Research Studies on Yoga for War-Related PTSD
by Lynn Stoller © 2019 SEYI. All Rights Reserved.
Following is a brief description of known published studies that evaluated the effectiveness of specific yoga protocols for addressing symptoms of war-related PTSD, organized by year of publication. Only studies that exclusively included active duty military personnel or veterans are listed. Key study limitations are highlighted and later discussed in the summary of this article in order to help inform future research in this area. Demographic data is not included in these brief summaries, but the reader should keep in mind that in the majority of studies male Caucasians are heavily represented and thus the results may not be generalizable to other demographic populations.
- Cushing, Braun, Alden, and Katz (2018) conducted a pilot study of 18 combat veterans from post 9/11 conflicts who had sub-threshold or diagnostic level PTSD. Veterans (median age= 43) were asked to participate in at least five of seven total weekly 60 minute Vinyasa-style, trauma-sensitive, military-informed yoga sessions, taught by a Warriors at Ease instructor. The Vinyasa protocol, advocated by Warriors at Ease and Meghan’s Foundation, included breath work, physical postures (with an emphasis on coordinating movements with the breath in a flowing yoga style), and a body scanning meditation. Following consent, the Post-traumatic Checklist, Military Version (PCL-M) was used to determine inclusion cut-off for PTSD symptoms. Several additional test batteries were used for the pre- and post-tests, including the Beck Anxiety Inventory (BAI), the Pittsburgh Sleep Quality Index (PSQI), the Patient Health Questionnaire (PHQ-8), and the Mindful Attention Awareness Scale (MAAS). Findings included significantly reduced symptoms in all three of the symptom clusters (hyperarousal, re-experiencing, and numbness/avoidance) of the PCL-M, as well as improved mindfulness, reduced depression, and reduced anxiety.
Study Limitations: This study was small and lacked a control group.
- Reinhardt and colleagues (2018) conducted a randomized controlled trial of Kripalu yoga with 51 military veterans in the Boston area, recruited both within and outside of the VA Healthcare System. This study was perhaps most notable for the large attrition rate, with just 10 veterans completing the 10-week yoga protocol despite 26 being initially randomized to the treatment group (10 withdrew and 6 were lost to follow-up). Those that participated in the yoga intervention did not report greater reductions in PTSD symptoms than those who were in the wait-list control group. The seven participants in the waitlist control group who went on to complete the yoga intervention demonstrated significant reductions PTSD symptoms as compared with their pre-test scores; however, the sample size was too small to reach any definitive conclusions. The researchers postulate that the dropout rate could have been caused by a) the time commitment required for the yoga intervention; b) increased anxiety or triggers due to “enhanced mind-body awareness or interoception”; and c) using a group yoga format.
Study Limitations: This study had a control group but it was not an active control plus the sample sizes were small due to a high attrition rate.
- Avery, Blasey, Rosen, and Bayley (2018) recruited nine veterans (7 males; 2 females) from an existing clinical yoga program and observed them over a 16-week period. Severity of PTSD symptoms and perceived stress were measured at baseline and at 4, 6, 8 and 16 weeks. Other psychological measures were administered at baseline and at week 6. Findings showed a significant reduction in self-reported symptoms of PTSD, while symptoms of stress did not show significant improvement. Psychological flexibility and set-shifting predicted changes in PTSD symptoms.
Study Limitations: This study was very small and did not have a control group.
- Bremner et al. (2017) conducted a PET study with 26 OEF/OIF combat veterans with PTSD who had recently returned from a combat zone. The participants were block randomized to receive eight sessions of MBSR (N=9) or present-centered group therapy (PCGT; N=8). All participants underwent PET brain imaging during exposure to neutral and Iraq combat-related slides and sound before and after treatment. Participants who received MBSR—but not PCGT—showed a) improvement in PTSD symptoms as measured with the CAPS that lasted for at least 6 months after treatment and b) an increase in mindfulness as measured with the Five Factor Mindfulness Questionnaire (FFMQ). MBSR did not produce any negative effects such as dissociation or flooding of traumatic memories. The PET study results showed increased anterior cingulate and inferior parietal lobule and decreased insula and precuneus function in the MBSR group (as compared with the PCGT group) in response to traumatic reminders. The researchers pointed out that these brain areas have been implicated in PTSD and are involved in extinction of fear responses to traumatic memories as well as stress response regulation.
