Archives of Physical Medicine and Rehabilitation
Volume 86, Issue 6 , Pages 1141-1146, June 2005

Factor Structure of the Pain Disability Index in Workers’ Compensation Claimants With Low Back Injuries

  • Raymond C. Tait, PhD

      Affiliations

    • Corresponding Author InformationReprint requests to Raymond C. Tait, PhD, Dept of Psychiatry, Saint Louis University School of Medicine, 1221 S Grand Blvd, St. Louis, MO 63104
    • ,
  • John T. Chibnall, PhD

Department of Psychiatry, Saint Louis University School of Medicine, St. Louis, MO.

Article Outline

Abstract 

Tait RC, Chibnall JT. Factor structure of the Pain Disability Index in workers’ compensation claimants with low back injuries. Arch Phys Med Rehabil 2005;86:1141–6.

Objective

To examine the factor structure of a telephone-administered Pain Disability Index (PDI) and the effects of race and sex on the PDI.

Design

Computer-assisted telephone interviews of a cohort with occupational low back injuries.

Setting

General community.

Participants

Missouri workers’ compensation claimants (N=1329) with low back injuries.

Interventions

Not applicable.

Main Outcome Measures

PDI, levels of pain severity, Social Security Disability Insurance status, and the Fear-Avoidance Behavior Questionnaire.

Results

Results for the total sample and by race/sex group indicated support for a 2-factor model of the PDI corresponding to voluntary activities (eg, social, occupational, recreational) and obligatory activities (eg, activities of daily living, eating, sleeping). Additional psychometric analyses of the voluntary and obligatory subscales indicated adequate reliability and construct validity overall and in each of the race/sex groups. African Americans reported more pain-related disability on both subscales than whites. Women reported more disability on the voluntary subscale than men.

Conclusions

The results support use of the PDI as a bidimensional measure of pain-related disability, with strong psychometric properties. They also support its administration by telephone.

Key Words:  Disability evaluation , Low back pain , Occupational medicine , Rehabilitation , Workers’ compensation

 

FUNCTIONAL ABILITY has been identified as a crucial component of the assessment of any chronic pain condition. Several converging lines of evidence have led to this recognition: (1) perhaps more than any other dimension of the pain experience, interference with activity is a crucial predictor of future adjustment among patients with chronic pain 1, 2, 3; (2) return of function drives a wide range of favorable treatment outcomes, so that functional restoration has become a linchpin of treatment 4, 5, 6, 7, 8; and (3) insurance carriers, especially workers’ compensation carriers, have prioritized return of function as an outcome to which they attach value relative to other possible outcomes, such as improved quality of life. 9, 10 Despite this evidence, clinical assessment of pain-related disability often does not occur in a medical encounter, 11 putatively because of time pressures. Hence, considerable attention has been given to the development of instruments that facilitate the assessment of pain-related disability.

The clinical context often drives decisions regarding which aspects of disability should be assessed. In the context of workers’ compensation, the criterion standard for evaluating return of function involves return to work. In the context of clinical treatment, return to work represents 1 metric against which to measure function, but it represents only 1 dimension of disability. For many patients, return to work is neither realistic nor relevant. Hence, pain treatment programs have considered disability more broadly, in terms consistent with those proposed by the Institute of Medicine: “… a disadvantage for a given individual (resulting from an impairment or a functional limitation) that limits or prevents the fulfillment of a role that is normal … for that individual.” 12 (p17) Numerous self-report measures have been developed that measure pain-related disability, including the Sickness Impact Profile, 13 the Multi-Dimensional Pain Inventory, 14 the Roland scale, 15 the Oswestry Disability Questionnaire (ODQ), 16 the Neck Pain and Disability Scale (NPDS), 17 and the Pain Disability Index (PDI). 18

The PDI has had widespread use since its introduction because it is brief and has strong psychometric properties, including evidence for its validity, 19, 20, 21, 22 reliability, 23 and sensitivity to change. 24 Further, unlike some inventories that are condition specific (eg, the NPDS is designed for neck pain, the ODQ for low back pain [LBP]), the PDI was constructed for use with multiple types of pain conditions. Consequently, it has been included in studies of mixed chronic pain conditions, 25, 26 mixed acute pain conditions, 27 chronic LBP, 28, 29 painful diabetic neuropathy, 30 postherpetic neuralgia, 31 osteoporosis, 32 traumatic extremity injuries, 33 and breast cancer. 34

