See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/41164086

The Health Benefits of Yoga and Exercise: A Review of Comparison Studies

Article  in  Journal of alternative and complementary medicine (New York, N.Y.) · January 2010
DOI: 10.1089/acm.2009.0044 · Source: PubMed

CITATIONS READS
571 20,973

2 authors:

Alyson Ross Sue A Thomas

National Institutes of Health University of Maryland, Baltimore
34 PUBLICATIONS   1,217 CITATIONS    111 PUBLICATIONS   5,669 CITATIONS   

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Alyson Ross on 12 December 2013.

The user has requested enhancement of the downloaded file.

THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE
Volume 16, Number 1, 2010, pp. 3–12 Original Articles
ª Mary Ann Liebert, Inc.
DOI: 10.1089=acm.2009.0044

The Health Benefits of Yoga and Exercise:

A Review of Comparison Studies

Alyson Ross, M.S.N., R.N., and Sue Thomas, F.A.A.N., Ph.D., R.N.

Abstract

Objectives: Exercise is considered an acceptable method for improving and maintaining physical and emotional
health. A growing body of evidence supports the belief that yoga benefits physical and mental health via down-
regulation of the hypothalamic–pituitary–adrenal (HPA) axis and the sympathetic nervous system (SNS). The
purpose of this article is to provide a scholarly review of the literature regarding research studies comparing the
effects of yoga and exercise on a variety of health outcomes and health conditions.
Methods: Using PubMed
and the key word ‘‘yoga,’’ a comprehensive search of the research literature from core
scientific and nursing journals yielded 81 studies that met inclusion criteria. These studies subsequently were
classified as uncontrolled (n ¼ 30), wait list controlled (n ¼ 16), or comparison (n ¼ 35). The most common
comparison intervention (n ¼ 10) involved exercise. These studies were included in this review.
Results: In the studies reviewed, yoga interventions appeared to be equal or superior to exercise in nearly every
outcome measured except those involving physical fitness.
Conclusions: The studies comparing the effects of yoga and exercise seem to indicate that, in both healthy and
diseased populations, yoga may be as effective as or better than exercise at improving a variety of health-related
outcome measures. Future clinical trials are needed to examine the distinctions between exercise and yoga,
particularly how the two modalities may differ in their effects on the SNS=HPA axis. Additional studies using
rigorous methodologies are needed to examine the health benefits of the various types of yoga.

Introduction chologic effects, primarily as a result of the release of cortisol

and catecholamines (epinephrine and norepinephrine). This

Y oga is an ancient discipline designed to bring balance

and health to the physical, mental, emotional, and spiri-
tual dimensions of the individual. Yoga is often depicted
response leads to the mobilization of energy needed to
combat the stressor through the classic ‘‘fight or flight’’
syndrome. Over time, the constant state of hypervigilence
metaphorically as a tree and comprises eight aspects, or resulting from repeated firing of the HPA axis and SNS can
‘‘limbs:’’ yama (universal ethics), niyama (individual ethics), lead to dysregulation of the system and ultimately diseases
asana (physical postures), pranayama (breath control), pratya- such as obesity, diabetes, autoimmune disorders, depression,
hara (control of the senses), dharana (concentration), dyana substance abuse, and cardiovascular disease.3,4
(meditation), and samadhi (bliss).1 Long a popular practice in As detailed in Figure 1, numerous studies have shown
India, yoga has become increasingly more common in Wes- yoga to have an immediate downregulating effect on both
tern society. In a national, population-based telephone survey the SNS=HPA axis response to stress. Studies show that yoga
(n ¼ 2055), 3.8% of respondents reported using yoga in the decreases levels of salivary cortisol,5,6 blood glucose,7,8 as
previous year and cited wellness (64%) and specific health well as plasma rennin levels, and 24-hour urine norepineph-
conditions (48%) as the motivation for doing yoga.2 rine and epinephrine levels.9 Yoga significantly decreases
A growing body of research evidence supports the belief heart rate and systolic and diastolic blood pressure.9–11 Stu-
that certain yoga techniques may improve physical and dies suggest that yoga reverses the negative impact of stress
mental health through down-regulation of the hypothalamic– on the immune system by increasing levels of immunoglob-
pituitary–adrenal (HPA) axis and the sympathetic nervous ulin A12 as well as natural killer cells.13 Yoga has been found
system (SNS). The HPA axis and SNS are triggered as a to decrease markers of inflammation such as high sensitivity
response to a physical or psychologic demand (stressor), C-reactive protein as well as inflammatory cytokines such as
leading to a cascade of physiologic, behavioral, and psy- interleukin-614 and lymphocyte-1B.15

School of Nursing, University of Maryland, Baltimore, MD.

3
4 ROSS AND THOMAS

FIG. 1. The impact of stress on the hypothalamic–pituitary–adrenal axis and the sympathetic nervous system. *Yoga has
been shown to have significant beneficial effects in these items.

