C u r re n t w o r k p l a c e re s e a rc h a s u p p l e m e n t t o a t w o r k

The Effectiveness of
Acupuncture in Treating
Low Back Pain

This Issue
The supplement:
Every issue of the Institutes At Work
newsletter will be be accompanied by
an issue of Infocus, which will serve to
highlight specific issues or research
findings as they relate to our stakeholders. Future issues will be directed
towards researchers, clinicians, policy
makers, as well as employers,
employees and other workplace parties.

For Clinicians:
Linkages
Transferring Research into Practice
Linkages has been a component of the
Institutes newsletter for the last few
years (this is #6). As part of the
redesign of At Work, the Infocus
section was created to address the need
for scholary commentary such as
Linkages provides. Future issues of
Linkages will be included with Infocus
when available.
Article by: Andrea Furlan, MD
E-mail: afurlan@iwh.on.ca
For additional copies of this newsletter,
please contact Laura Maniago by
e-mail, info@iwh.on.ca

Research Excellence
Advancing Employee Health

ow back pain affects 60 to 85% of

the population at least once in
their life.1 Of these, 1020% develop
chronic low back pain, which is
defined as continuous pain for more
than three months.2 In
1995, occupational low
back pain remained a
major problem in the
U.S.: an estimated $8.8
billion was spent on low
back pain claims, and the
rate of filing low back
pain claims was 1.8 per
100 workers.3 Nachemson
stated that disability from
back pain clearly is a
significant problem in
many countries, but his
report indicates that it is
greater in Canada, Great
Britain, The Netherlands,
and Sweden. In Canada,
2% of the work force is disabled
because of low back pain, accounting
for 20 days absent (per patient per
year)4. Because of the magnitude of
the disorder, a wide variety of

treatments have been used in the

management of low back pain.
Acupuncture is an ancient therapeutic
modality with its basis in traditional
Chinese medicine, using
the concept of vital
energy circulating
throughout the body,
along pathways which are
called meridians.
Acupuncture consists of
the insertion of needles at
specific skin points to
restore normal body
function. These points
can also be stimulated
using various other
techniques. The
mechanisms of
acupuncture analgesia are
partially explained by
conventional
physiological models. Needling creates
electrical impulses that inhibit pain
through connections between the
spinal cord and midbrain. It also
( Continued on page 2

The purpose of Linkages is to critically review the best available evidence in the literature
in the area of soft-tissue injury and to disseminate these reviews to clinicians, labour,
management, government, policy makers and the insurance industry. Articles reviewed in
Linkages will be topical English-language articles in the area of soft-tissue injuries whose
findings, we believe, will be useful and relevant to our stakeholders.

The Institute for Work & Health

is a knowledge-based change
organization that strives to research
and promote: new ways to prevent
workplace disability, improved
treatment, and optimal recovery and
safe return-to-work

Published by
Institute for Work & Health
250 Bloor Street East, Suite 702
Toronto, ON Canada M4W 1E6
Tel: (416) 927.2027
Fax: (416) 927.4167
Email: atwork@iwh.on.ca
Website: www.iwh.on.ca
Manager, Communications
Peter D. Birt
Editor
Chris St. Croix
Design
Vigeon Design & Associates
Issue #15a March 2000
ISSN # 1261-5148
Copyright 2000
The Institute for Work & Health
operates with the support of the

stimulates the release of endogenous

opioids and neurotransmitters such as
serotonin, by the central nerve
system.5
The NIH consensus group in 19976
found weak evidence to support the
use of acupuncture for low back pain
and stated that a more definite
recommendation should await the
results of a Cochrane systematic
review. This has now been published7
and is reviewed in this issue of
Linkages. Another review8 was
published around the same time, and
although the authors included almost
the same studies, they arrived at the
opposite conclusion. There are two
major differences between these two
reviews, which led to our choice of
van Tulders review for discussion.
First, different methods of assessing
methodological quality were used,
resulting in Ernst & White assigning
significantly higher ratings to most of
the trials than van Tulder. Van Tulder

