See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/221832317

Efficacy of Multidisciplinary Treatment for

Patients With Chronic Low Back Pain A
Prospective Clinical Study in 395 Patients
Article in Journal of clinical rheumatology: practical reports on rheumatic & musculoskeletal diseases
March 2012
Impact Factor: 1.08 DOI: 10.1097/RHU.0b013e318247b96a Source: PubMed

CITATIONS

READS

11

183

10 authors, including:
Sbastien Hagmann

Anita Zahlten

Universitt Heidelberg

Novartis Oncology

31 PUBLICATIONS 288 CITATIONS

27 PUBLICATIONS 475 CITATIONS

SEE PROFILE

SEE PROFILE

Marcus Schiltenwolf

Eva Neubauer

Universitt Heidelberg

Universitt Heidelberg

356 PUBLICATIONS 2,572 CITATIONS

37 PUBLICATIONS 334 CITATIONS

SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate,

letting you access and read them immediately.

SEE PROFILE

Available from: Marcus Schiltenwolf

Retrieved on: 06 June 2016

ORIGINAL ARTICLE

Efficacy of Multidisciplinary Treatment for Patients

With Chronic Low Back Pain
A Prospective Clinical Study in 395 Patients
Babak Moradi, MD, Sebastien Hagmann, MD, Anita Zahlten-Hinguranage, PhD,
Fernanda Caldeira, MSc, Cornelia Putz, MD, Nils Rosshirt, Eva Schonit, Alireza Mesrian, MD,
Marcus Schiltenwolf, MD, PhD, and Eva Neubauer, PhD

Background: The effectiveness of multidisciplinary treatment programs varies throughout the literature, and it remains controversial how
therapy outcome is affected by patients individual parameters and which
treatment settings work best.
Objectives: We set out to examine the impact of patient variables on the
effectiveness of a 3-week multidisciplinary treatment program in patients
with chronic low back pain. By presenting effect sizes, we aimed to enable
the comparison of our findings with other studies across disciplines.
Methods: Data on 395 patients were prospectively collected at study
entry, at the end of the program (T1) and after 6 months follow-up (T2).
Relevant therapy outcomes were analyzed by presenting effect sizes with
Cohens d. Group comparisons were performed for sociodemographic
and clinical features to determine the impact on therapy outcome.
Results: Medium effect sizes (d = j0.6 to j0.7) were shown for visual
analog scale (VAS) after treatment and at T2, indicating clinically relevant
pain relief. Significant changes in pain-related disability were observed
immediately at T1 with a strong treatment effect (d = 0.8). Functional
capacity was improved with low to medium effect sizes (0.4Y0.5). Qualityof-life subscales (36-item Short Form Health Survey) improved significantly at T1 for physical function, vitality, and mental health (d = 0.5Y0.8).
Center for Epidemiological Studies Y Depression Scale scores improved
significantly with strong effect sizes of d = 0.7. Sociodemographic parameters displayed a significant impact on effect sizes for visual analog
scale at T2, with females (d = j0.9), age group 30 to 39 years (d = j1),
and patients with low physical job exposure (d = j0.9) benefiting most.
An increase in number of pain locations (j0.7) and severity of accompanying pain (j0.7) in other body areas significantly impaired therapy
outcome and effect sizes of VAS.
Conclusions: Thus, multidisciplinary treatment ameliorates pain,
functional restoration, and quality of life with medium to high effect sizes
even for patients with a long history of chronic back pain. Effect sizes are
higher than for monodisciplinary treatments and treatment effects
remained stable at 6-month follow-up in a longitudinal uncontrolled
study design. Thus, we believe that multidisciplinary treatment is vital
for the treatment of patients with chronic low back pain. The impact of
sociodemographic and pain-related parameters needs to be taken into
account when including patients in an appropriate treatment program. We
emphasize the presentation of effect sizes as a vital treatment evaluation
to enable cross-sectional comparison of therapy outcomes.

From the Orthopaedic Clinic, University of Heidelberg, Germany.

The authors declare no conflict of interest.
Correspondence: Babak Moradi, MD, University Clinic of Heidelberg,
Department of Orthopaedics, Trauma Surgery and Paraplegiology,
Schlierbacher Landstrasse 200 a, 69118 Heidelberg, Germany.
E-mail: Babak.Moradi@med.uni-heidelberg.de.
Copyright * 2012 by Lippincott Williams & Wilkins
ISSN: 1076-1608/12/1802Y0076
DOI: 10.1097/RHU.0b013e318247b96a

76

www.jclinrheum.com

Key Words: chronic low back pain, multidisciplinary treatment, efficacy

of treatment, Cohens d, effect size, effectiveness
(J Clin Rheumatol 2012;18: 76Y82)

