SPINE (Phila Pa 1976) 2003 (Mar 15); 28 (6): 525–531 ~ FULL TEXT
Olav Frode Aure, PT, Jens Hoel Nilsen, PT, and Ottar Vasseljen, PhD
Patients complaining of lower back or radicular pain were randomized to either manual therapy or exercise for a period of two months. Both groups of patients improved with treatment, however the manual therapy group showed significantly greater improvement on both short and long- (1 year) term follow-up. The researchers in this study also observed a considerable reduction in sick leave for the manual therapy group.
Study Design: A multicenter, randomized, controlled trial with 1-year follow-up.
Objectives: To compare the effect of manual therapy to exercise therapy in sick-listed patients with chronic low back pain (>8 wks).
Summary and Background Data: The effect of exercise therapy and manual therapy on chronic low back pain with respect to pain, function, and sick leave have been investigated in a number of studies. The results are, however, conflicting.
Methods: Patients with chronic low back pain or radicular pain sick-listed for more than 8 weeks and less than 6 months were included. A total of 49 patients were randomized to either manual therapy (n = 27) or to exercise therapy (n = 22). Sixteen treatments were given over the course of 2 months. Pain intensity, functional disability (Oswestry disability index), general health (Dartmouth COOP function charts), and return to work were recorded before, immediately after, at 4 weeks, 6 months, and 12 months after the treatment period. Spinal range of motion (Schober test) was measured before and immediately after the treatment period only.
Results: Although significant improvements were observed in both groups, the manual therapy group showed significantly larger improvements than the exercise therapy group on all outcome variables throughout the entire experimental period. Immediately after the 2-month treatment period, 67% in the manual therapy and 27% in the exercise therapy group had returned to work (P < 0.01), a relative difference that was maintained throughout the follow-up period.
Conclusions: Improvements were found in both intervention groups, but manual therapy showed significantly greater improvement than exercise therapy in patients with chronic low back pain. The effects were reflected on all outcome measures, both on short and long-term follow-up.
From the FULL TEXT Article:
About 60% to 80% of the population in the western
world will experience low back pain (LBP) at some stage
in life. [4, 9, 17, 24, 34, 37] Due to a favorable prognosis in the
acute stages, 80% to 90% of the patients will improve
considerably within 6 to 8 weeks. [10, 26, 46] The prognosis
for chronic LBP is considerably less favorable, [8, 42, 45]
causing potentially long-lasting suffering to the patient
and significant socioeconomic costs. [7, 35]
A number of different conservative treatment methods
and methods for LBP have been studied, but controversy
remains as to the preferred treatment. [1, 14, 20, 21, 36, 44] It has
been proclaimed that various national guidelines for LBP
treatment in primary care are fairly consistent, but discrepancies
were emphasized with regard to exercise and
spinal manipulation. 
Several recent reviews claim a strong evidence of effectiveness
for exercise therapy in chronic LBP and moderate
evidence of ineffectiveness in acute LBP.  There is
some evidence for a dose-response effect of exercises for
chronic low back pain,  although the effect was dependent
on persistent exercise activity.  However, a recent
study of LBP patients found no significant difference between
graded medical exercise therapy and conventional
physiotherapy with lesser intense exercise regime.  In
the recently published report by the International Paris
Task Force on Back Pain, it was concluded that “there is
sufficient scientific evidence to recommend that patients
who have chronic low back pain perform physical, therapeutic,
or recreational exercises, keeping in mind that
no specific active technique or method is superior to another.”
1 This conclusion is supported by an extensive
report by The Swedish Council on Technology Assessment
in Health Care. [34, 38]
Most studies of manipulation in LBP focus on patients
in the acute or subacute stages. Review studies presented
over the past 13 years of randomized controlled trials
conclude that the effect of spinal manipulation is promising
but the results inconsistent. [20, 36, 44] In a review of
clinical trials in manual therapy including patients with
chronic LBP, van Tulder et al concluded that manual
therapy clearly had a positive effect on chronic LBP and
was better than medical treatment, bed rest, and instructional
information.  However, Andersson et al recently
published a study that compared osteopathic manipulation
combined with medical treatment and injections to
ordinary physiotherapy and standard medical care in subacute
and chronic LBP patients.  The only significant difference
found was a favorable outcome in medication consumption in the group receiving spinal (osteopathic)
manipulation at 12 weeks after the initiation of treatment.