Despite its widespread use, questions remain about the factor structure of the PDI. Early research using exploratory principal components analysis indicated that it was composed of 2 factors, one reflecting voluntary activities (eg, social, occupational, recreational) and the other reflecting obligatory activities (eg, activities of daily living [ADLs], sleeping, eating). 19, 20 Later research (also using exploratory factor analytic methods), based on much larger samples of patients presenting for multidisciplinary pain management, indicated a 1-factor solution. 22 Research using the PDI reflects this confusion: in some articles the PDI is treated as 1-dimensional, 35 whereas in others it is discussed in terms of its voluntary and obligatory subscales. 36

This study was conducted for 4 primary reasons: (1) to examine the factor structure of the PDI using confirmatory factor analytic techniques rather than exploratory analysis; (2) to perform this analysis on a large, nonclinical but disability-relevant sample of people with chronic pain—workers’ compensation claimants; (3) to ascertain whether the structure of the PDI varies as a function of race (African American vs white) and sex; and (4) to evaluate the reliability and construct validity of the scores indicated by the confirmatory factor structure. A secondary purpose was to examine the psychometric properties of the PDI when administered by telephone.

Back to Article Outline

Methods 

Sampling and Participants 

Data for the study were extracted from a larger study of racial and socioeconomic disparities in workers’ compensation treatment and outcomes. 37 All descriptions of the study population and measures are taken from the larger study. The study population was entirely first-incident workers’ compensation claimants with low back injuries whose claims were settled in Missouri over a 17-month period (January 1, 2001–June 1, 2002) in Jackson County (Kansas City metropolitan area), St. Louis County, and the city of St. Louis (N=3181). These regions account for 82.7% of all adult African Americans in Missouri and 30.1% of the adult whites in Missouri. 38 Across all claimants, 1942 (61.0%) were injured in St. Louis County, 622 (19.6%) in St. Louis, and 617 (19.4%) in Jackson County.

The eligible population (excluding from the original total of 3181 those who did not meet the race, injury, and language criteria and those who were either deceased, institutionalized, or incarcerated) totaled 2934 claimants. Of this group, 1475 (50.3%) completed the study and 432 (14.7%) refused to participate. An additional 1027 (35.0%) could not be located, despite intensive tracing efforts. Of those who completed the study, 1073 (72.8%) were located with contact information supplied by the State of Missouri or with simple tracing to update their current telephone numbers. This group did not differ significantly from the 432 people who refused to participate on any of the workers’ compensation treatment and settlement variables available from the State of Missouri for the whole claimant population. These variables included the duration of the workers’ compensation claim date of settlement, legal representation, treatment dollars spent, temporary total disability costs, settlement award, and residual disability rating. Intensive tracing of the remaining 402 people (27.3%) who completed the study was necessary to locate current telephone numbers. This group did not differ significantly on any of the workers’ compensation variables from the 1027 claimants who could not be located. Geographically, the percentages of study participants (n=1475) and nonparticipants (n=1459) represented in the 3 study regions did not differ significantly. Hence, study respondents appeared to be generally representative of the total workers’ compensation LBP population.

Procedures and Measures 

The institutional review board (IRB) of Saint Louis University approved the study, as did the IRB of Battelle Centers for Public Health Research and Evaluation (BCPHRE) in St. Louis. After receiving IRB approval, attempts were made to contact the original pool of 3181 claimants by telephone using contact information provided by the State of Missouri. Claimants were eligible if they were self-identified as either African American or non-Hispanic white, were fluent in English, and gave informed consent. Using computer-assisted telephone interviewing, BCPHRE personnel conducted interviews of 15 to 20 minutes’ duration. Participants were paid $25.