hronic low back pain (CLBP), with its associated disabilities

and costs, is the most common cause of long-term disability
in the working population and represents a serious problem for
our health care systems.1 The fact that 70% to 85% of the population in western societies experience musculoskeletal pain at
least once in their lifetime, with point prevalence of approximately 30%, underlines the epidemic proportions of this health
problem.1Y3 The socioeconomic impact is known to be enormous
and accounts for one third of all health care resources and costs.2,4
Only a small proportion of the costs are incurred by therapy; the
major costs result from loss of productivity, work absenteeism,
and disablement.2,4 Even after an intensive diagnostic workup,
the exact cause of musculoskeletal pain remains unestablished
in approximately 85% of cases; however, specific causes (e.g.,
tumors, infection, fractures, and herniated lumbar disks) have to
be excluded before treatment.3,5 Some 7% to 10% of all cases of
initial LBP become chronic, and these patients account for 80%
of the costs of LBP.5
Thus, CLBP is multifactorial with respect to its causes, and
the treatment is correspondingly diverse.6 A number of different
treatment options for patients with CLBP exist, but the reported
success rates vary dramatically.6Y9 Multidisciplinary therapy
incorporating multiple treatment components such as intensive
physical exercises and biopsychosocial and behavioral interventions is accepted to yield higher success rates than purely
biomedical treatment and is considered the criterion standard
by national and international pain societies in Europe and
elsewhere.8,10Y15
However, it is still unclear which treatment settings work
best and how therapy outcome is affected by patients individual
parameters.14,15 To tailor treatment settings and patient selection
for better therapy outcome, it is pivotal to compare different settings. The comparison of different rehabilitation programs within
1 institution remains difficult owing to ethical considerations
and the enormous costs of intensive multidisciplinary treatment.
Thus, it is of special interest to compare therapy outcome across
disciplines. Profound comparison remains difficult, however,
owing to inconsistent study settings, outcome parameters, and
statistical methods.
Estimates of effect size permit researchers to compare
findings among studies and across disciplines. This form of statistical evaluation allows closer comparison of therapy outcome
between multidisciplinary treatment and other treatment options,
for example, inpatient or outpatient nonmultidisciplinary treatments for patients with CLBP.

JCR: Journal of Clinical Rheumatology

&

Volume 18, Number 2, March 2012

Copyright 2012 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

JCR: Journal of Clinical Rheumatology

&

Volume 18, Number 2, March 2012

The purpose of this prospective study was to evaluate the

effectiveness of multidisciplinary treatment program in patients
with CLBP by determining effect sizes of different outcome
parameters. Further, we aimed to analyze the impact of sociodemographic and clinical parameters on therapy outcome.

METHODS
Study Design
A convenience sample of 395 patients was recruited from
our institution in a prospective cohort study with 6 months
follow-up. All patients underwent 3 weeks inpatient multidisciplinary therapy as described below. Assessment and treatment
of the patients were not altered for this study. Informed consent
was obtained from all patients. The ethics committee of the
University of Heidelberg approved the study. The treating physicians had referred all patients to the clinic without preselection
by an insurance company or the clinic.

Inclusion Criteria
Consenting patients of working age (18Y65 years) who had
an adequate command of the German language and were experiencing CLBP were enrolled into the study. Chronic low back
pain was defined as disabling pain of at least 3 monthsduration
that led to the patients being on sick leave for at least 6 weeks.
The patients generally had a long history of pain and had already
undergone several conventional forms of biomedical treatment.
In most cases, the patients were being referred to the multidisciplinary therapy program for the first time owing to failure of
standard therapy.

Exclusion Criteria
Excluded from the study were patients with specific etiology
of pain such as tumor disease; trauma/fracture; inflammatory
systemic disease or infection, for example, spondylodiscitis; severe degenerative changes; and rheumatological disease. Patients
with acute clinical symptoms due to structural pathology of the
lumbar spine such as nucleus pulposus prolapse with corresponding radicular pain, spinal stenosis, or spondylolisthesis with
claudicatio spinalis were also excluded. Further excluded were
patients with serious cardiopulmonary, vascular, or other internal
medical conditions; any sensorimotor and/or neurological deficits in the lower extremity; and surgery in the 12 months before
commencement of multidisciplinary therapy. Patients with pending litigation regarding pension demands and workers compensations were excluded.

Multidisciplinary Treatment
Patients underwent 3 weeks multidisciplinary therapy as
inpatients, with 8-hour sessions on 5 d/wk for a total of 120 hours
of treatment. This multidisciplinary approach embraces biological, social, and psychological aspects in one common explanation
of the patients pain. Its goal is to restore the patients physical and
psychosocial abilities, to increase their knowledge of back ergonomics and their adoption of protective behavior, to improve their
positive skills with regard to individual coping and emotional
control, and to increase their day-to-day activity levels and functioning at home so as to facilitate a return to the workplace. It
integrates physical exercises, ergonomic training, psychotherapy,
patient education, behavioral therapy, and workplace-based interventions on an individual basis and in group sessions (12
patients per group). Patients physical ability was assessed at study
entry by the treating physiotherapist through a comprehensive
physical examination measuring physical performance, force,
* 2012 Lippincott Williams & Wilkins

Efficacy of Multidisciplinary Treatment

endurance, and flexibility. Therapy goals were determined together with the patient, and a program for functional restoration
with an operant behavioral approach was performed. Exercise
included walking, medical training therapy, and free sports indoor
and outdoor. Continuous education by the medical staff gave
information on the goal of alleviating fear for movement and
explained increasing pain due to increasing workout as a physical
adaptation. Cognitive-behavioral therapy was conducted by an
interdisciplinary team of physician (orthopedic surgeon), psychologist, and physiotherapist to enhance patients understanding
of the nature and function of their pain, its social and psychological aspects, and its link to work and lifestyle. The primary goals
were to increase patients physical and mental coping strategies and
to encourage patients independence of medical treatment offers.
The psychotherapy embraced analysis of individual psychosocial
factors and conflicts contributing to persistent LBP. Stress behavior
was analyzed for situations in which pain and stress were experienced. Psychotherapeutic group sessions also included psychoeducation with behavioral therapy, problem solving, and stress
relaxation. Additional individual therapy consisted of daily sessions with physicians and psychologists, individual physiotherapy,
and cognitive-behavioral therapy.
After completion of their treatment programs, the patients
were discharged without further interventions by the hospital
team. They were allowed to contact the physician who had referred them for therapy, but they were encouraged to manage
similar further pain episodes on their own without immediately
contacting a physician. Further utilization of medical services
after completion of the therapy program was not monitored.