Study Limitations: The sample size was small which limited the statistical analyses and prevented adjustment for factors such as differences in severity of PTSD symptoms.
- Schulz-Heik and colleagues (2017) conducted the first published study to compare the effects of a clinical yoga program for veterans delivered in-person versus through a telehealth system. The in-person classes were delivered at the War-Related Illness and Injury Study Center (WRIISC) located in the VA Palo Alto Health Care System and were often simultaneously broadcasted to students at outlying community-based outpatient clinics through the agency’s video-conferencing system. Of the 64 veterans who completed the survey study, more than 80% who had reported a problem with pain, energy level, depression, or anxiety at baseline reported improvement following participation in the program. No significant difference was found between the telehealth and in-person groups on any measure of satisfaction, overall improvement, or improvement in any of 16 specific health problems. In addition to the problem areas listed above, improvements were seen in the areas of irritability/anger management, concentration, feeling jumpy or easily startled, repeated disturbing memories, falling asleep, and staying asleep.
Study Limitations: There was no control group; participants were not randomly assigned to in-person or telehealth conditions; the evaluations were subjective ratings rather than validated clinical measures; and all telehealth classes were chair-based which may confound the comparison between the in-person vs. telehealth programs.
- Stephenson, Simpson, Martinez, & Kearney (2016) studied the association between changes in mindfulness and changes in PTSD symptoms among 113 veterans with chronic conditions who participated in MBSR at a large VA hospital. All screened positive for PTSD. To conduct the study, secondary analyses were performed with combined data from two published and two unpublished studies of MBSR. The series consisted of 8 weekly 2.5-hour group sessions and a daylong retreat, following the curriculum developed by Kabat-Zinn (1990). Classes included a body scan meditation, breathing meditation, mindful movement consisting of gentle yoga, walking meditation or Qi Gong, and loving-kindness meditation. Participants were also assigned mindfulness meditation homework using CDs (30-45 minutes per day, 6 days/week). Increases in Non-Reactivity and Acting with Awareness were significantly associated with lower PTSD symptoms and were most strongly and consistently associated with PTSD changes. Increases in Observing were sometimes associated with worsening PTSD symptoms, whereas changes in other facets of mindfulness were non-significant. Hyperarousal showed the strongest association with changes in aspects of mindfulness, followed by Emotional Numbing, Re-Experiencing, and Avoidance. The researchers suggested that “acting with awareness” may have helped to draw participants to the present rather than ruminating over the past or worrying about the future, and that maintaining a “nonreactive” response toward internal stimuli could facilitate the process of habituation, making it easier for participants to further expose themselves to these experiences.
Study Limitations: The study did not include a control group and used only self-report measures. History of trauma exposure was not assessed for portions of the study, so it cannot be definitively concluded that symptoms represented PTSD in all cases. Also, the correlational methods used cannot measure causality or directionality.
- Johnston and colleagues (2015) conducted a single-arm study to investigate the effects of yoga on symptoms of PTSD and on resilience and mindfulness in 12 current or former military personnel who met the DSM-IV-TR criteria for PTSD. The yoga intervention consisted of twice-weekly sessions for 10 weeks, each lasting 90 minutes, in a group format. The Kripalu yoga methodology included an initial check-in, a group centering and breathing practice, a 10-15 minute warm-up period, 50-55 minutes of practice involving yoga forms, and ended with a 5-10 minute relaxation period. There were also general class themes such as increasing strength and flexibility and reducing stress, balancing the sleep-wake system, and developing a responsive rather than avoidant or hyperaware relationship with the environment. Within-subject analyses revealed a significant reduction in PTSD symptoms but no significant improve mindfulness or resilience in this population. The results were also benchmarked against other military intervention studies of PTSD using the Clinician Administered PTSD Scale (CAPS) as an outcome measure. As compared with this benchmark, this study’s treatment effect was visibly lower, though it was higher than the waitlist control benchmark.