These studies suggest that yoga has an immediate quieting the online database of biomedical journal citations produced
effect on the SNS=HPA axis response to stress. While the by the United States National Library of Medicine (NLM
).
precise mechanism of action has not been determined, it has Using the key word ‘‘yoga’’ and limiting the search to core
been hypothesized that some yoga exercises cause a shift clinical and nursing journals published in English, 183 arti-
toward parasympathetic nervous system dominance, possi- cles published after 1970 were identified. Although medita-
bly via direct vagal stimulation.16 Shapiro et al.17 noted sig- tion, one of the eight limbs of yoga, and yoga interventions
nificant reductions in low-frequency heart rate variability such as cleansing exercises arguably could be included in a
(HRV)—a sign of sympathetic nervous system activation—in scholarly review of yoga literature, studies solely focusing on
depressed patients following an 8-week yoga intervention. these modalities were excluded. Articles were eliminated if
Regardless of the pathophysiologic pathway, yoga has been they were editorials, anecdotal or single case studies, or of
shown to have immediate psychologic effects: decreasing extremely poor quality.
anxiety5,6,18,19 and increasing feelings of emotional, social, Studies were included in the review if they were of rea-
and spiritual well-being.20 sonably good quality and involved yoga asana as the primary
Several literature reviews have been conducted that ex- intervention modality. Quality of studies was determined
amined the impact of yoga on specific health conditions in- using the criteria outlined by Greenhalgh.26 Greenhalgh
cluding cardiovascular disease,21 metabolic syndrome,16 identified essential elements of quality research including:
diabetes,22 cancer,23 and anxiety.24 Galantino et al.25 pub- originality, appropriate subjects, sensible design, and mini-
lished a systematic review of the effects of yoga on children. mal bias. Much of the research regarding yoga interventions
These reviews have contributed to the large body of research has been done outside the United States with the majority of
evidence attesting to the positive health benefits of yoga. those studies, not surprisingly, done in India. Many of the
Many of the studies compared yoga to other treatment mo- early studies published in Indian journals prior to 1990 were
dalities, most commonly to exercise, meditation, and tradi- of questionable quality, with inadequate descriptions of
tional medicine. However, little has been written about what methodology and few randomized, controlled trials. How-
distinguishes yoga from other treatment modalities. The ever, the quality of more recent studies has improved no-
purpose of this article is to present a comprehensive review ticeably. Studies completed abroad were considered if they
of the literature regarding the impact of yoga compared to met the inclusion criteria and were available at the NLM.
exercise on a variety of health outcomes and conditions. Eighty-one (81) studies met the inclusion criteria and were
available at the NLM. Of these, more than half (n ¼ 46) were
published outside of the United States, with 29 of these
Methods
published in Indian journals. These 81 studies examined a
A comprehensive search for research articles focusing on wide range of outcome measures and included numerous
yoga interventions was completed from September until healthy and diseased populations. The studies were sepa-
December 2008. The articles were identified using PubMed
, rated into three categories: uncontrolled studies, controlled
HEALTH BENEFITS OF YOGA AND EXERCISE 5

studies, and comparison group studies. Thirty (30; 37.0%) of rosis,30 menopause,36 kidney disease,37 and schizophrenia.29
the studies were uncontrolled quasi-experimental studies Exercise has been recognized as having insulin-like effects on
typically comparing pretest and post-test scores on a variety blood glucose levels.38–40 Yoga has recently been found to
of outcome criteria following a yoga intervention. Sixteen (16; have beneficial effects on blood glucose levels in individuals
19.8%) were wait list or nonintervention controlled studies, of with diabetes and other chronic health conditions.7,8,10,41 In a
which 12 were randomized controlled trials. The remaining 35 blinded, randomized controlled trial involving 186 type 2
studies (43.2%) compared yoga to some other treatment mo- diabetics, Gordon et al. (2008) compared the effects of 6
dality. These 35 studies subsequently were classified accord- months of weekly classes plus home practice of yoga with
ing to the type of intervention being compared to yoga. The aerobic exercise plus stretching. Compared to baseline mea-
following categories of interventions were created: exercise, sures and a control group, both yoga and exercise led to sig-
relaxation response, usual medical treatment, psychotherapy= nificant reductions at 3 and 6 months in fasting blood glucose
cognitive interventions, and ‘‘other.’’ (29.48% and 27.43%, respectively, p < 0.0001).33 Both the ex-
The single largest category (n ¼ 12) of comparison studies ercise and yoga groups exhibited improvements in serum
involved the effects of yoga being compared to exercise, and total cholesterol ( p < 0.0001), and very low density lipopro-
it is this category that is the focus of this article. Several tein ( p ¼ 0.036) compared with controls. One indicator of
studies seemed to span multiple categories, such as usual oxidative stress—malondialdehyde—significantly decreased
cardiac care, which often utilizes an exercise component. in the yoga and exercise groups (19.9% and 18.1%, respec-
However, for the purposes of this article, only studies that tively, p < 0.0001 for both), and superoxide dismutase, a
listed exercise as the primary intervention were placed in measure of oxidative status, increased by 24.08% in the yoga
that category. group and 20.18% in the exercise group, ( p < 0.05 for both).
Yoga has been shown to be effective in relieving
Results symptoms of mental illness including depression,17,42,43
anxiety,6,44 obsessive–compulsive disorder,45 and schizo-
Table 1 details the populations, study methodology, and
phrenia.29 Duraiswamy et al.29 compared the effects of 4
outcome measures of the 12 studies comparing the effects of
months of daily yoga asana and pranayama with exercise on
yoga and exercise evaluated in this article. Five (5) of the 12
symptoms of psychosis in 61 schizophrenic patients receiv-
studies were conducted in the United States, 3 in India, and
ing antipsychotic treatment. The exercise intervention in-
1 each was completed in England, Germany, Turkey, and
volved walking, jogging, seated and standing exercises, and
Cuba. Eight of the studies (66.7%) were randomized con-
relaxation—activities that closely approximate yoga. Both
trolled trials. More than half of the studies (N ¼ 7) focused on
the yoga and exercise groups exhibited significant reductions
healthy populations, and the remainder focused on subjects
in psychotic symptom, but the yoga group improved sig-
with a wide variety of diseases and health conditions.
nificantly better (F ¼ 5.0, p ¼ 0.03). The yoga group scored
Table 2 provides a summary of the studies comparing
significantly better than the exercise group in social and oc-
yoga and exercise by outcomes measured. Nearly half of the
cupational functioning (F ¼ 7.98, p < 0.01) and on psycho-
studies have been conducted on healthy populations, and
logic, social, and environmental subscales of quality of life as
yoga interventions have yielded positive results in both
measured on the World Health Organization Quality of Life
healthy and diseased populations. Nearly every study uti-
BREF form (all p < 0.01).
lized a yoga intervention that combined physical asanas
Other studies using exercise interventions that closely
(standing, seated, or inverted) and restorative or relaxation
simulated the actions of yoga found clear differences be-
poses. Seven (7) of the 12 studies also incorporated medita-
tween yoga and exercise. Yurtkuran et al.26 conducted a
tion and=or breath work. Three (3) studies did not specify the
single-blind, randomized trial comparing the effects of yoga
type of yoga intervention used. The remaining studies uti-
with gentle range-of-motion exercises on symptoms related
lized Hatha yoga (N ¼ 4), Iyengar yoga (N ¼ 3), and In-
to hemodialysis in 37 renal failure patients. After 3 months of
tegrated yoga (N ¼ 2). While five of the studies provided
twice-weekly sessions consisting primarily of standing and
specific sequences of yoga poses used in the intervention, the
seated asanas and meditation, the yoga group exhibited sig-
remainder offered few details.
nificant reductions in pain (37%), fatigue (55%), and sleep
disturbance (25%) as measured by visual analog scales; these
Yoga and exercise
changes were significantly better than those in the exercise
In research involving the health benefits of yoga, exercise group ( p ¼ 0.03, p ¼ 0.008, p ¼ 0.04, respectively). The yoga
is the single most common intervention used as a compari- group also noticed significant beneficial changes from
son. Twelve (12) studies were found comparing the effects of baseline in grip strength (þ15%) and serum levels of urea
yoga and exercise (Table 1). Of these, nine focused on adults (
29%), creatinine (–14%), alkaline phosphatase (
15%), to-
and three on seniors. Excluding studies with no information tal cholesterol (
15%), erythrocytes (þ11%), and hematocrit
regarding gender or those involving exclusively one sex (þ13%). These changes also were better than those in the
(menopausal subjects), 597 (68.4%) of the 873 subjects who control group (all p < 0.05).
participated in the 12 studies were women. Most of the In addition to studies comparing the efficacy of yoga to
studies involved some form of aerobic exercise: walking, exercise in ill populations, studies have shown yoga to be
running, dancing, or stationary bicycling, plus some form of effective in relieving symptoms associated with natural life
stretching.5,27–35 Two (2) studies compared yoga with gentle, events in women such as pregnancy46,47 and menopause.36,48
nonaerobic exercises and stretching.36,37 Yoga appears to increase maternal comfort and shorten labor
Yoga appears to be equal or superior to exercise in relieving time in pregnant women,47 and decrease the number of hot
certain symptoms associated with diabetes,33 multiple scle- flashes in menopausal women.36,48 However, only one study
 Table 1. Studies Comparing Yoga to Exercise