Evaluation of Baseline Data

Various independent variables were assessed on the basis
of a questionnaire in German completed by the patients before
they started the multidisciplinary therapy. These baseline data
included sociodemographic variables, occupational, and workplace characteristics. Data on self-reported pain (visual analog
scale[VAS]), functional capacity (Funktionsfragebogen HannoverRuecken [FFbH-R]), pain disability (Pain Disability Index German
[PDI-G]), quality of life (36-item Short Form Health Survey [SF36]) and Center for Epidemiological StudiesYDepression Scale
(CES-D) were prospectively collected at baseline (day 1), at the
end of treatment on day 21 (T1), and at the 6-month treatment
follow-up (T2). All measures are valid, reliable, and widely used
instruments in medical outcome studies.

Visual Analog Scale

Patients rated their current pain intensity on a VAS from
0 to 10.

Pain Disability Index

Measures of pain-related disability in which greater levels
of disability are reflected by higher scores on a 10-point scale
were assessed with PDI-G.16,17 This validated questionnaire consists of 7 items representing the impact of pain on essential life
activities. The overall score is expressed by an aggregate value of
functional capacity. The resulting PDI-G scores can range between
0 (minimal disability) and 70 (maximal disability).

Functional Back Capacity

Measures of functional back capacity were assessed with
FFbH-R.18 This validated questionnaire consists of 12 items
representing physical activities of daily living (e.g., sitting on a
chair for 91 hour). The overall score is expressed by an aggregate
value of functional capacity. The resulting FFbH-R scores can
www.jclinrheum.com

Copyright 2012 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

77

JCR: Journal of Clinical Rheumatology

Moradi et al

range between 0 (minimal functional capacity) and 100 (maximal

functional capacity).

36-Item Short Form Health Survey

Generic health status was assessed by using the subscales
physical functioning, mental health, and vitality of the
German version of SF-36, a widely used instrument evaluating
quality of life.19

Center for Epidemiological Studies Y Depression Scale

Psychological evaluation, based on a screening for the
prevalence of depressive disorders, was performed using the
German equivalent of the Center for Epidemiological Studies Y
Depression Scale (CES-D).20 This inventory consists of 20
items evaluating the level of depressive symptoms. A score of
16 or higher on the questionnaire indicated a depressive episode.

Classification of Stage of Chronicity

According to the classification of von Korff et al.,21 a
widely used and validated instrument, patients were assigned to
groups with different stages of chronicity. Classification was
based on the evaluation of the following items: pain intensity
(0Y10; pain right now, average pain, worst pain), days of pain in
the previous 6 months, time since onset, disability due to pain
(three 0Y10 interference ratings), and disability days. The items
were summarized to a characteristic pain intensity score and to
disability points (disability score and disability days) and led to
a final classification into 5 grades of chronic pain:
Grade 0: pain free (no pain problem).
Grade 1: low disability, low intensity (characteristic pain
intensity G50; G3 disability points).
Grade 2: low disability, high intensity (characteristic pain
intensity Q50; G3 disability points).
Grade 3: high disability, moderately limiting (3Y4 disability
points, regardless of characteristic pain intensity).
Grade 4: high disability, severely limiting (5Y64 disability
points, regardless of characteristic pain intensity).

Evaluation of Outcome Variables and

Statistical Analysis
Patients health status was assessed at the 3 assessment
time points: before treatment (T0), after treatment (T1), and at the
6-month follow-up (T2). The therapy outcome after treatment
and at 6 months was measured according to the above-mentioned
outcome criteria. Descriptive statistics for each of the measures evaluated were computed. The pre j post comparisons
were conducted with Fishers test.
Cohens d effect sizes were calculated for VAS, CES-D,
PDI-G, FFbH-R, and SF-36 by subtracting the posttreatment
mean from the pretreatment mean and dividing the result by
the pretreatment SD. This strategy allows the generation of
standardized values for change based on dependent variables,
even in the absence of a control group.22 All data were analyzed using the SPSS statistical application for Windows
(version 17.0; SPSS, Inc., Chicago, IL).
effect size d

posttreatment mean
 pretreatment mean

pretreatment SD

The generally accepted regression benchmark for effect

size comes from Cohen23: 0.20 is a minimal effect (but significant in social science research), 0.50 is a medium effect, anything equal to or greater than 0.80 is a large effect size.23,24

