Stress and anxiety have been implicated as contributors to many chronic diseases and to decreased quality of life, even with pharmacologic treatment. Efforts are underway to find non-pharmacologic therapies to relieve stress and anxiety, and yoga is one option for which results are promising. The focus of this review is on the results of human trials assessing the role of yoga in improving the signs and symptoms of stress and anxiety. Of 35 trials addressing the effects of yoga on anxiety and stress, 25 noted a significant decrease in stress and/or anxiety symptoms when a yoga regimen was implemented; however, many of the studies were also hindered by limitations, such as small study populations, lack of randomization, and lack of a control group. Fourteen of the 35 studies reported biochemical and physiological markers of stress and anxiety, but yielded inconsistent support of yoga for relief of stress and anxiety. Evaluation of the current primary literature is suggestive of benefits of yoga in relieving stress and anxiety, but further investigation into this relationship using large, well-defined populations, adequate controls, randomization and long duration should be explored before recommending yoga as a treatment option. (A/tem Med Rev 2012;17:21-35.)
yoga, stress, psychological, anxiety
Although yoga has been practiced for over 5,000 years, it has only recently gained popularity in the United States and Europe. In America, the yoga market emerged as a 5.7 billion dollar industry in 2008, an increase of 87% from 2004. (1) The practice originated in India and has been implemented to alleviate both mental and physical ailments, including bronchitis, (2) chronic pain, (3) and symptoms of menopause. (4) Nonetheless, health care providers in the United States have not endorsed yoga as an alternative therapy. This was typified in a recent news item, published by the American Medical Association (AMA), in which yoga practice is included in a list of alternative therapies referred to as "unproven treatments". In it, a call is made for "rigorous research to study safety and efficacy" of these therapies. (5)
Research into the role of yoga in certain disease states and in improving overall health is ongoing. The specific focus of this review is the use of yogic principles and exercises for reducing anxiety and stress. A review of human trials on the relationship between yoga and reduction in stress and anxiety was performed, and the validity of these findings is presented in an effort to determine whether the existing primary literature supports the incorporation of yoga into the cognitive behavior therapy plan for stress and anxiety, or whether additional research in the field is warranted. Moreover, compliance with yoga practice is evaluated. If deemed beneficial, yoga could be a potential alternative or adjunctive option to pharmacologic therapy for patients with stress and anxiety disorders.
Yoga is a general term that encompasses breathing techniques, postures, strengthening exercises, and meditation. (6) Many types of yoga exist. One of the more popular forms in the United States is Hatha yoga, which incorporates postures, breathing techniques and meditation to benefit physical and mental well-being. Hatha yoga is further categorized into the Iyengar, Kundalini, Bikram, Ananda, Vivnoya, and Anusara styles. It is estimated that over 15.2 million Americans employed some form of yoga for health purposes in 2002. (1) With yoga's increasing popularity and emphasis on a spiritual connection between the mind and body, it is reasonable to explore its role in the treatment of mental disorders, such as anxiety and stress.
Anxiety and Stress
Anxiety and stress have unfavorable effects on the body that may progress into chronic conditions if left untreated. For example, psychological stress has been linked to deleterious effects on the immune system, (7) while anxiety has been connected to coronary heart disease, (8) decreased quality of life, (9) and suicidal behavior. (10)
Anxiety disorders are subdivided into panic disorder, obsessive-compulsive disorder, posttraumatic stress disorder, social anxiety disorder, phobias, and generalized anxiety disorder. (11) Although the clinical presentation of anxiety disorders ranges from chest pain and tachycardia to insomnia, all of these disorders center on an intense, unwarranted fear for a defined period of time (11) and affect approximately 18% of Americans in any given year. (12) Pharmacotherapy used to treat anxiety disorders, as outlined by the U.S. National Institute of Mental Health (NIMH), consists of benzodiazepines, beta-blockers, monoamine oxidase inhibitors, and antidepressants. (12)
In contrast, psychological stress is perceived and is less defined than anxiety disorders. (13) As a result, no recognized pharmacologic treatment algorithm exists for stress. Additionally, it should be recognized that some people experience stress and anxiety without having been diagnosed with a specific disorder by health care professionals; such individuals may also benefit from yoga exercises.
Assessment of Anxiety and Stress Disorders
The extent of stress and anxiety is subjective, but can be quantified using the scales summarized in Table 1. Of these, the State-Trait Anxiety Inventory (STAI), Perceived Stress Scale (PSS), the Hospital Anxiety and Depression Scale (HADS), and the Hamilton Anxiety (HAM-A) scales are the validated instruments used most frequently in the studies described herein. While most studies' results were based on validated questionnaires, one study used an unvalidated investigator-constructed questionnaire. Because stress and anxiety can manifest as physical symptoms, vital signs, such as blood pressure and heart rate, are sometimes utilized as indirect measures of stress and anxiety.
In addition to these surveys and vital signs, some other physiological indicators of stress and anxiety were used in the described studies and are listed in Table 3. For example, the hormones cortisol and dehydroepiandrosterone (DHEA) tend to be elevated in acute situations of psychobiological stress. (14) While hypersecretion of cortisol occurs when under stress, its secretion is also nonspecific; it is released from the adrenal glands not only when under physiological stress, but also with changes in circadian rhythm and imbalances in the hypothalamic-pituitary-adrenal (HPA) axis. Additional physiological markers have been used in some of the reviewed studies, including the neurohormone, melatonin, and the neurotransmitter, [gamma]-aminobutyric acid (GABA). In contrast with cortisol and DHEA, melatonin and GABA are implicated in relaxation, with increased levels associated with relaxation and decreased levels implying stress and anxiety. (15)
Literature searches were conducted using PubMed (from September, 1974, through Sept. 9, 2010). Two searches were conducted using the keyword 'yoga'; one was conducted in which 'yoga' was combined with the keyword 'stress', and the other was conducted in which the keyword 'yoga' was combined with the keyword 'anxiety'. Studies were limited to human subjects, English language, clinical trials, and randomized controlled trials, yielding 56 articles for yoga and stress and 42 articles for yoga and anxiety; many of these articles were common to both searches. Studies that focused on oxidative stress or disease states other than stress and anxiety were not considered for this review. Additional references were acquired by cross-referencing pertinent articles' bibliographies. This resulted in a total of 35 studies that met all of the above criteria and were utilized for this review.
The data regarding the usefulness of yoga and meditation in the short-term reduction of anxiety are inconsistent. While most studies demonstrated a significant decline from baseline in subjects' anxiety scale scores after intervention, these studies had many inherent weaknesses. Several small, preliminary studies showed significant reductions in stress and anxiety, but have at least one limitation in addition to their size, such as a lack of control group, non-randomization, lack of exclusion criteria, or large standard deviations associated with the data. (16-31) Despite these limitations, these studies merit discussion.
