Yoga in the workplace and health outcomes: a systematic review

Abstract

Background:
Health promotion in the workplace is intended to enhance employee health and well-being. Yoga programmes are easy to implement and have been effective in the management of various health conditions.

Aims:
To assess the evidence regarding the effectiveness of yoga programmes at work.

Methods:
A search of electronic databases of published studies up until the 1st of April 2017. Inclusion criteria for the systematic review were randomized controlled trials of adult employees and yoga in the workplace. Quality appraisal was carried out using the Cochrane Collaboration's tool for assessing risk of bias in randomized trials.

Results:
Of 1343 papers identified, 13 studies met the inclusion criteria. Nine out of 13 trials were classified as having an unclear risk of bias. The overall effects of yoga on mental health outcomes were beneficial, mainly on stress. Most of the cardiovascular endpoints showed no differences between yoga and controls. Other outcomes reported positive effects of yoga or no change.

Conclusions:
The findings of this study suggest that yoga has a positive effect on health in the workplace, particularly in reducing stress, and no negative effects were reported in any of the randomized controlled trials. Further larger studies are required to confirm this.

Full text

Occupational Medicine
doi:10.1093/occmed/kqz033
© The Author(s) 2019. Published by Oxford University Press on behalf of the Society of Occupational Medicine.
All rights reserved. For Permissions, please email: journals.permissions@oup.com
Yoga in the workplace and health outcomes: a
systematic review
L.M. PuertoValencia1,2, A . Weber2, H. Spegel2, R. Bögle3,4, A. Selmani2, S. Heinze2,5 and C. Herr2,5
1Institute for Medical Informatics, Biometry and Epidemiology (IBE) at the Ludwig-Maximilians-Universität München,
Marchioninistr. 15, 81377 Munich, Germany, 2Department of Occupational and Environmental Health/Epidemiology, Bavarian
Health and Food Safety Authority (LGL), Pfarrstr. 3, 80538 Munich, Germany, 3Yoga Forum München e.V, Steinstraße.
42, 81667 Munich, Germany, 4Center for Behavioural Medicine, 24/1, Sahajanand Society, Kothrud, Pune 411038, India,
5Institute and Outpatient Clinic for Occupational, Social and Environmental Medicine, Ludwig-Maximilians-University Munich,
Ziemssenstr. 1, 80336 Munich, Germany.
Correspondence to: L.M. Puerto Valencia, Bavarian Health and Food Safety Authority (LGL), Pfarrstr. 3, 80538 Munich,
Germany. Tel +49 09131 6808-0; e-mail: laurampv8@gmail.com
Background Health promotion in the workplace is intended to enhance employee health and well-being. Yoga
programmes are easy to implement and have been effective in the management of various health
conditions.
Aims To assess the evidence regarding the effectiveness of yoga programmes at work.
Methods A search of electronic databases of published studies up until the 1st of April 2017. Inclusion crite-
ria for the systematic review were randomized controlled trials of adult employees and yoga in the
workplace. Quality appraisal was carried out using the Cochrane Collaboration’s tool for assessing
risk of bias in randomized trials.
Results Of 1343 papers identied, 13 studies met the inclusion criteria. Nine out of 13 trials were classied
as having an unclear risk of bias. The overall effects of yoga on mental health outcomes were bene-
cial, mainly on stress. Most of the cardiovascular endpoints showed no differences between yoga and
controls. Other outcomes reported positive effects of yoga or no change.
Conclusions The ndings of this study suggest that yoga has a positive effect on health in the workplace, particu-
larly in reducing stress, and no negative effects were reported in any of the randomized controlled
trials. Further larger studies are required to conrm this.
Key words Employees; healthy population; randomized controlled trials; workplace; yoga.
Introduction
Initially, occupational health services were concerned
with the protection of workers from occupational haz-
ards. More recently, they have looked to produce changes
in individuals’ health practices [1]. Health programmes
at work aim to maintain and improve employees’ health
and well-being and reduce associated costs [2]. These
programmes usually include appraisal of health risks,
health education and stress management methods [1].
In recent years, interest in interventions for stress
reduction, improvement of mental health and promo-
tion of physical activity among workers has increased.
Job stress has risen in numerous countries [3], together
with a higher prevalence of mental health problems. For
example, in the UK at any given moment, an estimated
one in six working people have suffered symptoms
related to mental illness [4], and each year an estimated
175 million days of work are lost; around half of these
are due to stress [5]. Stress is associated with reduced job
performance and increased costs for employers [6,7]. In
addition, the National Institute for Health and Clinical
Excellence (NICE) recommends increased physical
activity within the workplace to improve well-being [5].
Yoga is a major area of interest in the eld of work-
place interventions and is also one of the traditional and
complementary medicines included in the World Health
Organization (WHO) traditional medicine strategy
2014–23 [8–10]. Yoga is an ancient Indian practice com-
bining postural exercises (Asana) with breathing tech-
niques (Pranayama) and meditation (Dyana) [11–13].
Downloaded from https://academic.oup.com/occmed/advance-article-abstract/doi/10.1093/occmed/kqz033/5382477 by guest on 20 March 2019