78

www.jclinrheum.com

&

Volume 18, Number 2, March 2012

RESULTS
Study Population and Evaluation of Baseline Data
Three hundred ninety-five patients (227 women and 168
men) who met the inclusion and exclusion criteria underwent
TABLE 1. Sociodemographic and Clinical Features of the
Study Population (n = 395) at Study Entry (T0)
Sociodemographic data
n
395
Age, mean (SD) [range], y
44.25 (9.1) [22Y64]
Male, %
42.5
Female, %
57.5
25.1 (4.5) [16.1Y47.6]
Body mass index, mean (SD)
[range], kg/m2
Marital status, n (%)
Single
85 (21.5)
Married
251 (63.5)
Divorced
41 (10.4)
Widowed
4 (1.0)
Smoker, n (%)
138 (34.9)
Regular daily sports activity, n (%)
265 (67.1)
Occupational status
Educational level, n (%)
Low
152 (38.5)
Intermediate
120 (30.2)
High
113 (28.7)
No data
10 (2.6)
Professional education, n (%)
335 (84.9)
Physical job exposure, n (%)
Low
123 (31.1)
Moderate
146 (37.1)
High
113 (28.5)
Job satisfaction, n (%)
High
188 (47.4)
Low
86 (21.7)
No data
47 (18.7)
Days of sick leave, n (%)
1Y10 d
48 (12.2)
11Y30 d
61 (15.4)
31Y65 d
41 (10.4)
965 d
44 (11.1)
No data
45 (11.4)
Present job security, n (%)
68 (17.2)
Evaluation of pain
Subjective pain perception (VAS),
51.8 (20.41)
mean (SD)
Duration of current pain, mean (SD), mo
16.4 (25.9)
Stage of chronicity, n (%)
Stage 2
125 (31.6)
Stage 3
167 (42.3)
Stage 4
89 (26.1)
Functional status
SF-36 physical functioning, mean (SD)
60.92 (19.9)
SF-36 mental health, mean (SD)
51.26 (16.30)
SF-36 vitality, mean (SD)
44.77 (19.02)
Functional disability (FFbH-R), mean (SD)
64.5 (19.1)
Existing depression (CES-D 916), n (%)
136 (34.4)
Pain disability (PDI-G), mean (SD)
26.66 (12.3)
* 2012 Lippincott Williams & Wilkins

Copyright 2012 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

JCR: Journal of Clinical Rheumatology

&

Volume 18, Number 2, March 2012

multidisciplinary treatment for CLBP at the Orthopedic Clinic,

University of Heidelberg. The patients mean age at study entry
was 44.25 (9.1) years, and their mean body mass index was 25.1
(4.5) kg/m2. All patients experienced CLBP as their main
symptom, with a mean pain duration of 16.4 (25.9) months.
Most patients (42.3%) displayed stage 3 of chronicity according
to von Korff et al.21 The mean duration of sick leave was 50.1
(95.4) days. A detailed description of baseline characteristics of
the study population is shown in Table 1.

Comparison of Outcome Variables at the End of

Treatment and at the 6-Month Follow-Up
Patients showed a significant improvement in all outcome
measures, including subjective pain perception (VAS), functional
back capacity (FFbH-R), and general health status (SF-36), after
multidisciplinary treatment (T1) and at the 6-month follow-up (T2)
compared with the pretreatment status (T0) (Tables 2 and 3). The
pre j post comparisons revealed significant changes in VAS
pain scores (F2,296 = 46.09, P e 0.0001). Medium effect sizes at T1
(d = j0.6) and T2 (d = j0.7) indicate clinically relevant pain
relief. Significant changes in pain-related disability were observed
after multidisciplinary treatment and at the 6-month follow-up
(F2,310 = 130.89, P e 0.0001), with a strong treatment effect
(d = 0.8) at T1, which remained almost stable at T2 (d = 0.7).
Functional capacity improved significantly at T1 (71.46
[19.35]) and at T2 (73.72 [21.68]) (F2,251 = 29.93, P e 0.0001),
with medium effect sizes between 0.4 and 0.5. Quality of life
improved at T1 for physical function (F1,303 = 54.8, P e 0.0001),
vitality (F1,247 = 41.15, P e 0.0001), and mental health (F1,247 =
41.58, P e 0.0001). Treatment effects were moderate for physical function and vitality (d = 0.5) and strong for mental health
(d = 0.8) at T1 and remained stable at T2. The depression scores
indicated significant improvement at T1 and at T2, with medium
effect sizes of d = 0.7.

Impact of Sociodemographic and Pain Covariates

on Effect Sizes of VAS
To further analyze the impact of sociodemographic parameters on therapy outcome, we evaluated effect sizes of subjective
pain perception grouped by sociodemographic parameters of
our study population (Table 4). Women displayed greater effect
sizes than men did (j0.9). There were significantly greater effect
sizes for the age group 30 to 39 years (j1) than for younger and
older age groups. Further, married patients (j0.9), patients with a
professional education (j0.7), and those with good job security
(j0.9) benefited most. Interestingly, the impact of duration of
sick leave seems to be U-shaped. Patients with short (G14 days)
and long (96 weeks) durations of sick leave displayed higher
effect sizes (j0.9 to 1.3) than those with 2 to 6 weeks. Patients
who wanted to negotiate a retirement plan had significantly lower
effect sizes (0.3) than those who did not (j0.8).

Efficacy of Multidisciplinary Treatment

TABLE 3. Cohens d Effect Sizes for Outcome Parameters

at T1 and T2
Outcome Variables
VAS
PDI-G
FFbH-R
SF-36 physical functioning
SF-36 mental health
SF-36 vitality
CES-D

T0 j T1 d

T0 j T2 d

j0.6
0.8
0.5
0.5
0.8
0.5
0.7

j0.7
0.7
0.4
0.5
0.8
0.5
0.7

Moreover, the impact of pain characteristics was evaluated.

Patients with 1 pain location (j1) and those with lumbar pain
(j0.9) displayed greater effect sizes than those with cervical
pain or pain in both locations. An increase in number of pain
locations (j0.7) and severity of accompanying pain (j0.7) in
other body areas significantly impaired therapy outcome and
decreased VAS effect sizes. Stage of chronicity showed no significant impact on therapy outcome. Patients with duration of
pain more than 2 years showed a poorer therapy outcome than
those with shorter pain duration. Patients with low frequency of
pain displayed the lowest treatment effect at T2 (0.1).

