FAILURE TO COMPLETE A FUNCTIONAL RESTORATION PROGRAM
FOR CHRONIC MUSCULOSKELETAL DISORDERS:
A PROSPECTIVE ONE-YEAR OUTCOME STUDY
Timothy J. Proctor, Ph.D.*
Tom G. Mayer, M.D.**
Brian Theodore, Ph.D. Candidate*
Robert J. Gatchel, Ph.D.***
*PRIDE Research Foundation, Dallas
**Department of Orthopedic Surgery, University of Texas Southwestern Medical Center, Dallas, Texas
***Department of Psychology, College of Science, University of Texas at Arlington, Arlington, TX; Supported in part by Grants 2R01 MH46452, 2R02 DE10713 and 2K05 MH0 1107 from the National Institute of Health
Corresponding Author:
Robert J. Gatchel, Ph.D., ABPP
Department of Psychology, College of Science University of Texas at Arlington
313 Life Science Bldg., Box 19528
Arlington, TX 76019-0528
Phone: (817) 272-2541
Fax: (817) 272-2364
E-mail: gatchel@uta.edu 
RUNNING HEAD: Functional Restoration Outcomes
ABSTRACT
Objective: Chronic disabling occupational musculoskeletal disorders (CDOMD) represent the highest medical/indemnity costs and greatest loss of productivity in the universe of occupational injury claims. Although a few studies have attempted to analyze those who fail to complete a “full dose” of rehabilitation, relative to those who do, these studies suffer from small sample size and limited assessment of outcome variables. This is the first large-scale study that comprehensively compares the likelihood of various socioeconomically relevant outcomes between functional restoration completers and non-completers, while simultaneously identifying a number of risk factors for non-completion.
Design: A prospective cohort study of consecutive CDOMD patients (n=1,440). They were divided into two groups based on program completion status. The Non-Completer group (NC; n=303) did not complete the prescribed treatment program, while the Completers (C; n=1,1 37) achieved program completion.
Setting: Chronic pain management facility.
Patients: CDOMD patients.
Intervention: Interdisciplinary Functional Restoration: Rehabilitation program.
Main Outcome Measures: Validated questionnaires of pain, disability and depression were added to results of a structured one-year post-treatment telephone interview on socioeconomic outcomes covering the dimensions of work status, health utilization, recurrent injury claims and resolution of financial disputes.
Results: The one-year post-treatment socioeconomic outcomes were most striking. The NC group was 7 times more likely to have post-rehab surgery to the same area, and nearly 7 times more likely to have greater than 30 visits to a new health provider in persistent healthcare seeking efforts. The NC group also had only half the rates of work return and work retention, being 9.7 times less likely to have returned to any type of work, and 7 times less likely to have retained work at the end of the year. Regression
analysis also revealed that work return, surgery to a compensable, injured area, more health care utilization from a new provider and more overall health care utilization (greater than 30 visits) were the most reliably predicted by rehabilitation completion status.
Conclusions: This large prospective study determined that non-completers of interdisciplinary tertiary rehabilitation for CDOMDs had comparatively poor socioeconomic outcomes in the year following treatment discharge, especially on work status and health utilization outcomes. These outcomes are of great relevance to societal medical/indemnity costs and future worker productivity. A number of risk factors of potential importance in identifying possible non-completers early in the treatment course were identified that may yield more effective interventions tailored to maintain compliance and decrease the drop-out percentage.