Page 2 of 9 OCCUPATIONAL MEDICINE
The practice of yoga may have benecial effects across
various health outcomes [11]. Moreover, yoga can be
learned independently of age or prior knowledge [14]
and can be practised with minimal equipment [7,15].
A considerable amount of literature, including sys-
tematic reviews, has been published on the effects of
yoga in the management of many health conditions, such
as chronic back pain, depression and insomnia [12]. In
contrast, fewer reviews on the effect of yoga work-based
programmes on general health outcomes are available
[7]. Gura’s review in 2002 indicated that Hatha yoga had
benecial effects on health and well-being [7].
The focus of this systematic review was on workplace
yoga programmes offered to employees. The objective
was to evaluate the available evidence from randomized
controlled trials (RCTs) assessing the effects of yoga at
work. The explicit research question was whether yoga
programmes offered to healthy employees in a work-
place setting produced better health and work perform-
ance-related outcomes compared to no yoga or other
interventions.
Methods
This systematic review was conducted according to
PRISMA (Preferred Reporting Items for Systematic
Reviews and Meta-Analyses) [16].
We conducted a systematic review up to 1st of April
2017 in the following electronic databases: PubMed,
Embase (platform Ovid), PsycINFO (platform
EBSCOhost) and the Cochrane Library (platform Wiley
Online Library). In addition, we searched in the IndMED
database (http://indmed.nic.in/), the WHO International
Clinical Trials Registry Platform (http://apps.who.int/tri-
alsearch/Default.aspx) and the US National Institutes of
Health ClinicalTrials.gov (http://www.clinicaltrials.gov).
We used Medical Subject Headings (MeSH) or equiv-
alent and text terms around the key words ‘yoga’ and
‘occupational’ or ‘employee’ or ‘organization’ or ‘work’
or ‘occupation’ or ‘worker’. The searches were adapted to
individual databases. No publication status restrictions
were applied. Reference lists of full-text articles were
manually searched for additional studies.
In terms of eligibility criteria, RCTs that compared
yoga offered at the workplace with no programme,
another mind–body practice (such as meditation and
relaxation techniques), physical activity (such as stretch-
ing) or minimal prevention programme (e.g. education
through booklets about healthcare) were included. The
publication had to be a peer-reviewed article and full
text had to be available in English. Studies including
exclusively employees older than 18years were selected.
The term employee involved only those workers who
hold the type of job dened as paid employment job;
therefore, students, retired adults and fulltime house-
wives were excluded. Studies that recruited volunteers
outside of workplace settings, patients with any medical
condition, pregnant women and yoga instructors were
also excluded. If the main intervention was conducted
at home, in the community or in clinical settings, the
study was excluded. The included studies had to spec-
ify that the main intervention was ‘yoga’. Studies were
excluded if yoga was not the main intervention but a part
of a multimodal intervention, for instance mindfulness-
based stress reduction programmes. Yoga traditions that
include any physical practice component (such as physi-
cal yoga postures) were included. Any frequency and
length orduration of the yoga programme were included.
Interventions based on yoga (e.g. stretching exercises
based on yoga) but not characterized as yoga, or stud-
ies examining meditation, yoga breathing or yoga lifestyle
without any physical practice, were excluded. Studies
Key learning points
What is already known about this subject:
• A small number of studies had reviewed the effect of yoga programmes in work settings.
• Especially, evidence from randomized controlled trials had not been comprehensively summarized, includ-
ing an overview of the efcacy of yoga on general health outcomes in employees.
• Consequently, this study aimed to collect and analyse this information.
What this study adds:
• Multiple trials reported mental health outcomes, principally stress, showing the benecial effects of yoga
compared with control groups.
• Yoga in the workplace had no negative effects on the variety of outcomes studied.
• Further studies with lower risk of bias, larger sample size and non-convenient sample selection are required.
What impact this may have on practice or policy:
• The information from this study could be used to develop yoga-targeted interventions to employees in order
to reduce stress levels.
• However, there is a denite need for more high-quality studies to support the evidence of yoga at workplace
scenarios.
Downloaded from https://academic.oup.com/occmed/advance-article-abstract/doi/10.1093/occmed/kqz033/5382477 by guest on 20 March 2019

