Spine (Phila Pa 1976). 2011 (Oct 1); 36 (21 Suppl): S120–130 ~ FULL TEXT
Christopher J. Standaert , MD , Janna Friedly , MD , Mark W. Erwin , DC, PhD,
Michael J. Lee , MD, Glenn Rechtine , MD , Nora B. Henrikson , PhD, MPH,
and Daniel C. Norvell , PhD
Department of Rehabilitation Medicine,
University of Washington,
Seattle, WA 98104, USA.
STUDY DESIGN: Systematic review.
OBJECTIVE: We sought to answer the following clinical questions:
(1) Is structured exercise more effective in the treatment of chronic low back pain (LBP)
than spinal manipulative therapy (SMT)?
(2) Is structured exercise more effective in the treatment of chronic LBP than acupuncture?
(3) Is SMT more effective in the treatment of chronic LBP than acupuncture?
(4) Do certain subgroups respond more favorably to specific treatments?
(5) Are any of these treatments more cost-effective than the others?
SUMMARY OF BACKGROUND DATA: Exercise, SMT, and acupuncture are widely used interventions in the treatment of chronic LBP. There is evidence that all of these approaches may offer some benefit for patients with chronic LBP when compared with usual care or no treatment. The relative benefits or cost-effectiveness of any one of these treatments when compared with the others are less well-defined, and it is difficult to identify specific subgroups of those with chronic LBP who may preferentially respond to a particular treatment modality.
METHODS: A systematic review of the literature was performed to identify randomized controlled trials comparing a structured exercise program, SMT, or acupuncture with one another in patients with chronic LBP.
RESULTS: Two studies were identified comparing the use of structured exercise with SMT that met our inclusion criteria. Although these studies utilized different approaches for the exercise and SMT treatment groups, patients in both groups improved in terms of pain and function in both studies. Using random-effects modeling, there was no difference between the exercise and SMT groups when the data from these studies were pooled. We identified no studies meeting our inclusion criteria that compared acupuncture with either structured exercise or SMT or that addressed the relative cost-effectiveness of these approaches in the treatment of patients with chronic LBP.
CONCLUSION: The studies identified indicate that structured exercise and SMT appear to offer equivalent benefits in terms of pain and functional improvement for those with chronic LBP with clinical benefits evident within 8 weeks of care. However, the level of evidence is low. There is insufficient evidence to comment on the relative benefit of acupuncture compared with either structured exercise or SMT or to address the differential effects of structured exercise, SMT, or acupuncture for specific subgroups of individuals with chronic LBP. There is also insufficient evidence regarding the relative cost-effectiveness of structured exercise, SMT, or acupuncture in the treatment of chronic LBP.
CLINICAL RECOMMENDATIONS: Structured exercise and SMT appear to offer equivalent benefits in the management of pain and function for patients with nonspecific chronic LBP. If no clinical benefit is appreciated after using one of these approaches for 8 weeks, then the treatment plan should be reevaluated and consideration should be given to modifying the treatment approach or using alternate forms of care. Strength of recommendation: Weak.There is insufficient evidence regarding the relative benefits of the acupuncture compared with either structured exercise or SMT in the treatment of chronic LBP.There is insufficient evidence to address differential effects of structured exercise, SMT, or acupuncture for specific subgroups of individuals with chronic LBP. There is insufficient evidence regarding the relative cost-effectiveness of structured exercise, SMT, or acupuncture in the treatment of chronic LBP.
Key words: acupuncture, chiropractic, comparative effectiveness,
cost-effectiveness, exercise, low back pain, spinal manipulative therapy, systematic review
From the FULL TEXT Article
Most people with chronic low back pain (LBP) are
managed nonoperatively, but it is very difficult to
know which, if any, of the myriad of treatment
options available will be the most effective for a given patient.
In part, this is related to the heterogeneity of the patient population.
LBP can arise from a number of different anatomic
structures or can be related to a large range of pathologic conditions
in the spine, and people have highly variable responses
to ongoing pain. Chronic pain can also be associated with
significant physiologic, neurobiologic, and psychosocial
changes that may vary substantially between individuals. If
these multiple factors are considered alongside the large range
of treatment options available, then it can be extremely challenging
to match the right treatment to the right patient. The
cost of this uncertainty is hardly insignificant. A recent review
by Dagenais et al  found that nonphysician care, including
physical therapy, chiropractic, and complementary alternative
medicine, accounted for 30% of the direct medical costs
for care of those with chronic LBP, equivalent to the amount
attributed to inpatient, outpatient, surgical, and emergency
medical care combined. Physical therapy alone accounted for
17% of the expenditures on chronic LBP, while the costs for
surgery accounted for only 5%. 
Although there is clearly a critical need to identify which
treatment options optimize clinical utility and cost-effectiveness
for specific patients, much of the literature on the nonoperative
management of chronic LBP consists of uncontrolled
trials or comparisons of a particular intervention to ineffective
forms of “usual care.” In addition, study patients are often
clustered into mixed groups that dilute the effect of treatment
on subsets of the population or they are selected by subjective
criteria that limit the external validity of the findings.
The purpose of this review is to assess the literature regarding
comparative studies of specific noninterventional, nonoperative
treatment approaches for chronic LBP, and to advance the
process of selecting between specific treatments.
