Yoga is an ancient mind-body practice which originated in India more than 2.000 years ago and is described systematically early on (Patanjali’s Yoga
According to United Nations [
The solution for dealing with stressors, anxiety, and learning disabilities certainly depends on many factors; however, evidence suggests that some of these problems may be eased by mind-body practices, which have been shown to redirect attention, improve concentration, increase self-control, and provide people with reliable and healthy coping mechanisms [
Studies utilizing yoga in school settings have been shown to benefit children and adolescents [
Therefore, the primary goal of this study was to systematically examine the available literature on yoga interventions exclusively in school settings. The objective of this report was to review methodological quality among selected studies, exploring the evidence of yoga-based interventions regarding academic, cognitive, and psychosocial benefits, and to contribute to the study of low-cost, health-focused alternative programs for children and adolescents in school settings.
Studies were identified by searching PubMed, PsycInfo, Embase, ISI, and the Cochrane Library. A wide search was conducted for studies published between 1980 and October 31, 2014, using the following terms or key words:
Peer-reviewed, published manuscripts were considered. Studies were selected if (1) they included a yoga or yoga-based intervention, (2) the intervention was restricted to school settings (integrated into the school schedule or after class), (3) they included children and adolescents (ages 5–18), (4) they included an evaluation of anxiety, depression, stress, or other psychological measures such as mood indicators, self-esteem, confidence, and quality of life at both preintervention and postintervention, (5) they included the assessment of academic or cognitive performance as a consequence of the yoga intervention (pre- and postintervention), (6) the research designs were pilot studies, quasi-experimental designs, or randomized designs and included control groups with no interventions or an active control (comparative intervention), and (7) they were written in English. Exclusion criteria comprised (1) studies that utilized only meditation or relaxation techniques without the physical components such as postures (
Authors screened abstracts to identify articles that meet inclusion criteria. Potential articles were then evaluated for inclusion. To conduct the study, all data was collected and evaluated in terms of selection criteria, procedure, participants, intervention, methodology, assessment tools, and outcomes. Follow-up and results were also assessed. To evaluate or discuss non-RCTs was not part of the review’s scope. Other reviews have mentioned non-RCTs, including methodology and possible bias [
The quality and reliability of the randomized control trials (RCTs) were evaluated according to the evidence levels recommended by the Oxford Center for Evidence-Based Medicine [
Classification of scientific evidence in systematic literature reviews according to the evidence levels recommended by the Oxford Center for Evidence-Based Medicine.
Level of evidence | Grading criteria |
---|---|
1a | Systematic reviews of RCTs including meta-analysis. |
|
|
1b | Individual RCT with narrow confidence interval. |
|
|
1c | Case of series “all or nothing.” |
|
|
2a | Systematic review of cohort studies. |
|
|
2b | Individual cohort study and low quality RCT. |
|
|
2c | Outcome research study. |
|
|
3a | Systematic review of case control studies. |
|
|
3b | Individual case control study. |
|
|
4 | Case series, poor quality cohort, and case control studies. |
|
|
5 | Expert’s opinion. |
Adapted from levels of evidence of the Oxford Center for Evidence-Based Medicine [
After the first appraisal, another examiner (J. Simões) evaluated the selected RCTs and
For the effect size calculation, the means and standard deviations (postintervention) from experimental and control groups were obtained directly from the studies. The comparison between experimental conditions was carried out after the analysis of the pooled effect size by the generic inverse variance method (random effect model) through standardized mean difference and Hedges’ g. Hedges’ g effect is the difference between the two means (postintervention for the experimental and the control groups) divided by the pooled standard deviation. Ninety-five percent confidence intervals were computed for all variables.
Two studies [
Of the 48 studies identified, 9 randomized control trials met criteria for inclusion in this review, as illustrated by Figure
PRISMA 2009 checklist.