L. M.PUERTO VALENCIA ETAL.: YOGA IN THE WORKPLACE Page 3 of 9
including any mental health- or physical health-related
outcome, as well as outcomes related to performance at
work, were included. If available, adverse effects of the
programmes were included.
Two authors independently reviewed the title, abstract
and full text of each paper selected using the eligibility
criteria. Discrepancies were rechecked, and consensus
was achieved by discussion. Thereafter, details of the
study design, setting, participant characteristics, inter-
ventions and outcomes (including methodological qual-
ity) were extracted from each selectedstudy.
Considering the risk of underestimation or overestima-
tion of the effect in any RCT, an assessment of the risk of
bias was conducted using the Cochrane Collaboration’s
tool [17]. The trials should ensure that participants in
intervention and comparison groups are comparable
regarding known and unknown predictive characteristics.
In consequence, one of the criteria included in the qual-
ity assessment is selection bias risk, which measures the
conduction of an adequate randomization and allocation
concealment in the study. Other criteria included the per-
formance bias risk, measuring the degree to which par-
ticipants and researchers were blinded in the trial. Further
criteria comprised detection, attrition, reporting and other
bias risks, which followed the guidance of the Cochrane
Collaboration’s. Additionally, we provided an appraisal of
the overall risk of bias within each trial taking into account
the relative importance of the different criteria [18].
A qualitative description of all the studies fullling
the eligibility criteria was produced. For each reported
outcome, the effect of yoga compared with the control
group was described based on the original study nd-
ings. We decided not to perform a meta-analysis due to
the diversity of the studies, regarding outcomes, study
populations and interventions.
Results
The literature search retrieved 1343 papers of which 253
were duplicates. Through screening of title and abstracts,
54 studies were assessed for full-text evaluation. Of these,
41 did not meet the inclusion criteria. The reasons for
exclusion are shown in Figure 1. Thirteen controlled tri-
als with 1297 participants met the eligibility criteria for
inclusion in the systematic review. Characteristics of the
setting and sample population, interventions, outcomes
assessment and main results are given in Table S1 (avail-
able as Supplementary data at Occupational Medicine
Online).
Five of the trials were conducted in India [19–22],
three in the USA [23–25], two in the UK [5,26] and
one study each in China [27], Taiwan [28] and Sweden
[11]. Three studies had interventions directed towards
mental health professionals (two specically to nurses)
[23,27,28]; in three studies, the interventions targeted
military personnel (air force, army, military base)
[20,25,29]; and one study each concerned university staff
[5], a software company [22] and a factory (industrial
workers) [21]. Four studies targeted a specic subgroup
of employees, with higher scores of stress [11,24,26] or
counterproductive work behaviour [19] from an infor-
mation technology rm, a local government authority,
and an insurance and a nancial company. Participants’
mean age varied from 22 to 46years. Women accounted
for between 0 and 100% of the study participants and
had sample sizes ranging from 28 to 205 participants.
An explicit type of yoga practice was stated in seven
studies. From these, two described Kundalini yoga
[11,23], two Dru yoga [5,26], two Hatha yoga [25,29]
and one Vini yoga [24] as the yoga subtype of prac-
tice. These yoga practices have some subtle differences.
Kundalini yoga incorporates additional sequences of
physical postures and has a strong focus on meditation
and breathing [30]. Dru yoga has distinctive soft and
owing movements mixed with breathing and visualiza-
tion awareness [31]. Hatha yoga is a greater aerobic yoga
style with a strong focus on physical tness [32]. Lastly,
Vini yoga emphasizes slow and comfortable breathing
during all physical movements and allows many move-
ment variations [33]. However, all three main compo-
nents of yoga, postural exercises, breathing control and
meditation [12], were included in the 13 RCTs. Duration
and frequency of the yoga sessions varied between the 13
included studies. Programme length and intensity ranged
from once weekly during 8 weeks [23,26] to six sessions
per week during 6months [29]. From the 13 studies, one
had two control arms [24]. Ten studies compared yoga
with no intervention [5,20–28]. In two studies, yoga was
compared with physical exercises [19,29], and in one
study each, yoga was compared with mindfulness-based
stress management [24] and cognitive behaviour therapy
[11].
The group of reported outcomes varied largely,
details of the included trials with their outcomes are
given in Table S1 (available as Supplementary data at
Occupational Medicine Online). Nine studies reported
diverse mental health outcomes using several self-report
measures [5,11,19,23–28], two studies reported quality
of life [11,25], while four trials reported cardiovascular
outcomes [11,24,28,29]. In addition, other outcomes
such as pain, biological stress markers and aerobic capac-
ity were reported in one study each [11,20–22,24,26,29].
The risk of bias for each study included in the system-
atic review was assessed using the Cochrane Risk of Bias
Tool [17] which includes seven criteria with rating ‘yes’,
‘no’ or ‘unclear’. The summary of the risk of bias assess-
ment for the 13 included studies is presented in Table 1.
In general, the risk of selection bias was unclear.
Ten trials [5,11,20–24,26–29] did not specify any
method for allocation concealment, while in six trials
[11,20,23,24,28,29], the random numbers generator
method was not stated, in consequence random sequence
Downloaded from https://academic.oup.com/occmed/advance-article-abstract/doi/10.1093/occmed/kqz033/5382477 by guest on 20 March 2019

Page 4 of 9 OCCUPATIONAL MEDICINE
generation was unclear. The risk of performance bias was
high in all 13 trials. However, blinding of participants and
personnel in trials with interventions such as yoga is very
unlikely, if not impossible to achieve. In consequence, the
risk of detection bias in the 10 trials with any subjective
self-reported outcome [5,11,19,22–28] was high as well.
Blinding of the researchers assessing objective outcomes
was not specied in any of the seven trials reporting this
type of outcomes [11,20–22,24,28,29], as a result the
risk of bias was unclear. Risk of attrition bias was mixed;
four trials [20,22,26,29] had a high risk because of no
intention-to-treat analysis when losses of follow-up were
highly likely to bias the results, ve trials [19,23–25,28]
had a lower risk and four trials [5,11,21,27] had an
unclear risk. Risk of reporting bias was unclear in all but
three trials [11,20,29] where the risk was high, due to
incomplete analysis of more than one outcome of inter-
est. Protocols of the trials were not retrievable through
any of the 13 studies. Other risks of bias are unclear
in nine trials [5,11,20,21,23,24,26,27,29] as a result
of insufcient information available to judge whether
blinding of researchers at any stage. Overall, 9 out of
13 trials were classied as having an unclear risk of bias
[5,11,21–24,26–28].
With respect to mental health outcomes, differ-
ent types of stress were reported in ve included tri-
als [11,24,26–28]. Four studies reported signicant
positive effects of yoga compared with no intervention
[24,26–28], while one study comparing yoga with cogni-
tive behaviour therapy did not report the results [11].
The two studies describing Perceived Stress Scale (PSS)
scores showed a benecial effect of yoga compared with
* References lists of full text articles manually searched for additional studies.
Records identified through main
databases searching (N = 1153):
Embase (Ovid) (n = 197)
Pubmed (n = 586)
PsycINFO (EBSCO) (n = 235)
Cochrane (Wiley) (n = 135)
Screening
Included Eligibility Identification
Additional records identified
through other sources (N = 190):
IndMED (n = 27)
WHO International Clinical Trials
Registry Platform (n = 79)
ClinicalTrials.gov (n = 80)
Cross references* (n = 4)
Records after duplicates removed
(n = 1090)
Records screened [Abstracts
retrieved for evaluation]
(n = 240)
Records excluded (N = 850):
[Topic not related to efficacy,
effectiveness of Yoga] (n = 433) or [Other
type of study or publication] (n = 44) or
[Focus on treatment, rehabilitation of
patients] (n = 254) or [Main intervention
not yoga] (n = 10) or [Participants not adult
employees] (n = 66) or [Setting not work
place scenarios] (n = 42) or [Duplicate]
(n = 1)
Full-text articles assessed for
eligibility (n = 54)
Articles excluded (N = 186):
[Other type of study or publication]
(n = 40) or [Focus on treatment,
rehabilitation of patients] (n = 4) or [Main
intervention not yoga] (n = 37) or
[Participants not adult employees] (n = 27)
or [Setting not work place scenarios]
(n = 68) or [Duplicate] (n = 9) or [Article
could not be obtained] (n = 1)
Studies included in systematic
review
(n =13)
Full-text articles excluded
(N = 41):
[Other type of publication] (n = 9) or [Not
a randomized controlled trial] (n = 17) or
[Participants not adult active employees]
(n = 4) or [Participants not exclusively
adult active employees] (n = 2) or [Main
setting not work place scenarios] (n = 6) or
[Language Chinese] (n = 1) or [Main
intervention not yoga] (n = 2)
Figure 1. Study ow diagram.
Downloaded from https://academic.oup.com/occmed/advance-article-abstract/doi/10.1093/occmed/kqz033/5382477 by guest on 20 March 2019