Three distinct, commonly utilized therapeutic modalities
were chosen for study: exercise, spinal manipulative therapy
(SMT), and acupuncture. These particular treatments were
selected on the basis of author consensus after initial assessment
of the available literature. Multiple systematic reviews
are available on these approaches that provide an appropriate
background for further evaluation. For each of these
treatments, there are significant variations in how they are
defined, applied, or practiced, as well as in the skill level and
training of providers. In addition, they are all accompanied
by a number of nonspecific effects through the very nature of
how they are delivered. This does not limit the importance of
understanding their role in clinical care, but it does make the
assessment of their benefits more challenging.
Exercise programs come in many forms and are delivered
in a variety of settings, frequently with multiple cointerventions.
Several recent systematic reviews have addressed the
role of exercise in the treatment of LBP. In the work of van
Middelkoop et al,  the authors concluded that the exercise
was effective at reducing pain and increasing function compared
with usual care with a small effect size. There was no
evidence to indicate that one exercise approach was superior
to others; however, and there were inadequate data on effects
within specific subgroups. Although they also concluded
that there was no difference between exercise and SMT in
short- or long-term benefits for LBP, the studies included in
their review did not all address comparisons between an isolated,
structured exercise approach and SMT, particularly the
delivery of high-velocity manipulation by trained providers.
Cochrane reviews on exercise as a treatment for chronic LBP 
and for preventing recurrences of LBP  have concluded that
exercise is “slightly effective” at reducing pain and increasing
function in adults with chronic LBP and that post-treatment
exercise programs are more effective than no treatment at
preventing recurrences of LBP. Overall, the level of evidence
is low in these reviews, and the quality of the literature is relatively
poor and hampered by heterogeneity of both treatment
approaches and patient populations.
The most recent Cochrane review of spinal manipulation in
chronic LBP  concluded that SMT results in a small, statistically
significant but not clinically significant improvement
in pain and function in patients with chronic LBP compared
with other treatments. The authors identified evidence of
varying levels of quality indicating that SMT has a short-term
effect on pain and disability when added to another treatment
and felt there was a significant need for cost data. As with
the exercise literature, this study and the work on which it is
based are affected by heterogeneity in the patient populations
studied and in the manipulative technique applied. What is
referred to as SMT may include high-velocity thrust techniques,
manual mobilization, or other specific techniques or
even broad treatment approaches such as osteopathy, which
cloud the actual treatment effect of specific manipulative techniques.
Although a number of studies have assessed the physiologic
effects of SMT, the proper role of this treatment modality
remains elusive, as does the most appropriate use of this
modality as compared with other nonoperative methods. [6–10]
Traditional acupuncture is based on ancient Chinese philosophical
beliefs regarding the flow of vital energy through the body
along the discrete pathways termed meridians. In acupuncture
treatment, specific points along these meridians are utilized to
balance the energy flows within the body. Many different styles
of acupuncture and adjunctive techniques have developed over
time as its use has disseminated into other cultures.  Although
its mechanism of action is poorly understood, acupuncture is
widely utilized in the treatment of LBP.  Acupuncture may be
superior or equivalent to medical care,  waiting list control, 
or usual care. [15, 16] A systematic review by Furlan et al  found
acupuncture to be more effective than no treatment or sham
in the management of chronic LBP but not more effective than
other conventional approaches. The authors of this study noted
low-methodologic quality for most of the studies identified.
Given dramatic increases in medical costs and the questionable
efficacy of a variety of treatments for LBP, [17, 18] it is
increasingly important to include cost-effectiveness analyses
in comparative effectiveness research. To date, the cost-effectiveness
of many treatments for LBP, including SMT
and acupuncture compared with structured exercise, has
not been well-defined. Two recent systematic reviews have
examined cost-effectiveness of a variety of treatments for
LBP and highlight the major limitations of the existing
literature on this topic. Dagenais et al  identified only 15
cost-utility analyses for any treatments associated with LBP,
including surgical and nonsurgical interventions. Of these,
no studies included SMT, two studies included acupuncture,
and only two were performed in the United States. The
authors concluded that there was not enough high-quality
evidence to determine the relative cost-effectiveness of most
treatments for LBP, including SMT and acupuncture. Most
analyses were performed in the United Kingdom or other
European countries with markedly different health care systems
than the United States.
A similar systematic review by Lin et al  evaluated treatments
endorsed by the clinical practice guidelines of the
American College of Physicians and the American Pain Society,
including interdisciplinary rehabilitation, exercise, acupuncture,
massage, SMT, yoga, cognitive behavioral therapy,
or relaxation for subacute or chronic LBP. This review identified 26 studies in total, although no additional studies were
identified comparing SMT or acupuncture with other treatments
for chronic LBP beyond those included in the review
of Dagenais et al.  Despite the overlap in the studies that
formed the basis of these reviews, Lin et al  reached conclusions
contrary to those of Dagenais et al  finding that a variety
of treatments for chronic LBP, including acupuncture and
SMT, are relatively cost-effective. The inconsistencies in these
reviews are emblematic of the need for high-quality cost-effectiveness
studies comparing treatments for LBP, including
exercise, SMT, and acupuncture, a need that is particularly
acute within the context of the US health care system.
The intent of this review was to evaluate the relative efficacy
of structured exercise, SMT, and acupuncture in the treatment
of chronic LBP and to assess the cost-effectiveness of
these approaches. We sought to answer the following clinical
Is structured exercise more effective in the treatment of chronic LBP than SMT?
Is structured exercise more effective in the treatment of chronic LBP than acupuncture?
Is SMT more effective in the treatment of chronic LBP than acupuncture?
Do certain subgroups respond more favorably to specific treatments?
Are any of these treatments more cost-effective than the others?
