Methodological issues in yoga therapy research among psychiatric patients

ABSTRACT

Yoga therapy has been widely accepted as part of the evidence-based modern healthcare system. Although research publications are growing in leaps and bounds, many methodological issues pose stumbling blocks. In this narrative review, various issues like standalone or add-on treatment, blinding, randomization, nature of dependent and intervening variables, duration of intervention, sustainability of effects, attrition bias, adherence and accuracy, all or none performance, diverse schools, heterogeneity and multidimensionality, assortment, permutations, and combinations of different components, neglect of essential ingredients, mindfulness, catch-22 situation, credentials of the instructors, cultural factors, naivety, multicentric studies, duration of collection of data, primary or standard treatment, interdisciplinary research, statistical lapses, qualitative research, biomedical research are discussed. There is a need to frame guidelines for conducting yoga therapy research and publication.

INTRODUCTION

Modern medical advancements justify integrating various conventional therapeutic systems like Ayurveda, Yoga, Naturopathy, and Siddha to support health, healing, and longevity. Yoga is a lifestyle modifier that outwardly involves physical postures, breathing exercises, and meditation techniques with mindfulness to promote the balance between the mind and the body. Yoga has now been extensively adopted as a method of relaxation and exercise to reduce stress and promote health and feelings of well-being. In this connection, the Indian Psychiatric Society, the national body of psychiatrists of India, released clinical practice guidelines for yoga and other alternative therapies for patients with mental disorders.[1] Several mental illnesses affect people for long periods or even a lifetime and are treated primarily by psychotropic medications. However, these medications are not always perfectly efficient. Some research suggest that yoga as an add-on treatment could be beneficial and help improve the quality of life of people with psychiatric illnesses.[2] Various methodological issues in yoga research of the past were analyzed, of the future were anticipated and discussed earlier.[3] The need of the hour is a symbiotic relationship between yoga and modern science. The link is “evidence-based data,” but not anecdotal adages or quotes from mythological texts. There were reviews on the efficacy of yoga in psychiatric patients as a package of care versus standard care and also reviews on nonstandard care for schizophrenia.[4] As an example of a general situation, while quoting the methodological difficulties, the authors expressed that no reliable inferences can be drawn concerning the effectiveness of yoga as a treatment modality for epilepsy.[5] In addition, the quality of the evidence to confirm outcomes is limited and of low quality. This article examines some of the methodological issues associated with yoga research with a special reference to psychiatric conditions.

STANDALONE OR ADD-ON TREATMENT

Standalone treatments have the advantage of one-to-one comparison with the other standard intervention or placebo. However, yoga has not been tried as a standalone in major psychiatric illnesses like schizophrenia, perhaps for the very ethical reason of depriving a group of patients of standard proven treatment. Hence, yoga techniques are used as add-on treatments for major psychiatric illnesses like schizophrenia.[2]

However, various yoga techniques are tried as standalone treatments for minor psychiatric illnesses like generalized anxiety disorder. There were many cross-over, placebo-controlled, and comparative studies with standard psychological interventions.[2]

DOUBLE BLIND

The studies are expected to be double-blind to get authenticity. A double-blind study uses a setup where neither the therapist nor the subject knows who receives a particular intervention. This method is effective because it prevents bias. It is used most often when there is a direct need to understand the benefits of an intervention against the placebo effect.

It is easy to blind both the patient and the doctor by masking the actual drug and placebo as far as medications are concerned. It is not so with yoga interventions.[6] As yoga is well known, patients can realize which is yoga and which is not. During evaluations, the patient may inadvertently mention this. Failure to blind the therapist or the person evaluating results in expectation bias.

RANDOMIZATION

After applying inclusion and exclusion criteria, participants (patients) are assigned to different groups. All have an equal chance of getting any particular assignment, and all are assumed to be equally competent to undertake any intervention. This holds true and is possible for drug trials. However, in the trials with yoga, all may not be equally competent to undertake all the interventions. A person with osteoarthritis of the knees may not be able to carry out Suryanamaskara or some asanas and may have to be slid to another group. Some persons may have religious objections to chanting sounds like Pranava or Omkara.