L. M.PUERTO VALENCIA ETAL.: YOGA IN THE WORKPLACE Page 5 of 9
no intervention (P<0.01) [26] and a difference in scores
between yoga, mindfulness-based stress programme and
no intervention group (P < 0.001) [24]. Each of the
studies assessing work-related stress in health profes-
sionals reported a signicant improvement in work stress
scores [27,28]. One reported a proportion difference of
41% (95% condence interval [CI] 22.25–56.12) and P
<0.001 [27], and the other one reported a mean differ-
ence of −27.78 with an SE of 8.64 and P <0.01 [28]. On
the contrary, no signicant effect of yoga on stress adap-
tation (changes to stress regulation, comprising problem-
atic and emotional adjustments) was reported [28].
The effects of yoga on sleep quality were benecial
in the two studies. One compared yoga with no inter-
vention in nurses and reported a mean difference of
−2.70 (95% CI −3.44 to −1.96) and P < 0.001 [27].
The second reported a difference in sleep quality scores
between yoga, mindfulness-based stress intervention and
no intervention with P <0.05 [24].
Two individual studies that included mindfulness as
an outcome reported no signicant ndings when com-
paring yoga with the control group. The rst study was
conducted in nurses and compared yoga with no inter-
vention describing a mean difference of 3.95 (95% CI
−0.66 to 8.56) [23]. The other study showed no signi-
cant difference between yoga, mindfulness-based stress
programme and no intervention [24].
In the group of subjective and psychological well-
being outcomes, positive affectivity reported by one
trial with an overall low risk of bias showed a sig-
nicant enhancement in the yoga group compared
with the physical exercise group (P < 0.001) [19].
Furthermore, psychological well-being scores reported
by one trial were higher in participants who practised
yoga compared to participants who did not receive
any intervention (P < 0.001) [26]. In the same way,
emotional well-being measured by six dimensions
showed improvement in all (P<0.05) but one dimen-
sion (agreeable–hostile) when comparing yoga with no
interventionin one trial [5].
Regarding further mental health outcomes, in one
trial of nurses, professional burnout measured by the
Maslach Burnout Inventory, which contains three
domains, showed an improvement in emotional exhaus-
tion by a mean difference of −7.65 (95% CI −14.41 to
−0.86) and in depersonalization by a mean difference
of −2.65 (95% CI −5.28 to −0.02) but not in personal
accomplishment when yoga was compared with no
intervention [23]. Aggression scores reported by one
trial were reduced in the yoga group compared with the
physical exercise group (P < 0.001) [19]. In another
trial with an overall low risk of bias, on the one hand,
no effects of yoga compared with no intervention were
found in sensory processing outcomes, whereas on the
other hand yoga was effective in reducing anxiety state
and trait (P<0.001) [25]. With reference to depression,
yoga did not produce better effects when compared with
mindfulness-based stress management intervention or
Table 1. Summary of risk of bias assessment of the 13 included studies using the Cochrane Risk of Bias Tool [17]
Bias Selection bias Performance bias Detection bias Attrition bias Reporting bias Other bias Overall risk
of bias within
the triala
Random sequence
generation
Allocation
concealment
Blinding participants
and personnel
Blinding of outcome
assessment
Incomplete
outcome
data
Selective
reporting
Other sources
of bias
Alexander (2015) [23] Unclear Unclear HighbHigh Low Unclear Unclear Unclear
Dwivedi (2015) [19]Low Low HighbHigh Low Unclear Low Low
Fang (2015) [27]Low Unclear HighbHigh Unclear Unclear Unclear Unclear
Lin (2015) [28] Unclear Unclear HighbHighc/Unclear Low Unclear Low Unclear
Pal (2015) [20] Unclear Unclear HighbUnclear High High Unclear High
Rajbhoj (2015) [21]Low Unclear HighbUnclear Unclear Unclear Unclear Unclear
Hartel (2012) [26]Low Unclear HighbHigh High Unclear Unclear Unclear
Stoller (2012) [25]Low Low HighbHigh Low Unclear Low Low
Wolever (2012) [24] Unclear Unclear HighbHighc, Unclear Low Unclear Unclear Unclear
Hartel (2011) [5]Low Unclear HighbHigh Unclear Unclear Unclear Unclear
Telles (2009) [22]Low Unclear HighbHighc, Unclear High Unclear Low Unclear
Granath (2006) [11] Unclear Unclear HighbHighc, Unclear Unclear High Unclear Unclear
Ray (2001) [29] Unclear Unclear HighbUnclear High High Unclear High
Unclear: unclear risk of bias in the specied criteria; mainly due to not enough available information.
Low: low risk of bias in the specied criteria; the study describes properly measures to avoid this type of bias.
High: high risk of bias; there is highly likely that correct measures to avoid the type of bias were not performed.
aCriteria unlikely to be avoidable (such as high risk of performance bias) were not included in the judgement of the overall risk of bias within the trial.
bIn studies with behavioural interventions such as yoga, it is difcult, if not impossible to blind participants and personnel [12]. In consequence, there is a high risk of
performance bias, but it is not likely to be avoidable.
cTwo main class of outcomes assessed separately; subjective (self-rated scale) and objective outcomes in the same trial with differences in the risk of bias assessment.
Downloaded from https://academic.oup.com/occmed/advance-article-abstract/doi/10.1093/occmed/kqz033/5382477 by guest on 20 March 2019