Section/topic | # | Checklist item | Reported on page # |
---|---|---|---|
|
|||
Title | 1 | Are There Benefits from Teaching Yoga at Schools? A Systematic Review of Randomized Control Trials of Yoga-Based Interventions. | 1 |
|
|||
|
|||
Structured summary | 2 |
|
1 |
|
|||
|
|||
Rationale | 3 | Yoga is a holistic system of varied mind-body practices that can be used to improve mental and physical health. Due to the well-known restorative effects of yoga on mental health, it has been utilized in a variety of contexts and situations beyond the standard therapy practice. Educators and schools in particular are looking to include yoga as a cost-effective, evidence-based component of urgently needed wellness programs for their students. However, there is no critically appraised evidence such as systematic reviews on potential benefits of yoga-based interventions in school settings. | 2 |
Objectives | 4 | The objective of this report was to review methodological quality among selected studies, exploring the evidence of yoga-based interventions regarding academic, cognitive, and psychosocial benefits, and to contribute to the study of low-cost, health-focused alternative programs for children and adolescents in school settings. | 2 |
|
|||
|
|||
Eligibility criteria | 5 | Peer-reviewed, published manuscripts were considered. Studies were selected if (1) they included a yoga or yoga-based intervention, (2) the intervention was restricted to school settings (integrated into the school schedule or after class), (3) they included children and adolescents (ages 5–18), (4) they included an evaluation of anxiety, depression, stress, or other psychological measures such as mood indicators, self-esteem, confidence, and quality of life at both preintervention and postintervention, (5) they included the assessment of academic or cognitive performance as a consequence of the yoga intervention (pre- and postintervention), (6) the research designs were pilot studies, quasi-experimental designs, or randomized designs and included control groups with no interventions or an active control (comparative intervention), and (7) they were written in English. Exclusion criteria comprised (1) studies that utilized only meditation or relaxation techniques without the physical components such as postures ( |
5 |
Information sources | 6 | Literature searches were conducted in PubMed, PsycInfo, Embase, ISI, and the Cochrane Library (1980–2014). An extensive search was conducted for studies published between 1980 and October 31, 2014, using the following terms or key words, |
5 |
Search | 7 | PubMed, EMBASE, and the Cochrane Central Register of Controlled Trials (CENTRAL) in the Cochrane Library | 5 |
Study selection | 8 | Four authors conducted the literature searching (C. Ferreira-Vorkapic, M. Marchioro, and S. Telles) and study selection (C. Ferreira-Vorkapic, M. Marchioro, and E. Kozasa). | 5 |
Data collection process | 9 | Authors (C. Ferreira-Vorkapic, M. Marchioro, S. Telles, and E. Kozasa) screened abstracts to identify articles that meet inclusion criteria. Potential articles were then evaluated for inclusion. To conduct the study, all data was collected and evaluated in terms of selection criteria, procedure, participants, intervention, methodology, assessment tools, and outcomes. Follow-up and results were also assessed. The review has been prepared using preferred reporting criteria for systematic review guidelines (PRISMA). J. M. Feitoza has performed all the statistics (effect size). For the effect size calculation, the means and standard deviations (postintervention) from experimental and control groups were obtained directly from the studies. | 5 |
Data items | 10 | Yoga, school, education, and children. | 5 |
Risk of bias in individual studies | 11 | After the first appraisal, another examiner (J. Simões) evaluated the selected RCTs and a |
6 |
Effect size calculation | 12 | The comparison between experimental conditions was carried out after the analysis of the pooled effect size by the generic inverse variance method (random effect model) through standardized mean difference and Hedges’ g. Ninety-five percent confidence intervals were computed for all variables. Two studies [ |
6 |
Summary measures | 13 | The quality and reliability of the randomized control trials (RCTs) were evaluated according to the evidence levels recommended by the Oxford Center for Evidence-Based Medicine (AHRQ, 2002). The items included study question, study population, randomization, blinding, interventions, outcomes, statistical analysis, results, discussion, and funding source. The quality of all the included trials was categorized into levels of evidence varying from 1 to 5, according to low, unclear, or high risk of bias. The quality and reliability of the randomized control trials (RCTs) were evaluated according to the evidence levels recommended by the Oxford Center for Evidence-Based Medicine. Table |
6 |
|
|||
|
|||
Study selection | 14 | A flow chart depicted the search process and study selection (Figure |