Banerjee et al. examined the effects of a yoga regimen (n = 35) on anxiety and on physiological and psychological stress in post-operative breast cancer patients. (16) A counseling program (n = 33) served as the control. After a 6-week intervention, the experimenters observed a significant decrease in HADS anxiety score in the yoga group, while the control group showed no improvement in anxiety. Although the authors controlled for other co-morbid illnesses and excluded the use of anti-anxiety medications, they did not account for other methods of relaxation in either the control or treatment group during the six weeks. Nevertheless, the results suggest a benefit of yoga in reducing anxiety in cancer patients.
Using a similar patient population and small numbers of patients, Rao, (17) Ulger, (18) Vadiraja (19,20) and their respective colleagues examined the effects of yoga on post-operative breast cancer patients undergoing treatment. Rao and co-workers randomized patients to either yoga or supportive therapy (control) intervention, but had a large drop-out rate; 45 patients enrolled in the yoga program, but only 18 completed it, and 53 patients enrolled in the supportive therapy group, but only 20 completed it. In addition, all patients received anxiolytics as co-medication while undergoing 6 weeks of radiation and 6 rounds of chemotherapy. While limited, as a result of these design issues, patients participating in the yoga program showed significantly decreased STAI anxiety scores compared to the supportive therapy, control patients. (17)
Ulger and Yagli studied the effects of a 4-week yoga program on breast cancer patients (n = 20) in a non-randomized, non-controlled trial and found improved quality of life and significantly decreased STAI anxiety scores after the yoga intervention when compared to baseline scores. (18) In a randomized, controlled study of breast cancer patients undergoing treatment, Vadiraja and colleagues found that 42 subjects who participated in a B-week yoga intervention had significantly increased positive affect and decreased negative affect on the Positive and Negative Affect Schedule when compared to 33 counseling control subjects. (19) These researchers also found that compared to controls, breast cancer patients who participated in a yoga intervention had decreased HADS anxiety and depression scores after intervention compared to counseling control subjects. (20)
Further affirmation that yoga can be effective in reducing anxiety was produced by others who studied different female populations than previously described. Waelde and Thompson (21) noted decreases in STAI scores after a B-week yoga intervention in 8 female caregivers of dementia patients. Michalsen and colleagues (22) studied women who were self-described as emotionally distressed, and found that the 16 who participated in 12 weeks of yoga had decreased STAI scores compared to the 8 controls who did not participate in yoga. As with the Banerjee, Ulger, Rao, and Vadiraja studies, (16-20) these results cannot be extrapolated to men; moreover, the suggested improvement in anxiety reported by Waelde and Thompson is undermined by the small subject population, large standard deviations associated with the STAI score means, lack of a control group, and lack of exclusion criteria. (21)
As with Waelde and Thompson, Javnbakht et al. (23) studied the effects of yoga on depression and anxiety in healthy women. Thirty-one women acted as controls, while 34 women received twice weekly 90-minute yoga sessions for 2 months. Using the Beck Depression Inventory (BDI) to assess the prevalence of depression within the experimental group before and after yoga intervention, no significant differences were found; however, when the yoga group was compared to the control group, the yoga group had significantly reduced STAI state and trait anxiety scores. (23) As with previous studies, this study had limitations, including short duration, a small number of subjects, and a study population limited to one gender.
A small study from Ando and colleagues avoided gender limitations, while examining the effects of mindfulness-based meditation (yoga, meditation and breathing) on cancer patients in treatment. (24) As noted, this small patient population (n = 28) was heterogeneous, in terms of both gender and primary tumor site: 24 were female, 4 were male, 21 had breast tumors, 2 had colon tumors, 2 had stomach tumors, and 2 had bladder tumors. While the researchers found statistically significant decreases in HADS anxiety and depression scores after 2 weeks of mindful meditation, this work suffers from a lack of a control group, lack of randomization, and a small number of participants with heterogeneous characteristics. (24)
Research on the effects of yoga intervention has been conducted in other small, gender-specific populations, such as postmenopausal women with rheumatoid arthritis, (25) pregnant women, (26) and male flood survivors. (27) Bosch and colleagues studied the effects of a 10-week yoga intervention in 16 women with rheumatoid arthritis (9 yoga subjects, 7 controls) and found significant decreases after invention in perceived pain and disability on the Health Assessment Questionnaire (HAQ), and in depression scores on the BDI in the yoga, but not the control, group. (25) No differences in cortisol levels were observed between the groups. (25) This work is limited by a lack of randomization of the subjects, lack of objective measures of disease state, use of a small study population of only postmenopausal female subjects, and a short period of intervention.
Many of these same limitations are inherent in Beddoe and colleagues' (26) work with pregnant women (n = 17) and in Telles et al.'s (27) work with male flood victims (n = 22). While Telles et al. (27) performed a 1-week randomized controlled study with the men who had survived a flood, Beddoe et al. (26) conducted a 7-week, non-randomized, noncontrolled study of pregnant women. Both groups found decreased anxiety scores in the subjects who received the yoga intervention; however, these studies were limited by small numbers of subjects of one gender, short duration of intervention, and, for the study of pregnant women, a lack of a control group. (26)
Other researchers examined yoga interventions in very specific populations, such as those that may have anxiety as a result of their jobs or training. For example, in 2 studies, Khalsa and coworkers examined a group of classical musicians. (28,29) In their first 8-week study, the researchers compared pre- and post-treatment Performance Anxiety Questionnaire (PAQ) scores during practice, a solo performance, and a group exhibition. Significant improvements from baseline in PAQ scores were seen in the yoga/meditation group (n = 10), whereas no differences from baseline were observed in the control group (n = 8) in the 3 categories. These promising results are tempered by limitations, including large standard deviations associated with the mean PAQ scores and lack of randomization. Moreover, to be enrolled in the study, each participant was required to submit an application to the yoga facility. Because the subjects were selected based on their application, bias is possible; it is unknown whether these subjects were picked because they would more likely yield favorable results.
Some of these concerns were addressed in a follow-up study by the same researchers, (29) though large standard errors remain. In this 2-month study, young adult male and female professional musicians (mean age ~25 years) were randomized to one of 3 groups: a yoga and meditation group (n = 15); a yoga lifestyle group (n = 15), which involved daily yoga practice and practical and spiritual teaching; and a non-intervention control group (n = 15). Multiple outcomes were assessed before and after the 2-month program, including performance anxiety, using the PAQ; mood, using the Profile of Mood States (POMS); perceived exertion during practice, using the Performance-Related Musculoskeletal Disorders (PRMD) questionnaire; stress, using the Perceived Stress Scale (PSS); and sleep quality and disturbance, using the Pittsburgh Sleep Quality Index (PSQI). (29) At the end of the study, there was significantly less tension/anxiety in the yoga lifestyle group compared to controls, and less anger/hostility in the yoga-only group compared to controls. Nearly significant reductions in depression/dejection (p = 0.07) and anger/hostility (p = 0.05) were noted for the yoga lifestyle group and in tension/anxiety (p = 0.07) for the yoga-only group when compared to controls. When scores from the yoga intervention groups were merged and compared to the control group, statistically significant reductions in tension/anxiety, and anger/hostility were noted. No significant differences in solo performance anxiety were noted among the groups, nor were there differences in PRMD, stress, or sleep. (29)
Yoga intervention has also been studied in another specific group, medical students, a cohort that often experiences high levels of stress and anxiety during training. In a non-randomized but controlled study of 50 medical students (25 yoga participants, 25 controls) over 3 months, Malathi and co-workers observed significantly reduced anxiety scores immediately following yoga practice and on the day of exams after yoga practice. (30) Similarly, in a non-randomized, non-controlled study, Simard and Henry found improved perceived stress and general health scores in 14 medical students after 8 and 16 weeks of yoga intervention. (31) As has been noted in many of the previous small studies, limitations in these studies include small numbers of participants, lack of controls and/or randomization, short study duration, and applicability to the general population.