First author Year Population (N) Yoga type Exercise type Duration Design Outcomes

Bowman 2006 Healthy seniors (26) Asana & pranayama Stationary cycling 6 weeks Randomized, VO2 max, BP, HRV,
controlled baroreflex sensitivity
trial (RCT)
Chattha 2008 Menopause (120) Asana & pranayama Walking & jogging, 8 weeks RCT Serum FSH, menopause
stretching, education symptoms (GCS),
PSS, personality traits (EPI)
Duraiswamy 2007 Schizophrenia (61) Asana & pranayama Walking, jogging 16 weeks Blinded RCT Psychotic symptoms (PANSS),
and exercises social functioning (SOFS),
QOL (WHOQOL-BREF)
Duren 2008 Healthy adults: Asana (inversions) Walking, cycling, N=A Retrospective Carotid artery distensibility,
yoga practitioners aerobics questionnaire pulse-wave velocity
& exercisers (26)
Gordon 2008 Type 2 diabetes (186) Asana & pranayama Aerobics & walking 24 weeks RCT repeated FBG, TC, LDL, VDL,
measures oxidative stress (MDA, SOD)
Hagins 2007 Healthy yoga Asana Treadmill walking Single session Repeated VO2 max, HR, %MHR, METs,
practitioners (20) measures and energy expenditure (kcals)
Khattab 2007 Healthy adults (22) Asana Walking 5 weeks Repeated HRV
measures
Oken 2004 Multiple sclerosis (57) Gentle Asana & Exercise 6 months Parallel group Cognitive attention, alertness,
meditation class þ stationary RCT POMS, STAI, fatigue (MFI),
biking, QOL (SF-36)
Oken 2006 Healthy seniors (134) Gentle Asana & Walking 6 months Parallel group Alertness, POMS, GHQ SF-36,
meditation RCT fatigue (MFI), balance
and flexibility
Sinha 2007 Healthy males (51) Asana, pranayama & Running & stretching 6 months RCT Reduced glutathione (GSS),
meditation oxidized glutathione (CSSG),
glutathione reductase (GR),
and total antioxidant status
West 2004 Healthy college Asana African dance Single class Quasi-experimental Salivary cortisol, PSS,
students (69) positive=negative
affect schedule
Yurtkuran 2007 Hemodialysis (40) Asana ROM exercises 12 weeks RCT Pain, fatigue, sleep (VAS),
grip strength, Urea, creatinine,
Ca, blood lipids, CBC

N=A, not applicable; ROM, range of motion; VO2 max, maximum oxygen consumption; HR, heart rate; %MHR, percentage maximum predicted heart rate; METs, metabolic equivalents; BP, blood
pressure; HRV, heart rate variability; FSH, follicle-stimulating hormone; GCS, Greene Climateric Scale; PSS, Perceived Stress Scale; EPI, Eysenck Personality Inventory; PANSS, Positive and Negative
Syndrome Scale; SOFS, Social Occupational Functioning Scale; QOL, quality of life; WHOQOL-BREF, World Health Organization Quality of Life Abbreviated form; FBG, fasting blood glucose; TC, total
cholesterol; LDL, low-density lipoprotein; VLDL, very-low-density lipoprotein; MDA, malondialdehyde; SOD, superoxide dismutase; POMS, Profile of Mood States; STAI, State-Trait Anxiety Inventory;
MFI, Multidimensional Fatigue Inventory; SF-36, 36-Item Short Form Health Survey; GHQ, General Health Questionnaire; GSS, glutathione synthetase; CSSG, cysteine-glutatione disulfide; VAS, visual
analogue scale; CBC, complete blood count.
HEALTH BENEFITS OF YOGA AND EXERCISE 7

Table 2. Outcomes of Studies Comparing Yoga to Exercise Based on Health Status

Health status Comparison of yoga to exercise

Outcome First author (year) Healthy Diseased Less beneficial than Similar or equal to More beneficial