6 |
Study characteristics | 15 | The methodological quality of most included trials was generally reduced. The details are as shown in Table |
6 |
Risk of bias within studies | 16 | The number of trials was too small to conduct any sufficient additional analysis of publication bias. | 6 |
Effect size results | 17 | After an overall effect size calculation of each study (except for Ramadoss and Bose, 2010 [ |
7 |
Synthesis of results | 18 | Regarding the effects of yoga on psychological well-being, of the six studies, three of them support the benefits of yoga or yoga-based programs for children in school settings. Khalsa et al. [ |
7 |
|
|||
|
|||
Summary of evidence | 19 | Although most of the studies were classified as 2b, according to AHRQ evidence level criteria, which means low quality randomized control trials, this review shows beneficial effects of yoga-based interventions at school on both psychological and cognitive functions (effect size was found for mood indicators, anxiety and tension, self-esteem and memory), but the negative effects of yoga were also observed in some of the studies and might be explained to some extent by the adaptation process by children, the absence of attentional control, and the inadequacy of yoga practice for children. Future research requires greater standardization and must deal with the problem of appropriateness: what type of yoga-based intervention is most suitable for children and at what frequency and duration and observed variables. | 16 |
Limitations | 20 | Not only is the number of RCTs low, but also the trials were of reduced methodology quality and had risk of bias in terms of design, reporting, and methodology. It is comprehensible that it is difficult to perform double-blinding studies with yoga, but blinding to the outcome assessors and data analyzer could be feasible and has not been reported. One limitation of this review is that it was not possible to calculate the effect size of all variables observed in the selected studies due to their heterogeneity. | 16 |
Conclusions | 21 | This review analyzed nine peer-reviewed RCT studies, in which yoga was taught to children in a school setting. Outcome measures included psychological well-being and cognitive functions, such as attention and memory. Effect size was found for mood indicators, anxiety and tension (POMS), self-esteem, and memory. While supportive in many studies, the utility of yoga in educational settings is inconclusive due to the small number of randomized control trials in the literature. Even though only RCTs were reviewed, methodological and statistical problems might have contributed to the uncertainty: inadequate sample sizes, absence of control groups, variability in the type of yoga being taught, long duration of yoga sessions, inappropriate psychometric tools for children, and failure to measure intervening variables such as mindfulness and body awareness, which are important parts of yoga practice. This review suggests valuable effects of yoga-based interventions at school on both psychological status and cognitive function in some studies but future research requires greater standardization and must deal with the problem of appropriateness; what type of yoga-based intervention is most suitable for children, specifically in terms of the frequency and duration? | 22 |
|
|||
|
|||
Funding | 22 | This research was funded by the FAPITEC Agency under Process no. 7838.UNI321.21944.25062013. | 22 |
Adapted from [
Selected randomized control trials (RCTs).
Study | Sample | Program | Intervention | Variables | Evaluation tools | Results | Evidence level |
---|---|---|---|---|---|---|---|
Hagins et al., 2013 |
Middle school students, ages 11 to 12 |
Vinyasa Yoga |
Yoga or physical education (c) | Blood pressure (BP), heart rate (HR), and behavioral stressor tasks (mental arithmetic and Mirror Tracing Tasks). | Automated blood pressure cuff, Mental Arithmetic Task (MAT), and Mirror Tracing Task (MTT). | There were no significant differences between groups. | 2b (not double blinded, no follow-up) |
|
|||||||
Haden et al., 2014 |
Middle school students, ages 11 to 12 |
Ashtanga Yoga |
Yoga or physical education (c) | Emotional (affect and self-perceptions) and behavioral variables (internalizing and externalizing problems and aggression). | PANAS, Child Behavior Checklist (CBCL), Revised Parent Rating Scale for Reactive and Proactive Aggression (R-PRA), and the Self-Perception Profile for Children (SPPC). | There were no significant changes between groups in self-reported positive affect, global self-worth, aggression indices, or parent reports of their children’s externalizing and internalizing problems. However, negative affect increased for those children participating in yoga when compared to the PE program. | 2b (insufficient blinding, no follow-up) |
|
|||||||
Khalsa et al., 2012 [ |
High school students, ages 12 to 13 |
Yoga Ed |
