The Safety of Yoga

By Traci Pantuso, ND

Adjunct Faculty, Bastyr University, Seattle; Owner Naturopathic Doctor, Harbor Integrative Medicine, Bellingham, WA

Dr. Pantuso reports no financial relationships relevant to this field of study.

To investigate the safety of yoga, the authors performed a meta-analysis and searched the following four electronic databases for trials assessing yoga: MEDLINE/Pubmed, Scopus, Cochrane Library and IndMED. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and the recommendations from the Cochrane Collaboration were followed. Studies were considered eligible if they were randomized, controlled trials (RCTs) and if adverse events were assessed. Participants of these studies included both healthy patients and individuals with a reported medical condition. Studies that compared yoga with no treatment, usual care, or any active treatment were eligible. No restrictions were made regarding yoga tradition, length, frequency, or duration of the trials. An adverse event was defined as any undesirable experience during the course of the study, and adverse events were classified into three groups 1) intervention-related adverse events, which were determined by the authors of the original studies as adverse events that most likely occurred as a result of the intervention, 2) non-serious adverse events, and 3) serious adverse events.

A total of 2520 articles were retrieved, with many of these articles being duplicates. After removing the duplicate studies, a total of 1531 studies were screened with 1041 of these studies excluded because they were not RCTs and/or did not include yoga interventions. Of the 490 full-text articles, 135 were excluded due to multiple reasons, including: not being full articles, non-randomized trial design, did not include a yoga intervention, or did not include a non-yoga control group. A total of 301 RCTs were screened for adverse events. Of those, 94 studies were represented in 111 full-text articles that reported data on at least one type of adverse event were included.

A total of 8430 participants were included in the 94 studies. Fifteen studies (n = 1316) included healthy participants and 77 studies (n = 7114) included individuals with a variety of physical and mental health conditions. Forty-three studies (n = 3514) were conducted in North America. Of the studies that reported gender and ethnicity, the participants were 75% female and 83% Caucasian. Thirty-two of the studies did not define the specific style of yoga used, while 19 of the studies used the Iyengar yoga technique and eight used Hatha (see Table 1). Fifty-three of the studies (n = 2221) compared yoga to usual care/no treatment, 26 of the studies (n = 1445) compared yoga to an exercise intervention, and 22 studies compared yoga to psychological or educational interventions (see Table 2).

The authors found no differences in the frequency of intervention-related adverse events, of non-serious or serious adverse events, or of dropouts due to adverse events found when comparing yoga with usual care/no treatment or with exercise.

Yoga compared to psychological/educational interventions did demonstrate increased intervention-related adverse events (odds ratio [OR], 4.21; 95% confidence interval [CI], 1.01-17.67; P = 0.05) and more non-serious adverse events (OR, 7.30; 95% CI, 1.91-27.92; P < 0.01) (see Tables 3 and 4). Overall, in the yoga groups there was a low frequency of adverse events with an incidence of 2.2% for intervention-related adverse events — 10.9% for non-serious and 0.6% for serious adverse events.

COMMENTARY

Yoga is one of the most popular types of meditative movements in the United States. Its popularity increased from 5.1% of the U.S. population regularly practicing yoga in 2002 to 9.5% in 2012, so assessing the overall safety of yoga is important.¹ The results of this meta-analysis confirm the results of a previous study based on data from the National Health Interview Survey (NHIS), which reported < 1% of adults who participated in yoga reported yoga-related adverse events leading to discontinued yoga use.^1,2

Overall, the results from this study demonstrate that yoga is relatively safe with a low frequency of adverse events and there were no differences in the frequency of adverse events when compared to a usual care/no treatment or exercise intervention.

This study has a number of strengths, including the meta-analysis study design, which allowed 94 trials and 8430 participants to be included. In addition, the studies that were selected had control groups and the risk of bias was low among studies.

The main limitation of this study is that many of the trials failed to report adverse events; 69.7% of the studies did not report quantifiable safety data. The types of yoga interventions that were reported were broad and included a variety of different yoga styles. A large percentage of the included studies did not mention what type of yoga was being used. Another limitation of this study is the lack of participant characteristic data reported in the trials. From the data that were reported, the participant population is 75% female and 83% Caucasian; however, 57 studies did not report ethnicity. The mean age was 48.8 years, but the age range was large and seven studies did not report age information. Such missing data limits the generalizability of the results of this study. Because this study explored the overall safety of yoga, there was no analysis of the health conditions for which the yoga intervention was being investigated. The majority of the participants (n = 7114) included in this meta-analysis reported a variety of physical and mental health conditions not evaluated in this study. Due to the lack of participant characteristic data, extrapolating these results to individuals with a specific health condition and a particular type of yoga practice is difficult.

Some follow-up to the results of this meta-analysis would be helpful to aid clinicians interested in incorporating these findings into their practices. For example, there are many types of yoga, which vary from seated breathing exercises with light stretching to more physically intensive types of yoga such as Vinyasa or hot yoga, but the general results of this safety analysis don't make a distinction based on yoga type. Which type(s) of yoga are the most safe or most unsafe, and for whom? It is difficult to answer these questions based on these results. In addition, we are left looking for an explanation for the increased intervention-associated adverse events in the yoga group compared to the educational/psychological intervention group. Is it simply that yoga adds a physical component, adding to risk, or are there other factors involved? This positive finding deserved more of an explanation as perhaps it hints at a mechanism of action or relevancy for certain patient populations. Until such unknowns are clarified, overall this meta-analysis seems to indicate that yoga is fairly safe for most people. Of course, any recommendations of yoga for our patients should be individualized, weaving in some of the evidence for use as reviewed in past issues of Integrative Medicine Alert, while the actual practice of yoga should be guided by a certified instructor.

REFERENCES

1. Clarke TC, et al. Trends in the use of complementary health approaches among adults: United States, 2002–2012. National Health Statistics Reports; no 79. Hyattsville, MD: National Center for Health Statistics. 2015.
2. Holton MK, Barry AE. Do side-effects/injuries from yoga practice result in discontinued use? Results of a national survey. Int J Yoga 2014;7:152-154.