Balance Oken (2006) x x

Baroreflex sensitivity Bowman (1997) x x
Carotid artery distensibility Duren (2008) x x
Energy expenditure (kcals) Hagins (2007) x x
Fasting blood glucose Gordon (2008) x x
Fatigue (MFI) Oken (2003) x x
Fatigue: reduced activity Oken (2006) x x
(MFI)
Fatigue (VAS) Yurtkuran (2007) x x
Flexibility Oken (2006) x x
Glutathione Sinha (2007) x x
reductase (GR)
HR Bowman (1997) x x
HRV Bowman (1997) x x
HRV Khattab (2007) x x
Kidney function (urea, Yurtkuran (2007) x x
creatinine)
Malondialdehyde Gordon (2008) x x
Menopausal symptoms Chattha (2008) x x
Metabolic equivalents Hagins (2007) x x
Mood: negative affect West (2004) x x
(PANAS)
Mood: positive affect West (2004) x x
(PANAS)
Pain (VAS) Yurtkuran (2007) x x
% MHR Hagins (2007) x x
Psychotic symptoms Duraiswamy (2007) x x
(PANSS)
Pulse-wave velocity Duren (2008) x x
Quality of Life (SF-36) Oken (2004) x x
Quality of Life Duraiswamy (2007) x x
(WHOQOL-BREF)
Reduced glutathione Sinha (2007) x x
(GSH)
Sleep disturbance (VAS) Yurtkuran (2007) x x
Social and occupational Duraiswamy (2007) x x
functioning (SOFS)
Strength (grip strength) Yurtkuran (2007) x x
Stress (PSS) Chattha (2008) x x
Stress (PSS) West (2004) x x
Stress (salivary cortisol) West (2004) x x
Superoxide dismutase Gordon (2008) x x
Total antioxidant status Sinha (2007) x x
Total cholesterol Gordon (2008) x x
Total cholesterol Yurtkuran (2007) x x
VO2 max Bowman (1997) x x

MFI, Multidimensional Fatigue Inventory; VAS, visual analogue scale; HR, heart rate; HRV, heart rate variability; PANAS, Positive and
Negative Affect Scale; %MHR, percentage maximum predicted heart rate; PANSS, Positive and Negative Syndrome Scale; WHOQOL-BREF,
World Health Organization Quality of Life Abbreviated form; SOFS, Social Occupational Functioning Scale; PSS, Perceived Stress Scale;
VO2max, maximum oxygen consumption.

compared yoga to exercise in healthy women. In a blinded, all three factors: psychologic, somatic, and vasomotor
randomized controlled trial involving 120 menopausal ( p < 0.001), while the exercise group exhibited significant
women, Chattha et al.36 compared the effects of an 8-week improvement only in the psychologic factor ( p < 0.05). The
regimen of daily asana and pranayama with an intervention yoga group also exhibited a significantly greater decrease in
that mimicked the activities of yoga by utilizing non- levels of stress, measured on the Perceived Stress Scale (PSS),
strenuous walking and stretching exercises. The yoga group than the exercise group ( p < 0.0001, effect size ¼ 1.10 and
scored significantly better compared to the exercisers on 0.27, respectively).
vasomotor symptoms ( p < 0.05) and neuroticism ( p < 0.05). These findings seem to indicate that both interventions
Analysis of data from the Greene Climacteric Scale revealed made subjects feel better, but yoga seemed to do better at
that the yoga group exhibited significant improvement in relieving physical symptoms and perceptions of stress.
8 ROSS AND THOMAS

Only one group of researchers has compared the efficacy of yoga group experienced significant reductions in levels of
yoga and exercise in both healthy and ill populations.30,31 salivary cortisol ( p < 0.05). Levels of salivary cortisol, an in-
Oken et al.30 compared the effects of 6 months of Iyengar yoga dicator of activation of the HPA axis response to stress,
and stationary cycling on attention, alertness, mood, anxiety, significantly increased in the African dance group
fatigue, and quality of life in 69 adult subjects (53 women) ( p < 0.0001). These findings indicate that yoga and exercise
with multiple sclerosis. Both interventions produced signifi- may both improve mood but affect the HPA axis differently.
cant improvement in fatigue compared with wait-list controls The study, while intriguing, utilized a convenience sample
( p < 0.01). No significant improvements were noted in either and based the results on a one-time intervention, limiting the
intervention in attention and alertness or quality of life. generalizability of the findings.
Similar negative results regarding cognitive function were In an interesting study, Khattab et al.34 used 24-hour
found in a later study involving the effects of Iyengar yoga Holter monitoring to compare HRV during a 60-minute yoga
and walking in 135 healthy seniors.31 Again, no changes practice versus 60 minutes of park walking in a small sample
were noted in cognitive outcomes or alertness. In this study, (N ¼ 11) of long-term Iyengar yoga practitioners as well as in
the yoga group performed significantly better than the ex- healthy, age- and sex-matched control subjects who had no
ercise group on levels of fatigue ( p ¼ 0.006) and on several prior experience with yoga or meditation. The yoga practi-
measures of quality of life including pain ( p ¼ 0.006) and tioners exhibited greater HRV, particularly in those measures
social functioning ( p ¼ 0.015). Only the yoga group exhibited associated with parasympathetic tone, during the yoga in-
significant improvements in flexibility ( p ¼ 0.05) and balance tervention than during walking ( p < 0.001), and during both
( p ¼ 0.05). yoga and walking than in the control group during yoga and
While it is possible that the differences in the findings of walking ( p < 0.001, p < 0.05, respectively). While no signifi-
the two studies by Oken et al. regarding quality of life was cant differences were found in the yoga practitioners and
due to differences in the type of exercise or the populations control subjects in HRV outside of the interventions, the
involved, it is also possible that differences were related to a authors of the study attributed this finding to the small
lack of power, as the number of subjects in the second study sample size.
(n ¼ 134) was nearly double the first (n ¼ 57). Further studies Bowman et al.,28 in a randomized, controlled trial in-
are needed to examine whether a larger dose of the inter- volving 26 healthy seniors, provided evidence supporting
ventions (both studies used a single 90-minute classes per the belief that exercise and yoga exert different effects on the
week) might be more likely to affect the cognitive outcomes. SNS. Heart rates significantly decreased following a 6-week
Secondary data analysis of the later study revealed that ad- (biweekly) yoga intervention, but not after aerobic cycling at
herence to yoga (77%) and exercise (69%) was not signifi- 70%–80% maximal heart rate. HRV, a measure of the heart’s
cantly different in a healthy elderly population (t ¼
1.95, resiliency or ability to respond to changes in demands, re-
p ¼ 0.056).49 mained unchanged in the cycling group. The yoga group
In research exclusively on healthy individuals, yoga has experienced significant increases in midfrequency (MF) HRV
been shown to be as effective as or superior to exercise on ( p < 0.01), but not high frequency (HF) HRV. HR and HRV
nearly every outcome measured (Table 2). Sinha et al.,35 in a were subsequently used to compute the a-index, a measure
study involving a convenience sample of 51 healthy males, thought to be indicative of sympathetic (at the MF level)
found yoga to be superior to running plus flexibility training versus parasympathetic (at the HF level) nervous system
in improving measures of antioxidant status. Serum reduced dominance, as well as a measure of baroreflex sensitivity. No
glutathione increased in the yoga group ( p < 0.05) and de- changes occurred in the a index at MF or HF following
creased in the exercise group. Similarly, glutathione reduc- aerobic training, but increased in the HF in the yoga group
tase, an indicator of oxidative stress, increased significantly ( p < 0.01), lending support to the notion that yoga acts on the
only in the exercise group ( p < 0.001), while total antioxidant SNS by increasing parasympathetic arousal.
status increased significantly in the yoga group ( p < 0.001) Teasing out the differences between yoga and exercise
and decreased significantly in the exercise group. While this with the current research has proven to be difficult. One
study seems to indicate yoga may be beneficial in reducing might expect that aerobic exercise would show greater im-
oxidative stress, additional research involving randomized provements in fitness outcomes. Indeed, measures of maxi-
clinical trials is warranted to provide stronger evidence. mum oxygen consumption (VO2max), an index of physical
In a study examining stress-related outcomes, West et al.5 fitness, were found to be significantly different in healthy
compared the effects of a single class of yoga to African seniors who participated in 6 weeks of Hatha yoga compared
dance and a college lecture. Perceptions of stress were to cycling at 70% of maximum heart rate ( p < 0.05).28 While
measured using the PSS, and affect was measured using the the aerobics subjects performed better than the yoga subjects,
Positive and Negative Affect Schedule. Both African dance, a VO2max increased in both groups: 11% in yoga ( p ¼ 0.01)
vigorous form of exercise, and yoga asana yielded significant and 24% following cycling ( p ¼ 0.01). In a retrospective study
improvements in perceived stress as measured on the PSS comparing long-term practitioners of yoga with aerobic ex-
( p < 0.001 and 0.0001, respectively) and in negative affect ercisers (running, walking, cycling) and sedentary subjects,
(both p < 0.05), with no significant changes noted in the lec- Duren et al.32 found no significant difference between the
ture group. The dance intervention led to significant im- yoga and aerobic groups in carotid artery distensibility (DC)
provements in positive affect ( p < 0.05), while the yoga or pulse wave velocity (PWV)—measures of arterial stiffness
group remained unchanged and the lecture group experi- that typically decrease (DC) or increase (PWV) with age, but
enced significant worsening ( p < 0.001). Both the yoga and improve with aerobic conditioning. The sedentary subjects
dance group perceived their stress levels to be reduced and had higher PWV and lower DC than compared to either the
their negative moods to be enhanced; however, only the yoga or aerobics group (both p < 0.001). While this study
HEALTH BENEFITS OF YOGA AND EXERCISE 9