Others researchers used larger sample sizes, but were still limited by their study designs. For example, although Gupta et al. reported significant improvements in STAI anxiety scores and a larger sample size was employed (n = 175 yoga subjects, 50 controls), the correlation between yoga and anxiety reduction is questionable due to a lack of exclusion criteria, use of a heterogeneous population of patients with an assortment of chronic ailments, large standard deviations, and short duration of the intervention (10 days). (32) Similarly, Khemka and colleagues studied a relatively large number of subjects (n = 86; 56 healthy male and 30 healthy female volunteers who had completed a one-month residential yoga course), who were randomized into either yoga-based deep relaxation or supine rest groups and found improved STAI anxiety scores in the yoga, but not the 'rest', group. (33) This work was limited, though, in that there was only one 20-minute session prior to assessment, no follow-up, and a heterogeneous mix of subjects.
Other studies have been conducted to examine the effects of meditation alone, rather than a full yoga program (one that includes breathing, postures and meditation), on lessening anxiety. While a full review of these studies is beyond the scope of this work, a few of these studies merit discussion, since yoga so often encompasses meditation. One of the earliest studies examining the relationship between meditation and anxiety was conducted by Kabat-Zinn and colleagues. (34) This pilot study showed that, among 20 participants, anxiety and frequency of panic attacks decreased significantly at 8 weeks and at a 3-month follow-up compared to their corresponding pretreatment scores. In a follow-up to the previous study, 22 subjects with generalized anxiety disorder and panic disorder participated in an 8-week meditation program and were followed for 3 years. Significant decreases in anxiety and reduced numbers of panic attacks were noted after intervention and maintained at the 3-year follow-up. (35) As was seen with the previously described research, these studies had some disadvantages, including small subject populations, no randomized control group, and no exclusion of patients taking anti-anxiety and other medications. (34,35)
Lee et al. also found favorable effects of a meditation-based stress management program in 46 patients with a known anxiety disorder. (36) After 8 weeks, the researchers found significant decreases from baseline in anxiety scores in both a meditation (n = 24) and an anxiety education group (n = 22), as well as a significant difference in the rate of change between the 2 groups. (36) It is difficult to determine the contribution of meditation to the declines in anxiety scores because anxiolytic medications were administered prior to the commencement of the study in order to standardize anxiety scores at baseline. As a result, the positive results may have been due to meditation or to residual effects of the medication or to a combination of the two.
Significant decreases in anxiety scores were also observed by Kjellgren and colleagues, (37) who measured changes in HADS score pre- and post-treatment with yoga. This 6-week pilot study included 48 participants in the yoga group and 55 in the relaxation group. There were some limitations in the study design, including a small study population and lack of both blinding and randomization.
In contrast to the aforementioned studies, which, despite their limitations, indicated that yoga and/ or meditation reduced anxiety, multiple other studies have failed to show the same positive results. For example, a study by Girodo and colleagues did not yield favorable effects of meditation in patients with a diagnosed anxiety disorder (n = 9) over a 6- to 8-month period. (37) In this study, subjects meditated twice daily while envisioning anxious situations and answered the Anxiety Symptom Questionnaire (ASQ, an unvalidated questionnaire) to assess changes in anxiety. Statistical shortcomings, including an extremely small sample size, no control group, the subjective nature of the intervention, and the use of an unvalidated questionnaire call into question the negative results of this study. (37)
Another negative outcome was found by Smith and co-workers, (38) who calculated the sample size necessary (n = 86 each, for the control and yoga intervention groups) to achieve an 80% power based on the State-Trait Personality Inventory
Sub-scale (STPI) for anxiety, but did not achieve this pre-established threshold (68 yoga, 63 control). Apart from the sample size, the authors' methodology was well-devised, with appropriate randomization and blinding of the researchers. No difference in anxiety scores was detected between the yoga and muscle relaxation control groups after a 10-week intervention. (38)
Similarly, Harinath et al. observed no differences in Institute for Personality and Ability Testing (IPAT) anxiety scores in 15 males following a 3-month yoga/meditation intervention. (40) These results may have been skewed by the small sample size, non-randomization, and the homogenous nature of the participants, who were healthy young men in the army.
As with anxiety, studies of the effects of yoga intervention on stress have yielded inconsistent results, some of which have resulted from limitations of the studies themselves, most notably small sample sizes with inadequate power, non-randomization of participants, and lack of blinding. In these studies, perceived stress was measured using scores from questionnaires, most commonly the Perceived Stress Scale (PSS), though some researchers used their own questionnaires, which may not have been validated, a further limitation. Significant improvements in stress scores were observed in several studies, (16,22,41,42) though no such improvements were noted by other researchers. (38,39)
PSS scores were the primary endpoints in assessing the role of yoga in stress reduction in post-op breast cancer patients, (16) distressed female patients, (22) and pregnant women. (42) Banerjee et al. observed a significant decrease in stress scores before and after radiation therapy in the yoga group (n = 35), while an increase in PSS scores was detected in the control group (n = 28). (16) The results of the Michalsen study were equally positive. (22) The yoga group (n = 16) experienced a significant reduction in stress, whereas the control group (n = 8) reported no change after 12 weeks. A main drawback of this study is a lack of randomization, which poses a possible bias.
In their 6-week pilot study, Kjellgren et al. reported significant decreases in stress scores from baseline for a yoga group, but not a relaxation group. (37) Weaknesses of this study were noted previously in regard to anxiety, and ought to be considered when evaluating these results.
In a randomized, controlled study, pregnant women had significant reductions in PSS scores and significant improvement in heart rate variability (an indication of improved adaptive autonomic responses) after a 16-week yoga intervention (n = 45), in contrast with significant increases in PSS scores and no improvement in heart rate variability in a deep relaxation control group (n = 45). (42)
Carlson, et al. described similar findings with the use of Symptoms of Stress Inventory (SOSI) scoring in prostate cancer patients, the only study in this review that looked exclusively at the male population. (41) The mean SOSI scores revealed significant stress relief with meditation (n = 59). Despite the positive correlation between meditation and stress relief, this report could have been strengthened by the incorporation of a control group and randomization.