Page 6 of 9 OCCUPATIONAL MEDICINE
no intervention [24]. Aself-care outcome measured in
one trial showed improvement in the yoga arm compared
to no intervention with a mean difference of 0.39 (95%
CI 0.14–0.64) [23].
In one trial with low risk of bias, in which the results
of quality of life were reported, yoga showed an improve-
ment (P > 0.05) in 16 of 18 mental health and qual-
ity-of-life factors when compared with no intervention
[25]. In another trial of university employees, the meas-
urement of positive psychological attitudes such as life
purpose satisfaction and self-condence during stress
showed improvement in the yoga group compared with
no intervention (P<0.01) [5].
In terms of cardiovascular outcomes, two studies
measured blood pressure [11,24]. A study conducted
on employees from an insurance company reported no
difference between systolic or diastolic blood pressure
values in the yoga group compared with mindfulness-
based stress management or no intervention group
[24]. Likewise, a study comparing yoga with cognitive
behaviour therapy in employees from the nancial sector
reported no different effect of yoga on diastolic blood
pressure [11].
The studies measuring heart rate variability showed
no improvement of the low-frequency range component
(LF) or the high-frequency range component (HF) in the
yoga intervention group compared with no intervention,
whereas the LF/HF ratio increased. To compare the groups
change of scores mean difference with a P cut value of 0.05
were used [28]. The other study measuring heart rate vari-
ability reported a change in the heart rhythm coherence
ratio between yoga, mindfulness-based stress and no inter-
vention (P<0.001), whereas no variations were found in
the RR interval (time between heart beats) between the
three comparison groups (P > 0.05) [24].
Other outcomes, for instance pain, biological stress
markers and aerobic capacity reported in only one study,
showed no effects, a positive effect of yoga (comparisons
resulted in P < 0.05) or the results were not reported.
With respect to pain, yoga had no effect or a positive
effect, depending on the type of pain measured [22,24].
Musculoskeletal exibility was positively affected by
yoga in comparison with no intervention [22]. No effects
of yoga were found on urinary catecholamine levels and
the results were not reported in the case of cortisol levels
[11]. In contrast, positive effects of yoga were found on
interleukin 1 beta and interleukin 10 measured in blood
[21]. Results from the comparison between yoga and the
control group in trials that measured aerobic capacity
and resting physiological conditions were not reported
[20,29]. No effects of yoga on productivity loss compared
with no intervention and mindfulness programme were
found [24]. Likewise, no effects on hand grip strength or
tapping speed were observed [22].
The detailed effects of yoga work-based programmes
for all the outcomes including the statistical comparisons
and the effect size when available are given in Table
S1 (available as Supplementary data at Occupational
Medicine Online). No adverse effect of yoga programmes
was reported in any of the trials.
Discussion
We identied 13 RCTs that examined the effect of yoga
offered exclusively to employees in workplace settings.
Male and female ofce employees, health professionals,
military personnel and industrial workers in India, USA,
UK, China, Taiwan and Sweden were studied. A large
variety of outcomes were reported; however, primar-
ily mental health outcomes were found. Risk of bias of
the trials was mainly unclear (9 out of 13). Stress was
the most frequently reported outcome in the trials (5 of
13). Four trials showed positive effects of yoga on stress
compared with control groups (measured by diverse
test comparisons). For other mental health outcomes,
results were varied: yoga in trials measuring sleep quality
reported better effects compared with controls, but in the
case of mindfulness, both groups showed similar results.
In terms of subjective and psychological well-being out-
comes, three trials showed a difference favouring yoga
over controls. However, most of the cardiovascular out-
comes resulted in no signicant difference between yoga
and control groups. No adverse effects of yoga were
reported in any trials.
The effects of yoga in workplace settings were meas-
ured using a wide range of outcomes. This was expected
as the population of the 13 trials varied greatly and the
potential effects of yoga are diverse, involving physi-
ological and mental health-related outcomes [11,34].
Furthermore, since certain groups of employees are
more prone to particular syndromes or diseases [9], the
outcome of interest depended on the study population.
For example, in the trials with employees from the army,
oxygen consumption and sensory processing were the
main outcomes [20,25,29].
Although a variety of outcomes were found, mostly
mental health endpoints, and in particular stress, were
reported in the 13 trials. Only recent investigations con-
sidered the study of yoga on physiological markers [35],
and the most commonly studied effects of yoga were in
the eld of mental health [36].
The benecial mental effects of yoga have been rec-
ognized and it has been used to reduce stress [12,36].
Moreover, stress is frequent and particularly important
in employees who have an occupation that requires low
physical activities (e.g. ofce workers) [37]. Stress was the
most common reported outcome in the included stud-
ies. The PSS employed in three of the studies [11,24,26]
is an extensively self-reported measure of psychological
stress during the last month and has been used as an out-
come measure in several studies [38,39]. The PSS has 10
questions with a total score ranging between 0 and 40
Downloaded from https://academic.oup.com/occmed/advance-article-abstract/doi/10.1093/occmed/kqz033/5382477 by guest on 20 March 2019