Study Limitations: This study had a small sample size and did not have a control group. Participants were allowed in the study who had co-morbid psychiatric conditions, a variety of medical concerns, were taking psychotropic medication, or were receiving mental health treatment; thus, the study findings cannot be definitely said to have resulted from the study intervention.
- Polusny et al. (2015) conducted a randomized clinical trial of 116 veterans to compare mindfulness-based stress reduction (MBSR) with present-centered group therapy for the treatment of PTSD. MBSR consisted of 8 weekly 2.5-hour group sessions and a daylong retreat, while the active control consisted of 9 weekly 1.5-hour group sessions focused on current life problems. The MBSR sessions included formal practice in three meditation techniques: body scan, sitting meditation, and mindful yoga (gentle stretches and movements practices with present-moment attention). Outcomes were assessed at baseline, during treatment, completion, and at a 2-month follow-up. Participants of MBSR were more likely to show clinically significant improvement in self-reported PTSD symptom severity; however, the effect was modest and they were no more likely to have loss of PTSD diagnosis.
Study Limitations: The sessions for the active control group were shorter in duration and also may not have accounted for all nonspecific factors present in MBSR, such as therapist expectations. Also, despite randomization, the control group showed less severe symptoms at baseline than the MBSR group, which may have affected results. Finally, the follow-up period was short which likely wasn’t enough time to capture relapse.
- Serpa, Taylor, and Tillisch (2014) conducted a pre/post questionnaire study of 79 veterans at a VHA medical facility who completed 9 weekly sessions of MBSR (mindfulness-based stress reduction) classes. The classes included seated and walking meditations, gentle yoga, body scans, and discussions of pain, stress, and mindfulness. Significant reductions in anxiety, depression, and suicidal ideation were found. Mental health functioning scores improved; pain intensity and physical health functionality did not. Increases in mindfulness were related to improvements in anxiety, depression, and mental health functionality.
Study Limitations: The study did not include a randomized control group or longitudinal follow-up and also did not consider relevant diagnoses, health care utilization data, or other veteran characteristics that might have affected treatment outcomes.
- Staples, Hamilton & Uddo (2013) conducted a single-arm yoga trial with 12 veterans with military-related PTSD. The yoga intervention was based on the yoga tradition of the Krishnamcharya Healing and Yoga Foundation and emphasized a therapeutic approach linking breath to movement with a meditative focus. The protocol was “fairly consistent” and commonly used the following forms: “dynamic side angle stretch (a gentle version of parsvottanasana) followed by chakravakasana (sunbird); virabhadrasana (warrior pose) followed by uttanasana (standing forward bend); and alternate leg urdhva prasarita padasana (upward stretched legs) followed by apanasana (knees to chest pose)” (p. 856). The 60-minute yoga sessions were held twice per week for 6 weeks. Following the 12-session intervention, the participants showed a significant improvement in hyperarousal symptoms and overall sleep quality, as well as a reduction in sleep-related daytime dysfunction. Quality of life was perceived by the participants as improved but was not statistically significant when measured.
Study Limitations: This study was small and lacked a control group. The 20% attrition rate also likely left the study underpowered, according to the researchers. In addition, the researchers suggested that the intervention may not have been long enough to result in measureable improvements in other areas that were tested.
- Carter et al. (2013) conducted a study of Sudarshan Kriya Yoga (SKY), adapted for veterans, with 31 male Vietnam veterans with PTSD who were randomized to either the treatment group or a 6-week wait-list control group. Twenty-five completed the study (14 treatment; 11 control). The treatment intervention consisted of 22 hours of guided group yoga instruction over a 5-day duration, followed by a 2-hour group session that was held weekly during the first month and monthly for the following 5 months. The SKY program was modified for veterans by adding joint mobility exercises, addressing warrior values, and by removing religious content. The treatment group showed a significant decrease in CAPS scores 6 weeks after completing the program. The control group showed no decline in scores at that time point, but improved significantly on the CAPS after they received treatment. Improvements were maintained six months following the treatment in both groups.