The focus in the present study was on chronic LBP
patients, a group with traditionally less favorable prognosis
both in physical and psychological terms compared
to acute LBP patients. [8, 40, 42, 45] The intention was to investigate
a manual therapy approach in tune with how
this regime is carried out in clinical settings, consisting of
spinal manipulation and mobilization techniques, specific
stretching, localized exercises, and information.
Based on the clinical examination, the treatment methods
are aimed to normalize function by means of spinal
or peripheral joint manipulation and mobilization techniques,
specific muscle stretching, and exercises to the
affected spinal segment or peripheral joints area. [11–13]
The aim of this study was to compare the effect of
manual therapy, consisting of specific exercises and segmental
techniques, to general exercise therapy in chronic
A letter with general information about the study
was mailed by the local Social Security Office (SSO) to all patients
in the area on sick leave for more than 8 weeks and less
than 6 months under ICPC codes L84 (back syndrome without
radiating pain) and L86 (back syndrome with radiating
pain).  This was done once for each patient and successively as
patients were registered by the SSO and took place in the period
from 1994 to 1997. Unfortunately, we do not know the exact
number of patients who were mailed the letter and thus do not
know the number of patients that refused to participate. Patients
willing to participate in the study on receiving the letter
from the SSO responded by phone to the physiotherapist in
charge of the study, who asked the patient to make contact
with their primary physician for referral. All responding patients
were then evaluated by the physiotherapist in charge of
the study according to the selection criteria, resulting in 49
included and 11 excluded patients, who were excluded for various
reasons based on the exclusion criteria given below. Patients
were then referred to the study’s collaborating physician
for the Schober test. Thereafter, the physician referred the patients
in consecutive order to the three participating clinics
(clinic 1, 2, 3, 1, 2, 3, etc.), where the randomization was
performed (see Design).
Inclusion criteria were men and women age 20 to 60 years
that had been sick-listed between 8 weeks and 6 months due to
LBP with or without leg pain. Exclusion criteria were unemployment
or early retirement because of LBP; prolapse with
neurologic signs and symptoms requiring surgery; pregnancy;
spondylolisthesis; spondylolysis; degenerative olisthesis; fractures;
suspicion of malignancy; osteoporosis; previous back
surgery; known rheumatic, neurologic, or mental disease; or
absence of pain aggravation on active, functional movement
tests (i.e., indicating nonorganic symptoms). Of the 49 patients
included in the study, 27 were randomized to manual therapy
(MT) and 22 to exercise therapy (ET). Fifteen patients (8 MT
and 7 ET) were treated at one clinic and 17 each at the other
two clinics (10 MT and 7 ET at one clinic and 9 MT and 8 ET
at the other). The three physiotherapy clinics were situated in a
town of 40,000 inhabitants. All patients were on 100% sick
leave at the start of the study. Two patients, 1 in each group,
who had recently received physiotherapeutic treatments for
LBP, were asked to wait 3 weeks before starting treatment. The
previous treatment was without effect and both were still on
100% sick leave.
Given the likely variability in the way treatment for LBP
patients is conducted in the physiotherapy clinic, the interest
was on detecting relatively large group differences, with an
effect size of 0.8 or larger. With the preselected α-level of 0.05
and the β-level of 0.20 (power: 1 – β = 0.8), power analysis
returned an estimate of 26 patients in each group.  As it turned
out, the actual results were 27 MT and 22 ET patients in the
two groups. A small imbalance in sample sizes is of little influence
on the power.  With the other premises unchanged, the
resulting sample sizes reduced the power from the planned 0.8
The study protocol was approved by the local ethics committee.
