j.apmr.2008.11.008

Archives of Physical Medicine and Rehabilitation
Volume 90, Issue 5 , Pages 768-777, May 2009

Article Tools

* Image Usage & Resolution