As part of the telephone interview for the larger study, participants completed the PDI. 18, 19, 20, 22 The PDI measures pain-related interference with role functioning in 7 areas (occupational, home/family, recreational, social, sexual, ADLs, life support), all rated on 11-point Likert-type scales (0, no disability; 10, complete disability). Other measures used to evaluate the construct validity of the PDI included pain intensity, which was assessed on 11-point scales that ranged from 0 (no pain) to 10 (pain as bad as it could be). Participants rated their worst, least, and usual pain levels over the previous 7 days 39; a mean of the 3 ratings was calculated. Internal consistency reliability of the 3 pain intensity indicators was .85. Participants also completed the Medical Outcomes Study 12-Item Short-Form Health Survey (SF-12), a short version of the SF-36 that yields a standardized scale score for general physical health that has established reliability and validity in patients with pain-related disorders. 40 The Fear-Avoidance Beliefs Questionnaire (FABQ) 41 measures activity fear-avoidance beliefs about the effect of physical and work activities on pain. It comprises 16 items rated on 7-point Likert-type scales (0, agree; 6, disagree). The mean of the 16 items was calculated for the total score, with higher scores indicating stronger belief that activity should be avoided. Internal consistency reliability of the FABQ was .88. A final variable was the status of the participant regarding Social Security Disability Insurance (SSDI). Participants were dichotomized into 2 groups: those who were (n=116, 8.7%) and were not (n=1211, 91.3%) receiving SSDI.

Statistical Analysis 

One- and 2-factor confirmatory factor models of the PDI were tested using Lisrel 8.5 a for the total sample and for African Americans, whites, men, and women separately. For the 2-factor model, PDI items were divided into a voluntary subscale (items 1–5: family/home, social, recreational, occupational, sexual activities) and an obligatory subscale (items 6 and 7: independent daily activities such as hygiene, driving, dressing; basic life-support behaviors such as eating, sleeping, breathing). Maximum likelihood estimation was used. Model fit was evaluated using the chi-square goodness-of-fit index (minimum fit function), the ratio of the chi-square value to the degrees of freedom, the root mean square error of approximation (RMSEA), and the cumulative fit index (CFI). For confirmatory models, chi-square values (and the ratio of this value to the degrees of freedom) that are closer to zero indicate better fit. RMSEA values of .08 or less are considered evidence of adequate fit, 42 and a CFI greater than .90 is desirable. Differences in fit between the 2 models were estimated using a chi-square difference test. PDI scores were compared as a function of race and sex using analysis of variance (ANOVA). Effect size was estimated using the Cohen d statistic (ratio of mean difference to pooled standard deviation [SD]). Conventionally, a Cohen d of 0.2 indicates a small effect, 0.5 indicates a moderate effect, and 0.8 indicates a large effect. Multiple linear regression with simultaneous entry of predictors was used to evaluate the construct validity of the PDI.

Back to Article Outline

Results 

Sample Description 

Complete PDI data were available for 1329 (90.1%) of the 1475 respondents. Of the sample, 797 (60.0%) were white, 532 (40.0%) were African American, 826 (62.2%) were male, and 503 (37.8%) were female. Participants averaged 13.1±2.3 years of education. As a result of the low back injury, 277 (20.8%) of the participants had undergone at least 1 low back surgery. At the time of data collection, 839 participants (63.1%) were working full time, 136 (10.2%) were working part time, and 354 (26.6%) were unemployed. In age, 304 (22.9%) were between 18 and 34 years, 828 (62.4%) were between 35 and 54, and 195 (14.7%) were 55 or older. Annual household income was less than $20,000 for 548 participants (35.0%), between $20,000 and $49,999 for 544 (41.6%), and $50,000 or more for 307 (23.5%). Participants were, on average, 21.4±12.4 months postsettlement of their workers’ compensation claim at the time of data collection.

Confirmatory Factor Analyses 

As shown in table 1, the 2-factor models of the PDI (voluntary vs obligatory) had a better fit to the data than the 1-factor models for the total sample and as a function of race and sex. The increase in fit from 1 factor to 2 factors was especially pronounced in the total sample, in whites, and in men. Internal consistency reliabilities were adequate for the 2-factor models, ranging from .78 to .93.

