Spine

Issue: Volume 22(17), 1 September 1997, pp 2016-2024

Copyright: (C) Lippincott-Raven Publishers.

Publication Type: [Functional Restoration]

ISSN: 0362-2436

Accession: 00007632-199709010-00016

Keywords: compensation, disability, litigation, low back pain, rehabilitation

[Functional Restoration]

The Effect of Compensation Involvement on the Reporting of Pain and Disability

by Patients Referred for Rehabilitation of Chronic Low Back Pain

Rainville, James MD*+; Sobel, Jerry B. MD*; Hartigan, Carol MD*+; Wright,

Alexander MD*++

Author Information

From *the New England Spine Care Center, Chestnut Hill, Massachusetts, and the

Departments of +Rehabilitation Medicine and ++Orthopedic Surgery, Tufts

University Medical School, Boston, Massachusetts.

Acknowledgment date: May 13, 1996.

First revision date: December 2, 1996.

Acceptance date: March 13, 1997.

Device status category: 1.

Address reprint requests to: James Rainville, MD; New England Spine Care

Center; 830 Boylston Street; Chestnut Hill, MA 02167.

------

Outline

Abstract

Materials and Methods

Results

Comparisons of Presenting Characteristics of All Study Subjects

Treatment Recommendations, Compliance, and Compensation Involvement

Comparisons of Patients Completing Spine Rehabilitation

Discussion

References

Abstract

Study Design. In this prospective, observational, cohort study of 192

individuals with chronic low back pain, the group of individuals was divided

based on compensation involvement, and their presentation pain and disability,

treatment recommendations, and compliance were compared. For 85 of these

individuals who completed a spine rehabilitation program, their pain and

disability at 3 and 12 months were compared.

Objectives. To test the theory that individuals with compensation involvement

presented with greater pain and disability and would report less change of pain

and disability after rehabilitation efforts.

Background.Previous studies have produced conflicting results concerning this

issue.

Methods.Individuals were recruited as consecutive patients referred for

consultation at a spine rehabilitation center. Pain, depression, and disability

were assessed using self-report questionnaires at evaluation and at 3 and 12

months. Rehabilitation services consisted of aggressive, quota-based exercises

aimed at correcting impairments in flexibility, strength, endurance, and

lifting capacity, identified through quantification of back function. Multifactoral

analysis of variance models were used to control for baseline differences

between compensation and noncompensation patients during analysis of target

variables.

Results.The compensation group included 96 patients; these patients reported

more pain, depression, and disability than the 96 patients without compensation

involvement. These differences persisted when baseline differences were

controlled for with multifactoral analysis of variance models. Treatment

recommendations and compliance were not affected by compensation. For patients

completing the spine rehabilitation program, length of treatment, flexibility,

strength, lifting ability, and lower extremity work performance before and

after treatment and patient satisfaction ratings were similar for the

compensation and non-compensation groups. At 3 and 12 months, improvements in

depression and disability were noted for both groups, but were statistically

and clinically less substantial for the compensation group. At the 12 month

follow-up visit, pain scores improved for the noncompensation group, but not

for the compensation group.

Conclusions.In chronic low back pain, compensation involvement may have an

adverse effect on self-reported pain, depression, and disability before and

after rehabilitation interventions.

------

Most experienced clinicians report a degree of pessimism when encountering

patients with chronic low back pain seeking or receiving compensation because

of their symptoms. Compared with noncompensation patients with similar

pathology, these patients seem to respond less well to interventions.

Previous medical research has supported the notion that compensation involvement

adversely influences the report of pain and disability, as well as outcomes

from treatments.4,5,16,19,21,23-26,43,44 Other studies have suggested that

those involved with compensation do not report increased pain or disability and

respond appropriately to treatments.1,7,12,15,31,32,40,42 The recruitment sites

of these studies have varied and include inpatient services, anesthesia and

multidisciplinary pain clinics, outpatient physician practices, and rehabilitation

services.1,4,5,7,16,19,21,23,25,27,31,40,42,43 Useful conclusions about this

subject remain unclear.