Eligible patients were given written information about
the study before giving their written consent to participate. The
information stated that the purpose of the study was to evaluate
two common physiotherapy treatment regimes, without
emphasizing the specific differences between the two regimes.
The study was carried out as a multicenter, controlled,
randomized trial with 1-year follow-up. A blocking design with
sex (male and female) and age (below and above 40 years of
age) as blocking factors was used with separate randomization
within each stratum. The randomization took place at each
clinic by presealed envelopes provided by an external research
corporation (Medstat Research AS, Lillestroem, Norway) and
was administered by one participating therapist at each clinic.
At each clinic, one specialist in manual therapy treated the
patients assigned to the MT group, and two physiotherapists
without this specialty were assigned to treat the patients in the
ET group. The patient’s therapist at start of treatment performed
a clinical evaluation in order to set up an individualized
treatment plan, which was an important aspect given the design
of this study. This evaluation was for the sole purpose of setting
up the treatment plan and not for scientific purposes. All patients
received 16 treatments, each lasting 45 minutes, with 2
treatments a week over a course of 8 weeks. The treatment dose
in terms of duration and number of treatments both in the
clinic and at home was equal in the two groups. However,
about one third of the treatment time in the clinic was devoted
to passive joint manipulation and mobilization techniques in
the MT group, leaving two thirds of the time to exercise.
Both groups were assigned a maximum of six individually
designed home exercises to be performed daily during the treatment
period. All patients were firmly encouraged to perform
walks, running, cycling, etc. at least 3 times a week. Brief information
on relevant anatomic and ergonomic topics was handed
out to all patients. The importance of self-responsibility and a
positive attitude towards their LBP in order to minimize fearavoidance
reactions and chronic symptoms of the problem was
emphasized to both groups. Patients were encouraged to return to
work as soon as possible. There was no restriction on medication
throughout the study. Other forms of treatment, such as acupuncture,
chiropractic, or alternative medicine, were not allowed during
the intervention period, but there were no restrictions in the
Forty-eight percent of the patients in the MT group and
46% of the patients in the ET group were women. No differences
in baseline characteristics between the two treatment
groups were observed (Table 1), nor was there any difference in
the L86 and L84 ICPC diagnosis classifications.
Only spinal manipulation, specific mobilization,
and stretching techniques described by Evjenth, Hamberg,
and Kaltenborn were allowed. [11–13] The following mobilization
or high-velocity, low-force manipulation techniques
Traction thrust to the thoracic-lumbar junction with the
patient sitting. 
Rotation-lateral flexion thrust to segments in the area from
T10 to L5 with the patient lying on his/her side. 
The sacroiliac manipulation/mobilization technique used in
the study was mainly a ventral or dorsal rotational technique
with the patient either prone or lying on his/her
side. [12, 13]
The patients also performed a subset of five general exercises
for the spine, abdomen, and lower limbs, and six specific
and localized exercises for spinal segments and the pelvic girdle
in each treatment session in order to normalize function (see Appendix). The purpose was mainly to mobilize hypomobile
areas or to stretch paravertebral muscle tissue depending on the
clinical findings. The exercises were performed by doing 2 or 3
sets of 20 to 30 repetitions for each exercise, with 30 seconds to
1-minute rest between each set. The patients were told not to
provoke pain during the exercises. Each manual therapy treatment
session lasted 45 minutes.
Patients assigned to general exercise therapy
were given 45 minutes of training. The first 10 minutes,
they performed warm-up on an exercise bicycle. The exercise
programs were individually designed based on the patient examination
and the clinical findings, and the programs consisted
of general training methods for LBP patients. Strengthening,
stretching, mobilizing, coordination, and stabilizing exercises
for the abdominal, back, pelvic, and lower limb muscles, suited
to the clinical findings, were allowed. The therapist was free to
choose type, number of repetitions, sets, and progression of
exercises. The training took place with or without training
equipment in the physiotherapy clinic. Group training, massage,
and methods were not allowed during the treatment period.
