Applications of Yoga in Psychiatry: What We Know

The Yoga Sutras of Patanjali conceptualizes the practice of yoga as comprising the following eight limbs (13): yamas (the five moral restraints—nonviolence, truthfulness, nonstealing, moderation, and nonhoarding); niymas (the five observances—purity, contentment, zeal/austerity, self-study, and devotion to a higher power); asanas (postures); pranayama (mindful breathing); pratyahara (turning inward); dharana (concentration); dhyana (meditation); and samadhi (union of the self with object of meditation) (13, 14). Most research studies focus on asanas (postures), pranayama (mindful breathing), and dhyana (meditation). The eight limbs were developed to be used together so that the ability to study them in isolation is difficult. This poses a challenge to the scientific method that is designed to measure changes in clearly defined dependent variables. Yoga was designed so that all eight limbs were to be practiced in sequence. The Western proclivity for studying component parts, such as postures, mindful breathing, and meditation, highlights the difference between Eastern and Western approaches to healing. There is, accordingly, a conflict between the Western view of dissecting healing modalities into component parts and the practice of yoga in which the component parts are designed to be used simultaneously. For example, it is hard to perform a posture correctly without simultaneously concentrating on the form and breath. This creates problems regarding a consistent means of describing YBTs in the scientific literature.

Breath, or pranayama, is essential to life. Breathing abnormalities are associated with distress and discomfort. For example, panic is associated with rapid shallow breathing. Rapid breathing can make one feel anxious by increasing sympathetic activity, the part of the nervous system responsible for the fight-or-flight response (15). Alternatively, slow breathing (two to four breaths per minute) with exhalation greater than inhalation, and holding of the breath at the end of inhalation or exhalation, increases parasympathetic activity and is associated with relaxation (16). Cultures all over the world have figured out the importance of the breath and breathing practices. Chants and singing are examples of “exhalation greater than inhalation” practices that improve balance in the autonomic nervous system (ANS) (17). Many disorders that mental health professionals treat, such as depression, anxiety, and posttraumatic stress disorder (PTSD), have low parasympathetic activity (18).

Hatha yoga is the most popular form of yoga in the United States and, frequently, the yoga practice used in research studies. Although the book The Yoga Sutras of Patanjali describes the postures as physical, the postures and breathing exercises are practiced so that the mind can be still for meditation and concentration. The schools of Hatha yoga vary in emphasis on breath, postures, and meditation. There is a range of physical intensity and spiritual focus across different forms of yoga practice (19). Some popular types of Hatha yoga include Iyengar yoga (alignment of the body postures, or asanas), Vinyasa yoga (from the word meaning “to flow,” breath-linked movement involving asanas and pranayama), Sudarshan Kriya yoga (SKY; almost exclusively breathing [pranayama]), Bikram yoga (physically intense and posture focused, with two breathing exercises; performed in a heated environment), and Ashtanga yoga (which involves synchronizing the breath with a progressive series of postures).

Research has shown that yoga users tend to be primarily female, Caucasian, younger, and college educated (1, 20). Mental health professionals, therefore, need to be aware of the potential barriers to entry for low-socioeconomic populations. Leisure time and access to a yoga studio (both financially and geographically) must be considered when working with marginalized populations, particularly considering the demands of working multiple jobs and child-care/family responsibilities. A study of chronic low back pain in low-income minority participants found that, although yoga was perceived as effective for pain relief, mood improvement, and stress management, lack of time represented the greatest barrier to practice (21). Given the increased burden of disease found in lower socioeconomic populations (22), this study highlights the need to make yoga-based interventions accessible to lower income populations.

The primary focus of this article is on the potential of yoga postures and/or breathing as a treatment for depression and anxiety disorders, because these disorders have been the most formally studied. The use of yoga for individuals with psychosis, bipolar disorder, eating disorders, and addiction is beyond the scope of this article. Although there is mounting evidence that mindfulness and other meditative practices are helpful for a variety of mental health conditions, these interventions are also beyond the scope of the present article.

Of the major psychiatric disorders, depression has a compelling literature supporting the use of yoga as a treatment, even though there are limitations in the current literature base. A 2013 meta-analysis of yoga for the treatment of depressive symptoms and major depressive disorder (MDD) included 12 randomized controlled trials (RCTs) and found evidence to support short-term ancillary effects, despite methodological limitations (23). The most recent systematic review of yoga as a treatment specifically for patients with MDD found 7 RCTs (N=240) that compared yoga to different control conditions: 2 studies used attentional controls (24, 25); 1 study used exercise as the control intervention (26); 1 study used electroconvulsive therapy (ECT) or antidepressant medication as the active control (27); 1 study used medication monotherapy as an active control (27); and 2 studies tested yoga as an add-on to antidepressant medication (28, 29). Yoga was comparable with exercise and medication only, although ECT produced stronger effects. For add-on medication and attentional control groups, one study found positive effects and the other did not (30). A recent RCT of yoga versus health education classes found that, for individuals with residual symptoms of depression (N=122), reductions in depression in the yoga group emerged at 3 and 6 months after 10 weeks of yoga but not at the primary 10-week endpoint (31).