Key Words: functional restoration; tertiary rehabilitation program; interdisciplinary chronic pain management; chronic occupational musculoskeletal disorders; socioeconomic outcomes; compliance; risk factors
INTRODUCTION
The high cost and prevalence of chronic disabling occupational musculoskeletal disorders (CDOMD) in industrialized countries is well established, and disability rates have been increasing over the past 20 years. 1,2 It is estimated that $27 billion is spent annually for diagnosis and care of musculoskeletal trauma, and this figure does not include other associated costs such as legal and dispute friction costs, vocational retraining, occupational modifications, Social Security disability payments and lost worker productivity. 3 Further, indirect costs associated with musculoskeletal disorders have been estimated to account for over $800 billion in expenditures, a cost up to 14 times greater than direct costs.4-7 Up to 85% of the adult workforce will miss time and seek professional care for musculoskeletal pain during the course of their careers. 8 For most individuals who experience musculoskeletal pain, symptoms resolve rapidly, allowing work return with little lost time. In the case of a small number of individuals, however, the pain develops into chronic pain with associated occupational disability. 9
Functional restoration, as originally developed by Mayer and Gatchel, 10 is a medically-directed, interdisciplinary pain management approach geared specifically for patients who develop CDOMD. This program places a strong emphasis on function, and combines quantitatively-directed exercise progression with disability management and psychosocial interventions such as individual and group therapy. Functional restoration is well-studied, and a wide body of research indicates a high rate of success as measured by socioeconomic outcome variables relevant to cost and worker productivity in occupational settings. These outcomes encompass work return, work retention, recurrent injury rates, new surgery rates, case settlement status, and additional healthcare utilization. 11-18 Indeed, in an earlier study, 16 comparably positive outcomes were found for work-related upper extremity disorders and spinal disorders when patients were treated with functional restoration. Replication studies, including a recent randomized
controlled trial, demonstrated the generalizability and effectiveness of the functional restoration programs in a variety of settings. 19-21
Failing to complete an effective tertiary rehabilitation program may represent a catastrophic terminal event in medical care for CDOM D leading to a determination of total and permanent disability. In a large number of these cases, failure to resolve productivity issues under workers’ compensation results in cost shifting of the disability problems from state workers’ compensation jurisdictions to the federal government within the United States. Social Security Disability Income (SSDI) and Supplemental Security Income (SSI) have seen remarkable growth in costs from $40 to $100 billion (now 5% of U.S. federal budget) per year over the past 15 years after stability through the 1980s.22 At the same time, the number of pre-retirement disabled workers on SSDI/SSI has more than doubled from 6 to 13 million workers, now representing a higher percentage of the potential U.S. workforce than the unemployment rate! While the percentage of acceptances for SSD I/SSI for musculoskeletal disorders has dropped over the past 20 years, there has been a corresponding increase in the number accepted for psychiatric conditions (depression, post-traumatic stress, anxiety, etc.), with a large number of CDOMDs persisting in disability creating eligibility based on both types of diagnoses. As such, tertiary rehabilitation represents tertiary prevention for younger working-age patients becoming lost to the workforce, with potentially catastrophic costs to the entire national retirement, social and economic systems.
While the effectiveness of functional restoration for CDOMD patients is well-examined, few studies have evaluated patients who fail to complete functional restoration relative to program completers. The few studies that have addressed this issue have faced a number of limitations, including very small sample sizes, the exploration of a limited number of variables, and the inclusion of only chronic low back pain patients. Outcomes of work status, health utilization and recurrent injury were assessed in several studies, but with sample sizes too small to draw conclusion on the degree of difference between the
groups, or to investigate a non-spinal disorder population. 11,12, 19 A number of physical and psychosocial variables have been studied, with differences in pre- and post-treatment performance for functional restoration completers and non-completers compared in very
23-25
small sample size studies.	Physical variables assessed involved measurements of
trunk mobility and strength, along with aerobic capacity and various isokinetic and isoinertial lifting tests. Psychosocial studies examined changes on the SF-36 health status inventory and the Structured Clinical Interview for DSM-IV diagnosis (SCID).
However, these variables were examined in separate studies. A comprehensive evaluation of all these physical and psychosocial measurements, now extended to a more generalized musculoskeletal disorder population, was the focus of the present large-scale study.
The purpose of the present study was to comprehensively compare functional restoration non-completers to completers on the broadest array of demographic, work adjustment/outlook, physical, psychological and objective one-year outcome variables yet attempted in the scientific literature. As noted, previous studies addressing this issue were associated with a number of limitations. The current study, which utilized a large, diverse, and temporally consistent sample of CDOMD patients sought to replicate the findings of previous small sample-size research evaluating limited variables, while also adding to the current body of knowledge by addressing previously unexamined variables.