Yoga or physical education (c) | Mood, anxiety, perceived stress, resilience, and other mental health variables. | The Self-Report of Personality (SRP) version of the Behavior Assessment Survey for Children Version 2 (BASC-2), POMS, the Resilience Scale (RS), and the Perceived Stress Scale (PSS). | Yoga participants showed statistically significant differences over time relative to controls on measures of anger control and fatigue/inertia. Most outcome measures exhibited a pattern of worsening in the control group over time, whereas changes in the yoga group over time were either minimal or showed slight improvements. | 2b (not double blinded, no follow-up) |
|
|||||||
Noggle et al., 2012 [ |
High school students, ages 16 to 17 |
Kripalu Yoga |
Yoga or physical education (c) | Psychological well-being, mood and negative affect (primary measures), |
Primary outcomes: Profile of Mood States-Short Form and Positive and Negative Affect Schedule for Children. |
Mood improved in yoga and worsened in controls. Negative affect worsened in controls and improved in yoga. | 2b (not double blinded, no follow-up) |
|
|||||||
Ramadoss and Bose, 2010 [ |
High school students, ages or grades not informed |
Niroga TLS |
Yoga once a week, yoga twice a week, yoga three times a week, yoga five times a week, or a waiting list group (c). | Stress and self-control. | Perceived Stress Scale (PSS-10) and Tangney’s Self-Control Scale (TSCS-13). | The intervention group demonstrated a slight decrease in stress and maintenance in self-control. In contrast, the control group, which received no classes, demonstrated no significant change in stress and a nonsignificant trend toward deterioration of self-control. | 2b (not double blinded, no follow-up) |
|
|||||||
White, 2012 |
Elementary school students, ages 9 to 11 (girls) |
The Mindful Awareness for Girls through Yoga |
Yoga or no intervention (c). | Stress, coping strategies, self-esteem, and mindfulness. | The Feel Bad Scale, the Schoolagers’ Coping Strategies Inventory, and the Healthy Self-Regulation Subscale of the Mindful Thinking and Action Scale for Adolescents. | No significant differences between groups were found. |
2b (not double blinded, no follow-up) |
|
|||||||
Sarokte and Rao, 2013 [ |
Middle and high school students, ages 10 to 16 |
Hatha Yoga |
Medhya Rasayana (Ayurveda treatment), Yoga, or no intervention (c) | Executive functions and mental status. | Short-term memory test, pictures, serial effects test, words, and mini mental state scale. | Group B showed highly significant and most effective changes in short-term memory test pictures and serial recall effects test using memory scope. Group C showed highly significant and most effective changes with respect to subjective and objective parameters in mini mental status scale. | 2b (not double blinded, no follow-up) |
|
|||||||
Telles et al., 2013 [ |
Elementary and middle school students, ages 8 to 13 |
Hatha Yoga |
Yoga or physical activity (c) | Self-esteem, attention, and physical fitness. | Stroop color-word task for children, Battle’s self-esteem inventory, and the teachers’ rating of the children’s obedience, academic performance, attention, punctuality, and behavior with friends and teachers. | Teachers’ rating of the children’s behavior. |
2b (not double blinded, no follow-up) |
|
|||||||
Verma et al., 2014 [ |
Middle and high school students, ages 11 to 15 |
Hatha Yoga |
Yoga and control groups | Cognitive functions. | Mental Ability Test and Battery and Memory Test. | Significant improvement was observed in measures of mental ability and memory in experimental group | 2b (not double blinded, no follow-up) |
Criteria for inclusion in the review.
Generally, the RCTs had low AHRQ evidence scores, with most studies receiving a score of 2b due to the lack of blindness and follow-up, reflecting the reduced quality of reporting in these studies.
All results and conditions of the studies are summarized in Table
Due to the limited number of RCTs and the great heterogeneity of the variables (ununiform constructs), the analysis of the effect size of specific measures could only be performed on studies that observed the same variables such as mood, tension, anxiety [
After an overall effect size calculation of each study (except for [
Plot of the general effect size in the selected studies.
Effect size from mood state indicators (POMS) was calculated from Khalsa et al., 2012 [
Profile of Mood States (POMS) general score effect size.
Profile of Mood States (POMS): subitems tension and anxiety effect size.
A third comparison was carried out for the variable self-esteem [
Self-esteem effect size.
Effect size for memory was also analyzed in [
Memory effect size.
Since the number of selected RCTs is reduced and other interesting results were observed during the review process (that could not provide for an effect size calculation), such as negative effects or different variables, the studies are individually described in the following sections in order to provide details, similarities, and differences between them.
Six RCT studies observed the effects of a yoga program on psychological well-being in school settings.