L. M.PUERTO VALENCIA ETAL.: YOGA IN THE WORKPLACE Page 7 of 9
[40]. Interpretation of the effect size is not straight for-
ward, taking into account that the PSS is not a diagnos-
tic instrument and cannot classify participants in groups
of stress levels as low, medium or high [41]. However, a
change in scores between the yoga group and controls
was reported in two of the trials (P<0.01), which means
that an overall reduction of stress in employees practis-
ing yoga was larger than in those not performing yoga.
It is important to consider that these two trials had an
unclear risk of bias and small sample size (n=143 and
n = 59). Nonetheless this reduction is promising, con-
sidering that yoga was only taught for 8 and 12 weeks.
Furthermore, this reduction is in line with the results of
a systematic review about yoga as a stress management
strategy [42]. In this review, three of four studies con-
ducted in a healthy population showed a reduction in
perceived stress scores in the yoga group [42].
In the case of work-related stress, a positive effect in
health professionals practising yoga was reported in two
studies with an unclear risk of bias [27,28]. The results of
one of the trials show that there was a higher proportion
of low-stressed participants in the yoga group. The scale
employed is validated and applied in various Taiwanese
healthcare settings [43]. The other trial reported a
mean difference of −27.78 (SE 8.64; P=0.01) between
yoga and control, and the work-related stress scale by
Lan (2004) was employed [28]; however, this scale
is a very specic instrument, and not widely applied.
Consequently, the interpretation of the meaning of this
reduction is uncertain.
Most of the mental health outcomes resulted in posi-
tive effects of yoga; for example, sleep quality improve-
ment in two trials with unclear risk of bias [24,27],
consistent with other studies showing a benecial effect
of yoga on sleep quality and a reduction of insomnia, has
been reported in special populations, such as postmeno-
pausal women and geriatric patients [44,45].
In contrast, most of the cardiovascular outcomes
did not result in distinct effects of yoga; no signicant
differences were found between comparison groups.
Furthermore, other outcomes such as pain reported
benecial or no effects of yoga programmes. In general,
therefore, it seems that yoga had no adverse effect but
not always showed an improvement of the outcomes. It
is important to consider that most of the trials included
in this systematic review had an overall unclear risk of
bias, mainly because of unclear allocation concealment
and selective reporting. Allocation concealment or pre-
venting knowledge of the next participant assignment
(into the intervention or the control group) is necessary
to avoid selection bias. The foreknowledge of interven-
tion assignment could cause selective enrolment of par-
ticipants. In the case of unclear selective reporting, the
specic concern is the possibility that only positive results
are reported in the publication leading to bias [18].
More information is needed to assess the quality of the
studies. In addition, blinding of participants and per-
sonnel in trials with behavioural interventions is almost
impossible. The validity of the results could be affected
by performance bias. Knowing the intervention group
may affect the outcomes, participants randomly allocated
to the intervention might put more effort on changes in
behaviour than participants from the control group [46].
However, as already mentioned, it is usually impossible
to blind participants in a study with an intervention like
yoga [18].
Of the 13 trials included in this systematic review,
most of the outcome variables were subjective. The
weight of this issue was indirectly addressed in the
risk of bias assessment; however, information about
reliability and validity of the instrument to measure
each outcome has to be considered in order to make a
deeper analysis. For example, the most common scale
to measure stress, the PSS, had appropriate reliability
and validity [47], but the evidence about validity and
measurement error of the instruments assessing mind-
fulness is insufcient [48].
In terms of external validity of the results, conveni-
ence samples to select potential participants were used
in all the included studies. The samples used in rand-
omized clinical trials are usually convenient [49,50]; as a
consequence, the external validity of the results is com-
promised. In addition to the convenience sample, gener-
alizability is restricted because of the population selected
in some of the trials. Four of 13 studies included a spe-
cic subgroup of employees, the ones with higher scores
of stress [11,24,26] or counterproductive work behaviour
[19]. Furthermore, overall sample size of the individual
trials was small, the biggest trial considered around 200
participants [24].
One of the strengths of the review is the method of
identifying relevant trials and the relative high sensitiv-
ity of the literature search. Use of additional databases
such as Indmed, clinicaltrials.org and WHO platform
increases sensitivity. In particular, the use of a database
from India (IndMED) involves a consideration about an
essential geographic source of data, taking into account
that yoga originated in India several centuries ago [12].
However, to have a broader view of the evidence, there is
a need to search other Indian literature, Indian languages
or other languages besides English.
Future randomized controlled studies, which specify
explicitly their methods, are needed in order to have a
clearer assessment of risk of bias. The analysis of more
studies with lower risk of bias, larger sample size and the
inclusion of studies with non-convenient sample selec-
tion would lead to a broader applicability of the results.
The results of this systematic review show that yoga
for employees has been measured according to a vari-
ety of outcomes. The systematic review results show that
yoga in the workplace had no negative effects on the vari-
ous outcomes studied.
Downloaded from https://academic.oup.com/occmed/advance-article-abstract/doi/10.1093/occmed/kqz033/5382477 by guest on 20 March 2019