MATERIALS AND METHODS
Subjects 
Subjects consisted of 1,440 consecutive, prospectively assessed CDOMD patients who consented to, and started, a prescribed course of functional restoration treatment from January 1996 to March 2000. All patients in the study cohort received extensive prior treatment for their injury; however, disability had not resolved at the time of admission to the tertiary rehabilitation program averaging 19 months since injury (Table
1). The participation criteria that had to be met prior to entering this program were: (1) more than 4 months elapsed since a work-related injury; (2) primary and secondary nonoperative care failed to resolve disability and pain; (3) surgery had not produced relief, resolution, or simply was not an option; (4) severe functional limitations remained; and (5) English or Spanish-speaking. This cohort of subjects was divided into the following two groups for purposes of this study:
Non-completer Group (NC). The 303 subjects (21% of entire cohort) in this group began, but did not complete, the functional restoration program. The most frequent primary causes of non-completion were treatment non-compliance (43.6%), treatment refusal (16.5%), failure to develop a work plan (10.9%), failure to progress physically during rehabilitation (6.3%), secondary gain issues (4.3%) or continued substance abuse (2.3%). These categories accounted for 83.5% of NC patients. Almost half of the remaining 15. 1 % left the program to either return towork (5.3 %) or because they were faced with financial issues (2.3%). Similarly, a small group of subjects left the program for reasons related to quality of life (3 %), for other medical problems (1%) or denial of insurance pre-certification (3%).,
Completer Group (C). The 1,137 subjects in this group successfully completed the functional restoration treatment program (79% of the total cohort).
Procedure 
After signing an IRB-approved informed consent, all patients received an initial evaluation consisting of a medical history, physical examination, psychological intake interview, medical case management disability assessment interview, and a quantitative functional capacity evaluation. The treatment program consisted of quantitatively-directed exercise progression, supervised by physical and occupational therapists, in
conjunction with components of a multimodal disability management approach, which included individual counseling, group therapeutics, stress management, biofeedback, coping skills training, and education focusing on disability management, vocational reintegration, and future fitness maintenance. 10,26,27
Demographic information was derived from the intake interviews noted above. A number of psychosocial instruments were administered at the initial interview and prior to the beginning of treatment, including: the Quantified Pain Drawing, which incorporates a visual analog scale (VAS) for self-report of perceived pain intensity; 28 the Million Visual Analogue Scale (MVAS),27,29 which is a visual analog questionnaire of disability; and the Beck Depression Inventory (BDI) .30 Finally, as earlier reviewed in the Dersh study,25 the SCID-NP31 was administered during the first week of the treatment program.
A structured telephone interview was administered at one-year post-discharge, and assessed outcome data including health utilization, recurrent injury, unresolved litigation, and work status. Multiple attempts were made to contact all patients at one year; however, it was not possible to collect outcomes on every individual. For Group C, at least partial outcome data were collected on 1,067 of the 1, 13 7 subjects (94%). For Group NC, at least partial data were collected on 257 of the 303 subjects (85%).
For the purpose of the present study, statistical analyses of group differences involving percentages were performed using a logistic regression, controlling for the potentially confounding effects of race, years of education, comorbid health conditions and smoking rate. Analysis of covariance (ANCOVA) was utilized with continuous data again controlling for potentially confounding effects of the same variables. A stepwise regression was performed on the data for one-year socioeconomic outcomes in order to determine the variables best predicted by completion status. Finally, in order to control for a potentially inflated Type I error rate, a Bonferroni correction was applied to 36
separate between-group comparisons (all the data from Table 1 and Tables 3 through 5 but not the demographic data in Table 2). The Bonferroni correction resulted in an alpha-value of .00 1.