In our recently completed doctoral research study,[7] the group allocation was planned based on the day of the week the subject attended the hospital for follow-up: Monday Suryanamaskara group, Tuesday Asanas group, Wednesday Pranayama group, Thursday Omkara Japa group, Friday Relaxation Yoga, and Saturday Energiser group. This may be considered quasi-randomization, as is done in earlier studies.[8] This procedure is akin to block randomization or cluster randomization. Any person who has reservations about any intervention is slid into the next group until he settles down comfortably. According to Patanjali, the basis of any asana is the posture of sukham and sthiram (comfort and stability). The same thing naturally applies even to other yogic interventions.

The selection of omni-fit individuals who can be assigned to any group without hassles is a utopia. Hundreds have to be screened to obtain dozens. Hence, the perfection for randomization has to be compromised with the “allocation of the fit” criteria. Randomization by strictly following a random sequence generator is a challenge with yoga. An occasional mention of randomization by the roll of a die or through a computerized randomization table leaves one skeptical about the genuineness for the reasons mentioned above.

DEPENDENT VARIABLE

Dependent variables are selected from some of the measurable attributes. If the negative symptoms are assessed through a scale like PANSS, the potential to respond positively to any intervention or the resistance against the intervention varies with other attributes like the duration of illness or premorbid personality. Lack of consistency in operational definitions is a major flaw.[9]

Key outcomes vary, including mental status, physical health, global state, wellness index, quality of life, social functioning, adverse effects, and care costs. In meta-analysis studies, it is not possible to compare different outcomes. If all outcomes are divided into short term (less than 6 months), medium term (seven to 12 months), and long term (over 1 year), most of the yoga studies confine to short-term outcomes.[10] The long-term effects of yoga are infrequently studied, although needed.

INTERVENING VARIABLES

There are many issues in the environment that influence the course of the illness. Stress-provoking life events and expressed emotions by significant others can influence the outcome of psychiatric illnesses. Diet, exercise, and lifestyle are a few factors to mention that have a substantial role in many illnesses. Yoga studies lack cognizance of these factors.

DURATION OF INTERVENTION

When it is concerned with a drug trial, the optimal dosage, whether a new drug or the gold standard, is used, and the help of assumed dosage equivalents is taken. However, in the case of yoga intervention, the dosage, in the form of intensity, duration of each session, and total duration and frequency vary.[11]

Frequencies ranging from daily yoga interventions to once or twice a week are seen. The length of each session may range from 15 min to 1 h. The total duration of intervention may range from 1 month to 6 months. Judging the efficacy with no uniform “dosage” has its limitations.[12]

SUSTAINABILITY OF EFFECTS

The studies usually stop short of evaluating the effects of yoga interventions at the end of the study period. Whether these interventions induce long-lasting effects even after discontinuation is a valid question. Yoga is a lifestyle. Some participants continue the practice, and some may not, but their attitude toward life may change. Long-term effects are poorly studied.

CONTROLS

A control should ideally simulate the intervention to maintain the “blinding” aspect of the participant. Different types of exercises are usually chosen as controls, and in some other studies, talk therapies are tried as controls. Higher the resemblance, the more effective the control is. It is wise to use a dismantling strategy where a comparison intervention is developed containing other ingredients as the original experimental intervention, with one key attribute eliminated.[13] Unlike the drug trials, where the placebo can be a look-alike replica, it is not easy to have perfect placebo controls for yoga.

The conventional controls are treatment as usual, waitlist, active other treatment, and dismantling or add-on control. The most common types of comparison groups are the usual care or waitlist controls, where no changes are made to the usual activities of the participants in those groups.[14]

In almost all of these studies, equivalent time and attention were not provided to the control group. Waitlist controls are akin to no-intervention controls, and the very feeling of no intervention deprives the faith-healing aspect of the “intervention.”

For cross-over studies, shifting from one intervention to the other will always have the risk of the carryover effect. Yoga, being a lifestyle modifier, is claimed to carry its beneficial effects even after the cessation of the formal intervention. If the initial intervention is yoga for a group and the cross-over to some other intervention like aerobic exercise can carry the spillover benefit to the other intervention resulting in erroneous interpretation.