Study Limitations: This study had a small sample size and had a wait-list vs. active control group. The researchers note that though it is difficult to prove which components of the program were responsible for the results, the participants had a 30-year history of treatment resistant disabling PTSD which had not responded to therapy in the past, and thus it was less likely that the positive results could be attributed to the small amount of time used for group processes.
- Kearney and colleagues (2013) conducted a randomized controlled MBSR study of 47 veterans (37 M; 10 F) who were diagnosed with PTSD. The control group received treatment as usual. Intention-to-treat analysis found no reliable effects of MBSR on PTSD or depression. Completer analyses (≥4 classes attended) showed medium to large between group effect sizes for depression, mental health quality of life, and mindfulness skills. This study was included in the mega-analysis by Stephenson et al. (2016), discussed above.
Study Limitations: The study only used self-report measures. A majority of participants were also receiving medication management and/or supportive treatment; thus, definitive conclusions cannot be made regarding the unique effects of MBSR.
- Kearney and colleagues (2012) conducted a MBSR study of 92 veterans (70 M; 22 F) who had a variety of physical and mental health conditions. The treatment was an adjunct to their usual care at a large VA Medical Center and consisted of 8 weekly 2.5 hr. sessions of MBSR plus a full day retreat and homework 6 days/wk. Participants were assessed at baseline, 2 mos., and 6 mos. after enrollment. At 6 mos., there were significant improvements in PTSD symptoms (<.001), depression (<.001), behavioral activation (<.001), mindfulness (<.001), and other measurements. Of the PTSD symptoms, hyperarousal showed the largest change. Almost 48% of veterans had clinically significant improvement in PTSD symptoms. This study was included in the mega-analysis by Stephenson et al. (2016), discussed above.
Study Limitations: There was no control group. Also, the study only used self-report measures; a clinical interview would have been helpful to confirm PTSD status. Some participants were lost to follow-up, which could have introduced bias.
- Khalsa (2011) conducted a 10-week, 2x/week, 90-minute yoga intervention (with daily 15 minute daily home practice) to determine whether yoga is a feasible and effective intervention to reduce PTSD symptoms in service men and women with PTSD. Ten subjects completed treatment as of October 2011, with results showing a significant reduction in PTSD symptoms with a 25% drop on CAPS scores. Baseline CAPS scores for the group (M= 70.40; SD = 21.60) fell into the severe PTSD symptom range, whereas post-intervention, the group mean (M= 52.20; SD = 24.10) fell into the moderate PTSD symptom range. The effect sizes fall into the range of existing treatments for PTSD. According to Khalsa, the data suggest that yoga is feasible and efficacious for PTSD treatment.
Study Limitations: This pilot study was small and did not include a control group.
- Stankovic (2011) conducted an 8-week study of iRest yoga nidra (a form of deep relaxation) with combat veterans diagnosed with PTSD which found subjective PTSD symptom relief in all 11 of the 16 initially enrolled participants who completed the program. Participants reported reduced rage, anxiety, and emotional reactivity, and increased feelings of relaxation, peace, self-awareness, and self-efficacy, despite the challenges of intrusive memories or focusing difficulties. The highest average ratings were for the phases of the annamaya kosha (body sensing) -74%, welcoming (awareness) – 60%, and exploring the anandamaya kosha (bliss and inner strengths) – 57%. The lowest rating was for the vijnanamaya kosha (beliefs) phase -26%. Since the aspects of iRest most valued by the participants were ones focused on “sensing physical pleasure, bliss, and essential qualities of inner strength”, the researchers surmised that veterans with severe, chronic combat-related PTSD symptoms may find the greatest relief through practices that “cultivate and sustain ‘opposite’ states of mind and body” (p. 34).
Study Limitations: This study did not have a control group and used subjective measures.