Table 1. Confirmatory Factor Analysis Results: 1- and 2-Factor Models of the PDI
Sampleχ2 (df)χ2/dfRMSEACFIχ2 Diff (df)Internal Consistency Reliability
Factor 1Factor 2
Total (N=1329)
1-factor model177(14)12.6.100.99 .93
2-factor model71(13)5.5.061.00106(1)⁎.92.80
White (n=797)
1-factor model137(14)9.8.100.98 .93
2-factor model62(13)4.8.070.9975(1)⁎.92.79
African American (n=532)
1-factor model61(14)4.4.080.99 .93
2-factor model31(13)2.4.061.0030(1)⁎.92.79
Male (n=826)
1-factor model146(14)10.4.110.98 .93
2-factor model57(13)4.4.060.9989(1)⁎.93.81
Female (n=503)
1-factor model57(14)4.1.080.99 .93
2-factor model35(13)2.7.061.0022(1)⁎.91.78

Abbreviation: Diff, difference.

⁎ P<.001.

Descriptive Data 

Summary scores for the voluntary and obligatory items, consistent with the confirmatory factor models, were calculated. The voluntary score consisted of PDI items 1 to 5 (range, 0–50); the mean of this subscale in the total sample was 24.0±13.5, with a mean item score of 4.8. The obligatory score consisted of PDI items 6 and 7 (range, 0–20); the mean in the total sample was 6.9±5.3, with a mean item score of 3.5. The correlation of the voluntary and obligatory subscales was .78 in the total sample, .75 in whites, .79 in African Americans, .77 in men, and .79 in women (all P<.001). As shown in table 2, race by sex ANOVAs indicated significant main effects for race for the voluntary subscale (F1,1325=31.9, P<.001) and for the obligatory subscale (F1,1325=47.1, P<.001), with African Americans reporting more disability than whites. The Cohen d for the race effect was in the small-to-moderate range: .36 for the voluntary subscale and .43 for the obligatory subscale. A main effect for sex emerged only for the voluntary subscale (F1,1325=6.0, P<.05), with women reporting more disability than men. The Cohen d was .23 for this effect. There were no significant interactions.

Table 2. PDI Descriptive Data by Race and Sex
VariablesMen (n=826)Women (n=503)Race Main Effect⁎
PDI voluntary subscale†
White (n=797)21.5±13.323.3±13.322.0±13.3
African American (n=532)25.8±13.327.9±13.226.9±13.3
Sex main effect‡22.8±13.525.9±13.4
PDI obligatory subscale§
White (n=797)5.9±4.96.4±5.26.0±5.0
African American (n=532)7.9±5.58.6±5.78.3±5.6
Sex main effect6.5±5.27.6±5.5

NOTE. Values are mean ± SD.

⁎ P<.001, voluntary and obligatory subscales.

† Subscale score is the sum of scores for items 1 to 5.

‡ P<.05, voluntary subscale.

§ Subscale score is the sum of scores for items 6 and 7.

Construct Validity 

Using multiple linear regression with simultaneous entry of predictors, the construct validity of the PDI voluntary and obligatory subscales was evaluated by predicting these scores from measures of general physical health (SF-12 physical), disability (SSDI), fear-avoidance beliefs (FABQ), and pain intensity. As shown in table 3, from 58% to 62% of the variance in the PDI voluntary subscale could be explained by the 4 predictors, with all predictors contributing significantly to the equation. This pattern was consistent across race and sex groups. As shown in table 4, the 4 predictors explained from 44% to 48% of the variance in the PDI obligatory subscale. All 4 predictors were significant for the total sample and for men. For whites, African Americans, and women, only SSDI failed to significantly predict the obligatory subscale.

Table 3. Multiple Regression Equations: Construct Validity of PDI Voluntary Subscale
PredictorsStandardized Regression Coefficients by Sample
TotalWhiteAfrican AmericanMenWomen
SF-12 physical health−.28⁎−.32⁎−.26⁎−.23⁎−.36⁎
SSDI (0=no; 1=yes).08⁎.08†.09‡.08†.09†
FABQ (fear avoidance).23⁎.24⁎.23⁎.22⁎.24⁎
Pain intensity.43⁎.39⁎.42⁎.47⁎.36⁎
Total equation R2.62⁎.63⁎.58⁎.62⁎.62⁎

⁎ P<.001.