The patients were observed and guided closely by the
therapist during each session.
Outcome measures were spinal range of
motion, pain, functional disability, general health, and return
to work as follows:
A. Spinal range of motion was measured by the modified
Schober test. 
B. Pain intensity due to LBP was recorded on a 100-mm
Visual Analogue Scale (VAS), 0 indicating no pain and 100
the worst pain ever.  The scale was continuous without
intermediate markings. Pain at the moment, worst pain the
last 14 days, and mean pain during the last 14 days were
scored. The final outcome measure used in the statistical
analyses was the mean of these three recordings.
C. Functional disability was recorded using the Oswestry
Low Back Pain Disability Questionnaire. [15, 33] Each item is
scored on a 0–5 scale, and the maximum score is 50. The
maximum score in this study was 45, because one item was
omitted (sex life). Relative values are reported (total score/
total possible score 100%). 
D. General health was measured by The Dartmouth COOP
Function Charts.  Maximum score was 35.
E. Return to work was self-reported by the patients based on
the status at each test session. Patients partly or fully sicklisted
were contained in the “sick leave” group, whereas all
those in the “returned to work” group had resumed fulltime
All outcome measures except for spinal range of motion
were scored on questionnaires administered by the patients and
carried out five times during the study: before and immediately
after the treatment period (i.e., within 3 days after the last
treatment session), and then at 4 weeks, 6 months, and 12
months after end of treatment. A collaborating physician who
was blinded to which group the patients were assigned recorded
spinal range of motion. This was carried out at pre- and
posttest only. All pretests were performed after randomization,
except for spinal range of motion, which was performed before
Intention-to-treat analysis was carried
out for all included patients. Patients who dropped out for
reasons other than the treatment to which they were randomized
(dropout type A) were given the baseline registration score
for missing data points during the follow-up period. Patients
dropping out because of the treatment to which they were assigned
(dropout type B) were given the worst score registered
for any patient in their treatment group. During the treatment
period, two patients dropped out in the MT group due to lack
of effect and one in the ET group due to reasons unrelated to the
treatment. They were all registered and statistically analyzed
after the intention-to-treat principle.
All registration forms were entered into the database by two
different persons. The two registrations were compared and
discrepancies corrected. Mean was used as the measure of central
tendency for normally distributed data and median for not
normally distributed data, and 95% confidence interval (CI)
was used to express the range estimated to contain the true
population mean/median with a probability of 95%. 
Repeated measures ANOVA was used to test differences
between groups (MT vs. ET) and within groups (within each
treatment group over time) for the pain, Dartmouth, and Oswestry
variables. The Tuckey-Kramer test was then used for
pair-wise comparisons of means. Variables showing significant
differences were retained for further post hoc analyses, and the
student t test was used on the three above-mentioned outcome
measures to test differences in improvement between the two
treatment groups at all the posttreatment test sessions. Paired t
test was used to investigate changes within groups, and the
results from the posttreatment and follow-up test sessions were
compared to the pretreatment result. Wilcoxon signed-ranks
test (within groups) and Mann-Whitney U test (between
groups) were used for the Schober test. The Fisher exact test
was used to test group differences in sick-leave status and risk
ratio (RR) used to estimate the risk of being sick-listed in the
MT versus the ET group at all follow-up sessions. The level of
significance was set to P ≤ 0.05. 
The NCSS (Number Cruncher Statistical System; Utah,
USA) was used for statistical analyses.
Significant improvements in pain, general health, and
functional disability were observed in both groups from
before to after treatment (P < 0.01), and this improvement
was maintained throughout the 1-year follow-up
(Table 2). Significantly larger improvements (P < 0.05)
were found in the MT group compared to the ET group
at all posttreatment test sessions (Table 2). The mean
reduction of pain from pre- to posttest in the MT group
was twice that of the ET group (33 vs. 17 mm), correspondingly
for general health (9 vs. 4 score point units)
and functional disability (21% vs. 9%). The effects
gained from the treatments were stable in the 1-year
posttreatment period in both groups.