ATTRITION BIAS

In any research, drop-outs are a problem, and yoga research is no exemption. However, the reasons for dropping out are not studied, though it is an essential part of the data for further refining the research or improving the intervention’s acceptability. The study of remaining cases alone and the generalization of the findings is a fallacy of attrition bias.[15]

ADHERENCE AND ACCURACY

Some of the studies are conducted in a supervised environment. However, in most cases, the initial sessions are conducted by the trainer, but the later sessions are done at home. It is only an assumption that they have adhered strictly to the instructions and followed the procedure to the dot.[16] Psychiatric patients, especially those with depression or chronic schizophrenia, are well known for their amotivation, and their compliance and accuracy of the activity are doubtful.

ALL OR NONE PERFORMANCE

The participants who could be assessed finally after their declaration of regularly undertaking the activity are considered subjects. Those who could not come for the final assessment or those who expressed their failure to comply are considered drop-outs. There are no intermediate grades or partial fulfillers. In our study mentioned earlier, which was of 3 months, duration-wise, among the subjects, only half of them did the practice regularly and accurately, but the rest of them did it for a lesser and variable period. All did not comply with fool-proof accuracy of performance. For instance, there are a series of five asanas in an intervention package. However, some practiced four or three asanas each day as they were not comfortable with the other asanas. Unless some provision exists for accommodating the partial compliers or performers, the results would be wrongly interpreted.

DIVERSE SCHOOLS OF YOGA

Although yoga is based on ancient scriptures like Vedas and Upanishads, or later works like Yoga Vasishtam, Bhagavad-Gita, or Patanjali Yoga Sutras, innumerable schools of yoga have developed during recent decades. To name a few: Acro, Aerial, Aqua, Anusara, Bikram, Integral, Iyengar, Jivamukti, Kripalu, Krishnamacharya, Kundalini, Laughter, Moksha, Power, Partner, Restorative, Sahaja, Silver Age, Sivananda, Sudarshan Kriya, Svaroopa, Tantric, Tibetan, Viniyoga, Vinyasa. The list seems endless. There are many shades, variations, and belief systems.[17] Orthodox yoga practitioners term the newer variations “Plastic yoga,” as these lack the real spirit of yoga but consist of only mechanical exercises. The study of different modes of yoga as one uniform intervention is a fallacy.[18]

OCEAN OF YOGA: HETEROGENEITY AND MULTIDIMENSIONALITY

Yoga is such a gigantic entity that it can be compared to an ocean, but not a puddle. It encompasses a wealth of information dating back to the Vedic period. It has philosophical, metaphysical, psychological, and therapeutic perspectives. If yoga shastra is a set, yoga therapy is a subset.[19]

Imagine somebody trying to conduct an experiment where one of the interventions is termed psychotherapy, and the other is termed psychotropic medication. There are dozens of varieties of psychotherapies and hundreds of psychotropic substances. Similarly, yoga intervention is not a single entity; there are a variety of interventions, like hundreds of yoga asanas, pranayamas, kriyas, and meditation techniques. It is not correct to compare different studies with different yogic activities by placing all of them under the same umbrella. Systematic categorization and comparison of the different yoga practices deserve attention.

ASSORTMENT, PERMUTATIONS, AND COMBINATIONS

Most studies follow a protocol consisting of a combination of yoga procedures like a few asanas and one or two varieties of pranayamas and may include Suryanamaskara or a few kriyas like kapalabhathi.[20] Protocols vary from study to study with different combinations and permutations. The relative effects of different aspects of yoga interventions are neglected. Lack of uniformity in the protocols and calling everything yoga is a fallacy.

NEGLECT OF ESSENTIAL INGREDIENTS OF YOGA

The fundamental stepping stones for other standard yogic practices are Yama and Niyama. These are the basic requirements and are termed “Adhikara yoga,” the eligibility criteria.[21] As yoga is not just maintaining postures or breathing, yoga is a lifestyle, the five Yama principles denote how the individual is expected to behave for societal good, and the five Niyama principles denote how he should conduct himself for his upliftment.

MINDFULNESS

It is the key concept in yoga practice. There cannot be any difference between yoga asana and physical exercise without mindfulness. The issues like the nature of controls and comparison conditions for research that includes mindfulness, its adequate description, related training and interventions, and the question of how mindfulness can be measured are uncertain.[22] It is not very clear how to evaluate participant compliance with mindfulness.[23]

CATCH-22 SITUATION

The most common symptom of psychiatric conditions, chronic schizophrenia, and depression is amotivation. Yoga interventions are sought to benefit the symptoms, including amotivation. Unless the participant is motivated, he will not cooperate with the intervention. So the very problem for which the intervention is meant would be a hindrance. A catch-22 situation!