† P<.01.

‡ P<.05.

Table 4. Multiple Regression Equations: Construct Validity of PDI Obligatory Subscale
PredictorsStandardized Regression Coefficients by Sample
TotalWhiteAfrican AmericanMenWomen
SF-12 physical health−.24⁎−.24⁎−.26⁎−.18⁎−.31⁎
SSDI (0=no; 1=yes).05†.06.06.07†.05
FABQ (fear avoidance).19⁎.22⁎.18⁎.18⁎.21⁎
Pain intensity.40⁎.33⁎.39⁎.44⁎.32⁎
Total equation R2.47⁎.44⁎.48⁎.47⁎.47⁎

† P<.05.

⁎ P<.001.

Back to Article Outline

Discussion 

Our results support a 2-factor structure for the PDI in people with chronic pain related to occupational low back injury. Not only did the pattern obtain to the total sample, but it also applied across race and sex subgroups. The factors representing voluntary and obligatory activities also demonstrated levels of internal consistency reliability that ranged from acceptable to excellent, which is particularly notable in light of the brevity of the obligatory subscale (2 items). Finally, associations between the subscales and measures of pain, disability, activity avoidance, and general physical health supported the construct validity of the subscales. A secondary consequence of these results is that the methods of administering the PDI can be expanded to include telephone administration, a potential advantage in prospective studies using telephone follow-up or community-based studies of pain-related disability. These results provide strong support, together with the literature on the PDI, for the usefulness of the PDI in outcomes research for pain.

Given the strong psychometric properties of the PDI, its bidimensional measurement of disability, its brevity, and its ease of administration, it is also useful in clinical settings. As a screening tool, the PDI can be used to identify patients reporting moderate to high levels of pain-related disability, who could then be targeted for more detailed clinical assessment to identify disability-augmenting factors, including psychologic and psychosocial issues. 43, 44, 45 As a tracking measure, the PDI could be used to facilitate evidence-based medical decision making. If administered routinely (either in the office or by telephone) to patients recovering from a painful injury, deviations in progress (eg, a slower-than-anticipated recovery) could be readily identified. Patients with self-limiting injuries could be easily discriminated from those with problematic recoveries, based on changes in PDI levels. Patients exhibiting elevated PDI levels after a reasonable recovery time could be flagged to receive intensive, functionally oriented treatment. This balanced approach could avoid costly overtreatment while delivering more aggressive treatment without delays that could impede good outcomes. 46, 47

The primary limitation of the current findings is generalizability. Our respondents were workers’ compensation claimants who had settled their claims for low back injury, so the results may not generalize to clinical pain populations or patients with other types of pain. Further, the study was conducted in Missouri, where workers’ compensation claimants cannot choose their initial treatment providers. The extent to which the results generalize to other states where claimants can do this cannot be assumed. Respondents were also either African American or white, so our results may not apply to other racial or ethnic groups, including Asians and Hispanics. Finally, the study respondents could have differed from the nonrespondents on variables that we did not measure or to which we did not have access. Thus, future research on the PDI in other populations and with other racial and ethnic groups is clearly needed to validate the current findings.

Back to Article Outline

Conclusions 

We recommend that the PDI be scored as 2 separate, correlated subscales composed of items 1 to 5 for the voluntary subscale and items 6 and 7 for the obligatory subscale. Research is needed to examine the differential validity of the 2 subscales and their respective sensitivities to treatment interventions. In addition, our own future research will be aimed at increasing the number of items on the obligatory subscale in order to enhance its validity and reliability.

Supplier

Back to Article Outline

Acknowledgment 

We are indebted to Richard Stickann and Lawrence D. Leip of the Missouri Division of Workers’ Compensation for their invaluable assistance with this project.