used a scale which was more

comprehensive and therefore identified
more flaws in the trials. Second, the
two groups used different methods to
summarize the results. Ernst & White
used a statistical approach (metaanalysis). Van Tulder refrained from
statistical pooling, choosing instead to
complete a qualitative analysis which
take into account the levels of
evidence. The latter approach is more
appropriate, given the poor quality of
studies included and the clinical
heterogeneity of subjects. Both set of
authors acknowledge that it is
problematic to form a firm judgement
based on the available evidence, and
recommend further studies. In spite of
this stated caution, Ernst & White
quantitatively pooled the results from
these trials and concluded in favour of
acupuncture.

ARTICLE REVIEWED
Tulder MW van, Cherkin DC, Berman
B, Lao L, Koes B. The effectiveness of

The Value of Systematic Reviews

Six million new articles reporting results of biomedical research are published
each year. For health professionals, consumers and policy-makers, it is an
overwhelming task to sift through this information efficiently and to be able
to make decisions on patient care, treatment choice and health policy in a
timely manner. Traditionally, research results have been summarized in nonsystematic narrative reviews. However, these are open to bias as they may
involve the authors subjective opinion of an articles quality. Systematic
reviews offer a better alternative to these narrative reviews as they apply
scientific strategies in ways that limit bias, to the assembly, critical appraisal,
and synthesis of all relevant studies that address a specific clinical question.1
Although systematic reviews should use methods which minimize bias and
error, flaws in their methodology may contribute to invalid results. Therefore,
critically reviewing systematic reviews is as important as reviewing the
individual trials in these reviews. This issue of Linkages reviews a systematic
review article on acupuncture for low back pain. The systematic review was
critically reviewed by two internal Institute reviewers using standard criteria.2
Two external clinical reviewers provided commentaries on the articles
relevance and applicability. We thank those who contributed to these
commentaries.

Ontario Workplace Safety &

Insurance Board

1. Cook DJ, Mulrow CD, Haynes RB. Systematic reviews: Synthesis of best evidence for clinical decisions. Annals of
Internal Medicine. Systematic Review Series. 1997; 126(5): 376380.
2. Oxman AD, Guyatt GH. Validation of an index of the quality of review articles. Journal of Clinical Epidemiology. 1991;
44: 12711278.

acupuncture in the management of acute

and chronic low back pain. A systematic
review within the framework of the
Cochrane Collaboration Back Review
Group. Spine 1999; 24(11):
11131123.
Objective To assess the effects of
acupuncture for the treatment of nonspecific low back pain.
Data Sources They searched the
Cochrane Trials Register, Medline,
Embase, Science Citation Index and
the reference lists of articles.
Study Selection Types of participants:
subjects with non-specific low back
pain were included, either acute or
chronic. Types of intervention: only
needling acupuncture either
traditional (classic meridian points) or
contemporary acupuncture (nonmeridian or trigger points). Types of
outcome measures: at least one of: pain
intensity, a global measure, functional
status or return to work. Types of
studies: only randomized controlled
trials (RCTs).
Methodological quality of the studies
was independently assessed by two
reviewers, blinded with respect to
authors, institution and journal, using
the criteria recommended in the
method guidelines for systematic
reviews by the Cochrane Back Review
Group.9
Analysis The authors decided not to
pool the data statistically, instead they
chose to perform a qualitative review
by assessing the methodological
quality and the outcome of the
original studies and attributing
levels of evidence to the
effectiveness of acupuncture. They
used the following categories:
Strong evidence: consistent findings
in multiple higher quality trials.
Moderate evidence: consistent
findings in one higher quality trial

and one or more lower quality trial.