- Engel et al. (2007) conducted a pilot study of iRest yoga nidra for combat-related PTSD at Walter Reed Medical Center with a cohort of 7 active duty military personnel who scored above 40 on the PTSD Checklist (PCL). The protocol was developed by Dr. Richard Miller, founder of the Integrative Restoration Institute (www.irest.us). The 9-week study consisted of 18 classes plus at-home study using a guided program on CD. Study findings showed a downward trend in scores on the Posttraumatic Stress Disorder Checklist (PCL). Participants reported decreased insomnia, depression, anxiety, and fear; improved personal relations; increased comfort with situations they can’t control; and increased sense of control in their lives.
Study Limitations: There was no control group as this was a prospective outcomes feasibility/pilot study. The sample size was too small to conduct inferential analyses.
- Carter and Byrne (2004) conducted a series of 6-week pilot trials of yoga with small groups of male Vietnam War veterans. The first trial consisted of 8 men, all of whom were diagnosed with PTSD, with or without Major Depressive Disorder. All participants were placed on SSRI (serotonin re-uptake inhibitors). The yoga intervention consisted of Iyengar Yoga for Depression as described by BKS Iyengar in his book, Yoga: The Path to Holistic Health. Classes occurred once per week for an hour, and due to the relatively short class duration, the recommended pranayama practices were rarely presented. Following completion of the series, both physician-administered and self-administered depression scales revealed significantly improved mood; however, there was no change in insomnia or anger management. The high focus on backbends caused some participants to experience back pain, which was addressed through counterforms. In a second trial, the asanas for anxiety were used, but the country was gearing up for war so there was a good deal of instability on the part of the participants. In a third trial, the yoga forms were taken from Gary Kraftsow’s book, Yoga for Wellness; yoga nidra was presented from Larry Payne’s Yoga Rx; and several forms of pranayama techniques were used, including TKV Desikachar’s progressive elongation of the out-breath as well as alternate nostril and Ujjayi breath. With the addition of the breathing techniques and meditation, considerable improvement was reported in sleep and anger management, though these results were not measured.
Study Limitations: There was no control group; sample sizes were small; some participants participated in more than one trial, which caused the baseline measures for depression to move lower with successive trials thus making it harder to register improvements; and the use of SSRI medication confounds the results for depression.
Lessons Learned and Future Directions for Research on
Yoga for War-Related Trauma
by Lynn Stoller © 2019 SEYI. All Rights Reserved.
The last decade and a half has produced the first published studies of the effects of yoga on symptoms of combat stress and war-related PTSD. Though the empirical evidence base is still under development, the findings thus far lend support for the idea that yoga can be a powerful adjunct treatment modality for these conditions. The studies presented here have produced evidence that yoga may have potential to reduce symptoms in any or all of the symptom clusters of PTSD as well as reduce symptoms of anxiety, depression, emotional reactivity, behavioral activation, pain intensity and/or suicidal ideation. Among the other findings are improved sleep quality, personal relations, feelings of peace and relaxation, better anger management, sense of control, as well as increased comfort in situations that are out of one’s control.
As the studies are highly heterogeneous, it is very unlikely that any one yoga intervention could possibly produce all of the above results. After all, if yoga was “one size fits all”, there would be no need for a skilled yoga therapy profession to exist. Carter and Byrne’s study, though mostly anecdotal, provides a good example of how different yoga practices can produce different effects (see previous section). Furthermore, as was pointed out by Yehuda and colleagues (2015), it is important to consider the different phenotypes of PTSD when designing studies for the condition, as a traumatized individual who fits the hyperaroused/re-experiencing phenotype may respond differently to an intervention than one who fits the detached/dissociated phenotype. Nevertheless, it is striking that almost all studies investigating yoga for combat-related PTSD have produced positive findings in one or more areas relevant to the condition, demonstrating yoga to be a fertile area of research deserving of priority status in the military community.