CREDENTIALS OF THE INSTRUCTORS

There is no uniformity in the training standards of the yoga instructors. Some obtain their certificate, diploma, or degree from a university, and some get trained in a nonacademic setting under a traditional guru. However, the problem is self-proclaimed instructors with a meager few days of attending a crash course, obtaining an instructor placement, and getting involved even in research projects. This lack of uniformity in the quality of training would matter in the intervention quality and jeopardize the comparative evaluations.[11]

CULTURAL FACTORS

Studying and teaching only sterile yoga asanas in cultural context is another concern. Misusing or reconstructing Sanskrit terms inaccurately is another issue. The natural fragrance of original yoga is lost, leaving “plastic yoga” behind. Some of the research gets stuck in such degenerated variety.[24]

NAIVETY REQUIREMENT

This is one of the usual exclusion criteria, and it is most reasonable to have clean-slate participants. However, nowadays, it is hard to find people who have not been exposed to some sort of yoga activity at one time or another. It is not easy to draw the line: between how recent the exposure was, and the nature of yoga activity, asanas or pranayama, or Suryanamaskara or meditation.

MULTICENTRIC STUDIES

Some studies are conducted at multiple centers, and among them, some in the same country and others in different countries. Multiple factors vary, including the setting, instructors, participants, and their relations. Yoga is an intervention requiring more subjective involvement, unlike many mechanical interventions like surgery or exercise. Subtle differences in the centers affect the results of the study.

CHANGES IN PRIMARY OR STANDARD TREATMENT

Yoga is generally used as an adjunct treatment. The assumption is that the primary treatment remains stationary throughout the study period, so it would not be a hurdle as an intervening variable. In a psychiatric illness like schizophrenia, in some cases, adjustment of antipsychotic drug dosage may be required. In cases of depression, apart from an antidepressant, an anxiolytic or an augmenting agent may be required in some cases in the middle of the study period. In illnesses like diabetes mellitus, adjustment of the doses may be required depending upon the glycemic control. It is required to maintain stationary not only the primary drug and its dosage but also other factors like diet and exercise. For meeting this criterion, many cases are required to be dropped out.

INTERDISCIPLINARY RESEARCH

Yoga is not a regular department in modern healthcare establishments. Yoga education and research institutes usually maintain tie-up with other medical disciplines located elsewhere. Yoga therapy researchers depend upon consultants from various departments. Although they start with an ambitious randomized sampling design, they end up with convenient sampling, as the case allocation mostly depends on the courtesy and convenience of the referring consultant.

STATISTICAL LAPSES

Meta-analysis reviews have pointed out certain statistical errors in yoga studies. Some of them are using parametric tests for nonparametric data, distortion of results by poor handling of skewed data, and dealing with continuous data as dichotomous data.[25]

NEED FOR QUALITATIVE RESEARCH

Research always seeks evidence, and the facts must be confirmed by experimentation and objective proof. Yoga envisages subjective well-being and ultimate Ananda or bliss. Very few studies focus on the subjective aspect.[26] Lack of objectivity is a drawback, and the study of the real benefit of yoga is an advantage with these qualitative parameters as outcome.

YOGA AND BIOMEDICAL RESEARCH

Finding biomedical markers to link with improvement is a breakthrough in yoga research. Hormonal assays like cortisol, oxytocin, and prolactin levels, fMRI-BOLD signals, P300waves, BDNF, telomerase activity, and levels of cytokines are a few to mention studied in response to yoga practice.[27] This is an attractive move for the scientific community to accept or reject authenticity. However, such sophistication would not be available to all yoga researchers except in advanced centers.

CONCLUSION

In conclusion, yoga therapy interventions pose unique difficulties in fitting into the evidence-based scientific rigor of modern medicine. Psychiatric patients present additional challenges as yoga interventions require active participation and mindfulness. Some challenges are associated with the very inherent nature of the yoga field, and some with the misconceptions and inappropriate practices of the therapists or researchers. In addition, there exist some hardships connected with research concerns like blinding and randomization. There is a need to refine existing guidelines and protocols for yoga research to promote uniformity and standardization. There are already some Delphi-based efforts on the way to developing guidelines for reporting yoga research.[28]

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.