MATERIALS AND METHODS
Electronic Literature Database
We conducted a systematic search in MEDLINE and the
Cochrane Collaboration Library for literature published
through December 2010. The search results were limited to
human studies published in the English language. Reference
lists of key articles were also systematically checked to identify
additional eligible articles. Considering the substantial
literature on interventions for LBP, we limited our search to
randomized controlled trials (RCTs) and meta-analyses that
studied the interventions of interest. We included studies
evaluating adult patients with chronic LBP with or without
radiating pain for at least 3 months. In contrast to other published
systematic reviews, we sought to identify studies where
a clear comparison could be made between the interventions
of interest (exercise, manipulation, and acupuncture) to avoid
heterogeneity that often leads to little or no observed effect
or effects that cannot be generalized to the clinical setting.
Hence, we excluded studies with combination treatments
(e.g. , manipulation and exercise, multimodality physical
therapy). To avoid bias, we also excluded studies with selection
criteria that were either too subjective, not applied in the
typical clinical setting, or too strict to allow generalization
to the larger population of patients with LBP (e.g., pain in a
specific area of the spine required for inclusion) and studies
whose data presentation did not allow separate analysis only
of those with chronic LBP (Table 1).
For question 4, we assessed the RCTs that met the study
criteria for questions 1 to 3 to identify those that included
subgroup analysis stratifying one or more subgroups. Papers
had to provide raw data for each treatment intervention by
subgroup so that the heterogeneity of treatment effects could
be evaluated, unless the authors reported specific tests for
For question 5, we searched for economic evaluation studies
that compared treatments defined in questions 1 to 3. We
limited our results to cost-effectiveness or cost-utility analyses
and excluded cost-only studies. We excluded studies if the
treatments of interest were not randomized, pediatric studies
(< 18 years of age), and patients with cancer, inflammatory or
systemic disease, instability, infection, or pregnancy. We also
excluded studies where the treatment arms included multiple
treatment modalities. We were interested in studies that evaluated
the costs and the effectiveness of the specific treatment
comparisons in questions 1 to 3.
Inclusion and Exclusion Criteria for Study Questions 1–3 (PICO)
Article Selection and Data Extraction
Two reviewers (N.H. and D.N.) independently reviewed each
retrieved citation. After title and abstract review, we selected
articles for full-text review that were
(1) likely eligible articles or
(2) could not be excluded unequivocally from the title or abstract.
Full-text articles were reviewed independently by four
reviewers (C.S., M.E., N.H., and D.N.); disagreements were
resolved by consensus. Final-included articles were decided by
consensus among the entire author team (Figures 1 and 2).
From the included articles, we extracted the following information:
study design, population, diagnosis, intervention description, outcome measures, and relevant results.
Flow chart showing the results of literature
search identifying RCTs comparing each of the
three treatment interventions.
Flow chart showing the results of literature
search identifying cost-effectiveness studies associated
with the three comparisons.
Level of evidence ratings were assigned to each included
article independently by two reviewers using criteria set by
The Journal of Bone and Joint Surgery, American Volume (J Bone Joint Surg Am) 21 for therapeutic studies and modified to delineate criteria associated with methodologic quality and
described elsewhere. 
We performed all analyses on a study level. For pain visual
analogue scales, we standardized all scales on a 0- to 10-point
scale. We calculated the change scores and the corresponding
standard deviations from pre- and postoperative pain and
function scores reported by the authors. Study level data were
pooled across studies if the treatment comparisons and outcomes
were the same or similar and if sufficient data were
provided to calculate effect sizes. The standardized mean differences
(SMDs) comparing the overall treatment effects were
calculated by subtracting the mean change scores and dividing
by the change score standard deviations. The reporting
of effect estimates facilitates the interpretation of the size of
the effect of a specific treatment as opposed to the statistical
significance. Both fixed-effect and random-effect estimates
were performed for pooled data. Fixed-effects methods are
based on the assumption that a single common effect underlies
each study being pooled with no significant heterogeneity
among the studies. Random-effects methods are based on
the assumption that individual studies are estimating different
treatment effects, reflective of heterogeneity. That is, random effects
methods allow that the true effect could vary from
study to study which accounts for both within- and between study
Overall Strength of Body of Literature
The initial strength of the overall body of evidence was considered
high if the majority of the studies were level 1 or 2 and
low if the majority of the studies were level 3 or 4. We downgraded
the body of evidence one or two levels based on the
(1) inconsistency of results,
(2) indirectness of evidence, or
(3) imprecision of the effect estimates (e.g., wide confidence intervals).
We upgraded the body of evidence
one or two levels based on the following criteria:
(1) large magnitude of effect or
(2) dose-response gradient.
The overall strength of the body of the literature was expressed in terms
of our confidence in the estimate of effect and the impact that
further research may have on the results. An overall strength
of “high” means we have high confidence that the evidence
reflects the true effect. Further research is very unlikely to
change our confidence in the estimate of effect. The overall
strength of “moderate” means that we have moderate confidence that the evidence reflects the true effect. Further research
may change our confidence in the estimate of effect and may
change the estimate. A grade of “low” means we have low
confidence that the evidence reflects the true effect. Further
research is likely to change the confidence in the estimate of
effect and likely to change the estimate. Finally, a grade of
“insufficient” means that evidence either is unavailable or
does not permit a conclusion. A more detailed description of
this process can be found in the “Materials and Methods”
Manipulation Versus Exercise
For question 1 (manipulation vs. exercise), our systematic
search identified 18 potential studies and 2 systematic reviews
(Figure 1). We identified two additional studies published
since the most recent review.  Among the 18 potential studies,
we selected 7 for full-text review. [23–29]
Among these, five were excluded for the following reasons: subjective selection criteria
(tenderness over facet joints required),  lack of a clearly
defined spinal manipulation group,  treatment intervention
not randomized,  no structured exercise group and inclusion
of patients with less than 3 months of LBP,  and lack of a clean
exercise versus manipulation comparison.  Consequently, this
question included two studies: one identified in the review by
van Middelkoop et al  (Ferreira et al ) and the other published
since that review (Cecchi et al. ).