Page 8 of 9 OCCUPATIONAL MEDICINE
Taken together, these ndings suggest a role for yoga
in promoting health in the workplace. However, the evi-
dence comes from few studies, with small sample sizes
and unclear risk of bias.
Competing interests
None declared.
References
1. Rantanen J, Fedotov IA. Standards, principles and
approaches in occupational health services 1992. In:
Encyclopaedia of Occupational Health and Safety [Internet].
4th edn. Hamilton, Ontario, Canada: International Labour
ofce. http://www.ilocis.org/documents/chpt16e.htm (3
July 2017, date last accessed).
2. ConnVS, HafdahlAR, CooperPS, BrownLM, LuskSL.
Meta-analysis of workplace physical activity interventions.
Am J Prev Med 2009;37:330–339.
3. KennyDT, CooperCL. Introduction: occupational stress
and its management. Int J Stress Manage 2003;10:275–279.
4. LelliotP, TullochS, BoardmanJ, HarveyS, HendersonM,
Knapp M. Mental Health at Work. London, UK: Royal
College of Psychiatrists, 2008. https://www.gov.uk/gov-
ernment/uploads/system/uploads/attachment_data/
le/212266/hwwb-mental-health-and-work.pdf (5 June
2017, date last accessed).
5. HartelN, HavenhandJ, KhalsaSB, ClarkeG, KrayerA.
The effectiveness of yoga for the improvement of well-
being and resilience to stress in the workplace. Scand J Work
Environ Health 2011;37:70–76.
6. Axén I, Follin G. Medical yoga in the workplace set-
ting–perceived stress and work ability—a feasibility study.
Complement Ther Med 2017;30:61–66.
7. GuraST. Yoga for stress reduction and injury prevention at
work. Wo rk 2002;19:3–7.
8. World Health Organization. WHO Traditional Medicine
Strategy 2014–2023. Geneva, Switzerland: World Health
Organization, 2013.
9. Sakuma Y, Sasaki-OtomaruA, IshidaS et al. Effect of a
home-based simple yoga program in child-care workers:
a randomized controlled trial. J Altern Complement Med
2012;18:769–776.
10. ChuAH, KohD, MoyFM, Müller-RiemenschneiderF.
Do workplace physical activity interventions
improve mental health outcomes? Occup Med (Lond)
2014;64:235–245.
11. GranathJ, IngvarssonS, vonThieleU, LundbergU. Stress
management: a randomized study of cognitive behavioural
therapy and yoga. Cogn Behav Ther 2006;35:3–10.
12. HartleyL, DyakovaM, HolmesJ etal. Yoga for the primary
prevention of cardiovascular disease. Cochrane Database
Syst Rev 2014;(5):CD010072.
13. Cramer H, LaucheR, HallerH, Dobos G. A systematic
review and meta-analysis of yoga for low back pain. Clin J
Pain 2013;29:450–460.
14. JeterPE, NkodoAF, Moonaz SH, DagnelieG. A system-
atic review of yoga for balance in a healthy population. J
Altern Complement Med 2014;20:221–232.
15. Cheema BS, Marshall PW, Chang D, Colagiuri B,
Machliss B. Effect of an ofce worksite-based yoga pro-
gram on heart rate variability: a randomized controlled
trial. BMC Public Health 2011;11:578.
16. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA
Group. Preferred reporting items for systematic reviews
and meta-analyses: the PRISMA statement. Ann Intern
Med 2009;151:264–269, W64.
17. HigginsJPT, AltmanDG, GøtzschePC etal. The Cochrane
Collaboration’s tool for assessing risk of bias in randomised
trials. Br Med J 2011;343:d5928.
18. Higgins JPT, Grenn S (eds.). Cochrane Handbook for
Systematic Reviews of Interventions Version 5.1.0. The
Cochrane Collaboration, 2011. http://handbook.cochrane.
org (27 September 2018, date last accessed).
19. Dwivedi U, Kumari S, Akhilesh KB, Nagendra HR.
Well-being at workplace through mindfulness: inuence
of yoga practice on positive affect and aggression. Ayu
2015;36:375–379.
20. Pal R, Singh SN, Halder K, Tomer OS, Mishra AB,
SahaM. Effects of yogic practice on metabolism and anti-
oxidant-redox status of physically active males. J Phys Act
Health 2015;12:579–587.
21. Rajbhoj PH, Shete SU, Verma A, Bhogal RS. Effect of
yoga module on pro-inammatory and anti-inammatory
cytokines in industrial workers of Lonavla: a randomized
controlled trial. J Clin Diagn Res 2015;9:CC01–CC05.
22. TellesS, DashM, NaveenKV. Effect of yoga on muscu-
loskeletal discomfort and motor functions in professional
computer users. Wo r k 2009;33:297–306.
23. AlexanderGK, RollinsK, Walker D, WongL, PenningsJ.
Yoga for self-care and burnout prevention among nurses.
Workplace Health Saf 2015;63:462–470; quiz 471.
24. Wolever RQ, Bobinet KJ, McCabe K et al. Effective
and viable mind-body stress reduction in the workplace:
a randomized controlled trial. J Occup Health Psychol
2012;17:246–258.
25. StollerCC, GreuelJH, CiminiLS, FowlerMS, KoomarJA.
Effects of sensory-enhanced yoga on symptoms of com-
bat stress in deployed military personnel. Am J Occup Ther
2012;66:59–68.
26. Hartel N, Burton C, Rycroft-Malone J et al. Yoga for
reducing perceived stress and back pain at work. Occup
Med (Lond) 2012;62:606–612.
27. Fang R, LiX. A regular yoga intervention for staff nurse
sleep quality and work stress: a randomised controlled trial.
J Clin Nurs 2015;24:3374–3379.
28. Lin SL, HuangCY, Shiu SP, Yeh SH. Effects of yoga on
stress, stress adaptation, and heart rate variability among
mental health professionals—a randomized controlled
trial. Worldviews Evid Based Nurs 2015;12:236–245.
29. Ray US, Sinha B, Tomer OS, Pathak A, Dasgupta T,
Selvamurthy W. Aerobic capacity and perceived exertion
after practice of Hatha yogic exercises. Indian J Med Res
2001;114:215–221.
30. SarkissianM, TrentNL, HuchtingK, SinghKhalsa SB.
Effects of a Kundalini yoga program on elementary and
middle school students’ stress, affect, and resilience. J Dev
Behav Pediatr 2018;39:210–216.
31. DiStasioSA. Integrating yoga into cancer care. Clin J Oncol
Nurs 2008;12:125–130.
Downloaded from https://academic.oup.com/occmed/advance-article-abstract/doi/10.1093/occmed/kqz033/5382477 by guest on 20 March 2019