might indicate that long-term yoga and exercise may have methodologies that compare the effects of the various types
similar cardiac benefits, it has several methodological of yoga on a variety of outcome measures in a variety of
weaknesses, including the use of a convenience sample and populations, both healthy and diseased.
not controlling for physical activity in the yoga group. It is possible that the differences in fitness outcomes found
While exercise has been shown to definitively improve in the comparison studies of yoga and exercise might not
parameters of fitness, the fitness effects of yoga have only have been found if exercise were compared to the more
been examined in a handful of studies.27,50 Significant vigorous forms of yoga. The differences that have been
increases in strength, muscle endurance, flexibility, and found between yoga and exercise interventions may be a
VO2max occurred in 10 healthy volunteers after 8 weeks of result of how the two differ in their effects upon the SNS and
biweekly asana and pranayama classes.50 Metabolic expendi- HPA axis. Different levels of intensity of exercise have been
ture in experienced yoga practitioners during a yoga session shown to affect the HPA axis response to acute stress dif-
was similar to that of walking at 3.2 km=hr on a treadmill— ferently. Low-intensity exercise repeatedly has been shown
significantly lower than the recommendations for moderate to lower cortisol levels,52,53 while intense exercise leads to
physical activity recommended at the time of the study by proportional increases in cortisol.54 The critical level of in-
the American College of Sports Medicine.27 Yoga practi- tensity that leads to release of cortisol is approximately 60%
tioners had a lower maximum predicted heart rate, burned VO2max, with the greater the exercise intensity, the greater
fewer calories per minute (kcals), and expended less energy the cortisol release.54 Perhaps this explains why yoga, in-
metabolic equivalents while practicing yoga than while volving slow and often nonstrenuous activities, positively
walking 4.8 km=hr ( p < 0.0001). The authors further con- affects the HPA axis response to stress. Exercise stimulates
cluded that only sun salutations, a more strenuous form of the SNS, raising plasma epinephrine and norepineph-
yoga practice involving continuous movement, were com- rine.55,56 Yoga on the other hand, has been shown to lower
parable to walking 4.8 km=hr on a treadmill and might sympathetic stimulation, significantly lowering levels of
provide enough intensity to improve cardiorespiratory fit- plasma norepinephrine and epinephrine.9
ness in sedentary individuals.27 Given that the eight limbs of yoga are so multidimensional
and include aspects of exercise (Asana), breath work (Pra-
nayama), concentration (Dharana), and meditation (Dyana), it
Discussion
is not surprising that researchers have found positive results
In the 12 studies that compared the effects of yoga and regarding yoga in so many diverse areas. In three studies
exercise, yoga interventions yielded positive results in both comparing yoga with meditation techniques such as pro-
healthy and diseased populations (Table 2). However, with gressive relaxation, yoga was found to be equal or superior
the exception of the studies by Oken et al.,28,38 no group of to progressive relaxation in lowering blood pressure57 and in
researchers has sought to compare the effects of yoga and improving perceptions of mood and anxiety.44,45 Yoga, when
exercise in a systematic fashion with variety of patient pop- compared with supportive psychotherapy in randomized
ulations. Nevertheless, the evidence presented in the table trials involving patients with cancer undergoing chemo-
suggests that yoga interventions appear to be equal or su- therapy, has been shown to be significantly better at de-
perior to exercise in nearly every outcome measured except creasing levels of nausea and vomiting58 and strengthening
those involving physical fitness. the immune system.13 While the previously discussed exer-
Nearly every study reviewed utilized a combination of cise comparison studies involving yoga’s effects on cognitive
different yoga therapies including vigorous physical asanas, function led to nonsignificant results,30,31 yoga clearly ap-
gentle restorative poses, breath work, and meditation. This pears to have multidimensional effects on brain chemistry
raises an important question that has not been adequately and this warrants further inquiry.
addressed in the literature. Just as there are different specialties Given the fact that clear evidence exists regarding the effi-
in the practice of medicine, there are several different styles of cacy of both exercise and yoga interventions in alleviating
yoga, each with distinctive challenges and varying levels of symptoms and improving outcomes of patients with coronary
difficulty. Some types of yoga may be gentle and meditative artery disease,59,60 it is somewhat surprising that researchers
(Integral, Svaroopa), vigorous (Ashtanga, Power Yoga), or both have not discriminated more clearly between the effects of the
(Iyengar, Kundalini). Some forms involve changes in the envi- two interventions in this population. Exercise has been rec-
ronment such as using heaters and humidifiers (Bikram). ognized as a key component in cardiac rehabilitation. Yoga,
Iyengar yoga frequently is used for therapeutics and incorpo- when added to the components of usual cardiac care in ran-
rates the use of props such as ropes, straps, and chairs to enable domized trials, has been shown to be significantly better than
students to achieve poses that might not be accessible other- usual cardiac care at improving blood lipid levels,61 decreas-
wise. Each style of yoga differs in the emphasis placed on the ing markers of inflammation14 and in reducing the number of
various components of yoga such as asana, pranayama, or revascular procedures.61 Mahajan et al.62 conducted the only
meditation. The relative effects of these different types of yoga clinical trial that exclusively examined the effects of a yogic
on the HPA axis and SNS in response to acute and chronic lifestyle (yoga, pranayama, meditation, and a vegetarian diet)
stress have not been adequately examined. in comparison to usual cardiac care in patients with one or
Only one study could be found comparing the various more cardiac risk factors and concluded that the yoga subjects
styles of yoga. In a convenience sample of 16 volunteers, only experienced significantly lower levels of triglycerides and low-
Ashtanga yoga resulted in significantly higher heart rates density lipoprotein cholesterol, in addition to lower body
than either Hatha or gentle yoga.51 This study examined only weight (all p < 0.05).
heart rate as an outcome variable. Clearly, additional studies It is possible that yoga might be not only an acceptable
are called for, using larger sample sizes and better research additive to care, but an effective, feasible, and acceptable
10 ROSS AND THOMAS