RESULTS
Basic Demographic/Health Variables 
The basic demographic data for the C and NC groups are presented in Table 2. A significant difference was found for race, with the NC group having fewer Caucasians and more Hispanics, as compared to the C group, X2(3)=8.9, p=.03 1. Educational level also differed significantly (p=.036). The NC group (41.2%) was also found to have a higher overall rate of comorbid health conditions (diabetes, cardiovascular condition, hypertension, gastrointestinal condition, cancer, asthma and hyperthyroidism), relative to the C group (32.4%), X2(1 )=7.4, OR=1.5(1.1, 1.9), p=.007. More specifically, the NC group had a significantly higher rates of cardiovascular conditions, X2 (1)=3.9, OR=1.8(1.0, 3.3), p=.048, and gastrointestinal conditions, X2 (1)=14.6, OR=2.1(1.4, 3.2), p<.001.
Pre-Treatment Work/Injury-Related Variables 
Data for the work/injury-related variables were previously presented in Table 1. The only variable on which the NC group differed from the C group was pre-rehabilitation Case Settlement Status. Those in the NC group were more likely to have unsettled cases, X2 (1) = 27.6, OR=2.1 (1.6, 2.9), p<.001. [BOB: WHAT ABOUT COMORBID HEALTH CONDITIONS ON TABLE 2?]
Work Adjustment and Outlook Variables 
Table 3 presents findings for a number of work adjustment and outlook variables collected on case management interviews as part of the multimodal disability management component of the program during early program participation. As shown, differences were found on each of the four variables that were considered. A lower percentage of the NC group believed they would have a job available following treatment, X2(1)=1 6.7, OR=2.2(1.4,3.4), p=.00 1, to have reported a positive relationship with their employer/supervisor, X2(1)=1 8.9, OR=2.4(1.8,4.8), p<.001, to have expressed the desire to return to the same employer, X2(1)=24.9,OR=2.2, (1.6,3.0), p<.001, and expressed a desire to return to the same type of work, X2(1 )=12.6, OR=1.8(1.4,2.9), p<.001.
Psychosocial Variables 
As shown in Table 4, the NC group displayed higher initial evaluation (PRE) scores on Pain Intensity, Million VAS and BDI. Also in this study, differences between the NC and C groups in terms of DSM Axis I psychiatric disorders (current, lifetime, post-injury, and pre-injury) were analyzed. Axis I comparisons were performed with Pain Disorder excluded from the analyses, as 99% of the subjects in this sample met criteria for the diagnosis. No significant differences in prevalence of DSM disorders between the NC and C groups were found in these assessed diagnostic categories.
One-Year Socioeconomic Outcomes 
Table 5 presents the findings for the one-year post-rehab socioeconomic outcome variables. These data were divided into the following three sections: Health Utilization;
Dispute Settlement; and Work Status. Table 5 also includes the individual sample sizes (n) utilized in the analyses of each of the one-year socioeconomic outcome variables.
Work Status. The two groups differed significantly on work return, with 90.4% of the C group having worked during the year following treatment in the program, as compared to only 48.7% of the NC group, X2(1)=100.1, OR=9.7(6.3,15.2), p<.001. The two also differed significantly in terms of work retention, as only 40.6% of the NC group was still working at the time of the follow-up interview, as compared to 84.0% of the C group, X2(1)=86.6, OR=7(4.7,10.5), p=.<.001. Post-treatment work status was further explored by examining the work situations that patients in each of the two groups entered into during the year following discharge from functional restoration treatment. A significantly lower percentage of the NC group (14.7%) returned to their original employer when compared to the C group (3 6.7%), X2(1)=27.4, OR=3.6(2.1,6. 1), p<.00 1. Similarly, 12.7% of the NC group returned to their identical job, as opposed to 33.5% of the C group, X2(1 )=28.6, OR=4.1(2.2,7.4), p<.001.
