Content uploaded by Arndt Büssing
Author content
All content in this area was uploaded by Arndt Büssing on Apr 17, 2018
Content may be subject to copyright.
Critical Review
Effects of Yoga Interventions on Pain and Pain-Associated
Disability: A Meta-Analysis
Arndt B€
ussing,*Thomas Ostermann,*Rainer L€
udtke,
y
and Andreas Michalsen
z
*Center for Integrative Medicine, Faculty of Medicine, University of Witten/Herdecke, Herdecke, Germany.
y
Karl und Veronica Carstens-Stiftung, Essen, Germany.
z
Department of Internal and Complementary Medicine, Charit
e-University Medical Centre, Immanuel Hospital
Berlin-Wannsee, Berlin, Germany.
Abstract: We searched databases for controlled clinical studies, and performed a meta-analysis on
the effectiveness of yoga interventions on pain and associated disability. Five randomized studies
reported single-blinding and had a higher methodological quality; 7 studies were randomized but
not blinded and had moderate quality; and 4 nonrandomized studies had low quality. In 6 studies,
yoga was used to treat patients with back pain; in 2 studies to treat rheumatoid arthritis; in 2 studies
to treat patients with headache/migraine; and 6 studies enrolled individuals for other indications. All
studies reported positive effects in favor of the yoga interventions. With respect to pain, a random
effect meta-analysis estimated the overall treatment effect at SMD = .74 (CI: .97; .52, P< .0001),
and an overall treatment effect at SMD = .79 (CI: 1.02; .56, P< .0001) for pain-related disability.
Despite some limitations, there is evidence that yoga may be useful for several pain-associated
disorders. Moreover, there are hints that even short-term interventions might be effective. Neverthe-
less, large-scale further studies have to identify which patients may benefit from the respective
interventions.
Perspective: This meta-analysis suggests that yoga is a useful supplementary approach with
moderate effect sizes on pain and associated disability.
ª2012 by the American Pain Society
Key words: Yoga, pain, disability, review, meta-analysis, mind body medicine, complementary
medicine.
Pain disorders are most commonly treated with potent
drugs such as nonsteroidal anti-inflammatory drugs,
opioid analgetics, systemic corticoids, tricyclic antide-
pressants, and several others.
8,24,26,32,33,40
However,
chronic pain is not exclusively a physical condition but
a complex syndrome including physical, psychological,
and social processes. With respect to the multifaceted
causes, there is need for interdisciplinary procedures in
diagnosis and pain management.
34
In fact, current pain
research considers also psychological and social factors
with a significant influence on chronic pain.
16,18,28,31
There is evidence that intensive multidisciplinary bio-
psycho-social rehabilitation improves pain and function
in patients with chronic disabling low back pain,
16
while
the evidence for short-term effects of behavioral therapy
in back pain is much weaker.
18
Astin
1
concludes that
mind-body approaches (including some combination of
stress management, coping skills training, cognitive
restructuring, cognitive-behavioral therapy, relaxation
therapy, imagery, hypnosis, etc.) may be appropriate ad-
junctive treatments inthe management of various chronic
pain conditions.
One of the best known and frequently used mind-
body interventions is yoga. Its conceptual background
is originated from Indian philosophy, and there are nu-
merous schools or types of yoga (ie, Iyengar, Viniyoga,
Shivananda, etc.) with distinct priorities in terms of spir-
itual and physical practices. A typical yoga session with
a specific sequence of postures (asanas of Hatha Yoga),
The authors did not receive external grants or funds to perform this
analysis. The authors declare no conflicts of interest.
Address reprint requests to Univ.-Prof. Dr. med. Arndt B €
ussing, Center
for Integrative Medicine, Faculty of Health, University of Witten/Her-
decke, Gerhard-Kienle-Weg 4, 58239 Herdecke, Germany. E-mail:
arndt.buessing@uni-wh.de
1526-5900/$36.00
ª2012 by the American Pain Society
doi:10.1016/j.jpain.2011.10.001
1
The Journal of Pain, Vol 13, No 1 (January), 2012: pp 1-9
Available online at www.sciencedirect.com
breathing techniques (pranayama), and mental concen-
tration/meditation (dhyana) lasts between 1 and 2 hours.
For yoga practitioners, there is no need to adopt specific
spiritual attitudes or specific religious behavior. Yoga
practices (particularly the asanas) may increase patient’s
physical flexibility, coordination, and strength;
35
the
breathing practices and mediation may calm and focus
the mind to develop greater awareness and diminish
anxiety;
25
thus resulting in higher quality of life. Other
beneficial effects involve a reduction of distress, blood
pressure, and improvements in metabolic regulation.
44
A recent review on the effects of various nonpharma-
cological interventions found good evidence that
cognitive-behavioral therapy, exercise, spinal manipula-
tion, and interdisciplinary rehabilitation are moderately
effective for chronic low back pain; moreover, there
was fair evidence for acupuncture, massage, or yoga.