Spinal range of motion was measured only at the preand
posttreatment sessions. Significant improvements
were found both within (P < 0.01) and between groups,
with the MT group showing significantly larger improvement.
The mean improvement in the MT group was 31
mm (95% CI: 26–36) and in the ET group 9 mm (95%
CI: 6–12; P < 0.01).
At each test session, the subjects were asked about
their sick-leave status (Table 3). All patients in both
groups were fully sick-listed at pretest. At posttest, 73%
in the ET versus 33% in the MT group were partly or
fully sick-listed (P < 0.01). The respective numbers at 4
weeks follow-up were 57% versus 30% (P = 0.08), at 6
months 62% versus 11% (P < 0.01), and at 12 months
59% versus 19% (P < 0.01).
Although significant improvements were found in both
intervention groups, the manual therapy group showed
better results than the general exercise therapy group on
all outcome measures, including pain, functional disability,
general health, spinal range of motion, and return to
work. The effects were largely gained during the 8-week
treatment period, and the results remained stable
throughout the 1-year follow-up period.
To the authors’ knowledge, this is the first controlled
trial using a manual therapy approach including specific
exercises, mobilization, and stretching techniques in addition
to spinal manipulation on patients with chronic
LBP. [20, 22, 36, 44] Objections could be made to the flexibility
and diversity of treatment items allowed in the intervention
groups, as both groups received a package of interventions
rather than a single regime. This design made us
unable report on the factors within the regimes that contributed
most significantly to the effects. The approach is,
however, in line with how such programs are carried out
in the physiotherapy clinic, which should raise the external
validity of the results.
Plausible explanations for the group differences observed
could be the impact of mobilization/manipulation
in itself or a positive influence of the more specific approach
used by the manual therapists in general. The
importance of “specificity” in both diagnostics and conservative
treatment deserves further attention in future
studies of chronic low back pain patients. Perhaps the ET
group could have reached results similar to the MT
group had a more specific exercise regime been advocated
in the ET group. [18, 31, 32]
The number of patients was low, and the strict inclusion
criteria might reduce the generalizability of these
results to other chronic LBP populations. Nor can the
study address whether patients with nonorganic signs,
e.g., low back pain due to mental or psychological reasons,
would behave differently. An added psychological
impact on outcome measures of specific techniques and
forceful manipulations as opposed to general exercise
cannot be disregarded, nor can the fact that the study
was initiated, planned, and partly conducted by physical
therapists with a specialty in manual therapy. Lack of
blinding in the study could also affect the outcome, particularly
in a small community where word of mouth and
the inhabitants’ acquaintance with the physiotherapists
could affect the outcome. However, the stable and persistent
effects observed during the 1-year follow-up period
weigh against such bias.
The Schober test was conducted
by a blinded physician not otherwise involved in
the study. It adds further credibility to the trial that these
results (Schober) were in conformity with the other outcome
measures. The fact that all included patients had
been sick-listed due to their LBP for at least 8 weeks
before inclusion — i.e., the patients experienced chronic
pain — makes spontaneous remission unlikely. [8, 42, 45] This
claim is further strengthened by the outcome pattern in
the study, with clear improvements in both groups during
the 8-week treatment period and a fairly stable situation
thereafter. No additional treatments were given
beyond the 8 weeks in the program. There was no control
of whether the patients sought other treatments after
the intervention period.
Baseline values in this study correspond to other studies
both on the VAS for pain [3, 25, 41] and Oswestry Disability
Index, [3, 29, 33] whereas the values on the modified
Schober test were somewhat lower.25 In the ET group,
the relative change in Oswestry score was similar to another
exercise study,  whereas the improvement in the
MT group was slightly greater than reported in a study
on manipulation.  Patients receiving both manual therapy
and cortisone injection showed greater relative reduction
on pain compared to the MT group in this
A considerable reduction in sick leave was found in
the MT group, as only 19% of patients were sick-listed at
1-year follow-up compared to 59% in the ET group. The
results differ somewhat from other Norwegian studies of
chronic LBP patients, which have indicated that 40% to
50% of the patients in control groups receiving minor
interventions were still on sick leave at 1-year followup.