Cowen and colleagues studied the effects of yoga intervention among firefighters, a population that typically experiences stress as a result of their work. Seventy-seven firefighters had significantly decreased PSS scores after 4 yoga classes over 6 weeks in this non-randomized, non-controlled study. (43) An additional benefit of the yoga intervention was improved functional movement in these subjects after intervention.
Interestingly, although significant decreases in PSS scores in those practicing yoga have been observed, differences between stress scores in studies in which a yoga intervention was compared to other forms of relaxation, such as African dance, (44) physical exercise, (45) and cognitive behavioral therapy (46) have been mixed. Subjects who participated in African dance (n = 21) (44) or cognitive behavioral therapy (n = 16), (46) had reductions in PSS scores equivalent to those involved in yoga intervention (n = 18 (44) and n = 15 (46)), and, in the case of West et al.'s work, significantly greater reductions than exhibited by controls (n = 30, in a biology lecture control group). In contrast, Chatta and colleagues found a significantly greater reduction in PSS scores over 8 weeks in perimenopausal women who participated in a yoga intervention (n = 60) than in women who participated in an exercise intervention (n = 60). (45) These results demonstrate that the use of yoga is at least equivalent to, and may be better than, other techniques for the reduction of symptoms of stress.
As with anxiety, not all studies of stress showed positive results. For example, Smith et al. reported no change in stress scores from baseline after 10 weeks of yoga or muscle relaxation. (38) In this case, psychological stress was quantified using the General Health Questionnaire-12 (GHQ-12) survey, a questionnaire designed by the researchers to assess the general mental health of a patient, rather than focusing on stress, as the PSS or SOSI scales do.
In a randomized, controlled pilot study, Cohen and colleagues examined the effects of a 10-week yoga intervention on blood pressure and PSS scores in adults with metabolic syndromes Though there were no significant differences in blood pressure or PSS scores in the yoga (n = 12) compared to the control (n = 12) group in this study, there were trends toward reductions in both parameters in the yoga compared to the control group. (47)
One final, unique study considered coping mechanisms in stressful situations by comparing the results of the Stress Coping Questionnaire. (48) The researchers followed 25 female participants (12 yoga, 13 control) during a 3-hour reading period and found that most physiological parameters (e.g., heart rate, blood pressure, cortisol levels) were not statistically different between the yoga practitioners and the controls. Differences in psychological parameters (e.g., aggressiveness, excitability, somatic complaints) were found, with the yoga practitioners scoring significantly lower on these measures than the controls. In addition, the control group tended to respond to stress with anger while the yoga group did not. These results suggest the possibility of long-term psychological benefits and short-term changes in coping response to stress resulting from yoga practice.
Biochemical markers of stress
Cortisol levels were assessed in multiple independent trials; (19,22,25,26,41,44,46,48-50) however, the results were inconsistent, with the majority of studies showing no effect of yoga practice on cortisol concentrations. This is due, at least in part, to limitations in the studies' statistics and lack of untreated controls. In addition, because cortisol levels fluctuate throughout the day, researchers must obtain and compare samples taken at the same time(s) every day, in addition to measuring levels over time to correlate an intervention with any changes in cortisol concentrations. For example, Carlson et al. accounted for diurnal changes in cortisol levels by taking the measurements in 59 participants at the same time daily, but found no change in salivary cortisol levels from baseline after 1 week of meditation intervention. (41)
Granath and colleagues also found no change from baseline in salivary cortisol levels in either a Kundalini yoga or a relaxation group after a 4-month intervention period. (46) The subjects in this study were men and women in finance, with self-reported stress. Because the study size was so small (15 yoga, 16 relaxation), the study had large standard deviations around the means, making statistical significance difficult to achieve.
Similarly, Schell's, (48) Robert-McComb's, (49) Beddoe's, (26) and Bosch's (25) groups demonstrated no differences in salivary cortisol levels after 3-hour, 8-week, 7-week, and 10-week interventions, respectively. As with many previous studies, design limitations make it difficult to establish the validity of these results. All groups used small sample sizes (n = 25 for Schell, (48) n = 18 for Robert-McComb, (49) n = 17 for Beddoe, (26) and n = 16 for Bosch (25)) and the studies were of relatively short duration.
Conversely, Vadiraja and colleagues reported significant decreases in 6 a.m. and pooled diurnal salivary cortisol concentrations in 42 breast cancer patients after a 6-week yoga intervention compared to 33 breast cancer patients in the control group. (20)
Similarly, West, et al. reported a significant decrease in salivary cortisol in 18 undergraduate students after a semester-long Hatha yoga course. (44) In contrast, significant increases in salivary cortisol following a semester of African dance (n = 21) and no change after a biology lecture (n = 30) were reported. It should be noted that the subjects included only university students, a group that is not representative of the anxiety and stress disorder population. Furthermore, this study demonstrates the inaccuracy of associating psychological stress with cortisol levels, as the dance group reported decreases in PSS scores, yet had increased cortisol levels.
Decreased serum cortisol concentrations were also found in 8 yoga instructors after 1 hour of yoga practice as compared to before practice. (50) While suggestive, this work suffers from a lack of randomization, lack of inclusion/exclusion criteria, short duration, and a small sample size, again rendering interpretation difficult and preventing determination of causality between yoga practice and cortisol hormone reduction. It is also worth noting that of all the populations used in these studies, this was one of only two groups composed of persons who regularly practiced yoga prior to the study.
The relationship between yoga practice/meditation and other hormonal markers of stress, melatonin and DHEA, was examined in two studies, one in cancer patients (41) and one in healthy subjects. (40) Both groups recorded melatonin levels in subjects pre- and post-intervention with a program incorporating both yoga and meditation. Each group independently described no correlation between melatonin levels and yoga/meditation. (40,41) Carlson's group also measured DHEA, but found no significant changes from baseline after the intervention. (41)
Biochemical changes in GABA levels were studied by Streeter et al. (51) after a single 1-hour period of yoga practice in experienced yoga practitioners (n = 8) or a single, 1-hour period of reading in the controls (n = 11). The experimenters observed significant increases in GABA levels in the yoga practitioners, but not in the readers. (51) Increases in GABA levels in the brain are indicative of relaxation. (15) Despite the promising data, the small sample sizes and single intervention period make drawing conclusions for a broader population from these results difficult.
Safety and Compliance
Safety was not proposed as a primary endpoint for any of the studies described in this review, and adverse events associated with yoga and meditation were not noted by the investigators. It is notable that adherence to yoga and meditation interventions in all of the trials was high. Compliance with yoga regimens, based upon dropout rates, was calculated for 5 of the studies with the greatest number of subjects. Based upon the high compliance seen in these studies, it seems reasonable to conclude that the interventions employed were generally easy to practice and maintain. Therefore, if the promising results for yoga and meditation in the relief of anxiety and stress demonstrated in the described studies are confirmed, these may be options that are well tolerated by the majority of patients.