Health Utilization. Post-treatment surgery to the original musculoskeletal area of injury was 12.2% of the NC group, as compared to only 2.3% of the C group,
X2(1 )=31.3, OR=7.1 (3.7,13.8), p<.001. Health care utilization from a new provider was defined as visits to providers that occur in addition to scheduled visits with the treating and rehabilitation supervising doctors. In this study, 40.7% of the NC patients sought health care services from a new provider over one year, as compared to only 20.8% of the C group, X2(1)=26.5, OR=2.3(1.9, 3.9), p<.001. The variable was broken down into the two groups based on number of visits to a new provider: 1 to 30 visits and greater than 30 visits. A remarkable difference was found for the unremitting health utilizers. Indeed, 19. 1 % of the NC group accrued 30 or more visits to a new health care provider in the year following treatment, compared to only 3.4% in the C group, X2(1 )=38.5, OR=6.8(3.8,12.2), p<.001.
Case Settlement. In this study, a lower percentage of the NC group (89.1 %) settled their case (i.e., financial or indemnity award/dispute) at one year, as compared to the C group (96.1 %), X2(1)=1 1.5, OR=3.2(1.7, 5.9), p<.001.
-------------------------------------------
INSERT TABLE 5 ABOUT HERE
-------------------------------------------
Results of Stepwise Regression. The best model indicated that work return, surgery to compensable area, health utilization from a new provider and healthcare utilization greater than 30 visits were the most reliably predicted by completion status, F (4,1037)=63.7, p<.001 indicated approximately 20% of the variance accounted for by this model.
DISCUSSION
The present study represents the first large-scale, comprehensive examination of functional restoration completion status in CDOMD patients. The methodology used included the advantage of a large sample of NC subjects from the same temporal cohort as the C subjects. It served to replicate the findings of previous small sample-size studies, while also adding to the current body of knowledge by addressing previously unexamined variables. Results clearly documented that the 21% of patients who fail to complete functional restoration treatment differ from program completers on a number of pre-treatment factors. Overall, a sizeable difference was found between the groups on objective post-rehab socioeconomic outcomes. The NC group displayed greater health utilization, with 7 times greater likelihood of having post-treatment surgery to the area of injury and higher rates of health care utilization from a new provider. It is also remarkable that NC patients were almost 7 times more likely to utilize excessive rates of health care from a new provider (i.e., 30 visits or more), as compared to the C group. In addition, the NC group was more likely to have unresolved financial workers’ compensation disputes relative to the C group.
Work status findings, which represent one of the most important outcomes of functional restoration treatment, are of particular note. The NC group displayed lower rates of work return and retention. These differences were particularly great, with C patients being almost 10 times more likely to return to work, and 7 times more likely to retain work at one-year. Further analysis of work status revealed that NC subjects were less likely to return to work with the same employer and return to the identical job. These subjects were also less likely to return to modified work, with either the original employer or through some alternative means. It is noteworthy that the NC group had a response rate of about 65% on work status questions, so that if it were assumed that all patients not contacted had never returned to work, then the work return rate would drop to about 32%, and the work retention rate to about 27%. Nonetheless, these would appear to be surprisingly high numbers to those unfamiliar with patient behaviors in work-related injuries. However, in two prior studies, Mayer 11,12 demonstrated that a comparison group denied functional restoration treatment by virtue of the policy of the insurance carrier that happens to represent the employer, showed about a 40% return to work rate in the same population 20 years ago. These generally represent patients who will return to work once the option of continuous medical treatment for their condition is curtailed, and disability benefits under workers' compensation are discontinued. In the remainder, however, a continued dependence on some type of financial benefit is common, be it from a U.S. federal source (Social Security Disability Income (S SDI), from the employer (long-term disability (LTD)), or from a state/local source (unemployment insurance or welfare benefits).