Limited evidence: consistent findings in
one or more lower quality trial. No
evidence: if there were no trial or if the
results were conflicting. A trial was
considered to be of higher quality if
more than 5 of the 10 validity items
scored positively.
Results Eleven trials were included in
this systematic review. Only two were
of high quality. In eight of the eleven
trials, the individual authors had
concluded that acupuncture was
better than the control group
(including the two high quality trials).
In the remaining three trials they had
concluded that acupuncture was
similar to the control group.
However, van Tulder and his
reviewers disagree with the original
authors conclusions in seven of the
eleven studies. According to van
Tulder, there was no difference
between acupuncture and control in
seven trials; acupuncture was superior
in only two trials (one high and one
low quality); and the results were
unclear in the remaining two trials.
In these 11 trials, acupuncture was
compared to three major control
groups: no treatment, other treatments
and placebo (or sham) acupuncture.
Van Tulder and his reviewers were not
able to draw conclusions about the
effectiveness of acupuncture over no
treatment because the trials provided
conflicting evidence and were of low
methodological quality. They found
moderate evidence that acupuncture is

QUESTIONS ABOUT
LINKAGES?
For more information or copies of
previous issues of Linkages, please
contact: Andrea Furlan at the Institute for
Work & Health, (416) 927-2027 ext
2171, afurlan@iwh.on.ca

The effectiveness of
acupuncture for the
treatment of low back pain
still under
under question.
question.
isis still
There is no clear rigorous
evidence to support its use
as a regular treatment for
low back pain, although
practicioners and patients
may elect a short course to
assess individual response.
not more effective than trigger point
injection or Transcutaneous Electrical
Nerve Stimulation (TENS). And
finally, they found limited evidence
that acupuncture is not more effective
than placebo (or sham) acupuncture
for the management of chronic low
back pain.
Conclusions Because this systematic
review could not clearly conclude that
acupuncture is effective in the
management of back pain, van Tulder
and his reviewers would not
recommend acupuncture as a regular
treatment for patients with low back
pain. There clearly is a need for more
high-quality randomized controlled
trials.

WHAT DOES
THIS MEAN?
The effectiveness of acupuncture for
the treatment of low back pain is
still under question. There is no
clear rigorous evidence to support
its use as a regular treatment for low
back pain, although practicioners
and patients may elect a short
course to assess individual
response.
( Reviewers Comments on page 4

COMMENTARIES
Van Tulders review was conducted according to high methodological
standards. Despite a thorough literature search, the majority of individual trials found and included in this review were judged to be of
low quality. In addition to the methodological problems, it was noted
that some trials did not use a valid acupuncture treatment and that
patients were heterogeneous across trials. For all these reasons Van
Tulder concluded, and we concur with him, that statistical pooling
across trials could be misleading, and that subgroup analyses such as:
acute versus chronic pain or presence versus absence of radiating pain
were innapropriate. On the basis of the evidence gathered, van Tulder
et al. concluded that the effectiveness of acupuncture for low back
pain remains unclear, and therefore they did not recommend it as a
regular treatment for this condition. We agree with this statement, and
concur that additional RCTs (using high-quality methods and
ensuring that a valid acupuncture treatment is used) are needed in
order to make an evidence-based judgement on the effectiveness of
acupuncture for low back pain.
Andrea Furlan, MD., Evidence-Based Practice Co-ordinator, IWH
Judy Clarke, MA, Research Associate, IWH
When a review of a treatment, such as acupuncture, yields insufficient evidence, we must ask how the results can translate into clinical
practice beyond the authors statement that they would not
recommend acupuncture as a regular treatment for patients with low
back pain because the results did not clearly indicate that acupuncture is effective. Considering safety, cost and basic science can provide
more information. Acupuncture is generally safe,10 therefore some
may wish to try it although benefit for low back pain is not proven.
Furthermore, documentation of a possible cumulative effect of
acupuncture for pain relief,11,12 raises a question regarding whether a