Of note in regard to compliance is an analysis of the long-term effects of yoga by Miller and coworkers, who conducted a 3-year follow-up of a previous trial. (34,35) The researchers initially investigated the effect of meditation on anxiety in 22 patients with a diagnosed anxiety disorder over a 3-month period. (34) HAM-A and Beck Anxiety scores associated with the initial intervention were significantly improved from baseline, and these results were maintained at the 3-year follow-up. (35) High adherence rates to a mindfulness-based stress reduction program (including meditation and yoga) after 3 years were seen, as was the sustained effect of this intervention on mood and anxiety states in this patient population. Aside from this work, none of the other short-term studies ([less than or equal to] 4 months) described previously were followed up by a long-term study of the effects of yoga on stress and anxiety, thus rendering long-term compliance difficult to assess.
Conclusions and Future Directions
While a direct relationship between yogic practices and declines in stress and anxiety is not yet supported by statistically sound randomized controlled trials, the studies described suggest that yoga as a supplement to pharmacologic therapy may improve stress and anxiety symptoms. Because the data are conflicting and many experiments thus far conducted have not been well designed, yoga should not replace conventional medical practice until further work has been conducted and strong statistical evidence of its benefit exists.
Many difficulties arise in drawing definitive conclusions from these data. For example, different researchers used different populations for study, ranging from patients with diagnosed anxiety disorders to healthy yoga instructors to cancer patients. This mix of heterogeneous study populations makes comparisons of the results problematic; it is unreasonable to extrapolate data obtained from healthy college students (44) and young musicians (28,29) to the general population.
Further confounding the question of yoga's efficacy are the questions of the validity of outcome measures, numbers of subjects and validity of statistical analyses. Measurement of objective primary endpoints (e.g., biochemical measurements, validated psychological instruments) in randomized controlled trials with a predetermined power and appropriate sample size is necessary to yield more definitive statistics.
Further difficulties arise from lack of appropriate controls in some studies. These are imperative to validate the effectiveness of yoga in reducing stress and anxiety symptoms. Control subjects should experience the same camaraderie as the yoga intervention group to determine whether the results obtained were from the practice of yoga alone or from the personal friendships and support developed during yoga practice. Another useful control group would be a group involved in a different form of exercise, as any effects seen in a yoga group may simply result from the effects of physical activity of any kind, as was noted by West and colleagues. (44) Standardization of outcomes and interventions is also necessary; it is difficult to compare the studies thus far, as the authors employed different outcome measures. For instance, some groups measured anxiety using the HAM-A and HADS scales, while others used questionnaires designed specifically for their research and for which validation may be lacking. Often, these questionnaires and scales were self-reported, and answers may have differed depending on the amount of time allotted to ponder responses.
Standardization of interventions would also be of benefit in determining what effects, if any, the various forms of yogic practice have on stress and anxiety. Studies evaluating the different styles of yoga would be informative in determining which, if any, form of yoga (or whether full yoga practice or meditation alone) would be the most effective at reducing stress and anxiety.
An additional consideration in assessing patient populations includes accounting for the stress and anxiety associated with learning anything new. Stress and anxiety in certain patients may be elevated at the start of intervention. Thus, researchers should consider recruiting experienced yoga practitioners as participants in future studies. Future research should also include economic considerations. The cost-effectiveness of managing anxiety with and without yoga practice compared to the use of medications alone should be investigated.
In summary, the data thus far are suggestive of beneficial outcomes from the use of yoga as an intervention for stress and anxiety; yoga may be considered as a possible adjunctive therapy for those experiencing stress and anxiety. Due to its good compliance and lack of drug interactions, yoga appears to be safe and could be encouraged to improve quality of life and, perhaps, the symptoms of stress and anxiety. Nonetheless, only when the benefits of yoga practice have been realized through thorough, valid research study should yoga be recommended as a method to decrease the pill burden or to replace pharmacologic treatment.
(1.) Not listed. 2008 Yoga in America Study. Yoga Journal 2008:93. http://www.researchandmarkets.com/ reportinfo.asp?report_id=836593&t=o&cat_id=#. [Accessed August, 25 2010]
(2.) Behera D. Yoga therapy in chronic bronchitis. J Assoc Physicians India 1998;46:207-208.
(3.) da Silva GD, Lorenzi-Filho G, Lage LV. Effects of yoga and the addition of Tui Na in patients with fibromyalgia. J Altern Complement Med 2007;13:1107-1113.
(4.) Booth-LaForce C, Thurston RC, Taylor MR. A pilot study of a Hatha yoga treatment for menopausal symptoms. Maturitas 2007;57:286-295.
(5.) Landers SJ. Alternative Therapy Use Documented in New Survey. American Medical News January 13, 2009.
(6.) Ospina MB, Bond K, Karkhaneh M, et al. Meditation practices for health: state of the research. Evid Rep Technol Assess (Full Rep) 2007;155:1-263.
(7.) Segerstrom SC, Miller GE. Psychological stress and the human immune system: a meta-analytic study of 30 years of inquiry. Psychol Bull 2004;130:601-630.
(8.) Roest AM, Martens EJ, de Jonge P, Denollet J. Anxiety and risk of incident coronary heart disease: a meta-analysis. J Am Coll Cardiol 2010;56:38-46.
(9.) Olatunji BO, Cisler JM, Tolin DR Quality of life in the anxiety disorders: a meta-analytic review. Clin Psychol Rev 2007;27:572-581.
(10.) Brezo J, Paris J, Turecki G. Personality traits as correlates of suicidal ideation, suicide attempts, and suicide completions: a systematic review. Acta Psychiatr Scand 2006;113:180-206.
(11.) American Psychiatric Association, American Psychiatric Association. Task Force on DSM-IV. Diagnostic and statistical manual of mental disorders: DSM-IV-TR. Fourth edition, text revision. Washington, DC: American Psychiatric Association; 2000:943.
(12.) Anxiety Disorders. National Institute of Mental Health Web site. http://www.nimh.nih.gov/health/ publications/anxiety-disorders/complete-index. shtml. Updated 2010. [Accessed August 25, 2010.]
(13.) National Institute of Mental Health. Cognition and Stress: Advances in Basic and Translational Research. National Institute of Medicine Web site. http:// www.nimh.nih.gov/research-funding/scientificmeetings/2007/ cognition-and-stress-advances-in-basic-and-translational-research/index.shtml. Updated 2009. [Accessed August 26, 2010.]
(14.) Feder A, Nestler EJ, Charney DS. Psychobiology and molecular genetics of resilience. Nat Rev Neurosci 2009;10:446-457.
(15.) Dusek JA, Benson H. Mind-body medicine: a model of the comparative clinical impact of the acute stress and relaxation responses. Minn Med 2009;92:47-50.