9
Recently, Haaz and Bartlett
17
published a scoping review
on yoga and arthritis, indicating a reduction of symptoms
and disability. Thus, yoga could in fact be a beneficial
supportive intervention, but there is currently a lack of
an adequate meta-analysis to assess its effectiveness
with respect to pain symptoms. To assess its putative
relevance in the treatment of patients with various
pain conditions, we performed a meta-analysis of the
current literature focusing on pain and pain-associated
disability.
Methods
Search Strategy
Until January 2010, we searched databases, ie,
PubMed/Medline, the Excerpta Medica Database
(EMBASE), and CAMbase for clinical studies focusing on
yoga interventions and pain. English language search
terms were ‘‘yoga * pain.’’ To increase the chance to
find all relevant publication describing the effects of
yoga interventions on pain, there were no limitations in
the initial search in terms of language, year, status, or de-
sign. Finally, we asked experts for gray literature not
listed in the above mentioned databases, and checked
the reference lists of relevant articles and authors.
Selection Criteria
All potentially eligible studies were retrieved and the
full-text articles were reviewed to determine if they
met the inclusion criteria.
Inclusion criteria were controlled clinical studies (ran-
domized or nonrandomized) addressing the effects of
yoga interventions on pain symptoms. The findings
were analyzed with respect to 2 main outcome cate-
gories: 1) pain intensity/frequency (as measured with vi-
sual analogue/numeric rating scales, McGill pain
questionnaire, or Cornell Musculoskeletal Discomfort
Questionnaire); and 2) pain-induced disability (as mea-
sured with the Oswestry Disability Index, Pain Disability
Index, Functional Disability Inventory, Roland Disability
Index, Cornell Musculoskeletal Discomfort Question-
naire, Health Assessment Questionnaire, Maternal Com-
fort Questionnaire, or Sit-and-Reach Test).
We excluded case series, case reports, studies without
a control group, expert statements, and theoretical
reflections. We also excluded studies with complex
interventions such as mindful-based stress reduction
programs (which include yoga practices), because the
contributing effects of the relevant elements are not dis-
tinguishable. Presentation of data follows the recom-
mendation of the Moher’s QUOROM and PRISMA
statements.
29,30
Data Extraction
Review authors (AB and TO) assessed studies for in-
clusion in the review. They took part in the extraction
of data and independent assessment of methodological
quality. Disagreements were resolved by consensus. We
extracted study data on the following topics: general
study design (prospective, multicenter etc.), treatment
allocation (randomization, matched pairs, etc.), treat-
ment concealment and blinding, treatments (yoga style
and practices, duration and frequency, type of control
intervention), patient characteristics (mean age, gen-
der distribution), diagnosis, adherence to therapy
(compliance, drop-outs, etc.), and outcome assess-
ments.
To assess the methodological quality of the respective
studies, we adopted the Jadad score, which refers to ran-
domization (0 to 2 points), blinding of the assessor (stat-
istician, physician, assessor or researcher, as cited in the
original publications; 0 to 1 points) and dropout report-
ing (0 or 1 point) as indicators of methodological quality
of a study.
22
Because it is impossible to also blind pa-
tients (double blind) in yoga studies, the maximum
achievable Jadad score was 4 in our review. However,
while it is clear what blinding of a statistician means, it
is not very clear what blinding of researcher may
mean; we assumed that this term refers to the outcome
assessor.
Allocation concealment was assessed in accordance
with the Cochrane guidelines
20
: A = adequate (tele-
phone randomization or using consecutively numbered,
sealed, opaque envelopes); B = uncertainty about the
concealment (method of concealment is not known);
C = inadequate (eg, alternate days, odd/even date of
birth, hospital number).
Statistical Analysis
All relevant outcome data were extracted as they
were given in the publication. They were converted
into standardized mean differences (SMD) and their
standard errors (SE) using standard formulas.
20
SMD < 0 indicate superiority of yoga treatment com-
pared with control. SMDs < .5 were regarded as puta-
tively clinically relevant, and < .8 as large effect.
43
We performed various subgroup analyses with re-
spect to condition, methodological quality, and dura-
tion of treatment. For pain outcomes we additionally
performed a subset analysis for those studies which in-
cluded a visual analogue scale (VAS) as an outcome
parameter. Here, all VAS were linearly rescaled to
0 to 100 scoring and the analysis was based on
2The Journal of Pain Yoga Effects on Pain and Associated Disability
weighted mean differences (WMD), calculated from
the published mean changes for each group, instead
of SMD to facilitate clinical interpretations.
Overall estimates of the treatment effect were obtained
from random effects meta-analysis.