alternative to exercise in heart disease populations and in 5. West J, Otte C, Geher K, Johnson J, et al. Effects of Hatha
other populations that have traditionally benefited from ex- yoga and African dance on perceived stress, affect, and
ercise such as diabetes and obesity. This is a potentially rich salivary cortisol. Ann Behav Med 2004;28:114–118.
area for research for a variety of reasons. First, strong evi- 6. Michalsen A, Grossman P, Acil A, et al. Rapid stress re-
dence in the form of the Whitehall epidemiological studies duction and anxiolysis among distressed women as a con-
suggests that there is a dose–response relationship between sequence of a three month intensive yoga program. Med Sci
obesity and stress.63 Evidence also suggests that chronic Monit 2005;11:555–561.
stress leads to changes in food-seeking behavior, including 7. Khatri D, Mathur KC, Gahlot S, et al. Effects of yoga and
increased consumption of foods high in sugar and fat, which meditation on clinical and biochemical parameters of meta-
bolic syndrome. Diabetes Res Clin Pract 2007;78:e9–e10.
may eventually lead to obesity.64,65 As yoga seems to pro-
8. Gokal R, Shillito L. Positive impact of yoga and pranayam on
vide many of the benefits typically associated with exercise
obesity, hypertension, blood sugar, and cholesterol: A pilot
and also strongly influences the SNS=HPA axis response to
assessment. J Altern Complement Med 2007;13:1056–1057.
stress, it is possible that yoga might be a particularly useful 9. Selvamurthy W, Sridharan K, Ray US, et al. A new physi-
weapon in the arsenal against obesity. In a recent population- ological approach to control essential hypertension. Indian
based telephone survey involving 11,211 Americans, 57.4% of J Physiol Pharmacol 1998;42:205–213.
the 372 respondents (N ¼ 208) who admitted using comple- 10. Damodaran A, Malathi A, Patil N, et al. Therapeutic po-
mentary and alternative medicine during the past year re- tential of yoga practices in modifying cardiovascular risk
ported using yoga as a form of weight control.66 Research is profile in middle aged men and women. J Assoc Physicians
needed to examine the efficacy, feasibility, and acceptability of India 2002;50:633–639.
yoga interventions for the prevention and treatment of obesity 11. McCaffrey R, Ruknui P, Hatthakit U, Kasetsomboon P. The
in both healthy and ill populations. effects of yoga on hypertensive persons in Thailand. Holist
Nurs Pract 2005;19:173–180.
Conclusions 12. Stuck M, Meyer K, Rigotti T, et al. Evaluation of a yoga-
based stress management training for teachers: Effects
Overall, the studies comparing the effects of yoga and ex- on immunoglobulin A secretion and subjective relaxation.
ercise seem to indicate that, in both healthy and diseased J Medit Medit Res 2003;1–8.
populations, yoga may be as effective or better than exercise at 13. Rao RM, Telles S, Nagendra HR, et al. Effects of yoga on
improving a variety of health-related outcome measures in- natural killer cell counts in early breast cancer patients un-
cluding HRV,28 blood glucose,33,35 blood lipids,35,67 salivary dergoing conventional treatment. Comment to: recreational
cortisol,3 and oxidative stress.27,35 Furthermore, yoga appears music-making modulates natural killer cell activity, cyto-
to improve subjective measures of fatigue,30,31 pain, and sleep kines, and mood states in corporate employees Masatada
in healthy and ill populations.37 However, future clinical trials Wachi, Masahiro Koyama, Masanori Utsuyama, Barry B.
are needed to further examine the distinctions between exer- Bittman, Masanobu Kitagawa, Katsuiku Hirokawa. Med Sci
cise and yoga, particularly how the two modalities may differ Monit 2007;13:CR57–CR70. Med Sci Mon 2008;14:3–4.
in their effects on the SNS=HPA axis. Additional studies are 14. Pullen PR, Nagamia SH, Mehta PJ, et al. Effects of yoga on
needed to distinguish between the different types of yoga and inflammation and exercise capacity in patients with chronic
their various techniques. All of these studies need to use rig- heart failure. J Card Fail 2008;14:407–413.
orous study methodologies, including the use of larger sam- 15. Schultz PE, Haberman M, Karatha K, et al. Iyengar Yoga
ple sizes, randomized samples, and blinding of researchers. Can Promote Well-Being in Women Breast Cancer Survi-
vors. Spokane, WA: Washington State University, 2007.
These studies need to be replicated in a variety of populations,
16. Innes KE, Bourguignon C, Taylor AG. Risk indices associ-
both sick and well, as the effects may vary depending upon
ated with the insulin resistance syndrome, cardiovascular
the health status of the population.
disease, and possible protection with yoga: A systematic
review. J Am Board Fam Pract 2005;18:491–519.
Disclosure Statement 17. Shapiro D, Cook IA, Davydov DM, et al. Yoga as a com-
No competing financial interests exist. plementary treatment of depression: Effects of traits and
moods on treatment outcomes. Evid Based Complement
Alternat Med 2007;4:493–502.
References
18. Gupta N, Shveta K, Vempati R, et al. Effect of yoga based
1. Iyengar BKS. Light on Yoga. 2nd ed. New York: Schocken lifestyle intervention on state and trail anxiety. Indian
Books, 1976. J Physiol Pharmacol 2006;50:41–47.
2. Saper R, Eisenberg D, Davis R, Culpepper L, Phillips R. 19. Telles S, Naveen K, Dash M, et al. Effect of yoga on self-rated
Prevalence and patterns of adult yoga use in the United visual discomfort in computer users. Head Face Med 2006;
States: Results of a national survey. Altern Ther Health Med 2:46.
2004;10:44–48. 20. Moadel AB, Shaw C, Wylie-Rossett J, et al. Randomized
3. Sterling P. Principles of allostasis: Optimal design, predictive controlled trial of yoga among a multiethnic sample of
regulation, pathophysiology, and rational therapeutics. In: breast cancer patients: Effects on quality of life. J Clin Oncol
Schulkin J, ed. Allostasis, Homeostasis, and the Costs of 2007;25:1–9.
Physiological Adaptation. Cambridge: Cambridge Uni- 21. Raub JA. Psychophysiologic effects of hatha yoga on mus-
versity Press, 2004:17–64. culoskeletal and cardiopulmonary function: a literature re-
4. McEwen BS. Allostasis and allostatic load: Implications for view. J Altern Complement Med 2002;8:797–812.
neuropsychopharmacology. Neuropsychopharmacology 2000; 22. Upadhyay AK, Balkrishna A, Upadhyay RT. Effect of pra-
22:108–124. nayama (voluntary regulated yoga breathing) and yogasana
HEALTH BENEFITS OF YOGA AND EXERCISE 11