L. M.PUERTO VALENCIA ETAL.: YOGA IN THE WORKPLACE Page 9 of 9
32. TranMD, HollyRG, LashbrookJ, AmsterdamEA. Effects
of Hatha yoga practice on the health-related aspects of
physical tness. Prev Cardiol 2001;4:165–170.
33. Birdee GS, Ayala SG, Tyree R, Buchowski M. Oxygen
consumption during Viniyoga practice in adults. Int J Yoga
2018;11:194–200.
34. Sharma M, Haider T. Yoga as an alternative and com-
plementary treatment for patients with low back pain: a
systematic review. J Evid Based Complement Altern Med
2013;18:23–28.
35. JonesSM, GuthrieKA, ReedSD etal. A yoga and exercise
randomized controlled trial for vasomotor symptoms: effects
on heart rate variability. Complement Ther Med 2016;26:66–71.
36. Deshpande RC. A healthy way to handle work place
stress through yoga, meditation and soothing humor. Int J
Environ Sci 2012;2:2143–2154.
37. CheemaBS, HouridisA, BuschL etal. Effect of an ofce
worksite-based yoga program on heart rate variability: out-
comes of a randomized controlled trial. BMC Complement
Altern Med 2013;13:82.
38. Morgan ES, Umberson K, Hertzog C. Construct vali-
dation of self-reported stress scales. Psychol Assess
2014;26:90–99.
39. Sharma M. Yoga as an alternative and complementary
approach for stress management: a systematic review. J
Evid Based Complementary Altern Med 2014;19:59–67.
40. RobertiJW, HarringtonLN, StorchEA. Further psycho-
metric support for the 10-item version of the Perceived
Stress Scale. J Coll Counsel 2006;9:135–147.
41. Smith KJ, RosenbergDL, TimothyHaightG. An assess-
ment of the psychometric properties of the Perceived Stress
Scale-10 (PSS10) with business and accounting students.
Account Perspect 2014;13:29–59.
42. Field T. Yoga research review. Complement Ther Clin Pract
2016;24:145–161.
43. See L, Chang H, Liu M, Cheng H. Development and
evaluation of validity and reliability of a question-
naire on medical worker’s stress. Taiwan J Public Health
2007;26:452–461.
44. Sarris J, Byrne GJ. A systematic review of insomnia and
complementary medicine. Sleep Med Rev 2011;15:99–106.
45. AfonsoRF, Hachul H, KozasaEH et al. Yoga decreases
insomnia in postmenopausal women: a randomized clini-
cal trial. Menopause 2012;19:186–193.
46. KypriK, WilsonA, AttiaJ, SheeranPJ, McCambridgeJ.
Effects of study design and allocation on self-reported
alcohol consumption: randomized trial. Trials 2015;
16:127.
47. Lee EH. Review of the psychometric evidence of the
Perceived Stress Scale. Asian Nurs Res (Korean Soc Nurs
Sci) 2012;6:121–127.
48. Park T, Reilly-Spong M, Gross CR. Mindfulness: a sys-
tematic review of instruments to measure an emer-
gent patient-reported outcome (PRO). Qual Life Res
2013;22:2639–2659.
49. Martínez-Mesa J, González-Chica DA, Duquia RP,
Bonamigo RR, Bastos JL. Sampling: how to select
participants in my research study? An Bras Dermatol
2016;91:326–330.
50. Deaton A, Cartwright N. Understanding and misun-
derstanding randomized controlled trials. Soc Sci Med
2018;210:2–21.
Downloaded from https://academic.oup.com/occmed/advance-article-abstract/doi/10.1093/occmed/kqz033/5382477 by guest on 20 March 2019