A substantial proportion of patients with MDD (approximately 30%–40%) do not achieve full remission with antidepressants (32, 33). Open-label studies suggest that the addition of a yoga-based intervention for those who are taking antidepressant medications but who are still symptomatic is worthy of consideration and additional research (18, 34, 35). Some advantages of using YBTs include the lack of medication side effects and interactions, subjects’ reluctance to take antidepressant medications, the potential application in pregnant populations where concerns about medication-related toxic effects on the fetus are common (36, 37), and the lack of stigma. This supports consideration of YBTs as an adjunct for symptomatic individuals.

There are few dosing studies related to the use of yoga as a treatment for mental health disorders. A recent randomized controlled dosing study of individuals with MDD found that both twice- or thrice-weekly 90-min sessions of Iyengar yoga plus coherent breathing of 5 breaths per minute in addition to homework were equally effective for both remission and response, as measured by the Beck Depression Inventory–II (BDI-II) (38); the thrice-weekly dose outperformed the twice-weekly regimen with respect to reducing BDI-II scores ≤10 (34). This was the first study to explore dosing in an MDD population, and it suggests that 90-min Iyengar yoga classes two to three times per week plus homework may be enough to reduce depressive symptoms. More evidence-based research is needed to define treatment parameters. This does not mean, however, that individuals who are currently suffering should be deprived of this potential form of treatment.

The transdiagnostic potential of yoga is in line with a shift of perspective, which focuses on symptoms versus discrete diagnoses (45). There is evidence that depression, anxiety, and PTSD (18), along with frequently comorbid medical disorders such as epilepsy (18), hypertension (46), and cardiovascular disease (47), are all associated with an imbalance of the ANS. This imbalance is manifested by decreased parasympathetic nervous system (PNS) activity and increased sympathetic nervous system (SNS) activity, which is associated with negative affective states and may be the final common pathway between negative emotions and ill health (48, 49). In addition, low PNS activity, manifested by low vagal tone or high frequency heart rate variability (HRV), is also associated with decreased activity in the GABA system in depression (50, 51), PTSD (52–54), anxiety (55, 56), and epilepsy (57–59). The vagal-GABA theory provides an explanation as to why yoga may decrease symptoms in disorders associated with stress (18). This theory is based on polyvagal theory, which identifies SNS activity and PNS activity in the myelinated vagus nerve as being associated with certain behavioral states (60). Although it is adaptive to have increased SNS activity during acute stress, it is not adaptive to maintain increased SNS activity chronically, as seen in chronic stress associated with noncommunicable disorders. This imbalance has implications for everyday functioning, since myelinated vagal states facilitate self-regulation and interpersonal interactions (61). The ability to socially interact or maintain social relationships is associated with improved mental and physical health, as seen in octogenarians (62). A transdiagnostic approach that considers the comorbidity of both mental and physical disorders is consistent with the results from a STAR*D report (N=2,541), which found that general medical comorbidity was 50% in a large sample of outpatients with MDD who also demonstrated greater rates of somatic symptoms, gastrointestinal symptoms, and sympathetic arousal (63).

Interventions that decrease stress and allostatic load (i.e., the physiological burden of stress on the body) and allow the individual to return to homeostasis and health are important. The ability to return from allostasis to homeostasis is a marker of resilience. The ability of YBPs to restore balance to the ANS, so that SNS activity is not too high and PNS is not too low (64, 65), should be considered as a part of a treatment plan, since it has relatively low cost (compared with conventional mental health treatments), has a low side-effect profile, and can potentially address the symptoms of mental and noncommunicable disorders (e.g., cardiovascular disease, cancer, chronic respiratory diseases, and diabetes).

Most pharmacologic treatments for MDD involve drugs that work via the monoamine system (e.g., norepinephrine, serotonin, and dopamine) so that interventions emphasizing ANS or the GABA system are novel. There is evidence that increased vagal stimulation, such as that produced by an implanted vagal nerve stimulator (VNS), improves symptoms in treatment-resistant depression (66) and epilepsy (67, 68). This is consistent with the vagal-GABA theory. There is additional evidence that low HRV and low activity in the GABA system are associated with numerous disorders exacerbated by stress such as epilepsy, depression, and PTSD (18).

There is very little empirical literature regarding the use of heated yoga for mental health conditions. One RCT (N=52) found that, for women with emotional eating, Bikram yoga (performed in a room heated to 105°F with 40% humidity) outperformed a waitlist control for cortisol reactivity in response a stress test (69). In this same RCT, distress tolerance increased for those in the yoga condition compared with the waitlist control group (70). Another recent RCT found that, for physically inactive and stressed adults (N=63), Bikram yoga, compared with a no-treatment control, was associated with improvements in perceived stress, self-efficacy, general health, and energy/fatigue (71). In an uncontrolled study (N=54) of healthy individuals, Bikram yoga improved mindfulness and decreased stress (72). There were acute effects for regular Bikram yoga participants (N=53) after a 90-min session, regarding negative affect and state anxiety, that were greater for those with higher stress (73). The lack of an active control is a weakness is this area of the literature and needs to be addressed in future studies.