DISCUSSION
In this prospective longitudinal study, we aimed to examine
in detail the effectiveness of our multidisciplinary treatment program with regard to pain perception, functional capacity, and
quality of life in patients with CLBP. Further, the impact of covariates on therapy outcome was evaluated. By determining effect
sizes, we enable a closer comparison of therapy outcome between
multidisciplinary treatment and other published treatment settings.
Three hundred ninety-five patients with CLBP were involved in this prospective cohort study, which was characterized
by clearly defined inclusion and exclusion criteria, a standardized
treatment setting, and analysis of different recommended outcome
parameters.25 We conceptualized this study as a noncontrolled
trial. Thus, there was no form of patient selection, and all consenting patients from our pain clinic who met the inclusion criteria
participated in the study. Regarding the demographic data, the
study population was comparable with previous studies reporting
on patients with CLBP.6,10,12,26
The multidisciplinary treatment approach resulted in significant benefits for patients by the end of the therapy, as shown
by a significant improvement in all outcome measures. These
results remained stable at the 6-month evaluation. This displays
that the improvements were not negated by normal-life conditions.
Our results demonstrate the effectiveness of this multidisciplinary

TABLE 2. Outcome Variables at the 3 Assessment Time Points T0, T1 (End of Treatment), and T2 (6-Month Follow-Up)
Outcome variables
VAS
PDI-G
FFbH-R
SF-36 physical functioning
SF-36 mental health
SF-36 vitality
CES-D
* 2012 Lippincott Williams & Wilkins

T0

T1

T2

51.8 (20.41)
26.66 (12.30)
64.5 (19.1)
60.92 (19.9)
51.26 (16.30)
44.77 (19.02)
20.71 (11.98)

39.2 (21.0)
17.34 (12.03)
71.46 (19.35)
67.58 (22.36)
71.57 (18.59)
59.99 (19.78)
14.17 (4.84)

37.4 (23.9)
17.55 (14.15)
73.72 (21.68)
75.38 (37.36)
61.46 (7.65)
54.09 (20.04)
13.21 (10.95)

F2,296 = 46.09
F2,310 = 130.89
F2,251 = 29.93
F1,303 = 54.8
F1,247 = 41.58
F1,247 = 41.15
F3,257 = 100.53

e0.0001
e0.0001
e0.0001
e0.0001
e0.0001
e0.0001
e0.0001

www.jclinrheum.com

Copyright 2012 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

79

JCR: Journal of Clinical Rheumatology

Moradi et al

TABLE 4. Evaluation of Sociodemographic Parameters on

Effect Sizes for VAS at T2
Variable
Age, y

Sex
Marital status

Educational level

Professional education
Physical job exposure

Days of sick leave within

last 6 mo

Present job certainty

Desire for retirement

VAS

T0 j T2 Cohens d

18Y29
30Y39
40Y49
50Y63
Men
Women
Single
Married
Divorced
Low
Intermediate
High
Yes
No
Low
Intermediate
High
G14 d
2Y6 wk
6Y12 wk
912 wk
Yes
No
Yes
No

j0.70
j1.0
j0.8
j0.6
j0.5
j0.9
j0.5
j0.9
j0.6
j0.5
j1.1
j0.7
j0.7
j1.1
j0.9
j0.8
j0.5
j0.9
j0.5
j0.9
j1.3
j0.9
j0.7
0.3
j0.8

therapy program immediately after treatment (T1) and at

the 6-month follow-up (T2) compared with the pretreatment
status (T0). To highlight the different results for the various
parameters and to demonstrate how they compared to one
another, we computed within-group effect sizes as described
above (Table 5). This strategy permits computation of standardized values for change based on dependent variables, even if data
from a control group are not available.22 The effect of the treatment was established by the effect sizes (Cohens d ranging from
0.4 to 0.8). According to the calculated effect sizes, treatment
results were best with regard to pain-related disability, depression scores, and mental healthYrelated quality of life. Overall,
almost all effect sizes were medium or high, demonstrating that
a multidisciplinary regimen is most effective in treating CLBP by
increasing patients activity, restoring physical functions, and
addressing psychological disorders.
To analyze which type of patient benefits the most, we
further analyzed the impact of sociodemographic parameters
on therapy outcome.27 To this end, we evaluated effect sizes of
subjective pain perception grouped by sociodemographic parameters and pain characteristics of our study population. Sex- and
age-related differences of the treatment effect were shown, with
lower effect sizes for patients older than 40 years. This is in
accordance with the study by Bucher et al.28 These researchers
evaluated the impact of age on therapy outcome and found that
patients in all age groups benefit from multidisciplinary therapy
but with less improvement in those older than 35 years. Higher
differences were shown for marital status and educational level.
Job-related parameters seem to be relevant, with lesser effect
sizes for patients performing work with a pronounced physical
component and those with lower job security. In accordance with