Details of study characteristics and patient populations are in the supplementary digital
material (see supplementary tables, Supplemental Digital Content 1, and
Supplemental Digital Content 2.
In the first included article, Ferreira et al  conducted an
RCT of 240 adults with chronic LBP comparing SMT (defined
as joint mobilization or manipulation applied to the spine or
pelvis by a physical therapist), general exercises performed
under supervision, and motor control exercises (focused on
specific trunk muscles thought to control intersegmental
motion in the lumbar spine).  Treatment was performed over
an 8-week period with initial follow-up at 8 weeks and final
follow-up at 12 months. Study outcomes were the patient specific functional scale, the global-perceived effect scale, pain
(visual analog scale), Roland-Morris Disability Questionnaire
score, and global-perceived effect scale. We summarized the
visual analog scales and the Roland-Morris Disability Questionnaire
scores as the other study in our review also reported
these measures. Clinical improvement was noted in all groups
at 8 weeks. At 12 months, there was no significant difference
when comparing the motor control exercise group and the
SMT group with respect to changes in pain (SMD = – 0.26;
95% confidence interval [CI]: – 0.57 to 0.51;) Table 2 and Figure 3). With respect to changes in function, the motor control
exercise group was favored (SMD = – 2.07; 95% CI:
– 2.46 to – 1.69) (Table 2 and Figure 4). Similar findings were
observed with general exercise compared with SMT, although
the relative change in function was less than that noted for the
motor control exercise group (Table 2). The authors reported
no significant difference between groups with respect to the
patient-specific functional scale and global-perceived effect
Mean Baseline, Change Scores, and Standardized Individual |
and Pooled Mean Differences for Pain and Functional Outcomes
Forest plot representing the SMD and
95% CI comparing the motor control exercise with
the manipulation with respect to pain using a VAS
in two studies meeting criteria (random-effects model).
Forest plot representing the SMD and
95% CI comparing the motor control exercise with
the manipulation with respect to function using
the Roland-Morris Disability Index in two studies
meeting criteria (random-effects model).
In the second included study, Cecchi et al  conducted
an RCT of patients with chronic LBP assigned to receive a
3-week intervention of either back school (including “individual
tailored” exercises) or mixed physiotherapy or a 4- to
6-week SMT intervention (performed via a “manual medicine
approach” by physical medicine and rehabilitation specialists
with approximately 10 years of experience).  Follow-up
was for 12 months; the primary study outcomes were the
Roland-Morris Disability Questionnaire score and secondary
outcomes were pain, pain recurrence, further back pain
treatment, and sick leave. Although change in pain (SMD =
5.32; 95% CI: 4.61–6.04) and function (SMD = 2.32; 95%
CI: 1.89–2.76) favored the SMT group (Table 2 and Figures 3
and 4), further treatment for LBP during the follow-up period
was significantly more frequent in the SMT group. Approximately
50% of this group received further SMT, and the exercise
program provided in the “back school” was limited in
scope and delivered only over the final 2 weeks of the program.
In addition, a greater proportion of patients in the
“back school” group were on sick leave due to their back
pain at study inception.
The pooled analysis used the motor control exercise group
instead of the general exercise group from the Ferreira et al
study  since the effects were larger. We chose not to pool
the two exercise groups from this study as they were distinct
from one another. Any effect added by the general exercise
group would move the difference closer to the null. When the
two studies were pooled, spinal manipulation was favored
slightly over motor control exercise (SMD = 0.632; 95% CI:
0.347–0.917) with respect to pain when using a fixed-effect
model and no significant difference was observed when using
a random-effects model (SMD = 2.52; 95% CI: – 2.95–8.0)
(Table 2 and Figure 3). We are relying on the random-effects
model for our conclusions due to the substantial differences
between both the SMT and exercise interventions in the two
studies and the significant heterogeneity in the treatment
effects. Neither intervention was favored with both the fixed and
random-effects models with respect to function (Table 2, Figure 4).
Acupuncture Versus Exercise
For question 2 (acupuncture vs. exercise), our initial search
yielded 12 potential articles, 10 of which were excluded on
the basis of our criteria (five did not include adequate exercise
comparison group; three were not RCTs; two assessed
pregnancy-related LBP). Two articles were selected for full text
review; however, neither was selected for inclusion in
our review [13, 30] (Figure 1). The study by Haake et al  included
1,162 patients and compared verum acupuncture, sham acupuncture,
and “conventional care.” This article was excluded
because the comparison group received a combination of
drugs, physical therapy, and exercise. Interestingly, this study
found that the “true” and “sham” acupuncture groups
had equivalent degrees of improvement compared with the
“conventional care” group. A smaller study of 52 patients
by Yeung et al  comparing exercise alone to exercise with
electro-acupuncture was excluded, because acupuncture was
delivered as part of a combined treatment approach.