(yoga postures) in diabetes mellitus (DM): A scientific re- 2 diabetic patients: A meta-analysis. Diabetes Care 2006;
view. J Complement Integr Med 2008;5:article 3. Online 29:2518–2527.
document at: www.bepress.com=jicm=vol5=iss1=3 Accessed 41. Bijlani RL, Rama MD, Vempati RP, et al. A brief but com-
January 7, 2010. prehensive lifestyle education program based on yoga re-
23. Bower J, Woolery A, Sternlieb B, Garet D. Yoga for cancer duces risk factors for cardiovascular disease and diabetes
patients and survivors. Cancer Control 2005;12:165–171. mellitus. J Altern Complement Med 2005;11:267–274.
24. Kirkwood G, Rampes H, Tuffrey V, et al. Yoga for anxiety: A 42. Krishnamurthy MN, Telles S. Assessing depression follow-
systematic review of the research evidence. Br J Sports Med ing two ancient Indian interventions: Effects of yoga and
2005;39:884–891. Ayurveda on older adults in a residential home. J Gerontol
25. Galantino ML, Bzdewka TM, Eissler-Russo JL, et al. The Nurs 2007;33:17–23.
impact of modified hatha yoga on chronic low back pain: A 43. Woolery A, Myers H, Sternlieb B, Zeltzer L. A yoga inter-
pilot study. Altern Ther Health Med 2004;10:56–59. vention for young adults with elevated symptoms of de-
26. Greenhalgh T. How to read a paper: Assessing the meth- pression. Altern Ther Health Med 2004;10:60–63.
odological quality of published papers. Br Med J 1997;315: 44. Smith C, Hancock H, Blake-Mortimer J, Eckert K. A ran-
305–308. domized comparative trial of yoga and relaxation to reduce
27. Hagins M, Moore W, Rundle A. Does practicing hatha yoga stress and anxiety. Complement Ther Med 2007;15:77–83.
satisfy recommendations for intensity of physical activity 45. Shannahoff-Khalsa DS, Ray LE, Levine S, et al. Randomized
which improves and maintains health and cardiovascular controlled trial of yogic meditation techniques for patients
fitness? BMC Complement Altern Med 2007;7:1–9. with obsessive-compulsive disorder. CNS Spectrums 1999;
28. Bowman AJ, Clayton RH, Murray A, et al. Effects if aerobic 4:34–47.
exercise training and yoga on the baroreflex in healthy el- 46. Narendran S, Nagarathna R, Narendran V, et al. Efficacy of
derly persons. Eur J Clin Invest 1997;27:443–449. yoga on pregnancy outcome. J Altern Complement Med
29. Duraiswamy G, Thirthalli J, Nagendra HR, Gangadhar BN. 2005;11:237–244.
Yoga therapy as an add-on treatment in the management of 47. Chuntharapat S, Petpicketchian W, Hatthakit U. Yoga during
patients with schizophrenia: A randomized controlled trial. pregnancy: Effects on maternal comfort, labor pain, and birth
Acta Psychiatr Scand 2007;10:226–232. outcomes. Complement Ther Clin Pract 2008;14:105–115.
30. Oken BS, Kishiyama S, Zajdel D, et al. Randomized con- 48. Booth-LaForce C, Thurston RC, Taylor MR. A pilot study of
trolled trial of yoga and exercise in multiple sclerosis. Neu- Hatha yoga treatment for menopausal symptoms. Maturitas
rology 2004;62:2058–2064. 2007;57:286–295.
31. Oken BS, Zajdel D, Kishiyama S, et al. Randomized, con- 49. Flegal K, Kishiyama S, Zajdel D, et al. Adherence to yoga
trolled, six-month trial of yoga in healthy seniors: Effects on and exercise interventions in a 6-month clinical trial. BMC
cognition and quality of life. Altern Ther Health Med 2006; Complement Altern Med 2007;7:37.
12:40–47. 50. Tran MD, Holly RG, Lashbrook J, Amsterdam EA. Effects of
32. Duren CM, Cress ME, McCully KK. The influence of phys- Hatha yoga practice on health-related aspects of physical
ical activity and yoga on central arterial stiffness. Dyn Med fitness. Prev Cardiol 2001;4:165–170.
2008;7:1–8. 51. Cowen VS, Adams TB. Heart rate in yoga asana practice:
33. Gordon LA, Morrison EY, McGrowder DA, et al. Effect of A comparison of styles. J Bodywork Move Ther 2007;11:
exercise therapy on lipid profile and oxidative stress indi- 91–95.
cators in patients with type 2 diabetes. BMC Complement 52. Few JD. Effect of exercise on the secretion and metabolism of
Altern Med 2008;8:article21. cortisol in man. J Endocrinol 1974;62:341–353.
34. Khattab K, Khattab AA, Ortak J, et al. Iyengar yoga increases 53. Davies CT, Few JD. Effects of exercise on adrenocortical
cardiac parasympathetic nervous modulation among health function. J Appl Physiol 1973;35:887–889.
yoga practitioners. Evid Based Complement Altern Med 54. Howlett TA. Hormonal responses to exercise and training: A
2007;4:511–517. short review. Clin Endocrinol 1987;26:723–742.
35. Sinha S, Singh SN, Monga YP, Ray US. Improvement of 55. Peronnet F, Cleroux J, Perrault H, et al. Plasma norepi-
glutathione and total antioxidant status with yoga. J Alternat nephrine response to exercise before and after training in
Complement Med 2007;13:1085–1090. humans. J Appl Physiol 1981;51:812–815.
36. Chattha R, Nagarathna R, Venkatram P, Hongasandra N. 56. Bloom SR, Johnson RH, Park DM, et al. Differences in met-
Treating the climacteric symptoms in Indian women with an abolic and hormonal response to exercise between racing
integrated approach to yoga therapy: A randomized control cyclists and untrained individuals. J Physiol 1976;258:1–18.
study. Menopause 2008;15:862–870. 57. Cusumano JA, Robinson SE. The short-term psychophysio-
37. Yurkuran M, Alp A, Yurtkuran M, Dilek K. A modified logical effects of Hatha yoga and progressive relaxation on
yoga-based exercise program in hemodialysis patients: A female Japanese students. Appl Psychol 1993;42:77–89.
randomized controlled study. Complement Ther Med 2007; 58. Raghavendra RM, Nagarathna R, Nagendra HR, et al. Ef-
15:164–171. fects of an integrated yoga programme on chemotherapy-
38. Heath GW, Gavin JR, Hinderliter JM, et al. Effects of exercise induced nausea and emesis in breast cancer patients. Eur
and lack of exercise on glucose tolerance and insulin sensi- J Cancer Care 2007;16:462–474.
tivity. J Appl Physiol 1983;57:1857–1864. 59. Ornish D, Scherwitz LW, Billings JH, et al. Intensive lifestyle
39. Boule NG, Haddad E, Kenny GP, et al. Effects of exercise on changes for reversal of coronary heart disease. JAMA 1998;
glycemic control and body mass in type 2 diabetes mellitus: 280:2001–2007.
A meta-analysis of controlled clinical trial. JAMA 2001;286: 60. Daubenmier JJ, Weidner G, Sumner MD, et al. The contribu-
1218–1227. tion of changes in diet, exercise, and stress management to
40. Snowling NJ. Effects of different modes of exercise training changes in coronary risk in women and men in the multisite
on glucose control and risk factors for complications in type lifestyle intervention program. Ann Behav Med 2007;33:57–68.
 12 ROSS AND THOMAS