Back to Article Outline

References 

  1. Engel CC , Von Korff M , Katon WJ . Back pain in primary care (predictors of high health-care costs) . Pain . 1996;65:197–204
  2. Von Korff M , Ormel J , Keefe FJ , Dworkin SF . Grading the severity of chronic pain . Pain . 1992;50:133–149
  3. Von Korff M , Deyo RA , Cherkin D , Barlow W . Back pain in primary care (outcomes at 1 year) . Spine . 1993;18:855–862
  4. Hazard RG . Spine update (functional restoration) . Spine . 1995;20:2345–2348
  5. Mayer TG , Gatchel RJ . Functional restoration for spinal disorders (the sports medicine approach) . Philadelphia: Lea & Febiger; 1988;
  6. Mitchell RI , Carmen GM . The functional restoration approach to the treatment of chronic pain in patients with soft tissue and back injuries . Spine . 1994;19:633–642
  7. Rainville J , Ahern DK , Phalen L , Childs LA , Sutherland R . The association of pain with physical activities in chronic low back pain . Spine . 1992;17:1060–1064
  8. Wright A , Mayer TG , Gatchel RJ . Outcomes of disabling cervical spine disorders in compensation. A prospective comparison to tertiary rehabilitation response for chronic lumbar spinal disorders . Spine . 1999;24:178–183
  9. Flor H , Fydrich T , Turk DC . Efficacy of multidisciplinary pain treatment centers (a meta-analytic review) . Pain . 1992;49:221–230
  10. Matheson LN , Brophy RG , Vaughan KD , Nunez C , Saccoman KA . Workers’ compensation managed care (preliminary findings) . J Occup Rehabil . 1995;5:27–36
  11. Turner JA , LeResche L , Von Korff M , Ehrlich K . Back pain in primary care (patient characteristics, content of initial visit, and short-term outcomes) . Spine . 1998;23:463–469
  12. In:  Osterweis M ,  Kleinman A ,  Mechanic D editor. Pain and disability (clinical, behavioral, and public policy perspectives) . Washington (DC): Natl Acad Pr; 1987;
  13. Bergner M , Bobbitt RA , Carter WB , Gilson BS . The Sickness Impact Profile (development and final revision of a health status measure) . Med Care . 1981;19:787–805
  14. Kerns RD , Turk DC , Rudy TE . The West Haven-Yale Multidimensional Pain Inventory (WHYMPI) . Pain . 1985;23:345–356
  15. Jensen MP , Strom SE , Turner JA , Romano JM . Validity of the Sickness Impact Profile Roland scale as a measure of dysfunction in chronic pain patients . Pain . 1992;50:157–162
  16. Fairbank JC , Couper J , Davies JB , O’Brien JP . The Oswestry Low Back Disability Questionnaire . Physiotherapy . 1980;66:271–273
  17. Wheeler AH , Goolkasian P , Baird AC , Darden BV . Development of the Neck Pain and Disability Scale (item analysis, face, and criterion-related validity) . Spine . 1999;13:1290–1294
  18. Pollard CA . Preliminary validity study of the Pain Disability Index . Percept Mot Skills . 1984;59:974
  19. Tait RC , Pollard CA , Margolis RB , Duckro PN , Krause SJ . The Pain Disability Index (psychometric and validity data) . Arch Phys Med Rehabil . 1987;68:438–441
  20. Tait RC , Chibnall JT , Krause SJ . The Pain Disability Index (psychometric properties) . Pain . 1990;40:171–182
  21. Jerome A , Gross RT . Pain Disability Index (construct and discriminant validity) . Arch Phys Med Rehabil . 1991;72:920–922
  22. Chibnall JT , Tait RC . The Pain Disability Index (factor structure and normative data) . Arch Phys Med Rehabil . 1994;75:1082–1086
  23. Gronblad M , Hupli M , Wennerstrand P , et al.   Intercorrelation and test-retest reliability of the Pain Disability Index (PDI) and the Oswestry Disability Questionnaire (ODQ) and their correlation with pain intensity in low back pain patients . Clin J Pain . 1993;9:189–195
  24. Strong J , Ashton R , Large RG . Function and the patient with chronic low back pain . Clin J Pain . 1994;10:191–196
  25. Geisser ME , Roth RS , Theisen ME , Robinson ME , Riley JL . Negative affect, self-report of depressive symptoms, and clinical depression (relation to the experience of chronic pain) . Clin J Pain . 2000;16:110–120