Acupuncture Versus Manipulation
For question 3 (acupuncture vs. manipulation), our initial
search yielded 17 potential articles, 14 of which were excluded
because they were not RCTs. Three were selected for full-text
review; one was excluded because it did not include interventions
of interest.  The remaining two articles were based on
the same RCT [32, 33] and were excluded because patients had
spine problems affecting the cervical, thoracic, and/or lumbar
regions, not isolated chronic LBP, and the data were not presented
in a manner that allowed for isolation of the patients
with only LBP (Figure 1).
Subgroups and Cost-Effectiveness
For question 4 (subgroups), we examined the studies
included for questions 1, 2, and 3 (total = two studies).
Neither study included subgroup analyses. For question
5, cost-effectiveness studies evaluating treatment of
chronic LBP, our initial search yielded 79 citations. Ten
were selected for full-text review. All 10 articles were
subsequently excluded from our review: two were review
or methods articles [19, 34]; two did not meet our inclusion
criteria for economic evaluations [35, 36]; two did not meet
our patient-group criteria [37, 38]; and four did not meet our
intervention criteria (Figure 2). [34, 39–41]
The overall strength of the evidence evaluating the comparative
effectiveness of the structured exercise versus SMT was
“low.” That is, we have low confidence that the evidence
reflects the true effect and further research is likely to change
the confidence in the estimate of effect and likely to change
the estimate (Table 3). The overall strength of the evidence
evaluating the comparative effectiveness of the structured
exercise versus acupuncture, SMT versus acupuncture, comparative
subgroup effects, and cost-effectiveness comparing
the aforementioned treatments was “insufficient.” That is,
evidence is either unavailable or does not permit a conclusion.
Rating of Overall Strength
of Evidence for Each Key Question*
Despite the high prevalence of LBP and the substantial societal
and individual costs associated with it, there is very little
clinical evidence from which to base comparative decisions
for non-operative therapeutic options for a given patient.
There are studies that support the benefit of many distinct
approaches to chronic LBP, although there are also many
studies that identify significant limitations for most treatments.
There may well be substantial benefit to simply intervening in
the pain state of individuals with chronic LBP, and the benefit
associated with many treatments may be related to nonspecific effects. When considered as a whole, the evidence regarding
the superiority of one treatment above others for specific
patients is lacking. The problem is not necessarily that all
non-operative approaches lack efficacy, but, rather, the problem
may lie in the complexity of LBP and in the methodologic
shortcomings of existing clinical studies.
Given the psychosocial complexities often associated
with chronic pain, treatment approaches that address the
psychologic functioning or fundamental belief systems of
the patient regarding injury and recovery may be an important
component of care for those who fail to respond to initial
measures or who clearly have difficulties in this regard.
Although multimodal approaches did not fall within the
scope of this review, cognitive behavioral therapy may be a
beneficial component of care for many patients when delivered
in conjunction with more active treatments. [42, 43] Cognitive
behavioral therapy is not necessarily a uniform therapeutic
approach and can be delivered in many settings but generally
has the primary goal of replacing maladaptive coping skills,
beliefs, and behaviors with more appropriate systems. It is
often a component of multidisciplinary pain programs and
can also accompany structured exercise or physical therapy
programs. [42, 44]
Overall, either comprehensive registries allowing for subject
stratification or comparative clinical studies with rigorous
design, identification of valid and identifiable subgroups and
treatment arms with clearly defined interventions are a very
high priority if we are to use nonoperative measures in the
most efficacious and cost-effective manner.
EXERCISE VERSUS SPINAL MANIPULATIVE THERAPY
Exercise as a specific therapeutic intervention to treat LBP
clearly could entail a broad range of approaches, and it
may well be the case that different exercise programs are
more appropriate for certain individuals. However, the current
literature is not supportive of one particular exercise
approach being superior to others in the treatment of chronic
LBP. There may well be generalized effects from any type of
structured exercise that infer benefits for those with chronic
LBP. [2, 3, 45] Numerous RCTs have demonstrated the overall
benefit of exercise to general health and a number of authors have
noted the importance of overall fitness in LBP care and prevention. [46–49]
Similarly, many studies of SMT use variable approaches
that are termed “manipulation,” often delivered with co-interventions
or within the framework of a specific method of
practice. In a more defined sense, SMT is a highly specific
maneuver that requires considerable training and experience.
A number of published studies indicate that SMT has
some effectiveness in the treatment of acute and chronic back
pain with respect to pain and return to function.  Clear
delineation of the relative benefits of SMT and exercise would
be beneficial for clinicians and patients.
We identified two studies meeting our inclusion criteria
that compared the structured exercise with the SMT in the
treatment of chronic LBP. Although the exercise and SMT
interventions differed substantially between the studies, both
treatment groups showed improvements in pain and function
in both studies, indicating some potential benefit of each
approach. However, the data provided by these studies are
insufficient to establish if one of these treatments is superior
to the other in the treatment of chronic LBP. As the clinical
benefits of these approaches were evident within 8 weeks
in the studies identified, it would seem reasonable to expect
some degree of clinical improvement within this time frame
for a given treatment. If there is no apparent benefit within 8
weeks, then the care plan should be reevaluated and consideration
should be given to modifying the specific treatment
approach or to utilizing alternate forms of care.