16,41 In groups receiving major intervention, the percentage
of patients still sick-listed at 1-year follow-up
was slightly in excess of 30%. The reason for these differences
is uncertain. Variations in pre-baseline painperiod
length and pain intensity might contribute to the
discrepancies. “Returned to work” in this study demanded
full-time (100%) employment, whereas the
“sick leave” group contained both fully and partly sick-listed
patients. This could be a reason for the larger number
of patients on sick leave at 1-year follow-up in the ET
It is not possible to define a meaningful cut-off point
for positive clinical outcome on the variables used in this
study, making discussions on clinical relevance difficult.
Conclusions are based on statistical significance. However,
the sick leave data are quite clear and in agreement
with the other outcome measures. Return to work has
been related to physical function and pain in patients
with back pain.  Getting people back to work may thus
serve as a relevant clinical goal.
The effects in this study do call for trials in extended
LBP populations, as well with alternative exercise therapy
regimes. Particularly, there is a need to investigate
the effect of specific exercise techniques, as was emphasized
in the manual therapy regime in this study. Pragmatic
treatment approaches in tune with how therapies
are conducted in the clinic are encouraged also in future
Descriptions of the Five General Exercises
Patient lying prone: contract the gluteal, spinal, and
interscapular muscles, then lift the upper trunk a few
inches, hold the position for 5 seconds, and relax between
“Four-foot” standing: lift the opposite leg and arm
up while keeping the rest of the body as stable as
possible. Relax and repeat with the other leg and arm.
“Four-foot” standing: change between flexing and
extending the entire spinal column as far as pain
Lying supine: with the knees flexed and heels on the
floor, the patient uses his/her abdominal muscles to
roll the upper trunk halfway upwards, making sure
the lumbar spine is in contact with the floor. Hands on
the chest or the belly.
Standing: upright position, lean forward by flexing
the hips, keep this position while flexing the knees to
Description of the Six Specific Exercises
Side-lying with the knees flexed: in order to limit
movements to the affected area of the spine, the trunk
was rotated starting from the head and all the way
down to the affected area or segments of the spine. In
this position, the affected segments were flexed and
extended alternatively and repeatedly.
Same position: using a firm pillow under the affected
area, the patient tilted the pelvis in lateral flexion
or rotation to the ipsilateral side, thereby creating
an active mobilizing movement.
“Four-foot” standing: pushing the trunk backward
towards the heels while simultaneously lowering the
head towards the floor. This was done to stretch and
flex the lower lumbar segments and muscles. More
flexion could be achieved by placing a firm pillow
under the knees.
Same position: leaning down on the elbows and, in
this position, flexing and extending the spine. This
position alters the area of lumbar movement in a cranial
direction, i.e., it creates a stronger stretching effect
in the upper lumbar segments.
Lying supine: knees flexed, heels on the floor, and
hands in the back on top of each other under the affected
area of the spine and, in this position, flexing and extending
the spine to create a localized movement.
Same position: both knees and the lower lumbar
spine were rotated from side to side. The movement
was limited to the first notion of movement in the
affected segments, as noticed by the hands placed under
the lower back.
A randomized, controlled clinical trial was performed
in patients with chronic low back pain.
Two treatments, manual therapy and exercise
therapy, were investigated.
Both groups improved, but the manual therapy approach resulted in significantly greater
improvements than exercise therapy on spinal range of motion, pain, function,
general health, and sick leave.
The authors wish to thank Ørnulf Andersen, MD, Paal
Brynildsen, PT, Harald Brynildsen, PT, Aashild Bergkaasa,
PT, Bernt Waale, PT, Grethe Thunold, PT, Torunn B. Sandvik, PT, and Daniel Langer, MSc, whose
efforts made this study possible. Special thanks to Stig E.