61. Manchanda SC, Narang R, Reddy KS, et al. Retardation of 66. Sharpe PA, Blanck HM, Williams JE, et al. Use of comple-
coronary atherosclerosis with yoga lifestyle intervention. mentary and alternative medicine for weight control in the
J Assoc Physicians India 2000;48:687–694. United States. J Altern Complement Med 2007;13:217–222.
62. Mahajan AS, Reddy KS, Sachdeva U. Lipid profile of coro- 67. Yurtkuran M, Alp A, Yurtkuran M, Dilek K. A modified
nary risk subjects following yogic lifestyle intervention. In- yoga-based exercise program in hemodialysis patients: A
dian Heart J 1999;51:37–40. randomized, controlled study. Complement Ther Med
63. Brunner EJ, Chandola T, Marmot MG. Prospective effect of 2007;15:164–171.
job strain on general and central obesity in the Whitehall II
Study. Am J Epidemiol 2007;165:828–837. Address correspondence to:
64. Schiffman SS, Graham BG, Sattely-Miller EA, Peterson- Alyson Ross, M.S.N., R.N.
Dancy M. Elevated and sustained desire for sweet taste in School of Nursing
African-Americans: A potential factor in the development of University of Maryland
obesity. Nutrition 2000;16:886–893. Baltimore, MD 21201
65. Oliver G, Wardle J, Gibson L. Stress and food choice: A
laboratory study. Psychosom Med 2000;62:853–865. E-mail: alyross1@verizon.net

View publication stats