ACUPUNCTURE VERSUS MANIPULATION OR EXERCISE
Although there is some evidence that acupuncture may
potentially offer benefit for those with LBP,  we did not
identify any RCTs fulfilling our inclusion criteria that compared
acupuncture with the manipulation or structured
exercise in the treatment of chronic LBP. In the course of
our literature review, we did encounter two RCTs comparing
acupuncture to sham or the simulated acupuncture with
“usual back care” [13, 51] that have been published since the
date of the literature review by Furlan et al.  Both of these
studies observed significant but equivalent benefit for acupuncture
and sham or simulated acupuncture over “usual
care” in the treatment of LBP. The lack of observed differences
between acupuncture and simulated or sham acupuncture
raises significant questions regarding the mechanism
of action of this treatment, including the possibility
of a dominant role of placebo response or other nonspecific effects. Additional high-quality comparative trials are
needed to define the role of acupuncture in the treatment
of chronic LBP, and these should likely include a sham or
simulated treatment arm.
As the number of comparative studies meeting our inclusion
criteria was small, there were insufficient data from which
to identify any subgroup that benefited from any specific
treatment. This finding speaks to the substantial limitations
of the medical literature in this regard. It may also
speak to substantial problems in our medical classification
of patients with chronic LBP. It is clear that the broad categorization
of patients as having “chronic LBP” lacks the
granularity needed for the researchers and the clinicians to
be able to adequately study specific therapies. Further work
on establishing clinically relevant, reliable methods of subgrouping
individuals with chronic LBP is critically necessary.
Appropriately structured registries may be one means
of acquiring data that allows for the identification of valid
Unfortunately, there are very few published studies on the
cost-effectiveness of treatments for LBP in general. The recent
systematic reviews on this topic [19, 20] included studies on a wide
variety of treatment modalities and came to conflicting conclusions
despite largely relying on the same limited literature
base. There is also controversy about the minimum effectiveness
threshold (or alternatively, the maximum cost per quality
adjusted life year gained) that is acceptable for the treatment
of pain. This threshold could differ substantially depending
on the severity and chronicity of symptoms, prevalence of disease,
and societal views about rationing care. [52, 53]
We were unable to identify any studies of cost-effectiveness
comparing exercise, SMT, or acupuncture that met our inclusion
criteria. Although not specifically addressing chronic
LBP, the UK back pain exercise and manipulation trial  was
an interesting four-arm RCT that compared “best care” (i.e. ,
a standard of care providing education) with “best care” with
structured group exercise, SMT, or SMT with group exercise.
This study found that SMT, SMT with exercise, and “best
care” could all potentially be considered cost-effective treatments
for LBP depending on the maximum cost per quality
adjusted life year threshold chosen (> £8700, between £8700
and £3800, or < £3800, respectively). The exercise-alone
group demonstrated the clinical improvements as compared
with “best care”, but to a lesser degree than the SMT groups
and at a higher cost than the SMT with the exercise group.
These higher costs for the exercise group were in large part
due to the costs of additional physical therapy sessions as well
as medical care that this group received. This study highlights
the importance of defining cost-effectiveness thresholds to
interpret comparative cost data. Clearly, there is a great need
for additional cost-effectiveness studies of structured exercise,
acupuncture, and SMT for chronic LBP.
Exercise therapy, SMT, and acupuncture may all have a
role in the clinical management of patients with the chronic
LBP. The overall evidence of benefit for any of these is not
profound however, and we were unable to identify any evidence
of comparative benefit of one of these over the others.
Given the limitations of the literature, we are also unable
to determine if they are truly equivalent, either. Acupuncture,
in particular, is poorly addressed in the current literature,
and evidence of equivalency between acupuncture and
sham or simulated acupuncture in the treatment of LBP has
implications regarding the use and mechanism of action of
this treatment as well as for future study.
Further data are clearly necessary, and the establishment
of the broad registries and the performance of high-quality
studies on the comparative effectiveness of exercise, SMT,
and acupuncture in the treatment of chronic LBP would
appear to be priorities for clinicians, payers, and patients.
Ideally, further clinical studies should include well-defined
treatment arms, an identifiable and clinically relevant study
population, appropriately trained providers, and enough
subjects to allow for subgroup analyses and power to detect
real differences. For exercise, a structured, active treatment
program without co-interventions should be considered.
SMT should consist of a clearly defined method of manipulation
performed by the appropriately trained and the experienced
providers. Given the literature discussed above on
acupuncture, treatment arms of established evaluation and
treatment methods as well as sham or simulated acupuncture
should be included. Either identifying valid subgroups
a priori or obtaining thorough baseline demographics and
diagnostic/clinical information in a large study group to
allow post hoc analysis will be essential for identification of
differential benefits for sub-populations. Cost data and long-term
outcomes should be included in such studies to reliably
assess clinical outcomes and cost-effectiveness over time.
Structured exercise and SMT appear to offer equivalent benefits in the treatment
of chronic LBP, although the level of evidence in this regard is low.
If there is no clinical improvement perceived within 8 weeks of initiating SMT
or a particular exercise treatment, then the care plan should be reevaluated.
There is insufficient evidence in the medical literature to determine the relative
effectiveness of the acupuncture compared with either exercise or SMT
in the treatment of chronic LBP.
There is insufficient evidence in the medical literature to determine the relative
cost-effectiveness of exercise, SMT, and acupuncture in the treatment of chronic LBP.
Comprehensive registries and comparative clinical studies with rigorous design,
identification of valid and identifiable subgroups, and treatment arms with
clearly defined interventions are necessary if we are to optimize the clinical
and cost-effectiveness of non-operative care for the chronic LBP.
Dagenais S , Caro J , Haldeman S. A systematic review of low back
pain cost of illness studies in the United States and internationally.