80

www.jclinrheum.com

&

Volume 18, Number 2, March 2012

other studies, we showed that the desire for retirement and

compensation significantly lowered the treatment effect.8,29Y31
Furthermore, the impact of pain characteristics was evaluated. Higher number of pain sites and greater severity of accompanying pain in other body areas significantly impaired
therapy outcome and effect sizes as assessed by VAS. This is in
accordance with previous studies by our group and other institutions.32 Stage of chronicity did not show any significant impact on therapy outcome. We previous showed that also patients
with higher levels of chronicity improved significantly in all
outcome criteria. Thus, concluding that multidisciplinary therapy for CLBP should not be limited to patients in lower stages of
chronicity.33 Although patients with a longer history of pain (92
years) also benefit from multidisciplinary treatment, therapy
outcome was better in those with pain of shorter duration.
A number of studies have evaluated the outcome of multidisciplinary treatment, but only a few have assessed the effect
sizes.27,28,34Y38 Effect sizes are especially useful in enabling
comparison of the magnitude of experimental treatments from
different settings. Pohlmann et al.,35 investigating the efficacy of
a multidisciplinary treatment program for chronic pain, found
effect sizes are comparable to our data, with significant improvement and medium to high treatment effects. Their program
was comparable to our treatment plan with high treatment intensity of 122 hours. Pfingsten et al.26 reported slightly higher
effect sizes after multidisciplinary treatment in a comparable
setting, with effect sizes of 1.45 for pain intensity and 2.47 for
pain-related disability. This might be due to differences in the
study population because we did not limit the study population
by age or duration of sick leave. In a quantitative meta-analysis
by Keller et al.36 in 2007, pooled effect sizes were presented for
nonsurgical treatments for CLBP. Low to moderate treatment
effects were found for exercise therapy (0.13Y0.52) and behavioral treatment (pooled effect size, 0.57). This might be because
the studies included mainly involved monodisciplinary treatments and, moreover, did not display high treatment intensity.
Furthermore, invasive treatment options (transcutaneous electrical nerve stimulation, manipulation) were inferior regarding
short- and long-term treatment effects, with only small effect
TABLE 5. Evaluation of Pain Characteristics on Effect Sizes for
VAS at T2
Variables

VAS

T0 j T2 Cohens d

Pain location

Lumbar
Cervical and lumbar
Stage of chronicity
1
2
3
No. pain sites
Single site
Two sites
Multiple sites
Duration of current pain
e6 mo
6Y12 mo
1Y2 y
2Y5 y
95 y
Frequency of pain
e1 times a day
Several times a day
Constantly
Accompanying pain
Yes
No

j0.90
j0.80
j0.90
j0.90
j0.80
j1
j0.9
j0.7
j1.1
j1.1
j1.4
j0.4
j0.4
0.1
j1
j1
j0.7
j1.2

* 2012 Lippincott Williams & Wilkins

Copyright 2012 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

JCR: Journal of Clinical Rheumatology

&

Volume 18, Number 2, March 2012

sizes. A comprehensive review of multidisciplinary treatment for

chronic pain by Scascighini et al.38 underlined the superior efficacy of multidisciplinary treatment relative to standard medical
treatment and other nonmultidisciplinary approaches. A comparison of different multidisciplinary programs did not reveal
any difference between settings; however, although only 4 of the
included studies looked into this.
Our treatment plan is characterized by high treatment intensity, with a total of 120 hours during 3 weeks. Therapy programs with at least 100 hours of treatment have yielded higher
success rates than monodisciplinary treatments.36,39,40 Programs
with less than 30 hours have proved ineffective.41 Smeets et al.37
recently performed a multicenter, randomized clinical trial of
172 patients with nonspecific CLBP to determine the effectiveness of 3 commonly used nonsurgical treatment approaches.
These outpatient treatments were based on different theoretical models of CLBPVthe active physical model, the cognitivebehavioral model, and a combination of bothVand were delivered
3 times a week during 10 weeks. The authors then compared
patients treated with each of these therapeutic modalities with a
group of patients on the waiting list. At 6 months and 1 year after
treatment, all 3 active treatments were more effective than no
treatment, but no one treatment group showed more clinical
benefit than the other two.37 The authors concluded that these
results could, at least partially, be ascribed to the fact that exposure
to the cognitive-behavioral treatment components was only 78
hours and did not reach the 100 hours originally recommended by
Guzman et al.8 in a Cochrane review. These authors concluded
that intensive multidisciplinary, biopsychosocial rehabilitation
ameliorates CLBP and improves physical functioning, whereas
interventions of less than 100 hours do not yield an additional
effect compared with nonmultidisciplinary treatment or conventional care.
Treatment costs were, on average, 180 per day per patient,
which results in an average total treatment cost of 3780 per
patient. Published treatment costs of other institutes range from
90 to 300 per day per patient.35,40,42 It has to be taken into
account that costs of therapy are only a small proportion of all
costs associated with CLBP and that the major costs result from
loss of productivity, work absenteeism, and disablement.2,4 A
recent study by Jensen et al.42 could demonstrate that multidisciplinary treatment significantly reduces sickness absence even
7 years after therapy. The estimated cost reduction was more than
90,000 per referred patient.
One important limitation of the study is the missing control
group. On the basis of ethical considerations a randomized control group (e.g., with lower treatment intensity) was not realized
because all our patients experienced severe forms of CLBP
for which standard treatment with lower treatment intensity has
already failed. Other studies have included a waiting list control
group, which would have been an appropriate solution for our
study design.23,40,42 Because the design of this study was uncontrolled, it remains to be demonstrated that the outcome
results were based predominantly on treatment procedures rather
than confounding variables or simply the influence of time. A
meta-analysis by Flor et al.22 included 11 controlled studies and
demonstrated that the return-to-work rate for patients undergoing multidisciplinary treatment for CLBP was 67% compared
with 24% in the control group. This was further confirmed by
Demoulin et al.40 in a comparable treatment setting. However,
the fact that the patients displayed a mean pain duration of 16
months without any improvement before multidisciplinary treatment is a strong indicator that time did not play a predominant role
in determining improvement. Furthermore, the maintenance of
the positive treatment effects during a period of 6 months argues
* 2012 Lippincott Williams & Wilkins