In [
Emotional and behavioral functioning was measured before and after intervention utilizing the following tools: Positive and Negative Affect Scales (PANAS), the Child Behavior Checklist (CBCL), the Revised Parent Rating Scale for Reactive and Proactive Aggression (R-PRA), and the Self-Perception Profile for Children (SPPC). Results show no significant changes between groups in self-reported positive affect, global self-worth, aggression indices, or parents’ reports of their children’s externalizing and internalizing problems. However, negative affect increased for those children participating in yoga when compared to the PE program. Authors offer a few explanations for these findings: (1) the first contact with yoga may be demanding for children at this age (sixth grade) and as a result may increase stress levels in the short term; (2) the “dose” and type of yoga may not have been satisfactory; (3) one of the outcomes of yoga practice may be greater self-awareness and mindfulness and these variables were not assessed in the current study; and (4) the sample size was small.
Khalsa et al., 2012 [
Outcome measures revealed that yoga participants showed statistically significant differences over time relative to controls on measures of anger control and fatigue/inertia. Most outcome measures exhibited a pattern of worsening in the control group over time, whereas changes in the yoga group over time either were minimal or showed slight improvements. Authors explain that while statistically significant differences between groups were found for only a few outcome measures, each of these favored the yoga group. On most measures, findings suggested relatively small positive effects in the yoga group but marked declines in the control group. A few study limitations such as inadequate psychometric tools and short duration of the program might be responsible for the observed outcomes.
Noggle et al. [
Ramadoss and Bose, 2010 [
White [
Authors reported no significant differences between groups. In addition, over time, the intervention group was more likely than the control group to report higher perceived stress scores (although no increase in the frequency of stressors was found) and greater frequency of coping with stress. As the authors state, as self-awareness progresses with yoga practice, children may become aware of difficult emotions. This may lead to increased perceived stress at first but subsequently may result in better means to cope with such feelings. Both groups reported significantly greater self-esteem and self-regulation. According to the authors, the negative outcomes regarding perceived stress might be due to (1) use of inadequate psychometric tools, (2) the fact that awareness of stress might have facilitated coping and that this increased awareness of stress might also have precipitated more stress, and (3) the fact that increase in stress might be transient and part of the process of becoming mindful. Study limitations include a homogeneous sample (only girls), the quality of intervention (due to a large sample size), and, especially, the fact that a greater part of the practice was done as homework and children were unattended during most of the time.
Lastly, in a 15-week study, Hagins et al. [
The authors stated that the two behavioral stressor tasks (Mental Arithmetic Task and Mirror Tracing Task) utilized in this investigation had been previously used successfully in studies of stress reactivity in children. The MAT consisted of simple arithmetic counts during a specific period of time. In the MTT, participants had to trace a star using only a mirror version of the star for guidance. Participants had to trace the star as many times as possible without any errors during three minutes. Systolic and diastolic BP and HR were obtained during the tests through an automated blood pressure cuff. The procedures and measures in pre- and posttests were the same. Pretesting occurred one to two weeks before the start of the intervention and posttesting occurred one to two weeks after the final class. After data analysis, authors concluded that the yoga program did not reduce stress reactivity compared to a physical education class. Furthermore, statistical analysis comparing the first stressors (MAT versus MTT) used during the pretest found that there were no significant differences in BP or HR values relative to the type of stressor used. Besides, the difficulty in finding differences in BP or HR in the participants might be due to their good health, especially after such a short time intervention. The authors concluded that the results do not support the idea that benefits from yoga are derived from a mechanism related to increased regulation of the autonomic nervous system. As in other studies, the results observed here may be related to the way yoga practice was applied and a failure to directly address the issue of reaction to perceived stress.
Three RCT studies observed the effects of a yoga program on different cognitive functions, such as attention, memory, and developmental abilities, in school settings.
A three-month study by Sarokte and Rao [
Telles et al. [
After testing for the differences between groups, social self-esteem was the only variable with significant changes, increasing in the physical exercise group. pre and post values within each group also showed significant changes in cognitive function. In the Stroop task, both groups showed an increase in word scores, color scores, and color-word scores. The physical exercise group showed reduced interference raw scores and an increase in interference scores. In addition, both groups showed an improvement in obedience, academic performance, attention, punctuality, behavior with friends, and behavior with teachers.