The existence of a premorbid health condition, and in particular cardiovascular and gastrointestinal conditions, was also associated with program non-completion. Evans (1999) earlier found higher rates of premorbid health conditions in recurrent injury CDOMD patients, and speculated that these conditions may lead to a more sedentary lifestyle (which may be either a cause or effect of the health conditions), that ultimately
results in the physical deconditioning syndrome originally proposed by Mayer and Gatchel .10 Evan s32 also proposed that patients with comorbid health conditions might begin to adopt a “sick role,” as described by Gatchel .33 It must also be considered that individuals with comorbid health conditions, as a result of their experience with the illness, are more focused on somatic sensations. No differences, though, were found between the two groups with respect to area of musculoskeletal injury or severity of injury. These findings are important, as they indicate that the differences found in this study were not confounded by an uneven distribution of musculoskeletal injury types or severity. Readers unfamiliar with prior functional restoration studies may be surprised at this extreme average length of disability, and that these levels of work return are possible under these circumstances. Jordan 13 studied the effect of extent of disability, and demonstrated outcomes were marginally improved at 4-8 months of disability over 18-24 months, and that there was a linear decline in socioeconomic outcome quality after 24 months. However, return to work rates for those completing treatment even with 5 years of disability were still of high quality, approaching 70% of the completer group. The difference in length of disability between the NC and C groups in this study were unlikely, in and of themselves, to have been a cause of lower work status for the NC group.
In terms of work adjustment and outlook variables, it was found that subjects in the NC group were 3 times less likely to report a positive relationship with their employer/supervisor (if employed), relative to the C group. These patients were also less likely to possess the desire to return to work for the same employer or to return to the same type of work, and indicated that they would not have a job available following rehabilitation. These work adjustment/outlook variables are tied to job satisfaction, which is the work-related variable that is most commonly found to be associated with chronic pain, with a large number of studies having reported this relationship.35-40 A study by Williams et al 41 indicated that job satisfaction may serve a protective role
against the development of chronic pain and disability following an initial acute onset of back pain. The findings of the present study indicate that, in CDOMD patients, work adjustment and satisfaction, as well as a positive outlook about future employment, may serve a protective role during rehabilitation as an incentive to follow through with treatment. On the other hand, it appears that poor work adjustment/satisfaction and a negative outlook toward future employment represent risk factors for program non-completion.
There were also a number of important psychosocial differences found between the two groups. These were addressed in a number of ways. In each case where differences were found, the NC group displayed higher levels of psychosocial distress. More specifically, the NC group reported higher pre-treatment levels of depression, pain and disability. Obviously, the presence of this somewhat increased level of psychosocial comorbidity is a potential risk for program non-completion that clinicians should be aware of.
Finally, in providing a more comprehensive documentation of the consequences of not completing the functional restoration program, a stepwise regression analysis of the socioeconomic outcome variables was conducted. Results produced a model that included the following poor one-year outcomes associated with patient non-completion: low return to work rates; greater healthcare utilization rates (more than 30 visits during the year); greater likelihood of seeking care from a new provider; and higher surgical rates to the compensable body area. Such results strongly argue for better efforts directed at early identification of potential program non-completers in order to provide more tailored interventions. These data, combined with the extreme length of disability for this CDOMD cohort suggests that a program of earlier recognition and earlier intervention is likely to result in improved outcomes. We are currently evaluating methods to achieve this important goal.
CONCLUSIONS
Taken together, the above findings clearly document the lower level of post-treatment functioning in patients who fail to complete interdisciplinary rehabilitation in the patient setting described herein. NC patients utilize more health care services, are involved in more financial disability disputes, and have poorer work return/retention outcomes. Overall, the present study clearly indicates that functional restoration non-completers do indeed differ from completers. As a group, non-completers differ on a variety of established pre-treatment factors, including psychosocial indices (i.e., self-reported depression, pain and disability. In addition, at the start of treatment, those destined to be NC patients were associated with greater prevalence of the following characteristics: more comorbid health problems; greater rates of smoking; unresolved financial disability disputes; a negative relationship with the employer; a lesser desire to return to work for the same employer or the same type of work; longer pre-referral disability time. These findings may serve as an important base from which to develop an evaluation methodology in future studies, which measures potential risk for program non-completion. A methodology of this type would be extremely valuable, as it would allow clinicians to identify patients who are a risk for non-completion and subsequent deficits in contributing to the social welfare. With this information, it is possible that clinicians could take additional steps during treatment to address the factors that make these patients a risk for non-completion. While many of these factors are already addressed with all patients as part of the functional restoration treatment model, an assessment instrument would lend additional aid in this cause. If successful, this instrument could represent a powerful clinical tool, as well as an excellent measure for use in future research addressing the causes and prevention of treatment non-completion.