Efficacy of Multidisciplinary Treatment

against placebo effects. On the other hand, the relevance of other

confounding variables such as encouragement by the professionals and enthusiasm for the program may have influenced
treatment outcome. Thus, longer follow-up times are necessary to
prove the long-term effect of this treatment and to verify if treatment effects remain stable without the support of treatment providers and after the initial enthusiasm and excitement of patients
has faded. Sociodemographic, clinical, and psychological parameters of the study population need to be taken into account when
discussing generalizability and reproducibility of our results. A
detailed assessment of these parameters (e.g., depression score,
pain catastrophizing score) is vital to define confounding variables. Moreover, apart from patient parameters, a number of
factors on the side of the treatment providers, for example, interdisciplinary agreement on treatment organization, attitude, and
approach of the professionals, may have a significant impact on
therapy outcome but are hard to identify and measure. Further
studies are necessary to evaluate which specific intervention is
responsible for treatment outcome.
In summary, comparisons with other treatments seem to
indicate that the intensity, duration, and frequency of a multidisciplinary treatment program may be the most important factors in its success. In this regard, it has to be taken into account
that a number of studies analyze a multicomponent treatment
rather than a multidisciplinary approach owing to a lack of interdisciplinary coordination. Future studies need to evaluate
which features are responsible for improvement with the goal of
individualizing treatment settings and addressing the variability of patients needs. A tailored treatment setting could help to
reduce treatment costs and eventually reduce treatment duration. Adherence to internationally defined standards is necessary to increase homogeneity of treatment settings and improve
comparability.

CONCLUSIONS
The results of this longitudinal not controlled cohort study
confirm that multidisciplinary pain therapy significantly improves
pain, return-to-normal function, and some aspects of healthrelated quality of life even for patients with a long history of
chronic back pain, with stable treatment effects at the 6-month
follow-up. Effect sizes are higher than for monodisciplinary treatments and comparable with those of other therapies with high
treatment intensity. Treatment results were highest in conjunction
with pain-related disability, mental health-related quality of life,
and CES-D scores. Overall, almost all effect sizes were moderate
to high, demonstrating that a multidisciplinary regimen is most
effective in treating CLBP by increasing patients activity, restoring physical functions, and addressing psychological disorders.
These treatment settings are vital for the treatment of patients with
CLBP. We emphasize the presentation of effect sizes to improve
comparison of different treatment settings across disciplines. This
could help to determine which components of the treatment setting matter most to further tailor treatment to the patients needs.
REFERENCES
1. Badley EM, Rasooly I, Webster GK. Relative importance of
musculoskeletal disorders as a cause of chronic health problems,
disability, and health care utilization: findings from the 1990 Ontario
health survey. J Rheumatol. 1994;21:505Y514.
2. Van Tulder MW, Koes BW. Chronic low back pain and sciatica.
Clin Evid. 2002;7:1032Y1048.
3. Kohlmann T. Musculoskeletal pain in the population. Schmerz.
2003;17:405Y411.

www.jclinrheum.com

Copyright 2012 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

81

JCR: Journal of Clinical Rheumatology

Moradi et al

4. van Tulder MW, Koes BW, Bouter LM. A cost-of-illness study of

back pain in The Netherlands. Pain. 1995;62:233Y240.
5. Hicks GS, Duddleston DN, Russell LD, et al. Low back pain.
Am J Med Sci. 2002;324:207Y211.
6. Van der Hulst M, Vollenbroek-Hutten MMR, IJzerman MJ.
A systematic review of sociodemographic, physical, and psychological
predictors of multidisciplinary rehabilitation- or, back school
treatment outcome in patients with chronic low back pain. Spine.
2005;30:813Y825.

&

Volume 18, Number 2, March 2012

spinal disorders: summary and general recommendations. Spine.

2000;25:3100Y3103.
26. Pfingsten M, Hildebrandt J, Leibing E, et al. Effectiveness of a
multidisciplinary treatment program for chronic low-back pain. Pain.
1997;73:77Y85.
27. van Hooff ML, van der Merwe JD, ODowd J, et al. Daily functioning
and self-management in patients with chronic low back pain after an
intensive cognitive behavioral program for pain management. Eur Spine
J. 2010;9;19:1517Y1526.

7. Karjalainen K, Malmivaara A, van Tulder M, et al. Multidisciplinary

biopsychosocial rehabilitation for subacute low back pain in
working-age adults: a systematic review within the framework of the
Cochrane Collaboration Back Review Group. Spine. 2001;26:262Y269.

28. Buchner M, Neubauer E, Zahlten-Hinguranage JA, et al. Age as a

predicting factor in the therapy outcome of multidisciplinary treatment
of patients with chronic low back painYa prospective longitudinal
clinical study in 405 patients. Clin Rheumatol. 2007;26:385Y392.

8. Guzman J, Esmail R, Karjalainen K, et al. Multidisciplinary

bio-psycho-social rehabilitation for chronic low back pain.
Cochrane Database Syst Rev. 2002;CD000963.

29. Bendix AF, Bendix T, Haestrup C. Can it be predicted which patients

with chronic low back pain should be offered tertiary rehabilitation
in a functional restoration program? A search for demographic,
socioeconomic, and physical predictors. Spine. 1998;23:1775Y1783.