Larsen, PhD, for his kind help in designing the study and
the Larvik Social Security Office for recruiting patients.
Abenhaim L, Rossignol M, Valat JP, et al.
The role of activity in the therapeutic management of back pain. Report of the International Paris Task
Force on Back Pain.
Practical statistics for medical research. 1st ed.
London: Chapman and Hall; 1991.
Andersson GB, Lucente T, Davis AM, et al.
A comparison of osteopathic spinal manipulation with standard care for patients with low back pain.
N Engl J Med 1999;341:1426–31.
A prospective study of low back pain in a general population. I. Occurrence, recurrence, and aetiology.
Scand J Rehabil Med 1983;
Physical measurements as risk indicators for low-back trouble over a one-year period.
Blomberg S, Hallin G, Grann K, et al.
Manual therapy with steroid injections– a new approach to treatment of low back pain. A controlled multicenter trial with an evaluation by orthopedic surgeons.
Spine 1994;19: 569–77.
Carey TS, Garrett J, Jackman A, et al.
The Outcomes and Costs of Care for Acute Low Back Pain Among Patients Seen by
Primary Care Practitioners, Chiropractors, and Orthopedic Surgeons:
The North Carolina Back Pain Project
New England J Medicine 1995 (Oct 5); 333 (14): 913–917
Carey TS, Garrett JM, Jackman AM.
Beyond the good prognosis. Examination of an inception cohort of patients with chronic low back pain.
Cassidy JD, Carroll LJ, Cote P.
The Saskatchewan health and back pain survey. The prevalence of low back pain and related disability in Saskatchewan adults.
Coste J, Delecoeuillerie G, Cohen DL, et al.
Clinical course and prognostic factors in acute low back pain: an inception cohort study in primary care
Br Med J 1994;308:577–80.
Evjenth O, Hamberg J.
Muscle stretching in manual therapy. Vol. 1.
Alfta: Alfta Rehab Forlag; 1984.
Evjenth O, Hamberg J.
Muscle stretching in manual therapy. Vol. 2.
Alfta: Alfta Rehab Forlag; 1984.
Evjenth O, Kaltenborn F.
Manuell mobilisering av ryggraden (in Norwegian).
Oslo: Forlag Olaf Nordlis Bokhandel; 1989.
Exercises: which ones are worth trying, for which patients, and when?
Fairbank JCT, Couper J, Davies JB, et al.
The Oswestry Low Back Pain Disability Questionnaire.
Hagen EM, Eriksen HR, Ursin H.
Does early intervention with a light mobilization program reduce long-term sick leave for low back pain?
Hillman M, Wright A, Rajaratnam G, et al.
Prevalence of low back pain in the community: implications for service provision in Bradford, UK.
J Epidemiol Commun Health 1996;50:347–52.
Hodges PW, Richardson CA.
Inefficient muscular stabilization of the lumbar spine associated with low back pain. A motor control evaluation of transversus abdominis.
Measurement of pain.
Koes BW, Assendelft WJ, van der Heijden GJ, et al.
Spinal manipulation and mobilisation for back and neck pain: a blinded review.
Br Med J 1991;303: 1298–303.
Koes BW, Bouter LM, Beckerman H, et al.
Physiotherapy exercises and back pain: a blinded review.
Br Med J 1991;302:1572–6.
Koes BW, Bouter LM, van Mameren H, et al.
The effectiveness of manual therapy, physiotherapy, and treatment by the general practitioner for nonspecific back and neck complaints. A randomized clinical trial.
Spine 1992; 17:28–35.
Koes BW, van Tulder MW, Ostelo R, et al.
Clinical guidelines for the management of low back pain in primary care: an international comparison.
Leboeuf-Yde C, Klougart N, Lauritzen T.
How common is low back pain in the Nordic population? Data from a recent study on a middle-aged general
Danish population and four surveys previously conducted in the Nordic countries.
Lindstro¨m I, O¨ hlund C, Eek C.