13
Heterogeneity be-
tween studies was assessed by standard c
2
–tests and the
I
2
coefficient,
19
which measures the percentage of total
variation across studies due to true heterogeneity rather
than chance. Heterogeneity was further investigated
by several subgroup analyses, formed by study quality
(high: Jadad score = 4 and allocation concealment = A),
moderate quality (scores 2–3), low quality (score 0–1),
treatment duration (short, up to 4 weeks; intermediate,
6 to 10 weeks; long, 12 to 24 weeks), and type of
control group (waiting list: routine care only; active
treatment: any other intervention given additionally to
routine care).
Funnel plot asymmetry was assessed by Egger’s test.
12
Results
Search Results
We found 23 potentially relevant studies addressing
the effects of yoga on pain (intensity, frequency) or
pain-associated disability (affected function). Among
them, 4 studies were excluded because they had no con-
trol group. Nineteen controlled studies were considered
eligible for inclusion (Fig 1). However, 3 randomized
studies had to be excluded: a feasibility study describing
just the pretreatment results
21
; a study with inadequate
control group comparing yoga with yoga plus tuina
touching
11
; and a study which presented the outcome
analyses as multilevel modeling data
7
and thus not
suited for data extraction. Finally, 16 controlled studies
provided sufficient data to extract SMDs and their stan-
dard deviations (Fig 1).
Description of Studies
All of the remaining 16 studies according to specifica-
tions in the articles had a prospective design. Five
studies were single-blinded (ie, statistician, physician,
assessor or researcher, as stated in the respective
studies) and randomized and thus had a higher meth-
odological quality (Jadad score 4), 7 studies were ran-
domized (without blinding) and had a moderate
(scores 2–3), while 4 were nonrandomized controlled
studies with low quality (score 0–1; Tabl e 1). Seven stud-
ies had a waiting-list design, other control interven-
tions were physical activity and lectures, routine care
and conversation, and in 1 study, anti-inflammatory
drugs.
The number of patients enrolled varied considerably
from 12 to 291 (mean 6SE: 63 666); only 1 study en-
rolled >100 individuals (Table 1). Most studies included
participants aged #50 years (among them 1 study enroll-
ing exclusively adolescents, 26); only 1 study enrolled
older patients (56 68 years in the yoga group and 67
66 years in the control group).
In 6 studies, yoga was used to treat patients with back
pain; in 2 studies to treat rheumatoid arthritis; in 2 stud-
ies to treat patients with headache/migraine; and 6
studies enrolled individuals for other indications.
Twelve studies described effects on pain intensity and
frequency, and 12 studies on pain-related disability as
outcome variables. Four studies were categorized as
short-term (up to 4 weeks of treatment), 7 as medium
term (6–10 weeks), and 5 as long-term treatment
(12–24 weeks).
Potentially relevant studies addressing
yoga and pain (n=23)
puorglortnocon:seidutsdedulcxE
(n=4)
Potentially appropriate controlled studies to
be included (n=19)
Excluded studies (n=3):
feasibility study
(without post treatment results)
19
inadequate control group
(yoga versus yoga + tuina)
10
no data extraction possible
(multilevel modeling)
6
Included controlled studies; suitable for data extraction
(n=16)
Randomized and
single-blinded
(n=5)
Randomized
without blinding
(n=7)
Non-
randomized
(n=4)
Figure 1. Flow diagram of study exclusion.
B€
ussing et al The Journal of Pain 3
Table 1. Overview on Identified Clinical Studies
FIRST AUTHOR
[STUDY-ID] YEAR
PAIN
CONDITION N
AGE
(YEARS)YOGA STYLE
CONTROL
INTERVENTION
DURATION OF
TREATMENT
METHODO
-LOGICAL
QUALITY*
PAIN DISABILITY
INSTRUMENT SMD SD INSTRUMENT SMD SD
Bhatia [1]
3
2007 Headache 12 18–50
(most 25–35)
Not specified Anti-
inflammatory
drugs
Short 0 VASy.94 0.57 / / /
Tekur [4]
38
2008 Back pain 91 49 64/48 64 LAYT Physical activity,
lecture
Short 4 / / / ODI 1.25 .23
Yurtkuran [6]
45
2007 Hemo-dialysis 40 38 614/10 610.0 Not specified Physical activity Inter-mediate 4 VASy.20 0.31 / / /
John [7]
23
2007 Headache 72 34 69/34 610 Not specified Educational
sessions
(briefing and
handout)
Long 3 MPQ
(T-PRI)
.62 0.24 / / /
Kuttner [8]
27
2006 Irritable bowel
syndrome
28 14 62/14 62 Hatha/Iyengar Waiting list Short 3 VAS,y
Checklist
.54 0.37 FDI .73 .38
Sherman [9]
37
2005 Back pain 71 44 612/42 615 Viniyoga Physical
activity, book,
lecture
Long 3 / / / RDS .33 .24
Williams [10]
42
2005 Back pain 44 49 611/48 62 Iyengar Physical activity,
lecture