(16.) Banerjee B, Vadiraj HS, Ram A, et al. Effects of an integrated yoga program in modulating psychological stress and radiation-induced genotoxic stress in breast cancer patients undergoing radiotherapy. Integr Cancer Ther 2007;6:242-250.
(17.) Rao MR, Raghuram N, Nagendra HR, et al. Anxiolytic effects of a yoga program in early breast cancer patients undergoing conventional treatment: a randomized controlled trial. Complement Ther Med 2009;17:1-8.
(18.) Ulger O, Yagli NV. Effects of yoga on the quality of life in cancer patients. Complement Ther Clin Pract 2010;16:60-63.
(19.) Vadiraja HS, Rao MR, Nagarathna R, et al. Effects of yoga program on quality of life and affect in early breast cancer patients undergoing adjuvant radiotherapy: a randomized controlled trial. Complement Ther Med 2009;17:274-280.
(20.) Vadiraja HS, Raghavendra RM, Nagarathna R, et al. Effects of a yoga program on cortisol rhythm and mood states in early breast cancer patients undergoing adjuvant radiotherapy: a randomized controlled trial. Integr Cancer Ther 2009;8:37-46.
(21.) Waelde LC, Thompson L, Gallagher-Thompson D. A pilot study of a yoga and meditation intervention for dementia caregiver stress. J Clin Psychol 2004;60:677-687.
(22.) Michalsen A, Grossman P, Acil A, et al. Rapid stress reduction and anxiolysis among distressed women as a consequence of a three-month intensive yoga program. Med Sci Monit 2005;11:CR555-561.
(23.) Javnbakht M, Hejazi Kenari R, Ghasemi M. Effects of yoga on depression and anxiety of women. Complement Ther Clin Pract 2009;15:102-104.
(24.) Ando M, Morita T, Akechi T, et al. The efficacy of mindfulness-based meditation therapy on anxiety, depression, and spirituality in Japanese patients with cancer. J Palliat Med 2009;12:1091-1094.
(25.) Bosch PR, Traustadottir T, Howard P, Matt KS. Functional and physiological effects of yoga in women with rheumatoid arthritis: a pilot study. Altern Ther Health Med 2009;15:24-31.
(26.) Beddoe AE, Paul Yang CP, Kennedy HP, et al. The effects of mindfulness-based yoga during pregnancy on maternal psychological and physical distress. J Obstet Gynecol Neonatal Nuts 2009;38:310-319.
(27.) Telles S, Singh N, Joshi M, Balkrishna A. Post traumatic stress symptoms and heart rate variability in Bihar flood survivors following yoga: a randomized controlled study. BMC Psychiatry 2010;10:18.
(28.) Khalsa SB, Cope S. Effects of a yoga lifestyle intervention on performance-related characteristics of musicians: a preliminary study. Med Sci Monit 2006;12:CR325-31.
(29.) Khalsa SB, Shorter SM, Cope S, et al. Yoga ameliorates performance anxiety and mood disturbance in young professional musicians. Appl Psychophysiol Biofeedback 2009;34:279-289.
(30.) Malathi A, Damodaran A. Stress due to exams in medical students--role of yoga. Indian J Physiol Pharmacol 1999;43:218-224.
(31.) Simard AA, Henry M. Impact of a short yoga intervention on medical students' health: a pilot study, ivied Teach 2009;31:950-952.
(32.) Gupta N, Khera S, Vempati RP, et al. Effect of yoga based lifestyle intervention on state and trait anxiety. Indian J Physiol Pharmacol 2006;50:41-47.
(33.) Khemka SS, Rao NH, Nagarathna R. Immediate effects of two relaxation techniques on healthy volunteers. Indian 3 Physiol Pharmacol 2009;53:67-72.
(34.) Kabat-Zinn J, Massion AO, Kristeller J, et al. Effectiveness of a meditation-based stress reduction program in the treatment of anxiety disorders. Am J Psychiatry 1992;149:936-943.
(35.) Miller JJ, Fletcher K, Kabat-Zinn J. Three-year follow-up and clinical implications of a mindfulness meditation-based stress reduction intervention in the treatment of anxiety disorders. Gen Hosp Psychiatry 1995;17:192-200.
(36.) Lee SH, Ahn SC, Lee YJ, et al. Effectiveness of a meditation-based stress management program as an adjunct to pharmacotherapy in patients with anxiety disorder. J Psychosom Res 2007;62:189-195.
(37.) Kjellgren A, Bood SA, Axelsson K, et al. Wellness through a comprehensive yogic breathing program--a controlled pilot trial. BMC Complement Altern Med 2007;7:43.
(38.) Girodo M. Yoga meditation and flooding in the treatment of anxiety neurosis. J Behav Ther Exp Psychiatry 1974;5:157-160.
(39.) Smith C, Hancock H, Blake-Mortimer J, Eckert K. A randomised comparative trial of yoga and relaxation to reduce stress and anxiety. Complement Ther Med 2007;15:77-83.
(40.) Harinath K, Malhotra AS, Pal K, et al. Effects of Hatha yoga and Omkar meditation on cardiorespiratory performance, psychologic profile, and melatonin secretion. J Altern Complement Med 2004;10:261-268.
(41.) Carlson LE, Speca M, Patel KD, Goodey E. Mindfulness-based stress reduction in relation to quality of life, mood, symptoms of stress and levels of cortisol, dehydroepiandrosterone sulfate (DHEAS) and melatonin in breast and prostate cancer outpatients. Psychoneuroendocrinology 2004;29:448-474.
(42.) Satyapriya M, Nagendra HR, Nagarathna R, Padmalatha V. Effect of integrated yoga on stress and heart rate variability in pregnant women. Int J Gynaecol Obstet 2009;104:218-222.
(43.) Cowen VS. Functional fitness improvements after a worksite-based yoga initiative. J Bodyw Mov Ther 2010;14:50-54.
(44.) West J, Otte C, Geher K, et al. Effects of Hatha yoga and African dance on perceived stress, affect, and salivary cortisol. Ann Behav Med 2004;28:114-118.
(45.) Chattha R, Raghuram N, Venkatram P, Hongasandra NR. Treating the climacteric symptoms in Indian women with an integrated approach to yoga therapy: a randomized control study. Menopause 2008;15:862-870.
(46.) 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.
(47.) Cohen BE, Chang AA, Grady D, Kanaya AM. Restorative yoga in adults with metabolic syndrome: a randomized, controlled pilot trial. Metab Syndr Relat Disord 2008;6:223-229.
(48.) Schell FJ, Allolio B, Schonecke OW. Physiological and psychological effects of Hatha-Yoga exercise in healthy women. Int J Psychosom 1994;41:46-52.
(49.) Robert-McComb JJ, Tacon A, Randolph P, Caldera Y. A pilot study to examine the effects of a mindfulness-based stress-reduction and relaxation program on levels of stress hormones, physical functioning, and submaximal exercise responses. J Altern Complement Med 2004;10:819-827.
