Yoga Research and Public Health

Abstract

Research on yoga is witnessing an unprecedented proliferation currently, partly because of great interest in yoga’s health utility. However, yoga research does not seem to be sufficiently public health oriented, or its quality corresponding to its quantity. Yoga research is falling short to enable key stakeholders like end users, prescribers, and payers to meaningfully, confidently, and fruitfully answer the questions like: Is it generalizable? Is it standardizable? Which yoga style should be used/recommended/paid for? Or will it be worth the money? Therefore, it is important to examine the alignment to purpose or value of yoga research from a public health point of view so as to make it more practical. The issues such as lack of clear definition of yoga, wide variation in its dosage, cacophony of lineage-based styles, no data about comparative effectiveness between the yoga components, confounders and biases clouding the evidence regarding its benefits, too little data on long-term adherence, equivocal results about its cost effectiveness, discussions lacking embrace of better methods in research, and absence of a theory of yoga are examined. This is not a detailed discussion of every issue yoga research faces, but a high-level overview of those that have direct practical bearing. In the end, a few pragmatic approaches are offered. The article suggests that yoga-component analysis, development of a theory of yoga, adoption of a health-aligned functional typology of yoga, development and testing of a simple universal basic prototype of yoga intervention, emphasis on research about long-term adherence, and discouragement for mere proof of concept research might make yoga research serve the stakeholders better. It urges the research community to practice “context cognizant scholarship” to disentangle health compatible yoga from its historical-cultural-social body before examining it for health or medical application.

Background

There is an unprecedented proliferation in research publications on yoga.^1,2 Probably, the trend of its medicalization is driving this surge. ³ While it is encouraging that medicalized yoga is being subjected to increasing scientific scrutiny, yoga research does not seem to be sufficiently public health oriented. Furthermore, its quality does not correspond to its quantity. A search for “yoga” AND “public health,” in the article title in PubMed, returns mere 2 citations, and a search for “yoga” AND “health economics” returns zero articles. This paucity provides a sense about the importance scholarship accords to these topics. Likewise, highlighting the quality-quantity chasm, Sherman laments that the current efficacy studies serve only as proofs of concept, ⁴ Cook-Cottone complains about the lack of understanding on therapeutic dosage, ⁵ Fischer-White and Taylor admonish that until a structured, uniform approach to yoga research exists, research results will have minimal impact, ³ Elwy et al ² report that the methodological quality of researches is low, ² Park et al ¹ observe that the literature is replete with contradictions and is inconclusive, and the Canadian Agency for Drugs and Technologies in Health ⁶ concludes that there is insufficient evidence to comment on the safety and long-term efficacy of yoga. These observations stand out, given the fact that modern research on yoga began 96 years ago. ⁷

End users of yoga, prescriber providers, and insurer payers are the 3 key stakeholders whose interest in yoga research is practical. They seek research to meaningfully, confidently, and fruitfully answer the following questions about medicalized yoga: (1) Is it useful for health promotion? (2) Is it useful for disease mitigation? (3) Can there be harm? (4) Is it generalizable? (5) Is it standardizable? (6) Which yoga style should be used/recommended/paid for? (7) Can it be clearly identified? (8) Will it be worth the money spent? and, (9) Will it work reliably? It seems that except for the first 3, yoga research is not very helpful with the rest of the questions.

Therefore, it is important and necessary to subject yoga research to critical scrutiny. There is no dearth of literature that discusses challenges related to methodology and methods; however, publications that discuss meaningfulness, alignment to purpose, or value are very few.^3,8,9 This article is not to say that yoga research is devoid of utility or quality, but it is a commentary on its practicability. With a focus on public health issues and practical application of medicalized yoga in the western societies, this article is an attempt to fill the gap. At the end of the article, suggestions are made for remedial approaches.