9. Cairns D, Mooney V, Crane P. Spinal pain rehabilitation: inpatient

and outpatient treatment results and development of predictors for
outcome. Spine. 1984;9:91Y95.
10. Guzman J, Esmail R, Karjalainen K, et al. Multidisciplinary
rehabilitation for chronic low back pain: systematic review. BMJ.
2001;322:1511Y1516.
11. Van Tulder MW, Koes B, Malmivaara A. Outcome of non-invasive
treatment modalities on back pain: an evidence-based review. Eur Spine
J. 2006;15(Suppl):64Y81.
12. Patrick LE, Altmaier E, Found EM. Long-term outcomes in
multidisciplinary treatment of chronic low back pain: results of a 13-year
follow-up. Spine. 2004;29:850Y855.
13. Jensen IB, Bergstrom G, Ljungquiat T, et al. A 3-year follow-up of a
multi- disciplinary rehabilitation program for back and neck pain.
Pain. 2005;115:273Y283.
14. Airaksinen O, Brox JI, Cedraschi C, et al. European guidelines for the
management of chronic nonspecific low back pain. Eur Spine J.
2006;15(Suppl 2):192Y300.
15. Gatchel RJ, Mayer TG. Evidence-informed management of chronic low
back pain with functional restoration. Spine J. 2008;8:65Y69.
16. Chibnall JT, Tait RC. The pain disability index: factor structure and
normative data. Arch Phys Med Rehabil. 1994;75:1082Y1086.
17. Tait RC, Chibnall JT, Krause S. The pain disability index: psychometric
properties. Pain. 1990;40:171Y182.
18. Kohlmann T, Raspe HH. Der Funktionsfragebogen Hannover zur
alltagsnahen Diagnostik der Funktionsbeeintrachtigung durch
Ruckenschmerzen (FFbH-R). Rehabilitation. 1996;35:1Y8.

30. Fredrickson BE, Trief PM, Van Beveren P, et al. Rehabilitation

of the patient with chronic back pain, a search for outcome predictors.
Spine. 1988;13:351Y353.
31. Gallagher RM, Rauh W, Haugh LD, et al. Determinants of
return-to-work among low back pain patients. Pain. 1989;39:56Y67.
32. Moradi B, Zahlten-Hinguranage A, Barie A, et al. The impact of pain
spread on the outcome of multidisciplinary therapy in patients with
chronic musculoskeletal painVa prospective clinical study in 389
patients. Eur J Pain. 2010;14:799Y805.
33. Buchner M, Neubauer E, Zahlten-Hinguranage A, et al. The influence of
the grade of chronicity on the outcome of multidisciplinary therapy
for chronic low back pain. Spine (Phila Pa 1976). 2007;32:3060Y3066.
34. Schutze A, Kaiser U, Ettrich U, et al. Evaluation of a multimodal
pain therapy at the University Pain Centre Dresden. Schmerz.
2009;23:609Y617.
35. Pohlmann K, Tonhauser T, Joraschky P, et al. The Dachau
multidisciplinary treatment program for chronic pain. Efficacy data of a
diagnosis-independent multidisciplinary treatment program for back
pain and other types of chronic pain. Schmerz. 2009;23:40Y46.
36. Keller A, Hayden J, Bombardier C, et al. Effect sizes of non-surgical
treatments of non-specific low-back pain. Eur Spine J. 2007;16:
1776Y1788.
37. Smeets RJ, Vlaeyen JW, Hidding A, et al. Chronic low back pain:
physical training, graded activity with problem solving training, or both?
The one-year post-treatment results of a randomized controlled trial.
Pain. 2008;134:263Y276.

19. Ware JE, Sherbourne CD. The MOS 36-item short form health survey
(SF-36). Med Care. 1992;30:383Y473.

38. Scascighini L, Toma V, Dober-Spielmann S, et al. Multidisciplinary

treatment for chronic pain: a systematic review of interventions and
outcomes. Rheumatology. 2008;47:670Y678.

20. Radloff LS. The CES-D scale: a self-report depression scale for
research in the general population. Appl Psychol Meas.
1977;3:385Y401.

39. Van Tulder M, Koes B, Bombardier C. Low back pain. Best Pract Res
Clin Rheumatol. 2002;16:761Y775.

21. von Korff M, Ormel J, Keefe FJ, et al. Grading the severity of chronic
pain. Pain. 1992;50:133Y149.

40. Demoulin C, Grosdent S, Capron L, et al. Effectiveness of

a semi-intensive multidisciplinary outpatient rehabilitation program in
chronic low back pain. Joint Bone Spine. 2010;77:58Y63.

22. Flor H, Fydrich T, Turk DC. Efficacy of multidisciplinary pain treatment

centers: a meta-analytic review. Pain. 1992;49:221Y230.
23. Cohen J. A power primer. Psychol Bull. 1992;112:155Y159.
24. Keppel G, Wickens TD. Design and Analysis: A Researchers
Handbook. 4th ed. Englewood Cliffs, NJ: Prentice-Hall; 2004.
25. Bombardier C. Outcome assessments in the evaluation of treatment of

82

www.jclinrheum.com

41. Mirza SK, Deyo RA. Systematic review of randomized trials

comparing lumbar fusion surgery to nonoperative care for treatment
of chronic back pain. Spine. 2007;32:816Y823.
42. Jensen IB, Busch H, Bodin L, et al. Cost effectiveness of two
rehabilitation programmes for neck and back pain patients: a seven year
follow-up. Pain. 2009;142:202Y208.

* 2012 Lippincott Williams & Wilkins

Copyright 2012 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.