(50.) Kamei T, Toriumi Y, Kimura H, et al. Decrease in serum cortisol during yoga exercise is correlated with alpha wave activation. Percept Mot Skills 2000;90:1027-1032.
(51.) Streeter CC, Jensen JE, Perlmutter RM, et al. Yoga Asana sessions increase brain GABA levels: a pilot study. J Altern Complement Med 2007;13:419-426.
Carroll-Ann Wo Goldsmith, DSc--Associate Professor of Pharmaceutical Sciences. Department of Pharmaceutical Sciences, Massachusetts College of Pharmacy and Health Sciences, 1260 Elm Street, Manchester, NH 03101. Corresponding Author: Email firstname.lastname@example.org
Amber W. Li, PharmD-Clinical Pharmacist at St. Elizabeth's Medical Center, 736 Cambridge Street, Brighton, MA 02135.
Table 1. Validated Scales of Anxiety and Stress States Used in Described Studies Scale Abbreviation Description Hamilton Anxiety Rating Scale HAM-A Assesses severity of cognitive and psychological symptoms of anxiety; 14 items Spielberger State-Trait Anxiety STAI Assesses Inventory state-anxiety & trait anxiety; 20 items each, self-rated scale State Trait Personality STPI Measures anxiety on a 4-point scale; 10-item Inventory subscale self-reported questionnaire Symptoms of Stress Inventory SOSI Measures psychological and behavioral responses to stressful situations; 94 items Perceived Stress Scale PSS Measures stress on a visual analog scale; 14 items Beck Anxiety Inventory BAI Measures anxiety on 4-point scale; 21 item questionnaire Positive Affect & Negative PANAS Measures positive Affect Schedule and negative emotions; 20-item visual analog scale Performance Anxiety PAQ Measure feelings and Questionnaire symptoms of anxiety related to practice Hospital Anxiety Depression HADS and group or solo Scale performance; 20 items Measures anxiety and depression; 14 items (7 each for anxiety and depression) General Health Questionnaire GHQ Assesses distress; measures minor psychiatric illness in a community setting; 12 items Screening Questionnaire for SQD Measures PTSD and Disaster Mental Health depression; 12 yes/no questions, 9 for PTSD, 6 for depression (3 questions common to both PTSD and depression) Institute for Personality and IPAT Measures anxiety; 40 Ability Testing items Anxiety Symptom ASQ * Measures changes in Questionnaire anxiety symptoms; 15 items * unvalidated Table 2. Indirect Measures of Stress and Anxiety * Blood Pressure * Heart Rate * Respiratory Function ** Forced Expiratory Volume in second (FEV1) ** Respiratory Rate ** Oxygen Consumption Table 3. Biochemical Indicators of Stress and Anxiety. Indicator Effect with Stress or Anxiety Stress Hormones * Cortisol [up arrow] * DHEA [up arrow] Neurotransmitters * Melatonin [down arrow] * GABA [down arrow] Table 4a. Summary of Studies Included in this Review First Author Design Duration Ando (24) Non-randomized, 2 weeks (2009) non-controlled Banjeree (16) Randomized, 6 weeks (2007) controlled Beddoe (21) Non-randomized, 7 weeks (2009) non-controlled Bosch (25) Non-randomized, 10 weeks (2009) controlled, pilot Carlson (41) Non-randomized, 1 week (2004) controlled, blinded, pilot Chatta (45) Randomized, 8 weeks (2008) controlled Cohen (42) Randomized, 10 weeks (2008) controlled, pilot Cowen (43) Non-controlled, 4 yoga (2010) non-randomized classes over 6 weeks First Author Study Population Ando (24) n = 28 (meditation, yoga, and (2009) breathing); cancer patients in treatment Banjeree (16) n = 68 (35 yoga, 33 (2007) control);post-op breast cancer patients Beddoe (21) n =17 (mindfulness-based (2009) yoga); pregnant women Bosch (25) n =16 (9 yoga, control); (2009) postmenopausal women with RA Carlson (41) n = 59 (meditation and gentle (2004) yoga); breast and prostate cancer patients Chatta (45) n =120 (60 yoga, 60 physical (2008) exercise); perimenopausal women Cohen (42) n = 24 (12 yoga, 12 control); (2008) adults with metabolic syndrome Cowen (43) n = 77 (yoga); firefighters (2010) First Author Results Ando (24) Significant decreases in HADS scores after (2009) intervention Banjeree (16) Significant decrease from baseline in HADS (2007) and PSS scores in yoga, but not control, group Beddoe (21) Significant decrease from baseline in STAI (2009) state-anxiety scores and near significant decrease in PSS scores after invention; no effects on salivary cortisol concentrations Bosch (25) Significant decreases from baseline in HAQ (2009) disability, BDI depression, and pain scores in yoga vs. control groups; no difference in salivary cortisol concentrations Carlson (41) No differences from baseline after (2004) meditation in SOSI scores, levels of salivary melatonin, levels of salivary cortisol Chatta (45) Greater decline in PSS scores in yoga vs. (2008) exercise group Cohen (42) No significant changes, though trends (2008) toward significant declines in blood pressure and PSS scores in yoga vs. control group Cowen (43) Significant decrease in PSS scores after yoga (2010) intervention Table 4b. Summary of Studies Included in this Review First Author Design Duration Girodo (38) Non-randomized, 4 months (1974) non-controlled Granath (46) Randomized, 4 months (2006) controlled Gupta (32) Randomized, 10 days (2006) controlled Harinath (40) Non-randomized, 3 months (2004) controlled Javnbakht (23) Randomized, 2 months (2009) controlled Kabat-Zinn (34) Non-randomized, 5 months (1992) non-controlled Kamei (50) Non-randomized, 1 hour (2000) non-controlled Khalsa (28) Non-randomized, 8 weeks (2006) controlled, pilot First Author Study Population Girodo (38) n = 9 (1974) (meditation) Granath (46) n = 31 (2006) (16 cognitive therapy, 15 yoga) Gupta (32) n = 225 (175 yoga, 50 control); (2006) patients with history of chronic ailments (e.g., hypertension, Crohn's disease) Harinath (40) n = 30 (15 yoga and meditation, (2004) 15 control) Javnbakht (23) n = 65 (34 yoga, 31 control); (2009) women Kabat-Zinn (34) n = 20 (meditation) (1992) Kamei (50) n = 7 (yoga) (2000) Khalsa (28) n =18 (10 yoga, 8 control); (2006) musicians First Author Results Girodo (38) No difference between pre- and post (1974) meditation ASQ mean scores Granath (46) Cognitive therapy and Kundalini yoga groups (2006) both showed significant decreases from baseline in PSS scores; no change in salivary cortisol in either group Gupta (32) Significant decrease from baseline in STAI (2006) scores in yoga, but not control, group Harinath (40) Significant increase in melatonin levels at 2, 3, (2004) and 4 a.m. in yoga, but not control, group; no significant change in anxiety scores in either group Javnbakht (23) Significant decreases in STAI state-anxiety (2009) and trait-anxiety in yoga vs. control group; no significant change in BDI depression scores Kabat-Zinn (34) Significant decreases