The effect of graded activity on patients with subacute low back pain. A randomized, prospective clinical study with an operant conditioning behavioral approach.
Phys Ther 1992;72:279–93.
Natural history and clinical role of the herniated disc.
In: Wiesel SW, Weinstein JN, Herkowitz HN, eds.
The lumbar spine. 2nd ed.
Philadelphia, PA: WB Saunders; 1996:473–91.
Manniche C, Hesselsoe G, Bentzen L, et al.
Clinical trial of intensive muscle training for chronic low back pain.
Manniche C, Lundberg E, Christensen I, et al.
Intensive dynamic back exercises for chronic low back pain: a clinical trial.
Meade TW, Dyer S, Browne W, et al.
Low back pain of mechanical origin: randomised comparison of chiropractic and hospital outpatient treatment.
Br Med J 1990;300:1431–7.
Nelson EC, Landgraf JM, Hayes JW.
The COOP Function Charts: a system to measure patient function in physician’s offices.
In: Lipkin M, ed.
WONCA Classification Committee: Functional Status Assessment Measurement in Primary Care.
New York, NY: Springer Verlag; 1990:79–131.
O’Sullivan PB, Twomey L, Allison GT.
Altered abdominal muscle recruitment in patients with chronic back pain following a specific exercise intervention.
J Orthop Sports Phys Ther 1998;27:114–24.
O’Sullivan PB, Twomey LT, Allison GT.
Evaluation of specific stabilizing exercise in the treatment of chronic low back pain with radiologic diagnosis of spondylolysis or spondylolisthesis.
Roland M, Fairbank J.
The Roland-Morris Disability Questionnaire and the Oswestry Disability Questionnaire.
SBU–The Swedish Council on Technology Assessment in Health Care.
Back Pain, Neck Pain. An evidence based review. Summary and conclusions.
Report 145. Stockholm: SBU; 2000.
Seferlis T, Lindholm L, Nemeth G.
Cost-minimisation analysis of three conservative treatment programmes in 180 patients sick-listed for acute lowback pain.
Scand J Prim Health Care 2000;18:53–7.
Statens beredning for medisinsk utva¨rdering (SBU).
Ont i ryggen, ont i nakken.
En evidensbaserad kunskapssammensta¨ llning Vol. I. 145/1: 7–417.
Stockholm: SBU; 2000.
Statens beredning fo¨ r medisinsk utva¨rdering (SBU).
Ont i ryggen, ont i nakken.
En evidensbaserad kunskapssammensta¨ llning Vol. II. 145/2: 11–389.
Stockholm: SBU; 2000.
Strand LI, Ljunggren AE, Haldorsen EM, Espehaug B.
The impact of physical function and pain on work status at 1-year follow-up in patients with back pain.
The Norwegian Social Department.
Trygd og rehabilitering [in Norwegian].
Torstensen TA, Ljunggren AE, Meen HD, et al.
Efficiency and costs of medical exercise therapy, conventional physiotherapy, and self-exercise in patients
with chronic low back pain. A pragmatic, randomized, single-blinded, controlled trial with 1-year follow-up.
van den Hoogen HJ, Koes BW, van Eijk JT, et al.
On the course of low back pain in general practice: a one-year follow up study.
Ann Rheum Dis 1998; 57:13–9.
van Tulder MW.
Treatment of low back pain: myths and facts.
van Tulder MW, Koes BW, Bouter LM.
Conservative treatment of acute and chronic nonspecific low back pain. A systematic review of randomized controlled trials of the most common interventions.
van Tulder MW, Koes BW, Metsemakers JF, et al.
Chronic low back pain in primary care: a prospective study on the management and course.
Fam Pract 1998;15:126–32.
A new clinical model for the treatment of low-back pain.
WONCA international classification committee.
ICPC-2: International classification of primary care. 2nd ed.
Oxford: Oxford University Press; 1998.
Return to the LOW BACK PAIN Section
Return to the COST EFFECTIVENESS JOINT STATEMENT