REFERENCES
1. Melhorn J. Occupational orthopaedics: Raising public awareness. Amer J Orthop 2002:441-2
2. Mayer T, Gatchel R, Polatin P. Occupational Musculoskeletal Disorders: Function, Outcomes, and Evidence. Philadelphia: Lippincott, Williams & Wilkins, 1999.
3. Gatchel R, Mayer T. Occupational Musculoskeletal Disorders: Introduction and overview of the problem. In: Mayer T, Gatchel R, Polatin P, eds. Occupational Musculoskeletal Disorders: Function, Outcomes, and Evidence. Philadelphia: Lippincott, Williams & Wilkins, 2000:3-8.
4. Brady W, Bass J, Royce M, et al. Defining total corporate health and safety costs - significance and impact. J Occup Environ Med 1997;39:227-31.
5. National Safety Council. Injury Facts. Washington, D.C.: National Safety Council. 2000;1-154.
6. United States Bureau of Labor Statistics. Survey of occupational injuries and illnesses, 1996, Washington, D.C., United States Government Printing Office, 1998; 1 -56.
7. United States Bureau of Labor Statistics. BLS Issues 1996 Lost-work-time injuries and illnesses survey. American College of Occupational and Environmental Medicine Report, 1998;98-5:6-7
8. Waddell G. Low back pain: A 20th century health care enigma. Spine 1996;21:2820- 5.
9. Reid S, Haugh L, Hazard R, et al. Occupational low back pain: Recovery curves and factors associated with disability. J Occup Rehabil 1997;7:1-14.
10. Mayer T, Gatchel R. Functional Restoration for Spinal Disorders: The Sports Medicine Approach. Philadelphia: Lea & Febiger; 1988.
11. Mayer T, Gatchel R, Kishino N, et al. Objective assessment of spine function following industrial injury: A prospective study with comparison group and one-year follow-up. Volvo Award in Clinical Sciences, 1985. Spine 1985;10:482-93.
12. Mayer T, Gatchel R, Mayer H, et al. A prospective two-year study of functional restoration in industrial low back injury: An objective assessment procedure. JAMA 1987;258:1763-7.
13. Jordan K, Mayer T, Gatchel R. Should extended disability be an exclusion criterion for tertiary rehabilitation?: Socioeconomic outcomes of early vs. late functional restoration in compensation spinal disorders. Spine 1998;23:2110-7.
14. Mayer T, McMahon M, Gatchel R, et al . Socioeconomic outcomes of combined spine surgery and functional restoration in workers’ compensation spinal disorders with matched controls. Spine 1998;23:598-606.
15. Wright, A, Mayer, T, Gatchel, R. Outcomes of disabling cervical spine disorders in compensation injuries. Spine 1999;24:178-83.
16. Mayer T, Gatchel R, Polatin P, Evans T. Outcomes comparison of treatment for chronic disabling work-related upper-extremity disorders and spinal disorders. JOEM 1999;41:761-70.
17. Mayer T, Gatchel R, Evans T. Effect of age on outcomes of tertiary rehabilitation for chronic disabling spinal disorders. Spine 2001;26:1378-84.
18. Proctor T, Mayer T, Gatchel R, McGeary D. Unremitting Health-Care-Utilization Outcomes of Tertiary Rehabilitation of Chronic Musculoskeletal Disorders. JBJS 2004;86:62-69.
19. Hazard R, Fenwick J, Kalisch S, et al. Functional restoration with behavioral support: A one-year prospective study of patients with chronic low back pain. Spine 1989;14:157-61.
20. Bendix T, Bendix A. Different training programs for chronic low back pain – A randomized, blinded one-year follow-up study. Presented at International Society for the Study of the Lumbar Spine. Seattle, 1994.