Studies have suggested that heat itself may have independent mood-enhancing properties (40–47), which is reflected by the use of heat around the world; for example, sweat lodges, saunas, and hot springs. A recent article showed promising results for whole-body hyperthermia as a treatment for MDD in individuals not taking antidepressant medications, when compared with a blinded sham condition (74). Although this study was not the first of its kind (75), it was the most robust in design and effect sizes thus far. There are no known studies comparing heated versus nonheated forms of yoga for mental health indications.

There are methodological limitations in the literature base with regard to yoga as treatment for mental health conditions. These include: the heterogeneity of yoga interventions and patient samples; variability in use of home practice; inadequate sample size and randomization procedures; lack of intention-to-treat analysis and blinding of outcome assessors; limited fidelity assessments; inadequate control groups; unreported dropout, relapse, and remission rates; and lack of conclusive and large-scale RCTs (3, 23, 76, 77).

There is also a lack of safety data across the board for use of yoga in mental health conditions. A recent systematic review explored the incidence of yoga-associated injuries and adverse events in international observational yoga studies (N=9,129 yoga practitioners; N=9,903 nonyoga practitioners). The lifetime incidence of adverse events during a yoga class was 22.7%, with a 12-month prevalence of 4.6%. The most common adverse events were musculoskeletal, most commonly sprains or strains and mostly of a mild or transient nature. No precautions were recommended for healthy individuals, although the authors state that studies of YBP may have lower rates of adverse events because they are highly supervised (78). This suggests that the inclusion of a qualified yoga instructor as part of a treatment plan has advantages over self-study with no supervision. A meta-analysis of RCTs of yoga found that most do not report safety data; in those studies reporting safety data, 2.2% of practitioners reported yoga-related adverse events (79). More safety data are warranted.

To date, there are no known RCTs comparing different styles of yoga against each other for different patient populations or symptoms. Providers are, therefore, without conclusive evidence regarding what form of yoga to recommend for different conditions and symptoms. As such, the optimal choice of modality or discipline of yoga cannot yet be inferred from RCTs. Clinicians will need to rely on a patient’s self-report or intuitive sense of the helpfulness or efficaciousness of a certain modality. Of course, clinician instincts and experience may be useful as well, as may discussion with yoga teachers.

The swath of yoga practices and disciplines is wide; thus, there is variability in the literature of studied forms of yoga. As with all behavioral interventions, individual differences between instructors, even with a manualized protocol, must be considered. It is difficult to standardize or know exactly what components are in each class or intervention (19). The state of yoga teacher licensure is another methodological complication of yoga research, because yoga certification is not regulated by the government. There is no consensus on standardized yoga protocols (76), although one has been detailed and published as effective for use in MDD (34). More details of YBTs would enrich future reports of rigorous trials.

There is mounting evidence that yoga is helpful for a variety of conditions and symptoms that are directly or indirectly related to mental health concerns (i.e., depression and anxiety) (3, 80), and there are no clear contraindications for the use of yoga as an adjunctive treatment. For patients who do not like to take antidepressant medications because of side effects or personal beliefs, yoga may offer an alternative. Despite limited safety data on yoga overall, and specifically across disorders/symptoms (79), it is likely that yoga has fewer side effects compared with some psychiatric medications that carry a heavy side-effect burden.

There are some barriers to the use of yoga in mental health conditions. Yoga requires a significant amount of motivation and effort, not only to attend class but also to complete a class and to maintain frequent enough practice to maximize yoga's effect on mental health. A severely depressed patient may first need to achieve at least partial symptomatic relief prior to initiating yoga practice. To maintain benefits, yoga likely requires frequent and consistent practice, which requires persistence and discipline. Postures are not easy in the beginning and can trigger frustration and self-criticism. A mental health practitioner could work through the anticipated responses to the initiation of a yoga practice, such as comparison of self to others and responding to harsh internal criticism. It may be useful to remind patients of mindfulness and noticing, or “tagging,” the self-critical thought and returning the attention back to the breath, muscles being worked, or instructor’s directions. Additionally, a clinician may want to use a standardized clinical measurement at each visit to track the patient’s progress in yoga practice to give validity to the progress being made and increase awareness around the gains that the patient has worked hard to achieve.

As it stands, patients tend to start yogic practices independently of their prescribed health care treatment plan. As a medical provider, recommending that patients alert yoga instructors to any injuries or specific needs before initiating a practice should be encouraged and may help patients feel validated in bringing their concerns to their yoga instructor. Patients could exacerbate old injuries by attempting to perform each posture without considering physical limitations, especially since a new practitioner may not know what kind of accommodations are possible or necessary. Although there are still significant gaps in the literature, it is hoped that further research will allow health care providers to prescribe dosages and modalities of yoga, modify the regimen as treatment progresses, and systematically track improvement. Yoga should not be considered a first-line treatment for any psychiatric condition at this time. Following the standard of care, especially for more acute conditions, is recommended. For example, psychotic disorders, bipolar disorder, and severe depression such as that with prominent suicidality need immediate and evidence-based approaches. That said, YBTs can likely be considered as an adjunctive intervention when designing a comprehensive treatment plan.