Analysis of the demographics of patients referred to the authors' spine

rehabilitation center revealed that nearly equal numbers of compensation and

noncompensation patients presented for evaluation and treatment of chronic low

back pain symptoms. The authors' had observed significant improvement in

measures of flexibility and strength and self-reported disability in a prior

setting in which they were exclusively treating patients receiving compensation

and using a functional restoration approach.35,36 Improvements also were noted

in patients' pain attitudes and beliefs, but despite all of these areas of

positive outcomes, reports of pain did not show any consistent improvement.36

One had to wonder if pain symptoms had actually lessened, but that reinforcements

for reporting pain within compensation systems had made patients reluctant to

report any improvements. In the current setting, compensation and noncompensation

patients are mixed in a spine rehabilitation program. Therefore, this was an

ideal site for observing potential differences between patients' presenting

characteristics and treatment outcomes based on compensation status.

This prospective, observational study was conducted to explore whether patients

presenting for evaluation and treatment of chronic low back pain differ, based

on compensation status, in their report of pain and disability. The authors

also were interested in whether compensation involvement affected treatment

recommendations, compliance, and responsiveness to rehabilitation efforts at 3

and 12 month follow-up visits. The authors theorized that compensation

involvement had a negative influence on the above areas.

Materials and Methods

Subjects. One hundred ninety-two consecutive patients with primary symptoms of

chronic (duration > 4 months) low back pain and/or sciatica evaluated by a

physicians at this center between October 1, 1993 and February 28, 1994 were

entered into a prospective data base if they: 1) were between the age of 18 and

70 years, 2) had no surgically correctable lesion as the cause of low back pain

or were not interested in pursuing surgical options, 3) were at least 3 months

postdiscectomy or 6 months postfusion, 4) possessed reading and writing

comprehension of the English language, and 5) had no concurrent medical illness

(cardiopulmonary, central nervous system, etc.) that produced significant

disability. Eighty-seven percent of patients were referred by physicians or

medical case managers, and the remaining 13 percent were self-referred.

Patients who were receiving or seeking financial compensation because of their

back pain were defined as having compensation involvement. This included

patients receiving Workers' Compensation, Social Security Disability, or

private disability policy benefits, or patients who were the plaintiff in an

unsettled personal injury suit. Ninety-six of the 192 study subjects were

involved with one or more of these areas of compensation, and 96 were not. The

types of compensation involvement are summarized in Table 1.

Of the 192 patients, 38 were evaluated only and did not receive any further

treatment. Three required advice only, seven were not interested in participating

in any recommended treatment secondary to travel distance, 21 were not

interested in pursuing recommended treatment options, and seven were denied

insurance authorization for further treatment. Twenty-three of the patients who

were evaluated but received no treatment had compensation involvement, and 15

did not.

The 154 remaining patients were referred to physical therapy for treatment of

their chronic back pain syndromes. Thirty-seven of the 154 patients had minimal

disability and were referred only for the development of independent back

exercise programs. Six of these patients had compensation involvement. The

remaining 117 patients had moderate to severe disability and were referred to a

spine rehabilitation program. Of these, 67 had compensation involvement. For

those referred to the spine rehabilitation program, 32 (27%) dropped out of the

treatment program; 20 of these patients (30%) had compensation involvement.

Reasons for discontinuing treatment included medical illnesses for six

patients, conflicts with work schedules for two patients, and lack of interest

in the treatment approach for 24 patients.

Eight-five patients completed the spine rehabilitation program, 47 of whom had

compensation involvement. These patients who completed the treatment program

will be the focus of the treatment outcome portion of this report.

Paper and Pencil Measures. At the initial evaluation, all patients completed a

questionnaire that inquired about demographic variables, symptoms, diagnostic

tests, prior treatments, compensation involvement, and work status. Self-reported

disability was measured with an Oswestry Disability Questionnaire (OSWESTRY)

with potential scores ranging from 0 (no disability) to 100 (severe disability).9

The intensity of pain was measured by 10-cm visual analog scales anchored with

0 (no pain) and 10 (worst possible pain) for back (BACK-VAS) and leg (LEG-VAS)

pain symptoms.14 Depression symptoms were measured by the Beck Depression

Inventory (BDI) with a range of scores from 0 (no depression) to 63 (severe

depression).2

Patient satisfaction was assessed with a 10-item questionnaire with potential

responses of 1 (excellent), 2 (very good), 3 (good), 4 (fair), and 5 (poor).