Issues in Yoga Research

Definition of Yoga

Yoga is defined as a mind-body practice composed of physical postures, breathing techniques, and meditation.^10,11 One would then expect that to call a practice as yoga, all the 3 components should be present in it. However, one study examines meditation-only (karma) yoga, with no components of breathing or postures ¹² ; another examines breathing exercises–only (pranayama) yoga, with no components of postures or meditation ¹³ ; another examines postures-only (Hat.ha) yoga, with no components of breathing or meditation ¹⁴ ; and still another studies laughter-only (laughter) yoga, with no components of breathing, postures or meditation. ¹⁵ Furthermore, all these 4 disparate scientifically well-designed studies draw a common generalized conclusion that “yoga works.” Such reports may potentially bewilder and confuse stakeholders. ¹¹

Dose

Usefulness of yoga depends on its dose. Dose may include, exposure during intervention, home practice, information gathering outside of intervention, and cumulative effect of prior exposures. Currently no data exist that capture the latter 2. A few studies offer diffuse glimpses into home practice component^16,17; however, a review found that 72% of the studies do not report on the duration or details about the home practice. ² On the magnitude of dosage, only the duration of sessions (60-90 minutes) provides data that are practically useful. ⁴ The frequency of sessions varies from once a week to 6 times a week and the total intervention durations vary from 1 day to 2 years, rendering these data impracticable.^2,5,11 Saper et al ¹⁷ find that once-weekly or twice-weekly yoga classes have practically no difference in efficacy. Overall dose data are too incoherent for guiding users.

Styles and Details of the Components

One bibliometric analysis of yoga reports that “more than 40 different yoga styles were used in their analyzed randomized controlled trials (RCTs)” and comments that with such diversity, comparison is difficult. ¹¹ A systematic scoping review states that 54% of the studies on yoga do not provide details about the components within the style. ² A study comparing styles says that there is little correlation between styles and positive conclusions. ¹⁸ A user cannot decide about the style based on these findings.

Relative Contribution of Components

Measurement of relative contribution of components (eg, postures, breathing exercises, meditation) is desired but is missing in the current literature.^11,19 Studies have shown that meditation has health benefits,^20-22 and studies have also shown that contribution of postures in maintaining health and fitness is small to insufficient.^23-26 However, with one exception, ²⁷ literature does not document research that compares and contrasts the utilities of postures, breathing exercises, and mediation in health promotion or disease management. These findings showcase vividly the research practice divide.

Confounders

A systematic review of yoga practices notes that most studies do not control for potential confounds and therefore do not inform much about causality. ²⁸ Similarly, an article comments that yoga effects often become insignificant when factors like education or income are taken into consideration. ²⁹ With reference to complementary medicine, Greco et al ³⁰ provide a detailed treatment of nonspecific factors in healing such as practitioner attitudes, outcome expectancies, and so on, which are scantily addressed by current yoga research. Confounders make claims about yoga efficacy dubious.

Adherence

For chronic diseases, long-term adherence to yoga is crucial. ²⁸ There are a very few studies that have examined adherence to yoga, and that too, over no longer than 6 months.^31-35 Currently, we do not know if yoga can be successfully applied for chronic conditions in the real world.

Health Economics

Fischer-White and Taylor ³ write, “In the current health care climate, every dollar spent on research is scrutinized. Thus, yoga/yoga therapy research that is not . . . cost savings will be replaced by . . .” Studies about cost-effectiveness of yoga are conflicting. Two trials for low back pain showed that there was potential cost-effectiveness based on quality-adjusted life years.^36,37 However, a trial for maintaining vitality in older workers showed that yoga was not cost-effective. ³⁸ No health modality is worth much if it fails to convincingly pass the litmus test of health economics.

Biases

Yoga research is inundated with biases. Cramer et al ³⁹ suggest that the yoga RCTs are highly vulnerable to selection bias. Smith and Pukall ⁴⁰ and Keshavan et al ⁴¹ allude to file-drawer problem in passing. Negative results are meagerly reported^42-44; even a popular article discusses about the publication bias in yoga research. ⁴⁵ A yoga “self-efficacy scale” ⁴⁶ or a “yoga questionnaire” ⁴⁷ suggests a possibility of ethnicity and gender biases in them making their universal applicability questionable.^11,41 Most yoga interventions and their outcomes measures are over short durations, which creates a time-term bias.^2,11 Nudging the issue of causality of these biases, it must be noted that these biases come at a cost.

Research Methodology and Methods

Despite a talk about the utility of whole system research,^48-50 complexity science, ⁵¹ Bayesian method, ⁵² or composite measures of healing, ⁵³ the actual use of these methods is little. Furthermore, due to methodological issues, sometimes results on the same outcome are contradictory.^26,54

Theory of Yoga

Despite numerous excellent publications pertaining to mechanism of action of yoga, a cogent theory of yoga stands out in yoga research by its absence.