Long 4 PPI, VASy.76 0.31 PDI 1.40 .33
Boyle [12]
5
2004 Muscle
soreness
24 22–53 (38 62.6) Hatha Physical activity Short 0 VAS .87 0.35 SRT 1.31 .46
Galantino [14]
14
2004 Back pain 22 30–65 Hatha Waiting list Inter-mediate 2 / / / ODI 1.18 .45
Garfinkel [15]
15
1998 Carpal tunnel
syndrome
51 17–70 (mean 49) Iyengar Waiting list Inter-mediate 2 VASy.63 .28 / / /
Telles [17]
39
2009 Healthy
PC user
291 33 69/32 610 Not specified Waiting list Inter-mediate 4 CMDQ 1.20 .13 CMDQ 1.01 .12
Saper [19]
36
2009 Back pain 30 44 613/44 611 Hatha Waiting list Long 2 VASy1.34 .40 RDS .70 .37
Bosch [21]
4
2009 Rheumatoid
arthritis
20 (16) 56 68/67 66 Hatha Other Inter-mediate 1 VASy.27 .43 HAQ .38 .43
Williams [22]
41
2009 Back pain 90 48 62/48 62 Iyengar Waiting list Long 4 VAS .52 .21 ODI .34 .21
Chuntharapat
[23]
10
2008 Labor pain 74 18–35 Not specified Routine care,
conversation
Inter-mediate 2 VAS .52 .23 VAS
maternal
comfort
.50 .23
Badsha [24]
2
2009 Rheumatoid
arthritis
47 44 610/ 46 611 Raj No treatment
(waiting list)
Inter-mediate 1 / / / HAQ .63 .29
Abbreviations: VAS, visual analogue scale; ODI, Oswestry Disability Index ; MPQ, McGill pain questionnaire; HAQ, Health Assessment Questionnaire; FDI, Functional Disability Inventory; RDS, Roland Disability Scale; PDI, Pain Disability Index;
PPI, Present pain Index; SRT, Sit-and-Reach Test; CMDQ, Cornell Musculoskeletal Discomfort Questionnaire; HAQ, Health Assessment Questionnaire.
*Jadad score.
yLinearly rescaled (0–100).
4The Journal of Pain Yoga Effects on Pain and Associated Disability
Effect Sizes Pain
As shown in Table 1, all studies reported positive ef-
fects in favor of the yoga interventions. With respect to
pain, the effect sizes ranged from .20 6.31 to
1.34 6.40 (Fig 2). Heterogeneity of study results was
moderate (I
2
= 44%). A random effect meta-analysis
estimated the overall treatment effect at SMD = .74
(CI: .97 to .52, P< .0001), indicating a moderate over-
all effect.
For pain outcomes, we additionally performed a subset
analysis for those studies which included a VAS as an
outcome parameter. As the outcome scales were
comparable in this subset, they were based on WMDs
calculated from the published mean changes for each
group instead of SMD to facilitate clinical interpreta-
tions. As measured on a 100-mm VAS, the group differ-
ence was estimated at WMD = 12 mm (CI: 17; 7;
P< .001; I
2
= 19%).
Further subgroup analyses indicate that neither dura-
tion of treatment nor methodological quality was associ-
ated with better or worse study outcome (Table 2).
However, studies with higher methodological quality
had a better outcome as compared with studies with
low quality; moreover, studies with a waiting-list design
had somewhat higher effect sizes (and higher heteroge-
neity) than studies with others controls (Table 2). Restric-
tions to randomized controlled trials only yielded a SMD
=.82 (CI: 1.20; .53; P< .0001; I
2
= 54%). With respect
to the pain conditions, the 3 studies with healthy
individuals had the highest effects sizes (SMD = 1.14)
as compared with studies enrolling patients with
chronic pain conditions (Table 3). In fact, the effect sizes
for back pain or rheumatoid arthritis (SMD = .69) were
better as compared with various other pain conditions
(SMD = .54).
Overall, larger effects were observed in studies with
higher methodological quality, passive wait list control,
and in studies enrolling healthy individuals.
Effect Sizes Pain Related Disability
With respect to the improvement of pain related
disability, the single studies’ effect sizes ranged from
.33 6.24 to 1.40 6.33 (Fig 3). Heterogeneity of study
results was high (I
2
= 54%). A random effect meta-
analysis estimated the overall treatment effect at
SMD = .79 (CI: 1.02 to .56, P< .0001), indicating
a moderate effect. Subgroup analyses showed that
short-term interventions yielded stronger effects than
long treatments, while the methodological quality of
the study or the respective control group had no remark-
able impact on study outcome (Table 3). With respect to
the pain conditions, the 3 studies with healthy individ-
uals had somewhat higher effects sizes as compared
with studies enrolling patients with pain conditions
(Table 3). Here, the effect size of studies with chronic
back pain and rheumatoid arthritis were similar to the
overall effect size of the whole sample (Table 3).
Overall, the strongest effects on pain-associated dis-
ability were observed in studies with shorter duration.
Effect Sizes Mood in Pain
Although it was not our primary aim to analyze, 6 stud-
ies also reported effects on patients’ mood stages (Fig 4).