minimum number of treatments are necessary to detect benefit. The

most extensive work on this topic suggests somewhere around six to
ten treatments.13 What then can I, as a practitioner, say to patients
interested in acupuncture for low back pain? It is safe, although not
proven to be effective. If you decide to try it, you are likely to know
within 610 sessions if it is helpful.
Jeanette Ezzo, MPH, PhD
Formerly systematic reviews coordinator,
Complementary Medicine Program,
University of Maryland School of Medicine, Baltimore MD
Presently, epidemiology teaching faculty, Project Lead
Analysing any treatment for low back pain, the definition of which
is problematic14, is not easy,15 but it is important to critically appraise
acupuncture, a low risk treatment with significant benefit.16 However,
the decision to critique the qualitative systematic review of the effectiveness of acupuncture for low back pain by van Tulder et al., rather
than the quantitative review by Ernst et al.,17 is puzzling. While both
reviewed virtually the same RCTs, Ernsts more rigorous analysis
showed that the odds ratio of improvement with acupuncture
compared with control interventions was 2.3. For sham-controlled,
evaluator-blinded studies, the odds ratio was 1.37. From an experienced clinicians perspective, van Tulders review offers no valid
information that would discourage the use of acupuncture for low
back pain. Prospective RCTs, using consistent and appropriate
methodology and experienced practitioners are clearly needed.
Linda M. Rapson, MD, CAFCI
Executive President,
Acupuncture Foundation of Canada Institute
Consultant, Acupuncture Clinic,
Toronto Rehabilitation Institute, Lyndhurst Centre

Sources
1. Skovron ML. Epidemiology of low back pain. Baillires Clinical Rheumatology. 1992; 6:
55973.
2. Waddell G. The Back Pain Revolution. Glasgow: Churchill-Livingstone, 1998.
3. Murphy PL and Volinn E. Is occupational low back pain on the rise? Spine 1999;
24(7): 691697.
4. Nachemson AL. Newest knowledge of low back pain: a critical look Clin. Orthop
1992; 279: 820.
5. Vickers A, Zollman C. ABC of complementary medicine: Acupuncture. BMJ 1999;
319: 97376.
6. Acupuncture. NIH Consensus Statement 1997. Nov 3-5; 15(5): 134.
7. Tulder MW van, Cherkin DC, Berman B, Lao L, Koes B. The effectiveness of acupuncture in the management of acute and chronic low back pain. A systematic review within
the framework of the Cochrane Collaboration Back Review Group. Spine 1999; 24(11):
11131123.
8. Ernst E, White AR. Acupuncture for back pain. A meta-analysis of randomized
controlled trials. Arch Intern Med 1998; 158: 22352241.
9. Tulder MW van, Assendelft WJ, Koes BW, Bouter LM, et al. Method guidelines for

Feedback

on

FUTURE ISSUES OF INFOCUS... will provide insight into

contempory issues of workplace safety and health as they relate

10.
11.
12.
13.
14.

15.
16.
17.

systematic reviews in the Cochrane Collaboration Back Review Group

for Spinal Disorders. Spine. 1997; 22(20): 23232330.
See Note 6.
Price D, Raffi A, Watkins L and Buckingham B. A psychophysical
analysis of acupuncture analgesia. Pain 1984; 19: 2742
Han JS. Acupuncture activates endogenous systems of analgesia. Presentation to the NIH
Consensus Panel on Acupuncture, Bethesda. November 3, 1997.
Birch S. An exploration with proposed solutions of the problems and issues in conducting
clinical research in acupuncture. University of Exeter, Exeter, U.K. 1997.
Cedraschi C, Nordin M, Nachemson AL, Vischer TL. Health care providers should use
a common language in relation to low back pain patients. Baillires Clinical
Rheumatology. 1998; 12 (1): 115.
Andersson GBJ. Epidemiology of low back pain. Acta Orthop Scand (Suppl 281) 1998;
69: 2831.
See Note 8.
Ibid.

to our audience. Along with healthcare providers, we will focus

on researchers and policy makers, as well as employers,
employees and other workplace parties.
If you would like to comment on this publication, please
contact the editor by e-mail at atwork@iwh.on.ca.