How Can Yoga Research Be Made More Practical?

Medicalized yoga’s contribution in public health is probably subpar mainly because of its cacophony and fuzziness. Based on the issues discussed above, the following approaches in research might facilitate to bring coherence and order in the field of yoga intervention.

Conduct Component Analysis of Yoga

Currently yoga is not subjected to component analysis, which is a method, different and distinct from the statistical procedure “principal component analysis,” and entails a systematic assessment of 2 or more components that comprise a treatment package. ⁵⁵ A 1981 component analysis of yoga showed that when postures and breathing exercises were compared side by side, postures did not measure up for the beneficial effects. ²⁷ Comparative component analyses of yoga should be conducted on the highest priority.

Develop a Theory of Yoga

Around 1993, the idea of theory-based evaluation became popular in the field of evaluation research. ⁵⁶ In that stride, Sechrest and Figueredo ⁵⁷ wrote, “Under many circumstances, strong theory can compensate for relatively weak method.” Twenty years later, theory based evaluation might prove handy in solving many a problem in yoga research. A good theory of yoga can provide a logical blueprint and a robust skeletal framework for yoga research and can spare numerous wasteful dashes into blind alleys.

Adopt a Health-Oriented Typology Instead of Lineage-Based Yoga Taxonomy

Attempts have been made in anthropology to create typologies of yoga.^58,59 Though better than lineage-based styles of yoga, which anyway do not matter for results, ¹⁸ those typologies are still insufficient for medicalized yoga. This article proposes a functional typology with 3 domains: (1) behavioral, (2) physical, and (3) mental. Under the proposed typology, Hat.ha yoga would fit under physical domain, Karma yoga would fit under behavioral domain, and Jñāna yoga would fit under mental domain. Similarly, limbs such as Yama and Niyama would fit under behavioral domain, postures and breathing exercises would fit under physical domain, and meditative exercises would fit under mental domain. This typology will reduce confusion and bring a common uniform language in practice and research of yoga.

Build a Simple Universal Basic Prototype of Yoga Intervention With a Provision for Modular Add-ons

A systematic review of yoga found that 89% of RCTs reached positive conclusion. ³⁹ These results, in spite of huge variability in styles, dosage, and contents suggest that beneath diversity there is a simple core model of yoga. One can imagine such a strawman model, 30 minutes a session thrice a week, composed of a handful of easy postures and a comfortable plain deep breathing exercise that lead up to a meditation session that aims at reducing thought density. Researchers can start testing efficacy with this base model in various settings and conditions. Thereafter, they can add or subtract components and durations in modular blocks as per the situational demands, while simultaneously collecting and refining evidence. This approach might provide a better algorithm for standardizing yoga.

Conduct In-depth Research About Home Practice and Long-Term Adherence to Yoga

Home practice plays a significant role in the effectiveness of yoga^16,17,31,60 to a point that it might be more important than intervention exposure. ⁴⁴ However, it is minimally accounted for in the yoga research.^2,5 Therefore this attribute should be examined in depth. Additionally, adherence to yoga is generally low and only measured over short durations. More research is required to assess long-term adherence to yoga.

Reduce the “Proof of Concept” Research

With dwindling research funding for complementary therapies, proof of concept researches that abound in yoga research, should be strongly discouraged.

Although every research is supposed to serve the needs of its stakeholders, yoga research seems clearly not measuring up. Yoga brings with it a substantial sociopolitical baggage. The author believes that the predicament of yoga research has a lot to do with it. Researchers are either ignorant of the historical, cultural, or sociopolitical context of yoga or they are reluctant to disentangle health-compatible yoga from its current muddle. Pathbreaking works like The Structure of Scientific Revolutions ⁶¹ or the “Dartmouth studies” ⁶² admonish that science needs to reckon with such contextual barriers, lest it should be accused of, in the words of Mark Lipsey, malpractice. ⁶³ In tumultuous times such as ours, yoga can be a gift. It is not so much a matter of talent, tools, techniques, or methods-methodologies, as it is that of will. For medicalized yoga, anything less will not do.