The moderate effects in favor of the yoga interventions
(SMD = .65 [CI: .89 to .42]) are in congruence with
the described effects on pain and pain disability.
Funnel Plot Analyses
Formal inspections of the funnel plot did not reveal
any significant asymmetry (Fig 5) which might indicate
a publication bias (pain: asymmetry coefficient AC =
.67, P= .48; mood: AC = 1.09, P= .42; pain-related dis-
ability AC = .47, P= .68).
Discussion
Our findings suggest yoga as a useful supportive inter-
vention for a broad range of pain-associated diseases.
Four studies described strong effects of yoga on patients’
pain intensity/frequency, 6 studies moderate effects, and
2 weak effects; moreover, 5 studies reported strong
effect sizes for pain-associated disability, 4 moderate
effects, and 3 weak effects in favor of the yoga interven-
tion (Table 1). With respect to chronic back pain and
rheumatoid arthritis, the respective studies had moder-
ate effects in favor of the yoga intervention (SME
Figure 2. Standardized mean differences on pain. The size of circles represents the weight of the study in meta-regression.
B€
ussing et al The Journal of Pain 5
Disability = .76; SME Pain = .69). However, particularly
healthy individuals with labor pain, personal computer
usage, or induced muscle soreness had the strongest ef-
fects. Interestingly, also in adolescents with irritable
bowel syndrome, the yoga intervention resulted in mod-
erate effect sizes.
27
Patients with headache/migraine had
effect sizes ranging from .62
23
to .94,
3
indicating
a beneficial effect in principle. Particularly, the study
with application of anti-inflammatory drugs as control
had a strong effect in favor of the yoga group—albeit
this headache study had a very low methodological
quality. Nevertheless, there is evidence that yoga may
influence pain and/or pain-associated disability.
In contrast, 1 high-quality study addressing the effects
the yoga intervention had on pain intensity in hemodial-
ysis patients with end-stage renal disease
45
revealed just
weak effect sizes in favor of the intervention (albeit sev-
eral other parameters significantly improved). Less posi-
tive conclusions were also reached in the treatment of
patients with rheumatoid arthritis during a 10-week pe-
riod, with weak effect sizes for both pain and disability
4
;
however, the study enrolled just 20 patients (16 com-
pleters) and thus we cannot draw any valid conclusion.
Several other indications with at least 1 study with posi-
tive outcome would encourage further clinical trials.
It is an important finding that the methodological
quality of the studies (which was, in general, moderate)
had no relevant impact on the study outcome; of note,
studies with higher quality had a better pain outcome
as compared with studies with low quality, an association
Table 2. Subgroup Analyses on Pain
SUBGROUP
ENROLLED STUDIES
(ID NUMBERS)
OVERALL
SMD 95% CI Iy
Overall 1,6,7,8,10,12,15,17,19,21,22,23 .74 .97; .52 44%
Randomized Controlled Trials 6,7,8,10,15,17,19,22,23 .82 1.20; .53 54%
Duration of treatment
Short 1,8,12 .75 1.21; .30 0%
Intermediate 6,15,17,21,23 .69 1.14; .23 70%
Long 7,10,19,22 .70 .98; .41 13%
Jadad score
High 17,22 .88 1.55; .21 87%
Intermediate 7,8,15,19,23 .73 1.01; .45 0%
Low 1,12,21 .69 1.17; .21 0%
Control group
Waiting list 8,15,17,19,22 .85 1.22; .48 65%
Other controls 1,6,7,10,12,21,23 62 .87; .37 0%
Pain condition
Chronic pain*10,19,21,22 .69 1.06; .32 30%
Othery1,6,7,8,15 .54 .82; .27 0%
Healthyz12,17,23 1.14 1.36; .91 0%
*Chronic back pain and rheumatoid arthritis.
yMigraine/headache, carpal tunnel syndrome, irritable bowel syndrome, hemodialysis, carpal tunnel syndrome.
zLabor pain, PC user, muscle soreness.
Table 3. Subgroup Analyses on Pain-Related Disability
SUBGROUP ENROLLED STUDIES (ID NUMBERS)EFFECT 95% CI Iy
Overall 4,8,9,10,12,14,17,19,21,22,23,24 .79 1.02; .56 54%
Duration of treatment
Short 4,8,12 1.14 1.50; .79 0%
Intermediate 14,17,21,23,24 .78 1.07; .49 37%
Long 9,10,19,22 .64 1.11; .18 65%
Jadad score
High 4,9,17,22 .75 1.18; .32 80%
Intermediate 8,9,14,19,23 .56 .83; .30 0%
Low 12,21,24 .72 1.19; .26 15%
Control group
Waiting list 8,14,17,19,22,24 .75 1.04; .47 44%
Other controls 4,9,10,12,21,23 .84 1.25; .43 66%
Pain condition
Chronic pain*4,9,10,14,19,21,22,24 .76 1.08; .43 59%
Othery8.73 1.47; .01 /
Healthyz12,17,23 .88 1.28; .47 57%
*Chronic back pain and rheumatoid arthritis.
yIrritable bowel syndrome.
zLabor pain, PC user, muscle soreness.