Spine J 2008; 8: 8–20.
van Middelkoop M , Rubinstein SM , Verhagen AP , et al. Exercise
therapy for chronic nonspecific low-back pain. Best Pract Res Clin
Rheumatol 2010; 24: 193–204.
Hayden JA , van Tulder MW , Malmivaara A , et al. Exercise therapy
for treatment of non-specific low back pain. Cochrane Database
Syst Rev 2005; CD000335.
Choi BK , Verbeek JH , Tam WW , et al. Exercises for prevention
of recurrences of low-back pain. Cochrane Database Syst Rev
Rubinstein SM , van Middelkoop M , Assendelft WJJ , de Boer MR ,
van Tulder MW. et al. Spinal manipulative therapy for chronic
low-back pain. Cochrane Database Syst Rev 2011;(2): CD008112.
Haavik Taylor H , Murphy B. The effects of spinal manipulation
on central integration of dual somatosensory input observed after
motor training: a crossover study. J Manipulative Physiol Ther
2010; 33: 261–72.
Bialosky JE , Bishop MD , Robinson ME , et al. Spinal manipulative
therapy has an immediate effect on thermal pain sensitivity in
people with low back pain: a randomized controlled trial. Phys
Ther 2009; 89: 1292–303.
Fritz JM , Koppenhaver SL , Kawchuk GN , et al. Preliminary
investigation of the mechanisms underlying the effects of
manipulation: exploration of a multi-variate model including spinal
stiffness, multifi dus recruitment, and clinical findings [published
online ahead of print March 15, 2011]. Spine (Phila Pa 1976). doi:
Webb AL , Collins P , Rassoulian H , et al. Synovial folds—a pain in
the neck? Man Ther 2011; 16: 118–24.
Dahm KT , Brurberg KG , Jamtvedt G , et al. Advice to rest in bed
versus advice to stay active for acute low-back pain and sciatica.
Cochrane Database Syst Rev 2010;(6): CD007612.
Furlan AD , van Tulder M , Cherkin D , et al. Acupuncture and dryneedling
for low back pain: an updated systematic review within
the framework of the Cochrane collaboration. Spine (Phila Pa
1976) 2005; 30: 944–63.
Cherkin DC , Sherman KJ , Deyo RA , et al. A review of the evidence
for the effectiveness, safety, and cost of acupuncture, massage
therapy, and spinal manipulation for back pain. Ann Intern Med
2003; 138: 898–906.
Haake M , Muller HH , Schade-Brittinger C , et al. German Acupuncture
Trials (GERAC) for chronic low back pain: randomized,
multicenter, blinded, parallel-group trial with 3 groups. Arch Intern
Med 2007; 167: 1892–8.
Brinkhaus B , Witt CM , Jena S , et al. Acupuncture in patients with
chronic low back pain: a randomized controlled trial. Arch Intern
Med 2006; 166: 450–7.
Witt CM , Jena S , Selim D , et al. Pragmatic randomized trial
evaluating the clinical and economic effectiveness of acupuncture
for chronic low back pain. Am J Epidemiol 2006; 164: 487–96.
Thomas KJ , MacPherson H , Thorpe L , et al. Randomised
controlled trial of a short course of traditional acupuncture
compared with usual care for persistent non-specific low back pain.
BMJ 2006; 333: 623.
Martin BI , Deyo RA , Mirza SK , et al. Expenditures and health status
among adults with back and neck problems. JAMA 2008; 299:
Weiner DK , Kim YS , Bonino P , et al. Low back pain in older adults:
are we utilizing healthcare resources wisely? Pain Med 2006; 7:
Dagenais S , Roffey DM , Wai EK , et al. Can cost utility evaluations
inform decision making about interventions for low back pain ?
Spine J 2009; 9: 944–57.
Lin CW , Haas M , Maher CG , et al. Cost- effectiveness of guidelineendorsed
treatments for low back pain: a systematic review. Eur
Spine J 2011;20: 1024–38.
Wright JG , Swiontkowski MF , Heckman JD. Introducing levels
of evidence to the journal. J Bone Joint Surg Am 2003; 85-A (1):
Norvell DC , Dettori JR , Fehlings MG , et al. Methodology for
the systematic reviews on an evidence based approach for the
management of chronic LBP. Spine 2011; 36: S10–S18.
Goldby LJ , Moore AP , Doust J , et al. A randomized controlled
trial investigating the effi ciency of musculoskeletal physiotherapy
on chronic low back disorder. Spine (Phila Pa 1976) 2006; 31:
Ferreira ML , Ferreira PH , Latimer J , et al. Comparison of general
exercise, motor control exercise and spinal manipulative therapy
for chronic low back pain: a randomized trial. Pain 2007; 131:
Chown M , Whittamore L , Rush M , et al. A prospective study of
patients with chronic back pain randomised to group exercise,
physiotherapy or osteopathy. Physiotherapy 2008; 94: 21–28.
Marshall P , Murphy B. Self-report measures best explain changes
in disability compared with physical measures after exercise
rehabilitation for chronic low back pain. Spine (Phila Pa 1976)
2008; 33: 326–38.
Gudavalli MR , Cambron JA , McGregor M , et al. A randomized
clinical trial and subgroup analysis to compare fl exion-distraction
with active exercise for chronic low back pain. Eur Spine J
2006; 15: 1070–82.
Petersen T , Larsen K , Nordsteen J , et al. The McKenzie method compared
with manipulation when used adjunctive to information and
advice in low back pain patients presenting with centralization or
peripheralization. A randomized controlled trial. Spine 2011; 24: 24.