The authors suggest that the improved scores might be due to better aerobic fitness (observed in both groups). Also, the increase in interference T scores in the physical exercise group suggests reduced flexibility and ability to respond to the task demands after this intervention. Physical activity and yoga also separately improved emotional well-being in both groups, but the underlying mechanisms are not clear. The study has limitations such as the fact that the yoga and the physical exercise programs had to fit in the school schedule, which could have produced differences in outcomes. There was also no follow-up.
Finally, Verma et al., 2014 [
This review systematically examined the literature on yoga in school settings, exploring the evidence of yoga-based interventions on different psychological variables and cognitive functions.
Forty-eight peer-reviewed, published studies in which yoga was taught to school-aged children in a school setting were identified. Inclusion criteria included only randomized control studies (i.e., the control group had no intervention or an active control) in which yoga (and not just meditation) was taught and the effects on psychological well-being or cognitive functions were analyzed. After wide qualitative and quantitative synthesis, nine studies were selected.
Regarding the effects of yoga on psychological well-being, of the six studies, three of them support the benefits of yoga or yoga-based programs for children in school settings. Khalsa et al. [
In contrast, Haden et al., 2014 [
The analysis of the effect size for psychological well-being showed that an effect size was found for mood state indicators (POMS), demonstrating that the yoga group showed significant better scores in the postintervention condition. The same scale showed a second significant effect for the subitems tension and anxiety. The variable self-esteem also showed effect size and better results in the postintervention condition for the yoga group.
The evidence for the benefits of yoga in adults, whether healthy or suffering from mental disorders, is significant [
The negative effects of yoga observed in some of the studies here might be explained to some extent by (1) the adaptation process, (2) attentional control, and (3) inadequacy of yoga practice for children.
The practice of yoga requires effort and discipline. A child’s first contact with yoga is often demanding. When yoga is added to a child’s already existing academic and extracurricular activities, the child may experience higher levels of stress in the short term. According to Hayes and Feldman [
Yoga techniques such as breathing and meditation require attentional control, an executive function that is still not mature in children and adolescents. As the frontal lobes mature [
The duration of yoga practice observed in some of the studies might not be suitable for children due to their inability to control attention and reduced discipline. In studies where the yoga practice has been found beneficial for the students [
One of the outcomes of yoga practice may be greater self-awareness and mindfulness, a primary difference between yoga and standard physical education, and these variables were not assessed in any of the studies. Actually, physical education and yoga can be considered as complementary and it is therefore inappropriate to try and compare one as better than the other.
Although the number of RCT studies observing the effects of yoga on psychological and cognitive functions in school settings is very limited, the results seem promising. Effect size was found for mood indicators, tension and anxiety in the POMS scale, self-esteem, and memory.
This review identified three RCT studies that observed the effects of yoga-based interventions on different cognitive functions, such as attention, memory, and developmental abilities. Overall, participation in a yoga program was associated with improvements in subjective and objective parameters in mini mental status scale [
Lowered mood is associated with declines in cognitive function and, at least in adults, yoga has been reported to produce improvements in mood [
In addition, yoga practice has been shown to reduce anxiety based on reductions in psychological arousal [
Two of the studies reviewed here showed significant improvements in memory tasks after a few weeks of yoga-based interventions [
This review analyzed nine peer-reviewed RCT studies, in which yoga was taught to children in a school setting. Outcome measures included psychological well-being and cognitive functions, such as attention and memory. While supportive in some studies and different variables, the utility of yoga in educational settings is uncertain due to the small number of randomized control trials in the literature. Even though only RCTs were reviewed, methodological and statistical problems might have contributed to the uncertainty: inadequate sample sizes, absence of control groups, variability in the type of yoga being taught, long duration of yoga sessions, inappropriate psychometric tools for children, and failure to measure intervening variables such as mindfulness and body awareness, which are important parts of yoga practice. This review suggests important effects of yoga-based interventions at school on both psychological status and cognitive function in some studies, but future research requires greater standardization and must deal with the problem of appropriateness; what type of yoga-based intervention is most suitable for children, specifically in terms of the frequency and duration?
None of the authors have any conflict of interests regarding the publication of this paper.
This research was funded by the FAPITEC Agency under Process no. 7838.UNI321.21944.25062013. The authors would like to thank Andrew Jungkuntz for English revision.