6The Journal of Pain Yoga Effects on Pain and Associated Disability
which did not occur with respect to disability. Also re-
markable is the finding that short-term interventions
yielded stronger effects on pain-related disability than
longer treatments. One could argue that patients’ en-
thusiasm may decrease over long treatment periods;
however, this effect cannot be verified with respect to
pain intensity/frequency. On the other hand, Williams
et al
41
reported significant improvement of pain inten-
sity and disability within 12 weeks, which further im-
proved after 24-week intervention. This could indicate
that longer interventions would improve skills and abili-
ties (training) of the completer, while the dropouts
might be due to a loss of enthusiasm. The problem of ad-
herence, which may contribute to the effect that studies
with longer duration of yoga interventions were less
effective than shorter studies, should be addressed in
future studies.
It is an important issue whether some yoga styles might
be more effective than others, particularly because some
styles were designed for individuals with physical limita-
tions. Due to the heterogeneity of the studies with their
different yoga traditions, we cannot draw any valid con-
clusions on this topic. This and also the impact of the
qualification of yoga trainers has to be addressed in
future studies.
Our analysis has several potential limitations. For ex-
ample, the pooled estimates were based on heteroge-
neous data, with respect to indications, control groups,
and methodological quality of these studies, although
the methodological quality overall was moderate. The
analysis was limited by small sample sizes (in average
63 666) and evaluation of younger populations, and
thus it is unclear whether yoga is effective also in the el-
derly. Most studies had adequate control interventions,
ie, physical activity or at least a wait-list design, while 1
study compared a 3-month yoga intervention with edu-
cation sessions (briefing about medication overuse and
migraine modifications, and handouts emphasizing
self-care strategies and lifestyle modification)
23
and
thus is less appropriate. In a pilot study with weak qual-
ity, the putative activities of the control group were un-
clear; the authors by themselves stated that the ‘‘lack of
randomization may have led to a self-selection bias.’’
4
There was just 1 study which compared the effects of
yoga with a control group using conventional anti-
inflammatory drugs; this study described strong effects
in favor of a short yoga intervention, but had the lowest
methodological quality.
2
Although formal inspections of the funnel plots did
not reveal any significant asymmetry which might indi-
cate a publication bias (Fig 5), we cannot exclude the pos-
sibility that our analysis could lack studies with negative
results which were never published at all.
Having in mind these limitations, we found good evi-
dence that yoga interventions might be useful for several
pain-associated disorders. There are hints that even
short-term interventions might be effective. Neverthe-
less, further studies have to identify which patients may
benefit from the interventions,
6
and which aspects of
the yoga interventions (ie, physical activity and/or
Figure 3. Standardized treatment effects (SMD and confidence intervals) on pain-related disability. The size of circles represents the
weight of the study in meta-regression.
Figure 4. Standardized treatment effects (SMD and confidence intervals) on mood. The size of circles represents the weight of the
study in meta-regression.
B€
ussing et al The Journal of Pain 7
meditation and subsequent life-style modification) or
which specific yoga styles were more effective than
others.
The beneficial effects of yoga can be explained, in
part, by an increased physical flexibility, coordination,
and strength,
35
by calming and focusing the mind to
develop greater awareness and diminish anxiety,
25
re-
duction of distress,
9
improvement of mood (Fig 4), etc.
Because patients may recognize that they are able to
be physically active, even despite persisting pain symp-
toms, they experience higher self-competence and self-
awareness, which contributes to higher quality of life.
Apart from the above-mentioned training effects, it is
difficult to judge the relevance of contributing unspe-
cific effects (ie, due to positive expectation, attention,
conditioning, etc.). Looking at the studies with passive
(waiting list) controls, the treatment effects with respect
to pain were higher than those with an active control (ie,
physical activity), while with respect to disability, there
were no relevant differences between the control
groups. Because positive expectations are a relevant vari-
able in all interventional studies, further studies on the
effects of yoga should address this, too.
In conclusion, large-scale and sound research is highly
encouraged because yoga may have potential to be im-
plemented as a safe and beneficial supportive treatment
which is relatively cheap and requires just the motivation
of patients.