  26. Moulin DE , Iezzi A , Amireh R , Sharpe WK , Boyd D , Merskey H . Randomised trial of oral morphine for chronic non-cancer pain . Lancet . 1996;347:143–147
  27. Unruh AM , Ritchie J , Merskey H . Does gender affect appraisal of pain and pain coping strategies? . Clin J Pain . 1999;15:31–40
  28. Simpson RK , Edmondson EA , Constant CF , Collier C . Transdermal fentanyl as treatment for chronic low back pain . J Pain Symptom Manage . 1997;14:218–224
  29. Rittweger J , Just K , Kautzsch K , Reeg P , Felsenberg D . Treatment of chronic lower back pain with lumbar extension and whole-body vibration exercise . Spine . 2002;27:1829–1834
  30. Eisenberg E , Lurie Y , Braker C , Daoud D , Ishay A . Lamotrigine reduces painful diabetic neuropathy (a randomized, controlled study) . Neurology . 2001;57:505–509
  31. Harke H , Gretenkort P , Ulrich Ladleif H , Koester P , Rahman S . Spinal cord stimulation in postherpetic neuralgia and in acute herpes zoster pain . Anesth Analg . 2002;94:694–700
  32. Kerschan K , Alacamlioglu Y , Kollmitzer J , et al.   Functional impact of unvarying exercise program in women after menopause . Am J Phys Med Rehabil . 1998;77:326–332
  33. Durham RM , Mistry BM , Mazuski JE , Shapiro M , Jacobs D . Outcome and utility of scoring systems in the management of the mangled extremity . Am J Surg . 1996;172:569–574
  34. Bishop SR , Warr D . Coping, catastrophizing and chronic pain in breast cancer . J Behav Med . 2003;26:265–281
  35. Roth RS , Geisser ME . Educational achievement and chronic pain disability (mediating role of pain-related cognitions) . Clin J Pain . 2002;18:286–296
  36. Gronblad M , Hurri H , Kouri JP . Relationships between spinal mobility, physical performance tests, pain intensity and disability in chronic low back pain patients . Scand J Rehabil Med . 1997;29:17–24
  37. Tait RC , Chibnall JT , Andresen EA , Hadler NM . Racial disparities in the management of occupational back injuries (differences among African Americans and Caucasians) . Pain . 2004;112:389–396
  38. US Census Bureau. Census 2000. Available at: http://www.factfinder.census.gov. Accessed August 27, 2004.
  39. Jensen MP , Turner LR , Turner JA , Romano JM . The use of multiple-item scales for pain intensity measurement in chronic pain patients . Pain . 1996;67:35–40
  40. Gandhi SK , Salmon JW , Zhao SZ , Lambert BL , Gore PR , Conrad K . Psychometric evaluation of the 12-item short-form health survey (SF-12) in osteoarthritis and rheumatoid arthritis clinical trials . Clin Ther . 2001;23:1080–1098
  41. 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
  42. Browne MW , Cudeck R . Alternative ways of assessing model fit . In:  Bollen KA ,  Long JS editor. Testing structural equation models . Newbury Park: Sage; 1993;p. 136–162
  43. Gatchel RJ , Polatin PB , Mayer TG . The dominant role of psychosocial risk factors in the development of chronic low back pain disability . Spine . 1995;24:2702–2709
  44. Linton S . A review of psychological risk factors in back and neck pain . Spine . 2000;25:1148–1156
  45. Sterling M , Kenardy J , Jull G , Vicenzino B . The development of psychological changes following whiplash injury . Pain . 2003;106:481–489
  46. Frank JW , Brooker AS , DeMaio SE , et al.   Disability resulting from occupational low back pain. Part II: What do we know about secondary prevention?. A review of the scientific evidence on prevention after disability begins . Spine . 1996;21:2918–2929
  47. Gallagher RM , Myers P . Referral delay in back pain patients on worker’s compensation (costs and policy implications) . Psychosomatics . 1996;37:270–284
  • a SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606

 Supported by the Agency for Healthcare Research and Quality (grant no. R01 HS014007-01).No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated.

PII: S0003-9993(05)00080-8

doi:10.1016/j.apmr.2004.11.030

Archives of Physical Medicine and Rehabilitation
Volume 86, Issue 6 , Pages 1141-1146, June 2005

Article Tools