21. Jousset N, Fanello S, Bontoux L, et al. Effects of restoration versus 3 hours per week physical therapy: a randomized controlled study. Spine 29; 487-494, 2004.
22. Charting the future of Social Security Disability Programs: the need for fundamental change, Social Security Advisory Board Report, Congressional Printing Office, January 2001.
23. Hazard R, Bendix A, Fenwick J. Disability exaggeration as a predictor of functional restoration outcomes for patients with chronic low-back pain. Spine 1991;16:1062-7.
24. Gatchel RJ, Mayer T, Dersh J, et al. The association of SF-36 health status survey with one-year outcomes in a chronically disabled spinal disorder population. Spine 1999;24:2162-70.
25. Dersh, J. A comprehensive evaluation of psychopathology in chronic musculoskeletal pain disability patients. Unpublished doctoral dissertation. The University of Texas Southwestern Medical Center at Dallas, Dallas, TX, 2000.
26. Mayer T. Functional restoration of patients with chronic spinal pain. In: Fardon D, Garfin S, eds. AAOS Spine Orthopedic Knowledge Update, Ed. 2. Chicago: AAOS Press; 2002:229-38.
27. Anagnostis C, Mayer T, Gatchel R, et al. The Million Visual Analog Scale: Its’ utility for predicting tertiary rehabilitation outcomes. Spine 2003;28:1051-60.
28. Capra P, Mayer T, Gatchel R: Adding Psychological Scales to Your Back Pain Assessment. J Musculoskel Med 1985;2:41-52.
29. Million S, Hall W, Haavik-Nilsen K, et al. 1981 Volvo Award in Clinical Science: Assessment of the progress of the back-pain patient. Spine 1982;7:204-12.
30. Beck AT, Ward CH, Mendelson MM, et al. An inventory for measuring depression. Arch Gen Psychiat 1961;4:561-71.
31. First MB, Spitzer RS, Gibbon M, et al. Structured Clinical Interview for DSM-IV Axis I Disorders – Nonpatient Version 2.0. New York: New York State Psychiatric Institute; 1995.
32. Evans, T. Recurrent versus non-recurrent work-related injuries in the chronic disabling spinal disorder patients presenting for tertiary rehabilitation treatment: A psychosocioeconomic comparison. Unpublished doctoral dissertation. The University of Texas Southwestern Medical Center at Dallas, Dallas, TX, 1999.
33. Gatchel R. Psychological disorders and chronic pain: Cause-and effect relationships. In: Gatchel R, Turk D, eds. Psychological Approaches to Pain Management: A Practitioners Handbook. New York: Guilford Publications, Inc.; 1999;33-52.
34. Carosella A, Lackner J, Feuerstein M. Factors associated with early discharge from a multidisciplinary work rehabilitation program for chronic low back pain. Pain 1994;57:69-76.
35. Bigos S, Spengler D, Martin N, et al. Back injuries in industry: A retrospective study. III. Employee-related factors. Spine 1986;11:252-6.
36. Bigos S, Battie M, Spengler D, et al. A prospective study of work perceptions and psychosocial factors affecting the report of back injury. Spine 1991;16:1-6.
37. Cats-Baril W, Frymoyer J. Identifying patients at risk of becoming disabled because of low-back pain: The Vermont Rehabilitation Engineering Center predictive model. Spine 1991;16:605-7.
3 8. Linton SJ, Warg L. Attributions (beliefs) and job satisfaction associated with back pain in an industrial setting. Perceptual and Motor Skills 1993;76:51-62.
39. Skovron M, Szpalski M, Nordin M, et al. Sociocultural factors and back pain: A population-based study in Belgian adults. Spine 1994; 19:129-37.
40. Croft P, Papageorgiou A, Ferry S, et al. Psychological distress and low back pain. Evidence from a prospective study in the general population. Spine 1996;20:2731-7.
41. Williams R, Pruitt S, Doctor J, et al. The contribution of job satisfaction to the transition from acute to chronic low back pain. Arch Phys Med Rehabil 1998;79:366-74.