Questions assessed scheduling, business services, physician services, therapist

services, team-work, communication, education, home exercise instruction,

quality of care, and effectiveness of care. Responses to the 10 questions were

totaled with a range of scores from 10 to 50.

Follow-up Questionnaires.Three months after the initial evaluation, a research

assistant mailed a follow-up packet to all patients. Packets included an

OSWESTRY, BDI, BACK-VAS, LEG-VAS, and a Patient Satisfaction Questionnaire.

Twelve months after the initial evaluation, another follow-up packet was mailed

to all patients in the study. Packets included a stamped return envelope. If no

response was received in 3 weeks, a second packet was mailed. After 3 more

weeks, phone calls (maximum of three) were placed to all nonresponding patients.

If contacted, patients were given the options to complete the questionnaires on

the phone. When not at home, instructions were left with household members or

on answering machines to contact the research assistant or to complete the

mailed questionnaires.

Quantification of Physical Function. Trunk flexibility, straight leg raising,

trunk strength, lifting ability, and lower extremity work endurance were

quantified before and at the completion of spine rehabilitation.

Maximum trunk flexion and extension in the standing posture was measured with a

single inclinometer (AngleLevel, Dejon Tool Co, Covington, OH) placed over the

T12 spinousprocess.37 Maximum straight leg raising (SLR) was measured by

placing the inclinometer over the bony surface just below the tibialtuberosity

and raising the leg until significant back or leg pain occurred or until the

pelvis was observed to rotate.22,28

Trunk extension strength was quantified using Cybex back extension exercise

equipment (Lumex Corp., Ronkonkoma, NY). The testing procedure used a four-repetition

to maximum-weight-successfully-lifted protocol that is described in detail

elsewhere.41

Lifting ability was evaluated using the progressive isoinetrial lifting

evaluation (PILE) described by Mayer et al.30 The maximum weight lifted for

four repetitions from floor to waist was recorded as lumbar PILE, and that

lifted from waist to shoulder was recorded as cervical PILE.

To correct for anthropomorphic differences between patients, maximum back

strength and PILE weights were converted into percent ideal body weight before

data analysis. Ideal body weight was calculated for patients by measuring

height and actual weight and estimating body type and comparing these

measurements with those on standardized, sex-specific charts.33

Lower extremity work performance was quantified using a 9-minute protocol on an

isokinetic exercise bicycle (Cybex Fitron, Lumex Corp., Ronkonkoma, NY) during

which kilogram-meters (kg-m) per minute of work were recorded.41

Treatment. The spine rehabilitation program uses exercise to eliminate

impairments in flexibility, strength, endurance, and lifting capacity. The

occurrence of pain with exercises was expected, and the patients were

encouraged to work through pain symptoms to complete the established exercise

quota. Exercise quotas were increased throughout treatment. Treatments occurred

in groups consisting of a maximum of eight patients and were led by a physical

therapist and an exercise physiologist. Behavioral techniques were used to

promote wellness behaviors (exercise) and to extinguish pain behaviors

(limping, moaning, declining to perform exercises, etc.). Hands-on treatments

and applied modalities were not offered. Nonsteroidal anti-inflammatory and

antidepressant medications were prescribed at the discretion of the evaluating

physician, but narcotic analgesics were not prescribed, and their use was

discouraged. Referrals for psychological counseling occasionally were

requested.

The first week of treatment included a comprehensive physical therapy

evaluation, quantification of physical function, instruction in a home

stretching program, and orientation to the treatment facility.

Patients then were enrolled in a Level 1 treatment group that consisted of 2

hours of physical therapy three times per week. Sessions included 1 hour of

stretching and 1 hour of exercise on strengthening and endurance equipment.