References
1. Astin JA: Mind-body therapies for the management of
pain. Clin J Pain 20:27-32, 2004
2. Badsha H, Chhabra V, Leibman C, Mofti A, Kong KO: The
benefits of yoga for rheumatoid arthritis: Results of a prelim-
inary, structured 8-week program. Rheumatol Int 29:
1417-1421, 2009
3. Bhatia R, Dureja GP, Tripathi M, Bhattacharjee M,
Bijlani RL, Mathur R: Role of temporalis muscle over ac-
tivity in chronic tension type headache: Effect of yoga
based management. Indian J Physiol Pharmacol 51:
333-344, 2007
4. Bosch PR, Traustadottir T, Howard P, Matt KS: Functional
and physiological effects of yoga in women with rheuma-
toid arthritis: A pilot study. Altern Ther Health Med 15:
24-31, 2009
5. Boyle CA, Sayers SP, Jensen BE, Headley SA, Manos TM:
The effects of yoga training and a single bout of yoga on de-
layed onset muscle soreness in the lower extremity. J
Strength Cond Res 18:723-729, 2004
6. B€
ussing A, Edelh€
auser F, Weisskircher A, Fouladbakhsh
JM, Heusser P: Inner correspondence and feelings of
peaceful easiness in participants practicing Eurythmy
Therapy and Yoga: A validation study. Evid Based
Complement Alternat Med 2010 Sep 7; [Epub ahead of
print]
7. Carson JW, Carson KM, Porter LS, Keefe FJ, Seewaldt VL:
Yoga of Awareness program for menopausal symptoms in
breast cancer survivors: Results from a randomized trial.
Support Care Cancer 17:1301-1309, 2009
8. Chou R, Huffman LH: Medications for acute and chronic
low back pain: A review of the evidence for an American
Pain Society/American College of Physicians clinical practice
guideline. Ann Intern Med 147:505-514, 2007
9. Chou R, Huffman LH: Nonpharmacologic therapies for
acute and chronic low back pain: A review of the evidence
for an American Pain Society/American College of Physicians
clinical practice guideline. Ann Intern Med 147:492-504,
2007
10. Chuntharapat S, Petpichetchian W, Hatthakit U: Yoga
during pregnancy: Effects on maternal comfort, labor pain
and birth outcomes. Complement Ther Clin Pract 14:
105-115, 2008
11. da Silva GD, Lorenzi-Filho G, Lage LV: Effects of yoga and
the addition of Tui Na in patients with fibromyalgia. J Altern
Complement Med 13:1107-1113, 2007
12. Egger M, Davey Smith G, Schneider M, Minder C: Bias in
meta-analysis detected by a simple, graphical test. BMJ 315:
629-634, 1997
13. Fleiss JL: The statistical basis of meta-analysis. Stat
Methods Med Res 2:121-145, 1993
14. Galantino ML, Bzdewka TM, Eissler-Russo JL,
Holbrook ML, Mogck EP, Geigle P, Farrar JT: The im-
pact of modified Hatha yoga on chronic low back
pain: A pilot study. Altern Ther Health Med 10:56-59,
2004
15. Garfinkel MS, Singhal A, Katz WA, Allan DA, Reshetar R,
Schumacher HR Jr: Yoga-based intervention for carpal tun-
nel syndrome: A randomized trial. JAMA 280:1601-1603,
1998
16. Guzman J, Esmail R, Karjalainen K, Malmivaara A, Irvin E,
Bombardier C: Multidisciplinary bio-psycho-social rehabili-
tation for chronic low back pain. Cochrane Database Syst
Rev;CD000963, 2002
17. Haaz S, Bartlett SJ: Yoga for arthritis: A scoping review.
Rheum Dis Clin North Am 37:33-46, 2011
18. Henschke N, Ostelo RW, van Tulder MW, Vlaeyen JW,
Morley S, Assendelft WJ, Main CJ: Behavioural treatment
for chronic low-back pain. Cochrane Database Syst Rev 7:
CD002014, 2010
19. Higgins JP, Thompson SG: Quantifying heterogeneity in
a meta-analysis. Stat Med 21:1539-1558, 2002
Figure 5. Funnel plot asymmetry of studies addressing pain.