from baseline in HAM-A (1992) scale and BAI at 8 week and 3 month follow-up Kamei (50) Significant decreases from baseline in serum (2000) cortisol levels after yoga session Khalsa (28) Significant decreases in PAQ performance (2006) anxiety scores from baseline in yoga, but not control, group; significant difference in solo performance anxiety scores pre- and post-intervention in yoga vs. control group Table 4c. Summary of Studies Included in this Review First Author Design Duration Khalsa (29) Non-randomized, 2 month (2009) controlled interven- tion and 1-year follow-up Khemka (33) Randomized, 1 x 20 - (2009) controlled minute session Kjellgren (37) Non-randomized, 6 weeks (2007) controlled, pilot Lee (36) Randomized, 8 weeks (2007) controlled Malathi (30) Non-randomized, 3 months (1999) controlled Michalsen (22) Non-randomized, 12 weeks (2005) controlled Miller (35) Non-randomized, 3 years (1995) non-controlled First Author Study Population Khalsa (29) n = 45 (15 yoga lifestyle, 15 yoga (2009) and meditation only, 15 control); musicians Khemka (33) n = 86 (43 yoga-based deep (2009) relaxation, 43 supine rest); male and female healthy volunteers Kjellgren (37) n =103 (48 yoga, 55 control) (2007) Lee (36) n = 46 (24 meditation, 22 anxiety (2007) education) Malathi (30) n = 50 (25 yoga, 25 control); (1999) medical students Michalsen (22) n = 24 (16 yoga, 8 control) (2005) Miller (35) n =18 (meditation); (1995) patients with anxiety disorder First Author Results Khalsa (29) Yoga lifestyle and yoga meditation, but not (2009) control, groups had or trended toward significant decreases from baseline in PAQ performance anxiety and PSS stress scores at end of intervention; yoga lifestyle group maintained significantly decreased PAQ group and solo performance scores at one-year follow-up Khemka (33) Significant improvement in STAI-A state (2009) anxiety scores after yoga-based deep relaxation, but not supine rest Kjellgren (37) Significant decreases in stress and anxiety (2007) HAD scores and significant increase in optimism from baseline in yoga, but not control group Lee (36) Significant improvements in HAM-A and STAI (2007) anxiety scores from baseline in meditation, but not education control group Malathi (30) Significantly decreased basal STAI anxiety (1999) scores after yoga practice; significantly decreased STAI anxiety scores on day of exams after yoga practice Michalsen (22) Significant decreases from baseline in PSS and (2005) STAI scores and in salivary cortisol levels in yoga group when compared to control group Miller (35) Significant decreases in HAM-A, BAI, and (1995) number of panic attacks maintained over 3 years Table 4d. Summary of Studies Included in this Review First Author Design Duration Rao (17) Randomized, 6 weeks (2009) controlled Robert-McComb (49) Non-randomized, 8 weeks (2004) controlled, pilot Satyapriya (42) Randomized, 16 weeks (2009) controlled Schel1 (48) Non-randomized, 3 hours (1994) controlled Simard (31) Non-randomized, 16 weeks (2009) non-controlled, pilot Smith (39) Randomized, 10 weeks (2007) controlled Streeter (51) Non-randomized, 1 hour (2007) pilot Tellesz (27) Randomized, 1 week (2010) controlled First Author Study Population Rao (17) n = 38 (18 yoga, 20 supportive (2009) therapy); breast cancer patients in treatment Robert-McComb (49) n =18 (9 control, 9 stress (2004) reduction program participants) Satyapriya (42) n = 90 (45 yoga, 45 deep (2009) relaxation); pregnant women Schel1 (48) n = 25 (12 yoga, 13 control) (1994) Simard (31) n =14 (yoga); medical students (2009) Smith (39) n =131 (68 yoga, 63 relaxation) (2007) Streeter (51) n =19 (8 yoga, 11 reading) (2007) Tellesz (27) n = 22 (11 yoga, 11 control); (2010) male, flood survivors First Author Results Rao (17) Significant decreases in STAI state- and (2009) trait-anxiety and symptom distress in yoga vs. control group after surgery and during and after radiation and chemotherapy Robert-McComb (49) No difference from baseline in cortisol levels in (2004) either group Satyapriya (42) Significant decrease from baseline in PSS (2009) scores in yoga group, but significant increase in control group; PSS scores in yoga group significantly lower than control group; heart rate variability showed improved adaptive autonomic responses to stress in yoga, but not control group Schel1 (48) Control, but not yoga, group reacted with (1994) anger to stress; no difference in cortisol levels from baseline in either group; no trend in changes in prolactin levels in either group Simard (31) GHQ and PSS scores significantly improved (2009) from baseline at 8 and 16 weeks with yoga intervention Smith (39) No changes from baseline in STPI or GHQ (2007) scores in either group Streeter (51) Significant decreases from baseline in GABA (2007) levels in yoga, but not reading group Tellesz (27) Significant decreases from baseline in anxiety (2010) and sadness in yoga group one month after flood; controls showed increased anxiety and sadness Table 4e. Summary of Studies Included in this Review First Author Design Duration Ulger (18) Non-randomized, 4 weeks (2010) non-controlled Vadiraja (19) Randomized, 6 weeks (2009) controlled Vadiraja (20) Randomized, 6 weeks (2009) controlled Waelde (21) Non-randomized, 1 month (2004) non-controlled, pilot West (44) Non-randomized 1.5 hours (2004) First Author Study Population Ulger (18) n = 20 (yoga and meditation); (2010) breast cancer patients in treatment Vadiraja (19) n = 75 (42 yoga, 33 counseling); (2009) breast cancer patients undergoing radiotherapy Vadiraja (20) n = 75 (42 yoga, 33 counseling); (2009) breast cancer patients undergoing radiotherapy Waelde (21) n =12 (yoga and meditation) (2004) West (44) n = 69 (21 dance, 18 yoga, 30 (2004) lecture) First Author Results Ulger (18) Significant decrease in STAI scores after yoga (2010) intervention; significant increase in quality of life scores Vadiraja (19) Compared to control, yoga group significantly (2009) improved positive affect, emotional and social function and significantly decreased negative affect on PANAS Vadiraja (20) Significant decreases from baseline in HADS (2009) anxiety and depression scores and in PSS perceived stress after intervention and compared to controls; decreased 6 a.m. and pooled diurnal salivary cortisol levels in yoga group compared to controls Waelde (21) Significant decreases from baseline in STAI (2004) scores after intervention West (44) Significant decreases from baseline in PSS (2004) scores in yoga and dance, but not lecture group; no change in cortisol levels from baseline in lecture group, significant decrease from baseline in yoga group, significant increase from baseline in dance group