Strengthening equipment included Cybex Eagle Strength System (Lumbex Corp.,

Ronkonkoma, NY) back extension, rotary torso, multihip, and lat pull-down

machines, Roman chair (back hyperextension), PILE lifting station, dumb bells,

andtheraband. Endurance equipment included isokinetic bicycles (CybexFitron,

Lumbex Corp., Ronkonkoma, NY) and upper body ergometer (Cybex UBE, Lumbex

Corp., Ronkonkoma, NY). Initial strength, lifting, and endurance exercise

levels were based on evaluation test results. Patients usually attended five or

six Level I sessions.

Patients then entered Level II, where group sessions were increased to 2 hours

and 45 minutes and occurred three times per week. These sessions included 45

minutes of stretching, 1 hour of strengthening, and 1 hour of aerobic training

(including low-impact step aerobic exercise). Requantification of physical

function was performed biweekly and at the completion of treatment. Compliance,

behavioral problems, and treatment goals were discussed at biweekly team

conferences between therapists and physicians. In general, 5 weeks of Level II

was projected for patients, with a total treatment time averaging 8 weeks.

At discharge, all patients in the study received individualized, written

recommendations for exercise routines that were to be performed at home and/or

a fitness facility.

Statistical Methods. Statistical analyses were performed using SPSS statistical

software for personal computers (SPSS, Chicago, IL). Summary statistics were

computed. Comparisons between those with and without compensation involvement

were done using chi-square tests for nominal variables and independent-sample t

tests for ordinal, interval, and ratio values.

Multivariate analysis of variance (MANOVA) techniques were used to determine

whether compensation involvement maintained its effect on self-reported pain

and disability when other differing characteristics were controlled. The first

step was to use multiple regression analyses, correlations, and two-way ANOVA

equations to identify cofactors and variables that had significant influence on

BACK-VAS, LEG-VAS, and OSWESTRY. These cofactors and variables then were entered

into MANOVA equations, which included compensation involvement as a cofactor.

Chi-square anddiscriminant analyses were used to explore the relationships of

compensation involvement with treatment recommendations and compliance.

Paired-sample t tests were used to compare physical function results at the

beginning and at the end of spine rehabilitation and questionnaire scores at

evaluation with scores at 3 and 12 month follow-up visits.

Results

Comparisons of Presenting Characteristics of All Study Subjects

For the entire cohort of 192 patients, substantial differences were noted in

many demographic, subjective, and objective characteristics when patients were

compared based on compensation involvement (Table 2). On average, those with

compensation involvement were less educated, were more likely to work jobs with

medium or heavy labor, were more likely to relate a compensatable event to the

onset of symptoms, had more reports of leg pain, had more neurologic signs and

symptoms, were less flexible, and were more likely to report pain with trunk

movements and straight leg raising. They also reported higher levels of pain,

depression, and disability. These findings suggest that patients with

compensation involvement had more serious chronic back pain syndromes.

Further analyses identified depression (BDI), education, and age as variables

that might influence the differences in initial pain scores (BACK-VAS and

LEG-VAS) between patients with and without compensation involvement. These

variables were entered into MANOVA equations with compensation involvement as a

factor and BACK-VAS, then LEG-VAS as the dependent variables. The results

indicated that compensation involvement maintained a strong influence on

BACK-VAS scores (F = 12.67, df = 1, P PP

Treatment Recommendations, Compliance, and Compensation Involvement

Chi-square analysis revealed that compensation involvement was not associated

with membership in the evaluation only group. Discriminantanalysis revealed

that referral to an individual physical therapy program was determined mainly

by the characteristics of low OSWESTRY and BDI scores, mild impairments in trunk

flexibility, and lack of work disability (Wilks Lambda = 0.728, P

Chi-square analysis revealed that treatment compliance for the spine rehabilitation

program did not differ based on compensation involvement. The drop-out group

demonstrated higher mean scores for OSWESTRY (57 vs. 42, t value = 4.6, P vs.

5.9, t value = 2.36, P = 0.02), and age (45 vs. 39 years, t value = 2.22, P =

0.03). The drop-out group contained a greater percentage of patients who had

failed pain programs (29% vs. 10%, chi-square = 3.9, df = 1, P = 0.05) and who