8The Journal of Pain Yoga Effects on Pain and Associated Disability
20. Higgins JPT, Green S: Cochrane Handbook for Systematic
Reviews of Interventions. Version 5.0.1 ed. The Cochrane
Collaboration, 2008
21. Jacobs BP, Mehling W, Avins AL, Goldberg HA, Acree M,
Lasater JH, Cole RJ, Riley DS, Maurer S: Feasibility of conduct-
ing a clinical trial on Hatha yoga for chronic low back pain:
Methodological lessons. Altern Ther Health Med 10:80-83,
2004
22. Jadad AR, Moore RA, Carroll D, Jenkinson C,
Reynolds DJ, Gavaghan DJ, McQuay HJ: Assessing the quality
of reports of randomized clinical trials: Is blinding neces-
sary? Control Clin Trials 17:1-12, 1996
23. John PJ, Sharma N, Sharma CM, Kankane A: Effective-
ness of yoga therapy in the treatment of migraine without
aura: A randomized controlled trial. Headache 47:654-661,
2007
24. Katz JD, Shah T: Persistent pain in the older adult: What
should we do now in light of the 2009 American geriatrics
society clinical practice guideline? Pol Arch Med Wewn
119:795-800, 2009
25. Kirkwood G, Rampes H, Tuffrey V, Richardson J,
Pilkington K: Yoga for anxiety: A systematic review of the re-
search evidence. Br J Sports Med 39:884-891, 2005
26. Kroenke K, Krebs EE, Bair MJ: Pharmacotherapy of
chronic pain: A synthesis of recommendations from system-
atic reviews. Gen Hosp Psychiatry 31:206-219, 2009
27. Kuttner L, Chambers CT, Hardial J, Israel DM, Jacobson K,
Evans K: A randomized trial of yoga for adolescents with ir-
ritable bowel syndrome. Pain Res Manag 11:217-223, 2006
28. Leyshon RT: Coping with chronic pain: Current advances
and practical information for clinicians. Work 33:369-372,
2009
29. Moher D, Cook JC, Eastwood S, Olkin I, Rennie D,
Stroup DF: Improving the quality of reports of meta-
analyses of randomized controlled trials: The QUORUM
statement. Lancet 354:1896-1900, 1999
30. Moher D, Liberati A, Tetzlaff J, Altman DA, for the
PRISMA Group: Preferred reporting items for systematic re-
views and meta-analyses: The PRISMA statement. BMJ 339:
b2535, 2009
31. Molton IR, Graham C, Stoelb BL, Jensen MP: Current psy-
chological approaches to the management of chronic pain.
Curr Opin Anaesthesiol 20:485-489, 2007
32. Passik SD, Weinreb HJ: Managing chronic nonmalignant
pain: Overcoming obstacles to the use of opioids. Adv Ther
17:70-83, 2000
33. Pergolizzi J, Boger RH, Budd K, Dahan A, Erdine S,
Hans G, Kress HG, Langford R, Likar R, Raffa RB,
Sacerdote P: Opioids and the management of chronic severe
pain in the elderly: Consensus statement of an International
Expert Panel with focus on the six clinically most often used
World Health Organization Step III opioids (buprenorphine,
fentanyl, hydromorphone, methadone, morphine, oxyco-
done). Pain Pract 8:287-313, 2008
34. Pfingsten M: [Chronic low back pain–need for an inter-
disciplinary approach]. Anasthesiol Intensivmed Notfallmed
Schmerzther 44:40-45, 2009
35. Raub JA: Psychophysiologic effects of Hatha Yoga on
musculoskeletal and cardiopulmonary function: A literature
review. J Altern Complement Med 8:797-812, 2002
36. Saper RB, Sherman KJ, Cullum-Dugan D, Davis RB,
Phillips RS, Culpepper L: Yoga for chronic low back pain
in a predominantly minority population: A pilot random-
ized controlled trial. Altern Ther Health Med 15:18-27,
2009
37. Sherman KJ, Cherkin DC, Erro J, Miglioretti DL, Deyo RA:
Comparing yoga, exercise, and a self-care book for chronic
low back pain: A randomized, controlled trial. Ann Intern
Med 143:849-856, 2005
38. Tekur P, Singphow C, Nagendra HR, Raghuram N: Effect
of short-term intensive yoga program on pain, functional
disability and spinal flexibility in chronic low back pain: A
randomized control study. J Altern Complement Med 14:
637-644, 2008
39. Telles S, Dash M, Naveen KV: Effect of yoga on musculo-
skeletal discomfort and motor functions in professional
computer users. Work 33:297-306, 2009
40. Varrassi G, Muller-Schwefe G, Pergolizzi J, Oronska A,
Morlion B, Mavrocordatos P, Margarit C, Mangas C,
Jaksch W, Huygen F, Collett B, Berti M, Aldington D,
Ahlbeck K: Pharmacological treatment of chronic pain -
the need for CHANGE. Curr Med Res Opin 26:1231-1245,
2010
41. Williams K, Abildso C, Steinberg L, Doyle E, Epstein B,
Smith D, Hobbs G, Gross R, Kelley G, Cooper L: Evaluation
of the effectiveness and efficacy of Iyengar yoga therapy
on chronic low back pain. Spine (Phila Pa 1976) 34:
2066-2076, 2009
42. Williams KA, Petronis J, Smith D, Goodrich D, Wu J,
Ravi N, Doyle EJ Jr, Gregory Juckett R, Munoz Kolar M,
Gross R, Steinberg L: Effect of Iyengar yoga therapy for
chronic low back pain. Pain 115:107-117, 2005
43. Wolf FM: Meta-Analysis: Quantitative Methods For Re-
search Synthesis. Beverly Hills, CA, Sage Publications, 1986
44. Yang K: A review of yoga programs for four leading risk
factors of chronic diseases. Evid Based Complement Alternat
Med 4:487-491, 2007
45. Yurtkuran M, Alp A, Yurtkuran M, Dilek K: A modified
yoga-based exercise program in hemodialysis patients: A
randomized controlled study. Complement Ther Med 15:
164-171, 2007
B€
ussing et al The Journal of Pain 9