Cecchi F , Molino-Lova R , Chiti M , et al. Spinal manipulation compared
with back school and with individually delivered physiotherapy
for the treatment of chronic low back pain: a randomized trial
with one-year follow-up. Clin Rehabil 2010; 24: 26–36.
Yeung CK , Leung MC , Chow DH. The use of electro-acupuncture
in conjunction with exercise for the treatment of chronic low-back
pain. J Altern Complement Med 2003; 9: 479–90.
Eisenberg DM , Post DE , Davis RB , et al. Addition of choice of
complementary therapies to usual care for acute low back pain: a
randomized controlled trial. Spine (Phila Pa 1976) 2007; 32: 151–8.
Muller R , Giles LG. Long-term follow-up of a randomized clinical
trial assessing the efficacy of medication, acupuncture, and spinal
manipulation for chronic mechanical spinal pain syndromes. J
Manipulative Physiol Ther 2005; 28: 3–11.
Giles LG , Muller R. Chronic spinal pain: a randomized clinical
trial comparing medication, acupuncture, and spinal manipulation.
Spine (Phila Pa 1976) 2003; 28: 1490–502. Discussion: 1502–3.
van der Roer N , van Tulder M , van Mechelen W , et al. Economic
evaluation of an intensive group training protocol compared with
usual care physiotherapy in patients with chronic low back pain.
Spine (Phila Pa 1976) 2008; 33: 445–51.
Moffett JK , Torgerson D , Bell-Syer S , et al. Randomised controlled
trial of exercise for low back pain: clinical outcomes, costs, and
preferences. BMJ 1999; 319: 279–83.
Torstensen TA , Ljunggren AE , Meen HD , et al. Effi ciency 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.
Spine (Phila Pa 1976) 1998; 23: 2616–24.
Sogaard R , Bunger CE , Laurberg I , et al. Cost-effectiveness
evaluation of an RCT in rehabilitation after lumbar spinal fusion:
a low-cost, behavioural approach is cost-effective over individual
exercise therapy. Eur Spine J 2008; 17: 262–71.
Beam U. United Kingdom back pain exercise and manipulation (UK
BEAM) randomised trial: cost effectiveness of physical treatments
for back pain in primary care. BMJ 2004; 329: 1381.
Henchoz Y , Pinget C , Wasserfallen JB , et al. Cost-utility analysis
of a three-month exercise programme vs usual care following
multidisciplinary rehabilitation for chronic low back pain. J Rehabil
Med 2010; 42: 846–52.
Johnson RE , Jones GT , Wiles NJ , et al. Active exercise, education,
and cognitive behavioral therapy for persistent disabling low
back pain: a randomized controlled trial. Spine (Phila Pa 1976)
2007; 32: 1578–85.
Niemisto L , Rissanen P , Sarna S , et al. Cost-effectiveness of combined
manipulation, stabilizing exercises, and physician consultation
compared to physician consultation alone for chronic low back
pain: a prospective randomized trial with 2-year follow-up. Spine
(Phila Pa 1976) 2005; 30: 1109–15.
Gatchel RJ , Rollings KH. Evidence-informed management of
chronic low back pain with cognitive behavioral therapy. Spine J
2008; 8: 40–4.
Lamb SE , Hansen Z , Lall R , et al. Group cognitive behavioural
treatment for low-back pain in primary care: a randomised
controlled trial and cost-effectiveness analysis. Lancet 2010; 375:
Fairbank J , Frost H , Wilson-MacDonald J , et al. Randomised
controlled trial to compare surgical stabilisation of the lumbar
spine with an intensive rehabilitation programme for patients with
chronic low back pain: the MRC spine stabilisation trial. BMJ
2005; 330: 1233.
Cairns MC , Foster NE , Wright C. Randomized controlled
trial of specific spinal stabilization exercises and conventional
physiotherapy for recurrent low back pain. Spine (Phila Pa 1976)
2006; 31: E670–81.
Cady LD , Bischoff DP , O’Connell ER , et al. Strength and fitness
and subsequent back injuries in fi refi ghters. J Occup Med 1979; 21:
Duque I , Parra JH , Duvallet A. Maximal aerobic power in patients
with chronic low back pain: a comparison with healthy subjects.
Eur Spine J 2011; 20: 87–93.
Hoch AZ , Young J , Press J. Aerobic fitness in women with chronic
discogenic nonradicular low back pain. Am J Phys Med Rehabil
2006; 85: 607–13.
Pedersen MT , Blangsted AK , Andersen LL , et al. The effect of
worksite physical activity intervention on physical capacity, health,
and productivity: a 1-year randomized controlled trial. J Occup
Environ Med 2009; 51: 759–70.
Senna MK , Machaly SA. Does maintained Spinal manipulation
therapy for chronic non-specific low back pain result in better long
term outcome? [published online ahead of print January 17, 2011 ].
Spine (Phila Pa 1976). doi: 10.1097/BRS.0b013e3181f5dfe0.
Cherkin DC , Sherman KJ , Avins AL , et al. A randomized trial
comparing acupuncture, simulated acupuncture, and usual care for
chronic low back pain. Arch Intern Med 2009; 169: 858–66.
Edejer TT. Improving the use of research evidence in guideline
development: 11. Incorporating considerations of cost-effectiveness,
affordability and resource implications. Health Res Policy Syst
2006; 4: 23:2006.
Buyx AM , Friedrich DR , Schone-Seifert B. Ethics and effectiveness:
rationing healthcare by thresholds of minimum effectiveness. BMJ
2011; 342: d54.
Return to the ACUPUNCTURE Page