These studies have also illuminated several challenges that are commonly faced when conducting yoga with military populations. Most of the studies lacked a control group, and of the controlled studies, only Polusny et al. (2015) and Bremner et al (2017) included an active control. As for the others, the control group was simply offered usual care or was wait-listed. Staples and colleagues advised that a control arm should be matched for “time, attention, and expectancy” (2013, p. 858), implying that the activity presented to the control group should equal the duration of the yoga treatment; should inspire sustained attention on the part of the participants; and should create the expectancy of treatment success. When the control group simply receives usual care, then one can not be sure whether any changes are due to the yoga or due to other factors such as “support, attention, and engaging in simple movements” (Khalsa, Telles, Cohen, & McCall, 2016, p. 521, citing Park et al., 2014).
There are many types of active control groups that could be used depending upon the comparison sought. For example, one would need a movement-related control group to compare the difference between pure exercise and a mindfully-based movement practice such as yoga, whereas a comparison of two yoga meditation practices may or may not involve movement. As Staples and colleagues pointed out, yoga consists of multiple elements (yoga forms, breathing practices, relaxation techniques, and meditation) and thus it would be helpful to investigate which are more important for the desired treatment effect and to measure whether the combined effect is “additive or synergistic” (2013, p. 858). When these variables are not controlled, it is difficult to determine which of the practices are mostly responsible for therapeutic changes or to compare findings from different studies (Pilkington, Gerbarg, & Brown, 2016). Another study variable that is frequently left uncontrolled is the presence of co-morbid mental health conditions and the allowance of psychotropic medication, which may also confound results.
Another major limitation of most studies investigating yoga for war-related PTSD is the lack of physiological measurements to support subjective reports from study participants. There are several easily accessible ways to measure physiological responses, such as simply measuring heart rate, blood pressure, rate of respiration, and serum cortisol, all indicators of autonomic nervous system functioning. EEG has been successfully used to study the effects of meditation on brain activity (e.g. see Krishnakumar, Hamblin, & Lakshmanan, 2015) and could be used to measure the effects of yoga practices. Khalsa, Telles, Cohen, and McCall (2016) describe several cutting-edge quantitative outcome measures that have recently come into use in yoga research, including brain imaging and molecular biology techniques. MBSR researchers in particular (e.g. Bremner et al., 2017) have taken advantage of brain imaging technology, which obviously requires collaboration with scientists who specialize in them. Neuropsychologists and trauma psychiatrists are more apt to have developed these collaborative relationships, whereas other health professions such as occupational therapy could benefit from getting on the bandwagon.
Recruitment difficulties and a high dropout rate are other major issues confounding yoga research within the military community. From my own experience and discussions with others who have led veteran yoga classes in different organizations, growing a class in a community setting can take a bit of time unless it is directly fed by a veterans’ organization. As one founder of a veteran yoga program pointed out to me years ago, many combat veterans suffering from post-traumatic stress initially deny that they need help, and then when they are finally motivated to seek help, may still find it very difficult to follow-through due to depression, substance abuse, and/or logistical challenges. Thus it makes sense that the studies with the largest numbers of participants (and thus the strongest studies statistically) are ones that were conducted in the participants’ living environment such as a military base or VA medical setting. In these settings, transportation is not an issue and those who choose to volunteer in a study may be looking for something interesting to do to stave off boredom, such as after working hours on a military base. (This is another reason why it is important to have an active control group, i.e. to help rule out novelty as a confounding factor in the results.)
Despite the recruitment challenges of conducting yoga research in community vs. institutional settings, this is a nut we need to crack, since most veterans who have PTSD do live in the community and don’t seek help from the VA system. A good number of veteran yoga organizations have been established in recent years and it would greatly benefit the veterans living in our communities if these organizations collaborated with one another to conduct research as well as provide a more organized network of services. As the evidence builds with regard to what techniques work and don’t work to address combat-related PTSD, these organizations are beginning to look more and more like one another in practice, but this is a good thing! While some organizations may fear that this makes it harder to set themselves apart from the pack, the fact is, there are many hundreds of thousands of veterans with full or sub-threshold PTSD in the U.S. alone, so there is plenty of work for everybody in this field of practice and there will be for years to come (sadly to say). It is only through well-designed, unbiased research that we can illuminate the most salient elements in a yoga practice for promoting healing for combat stress and war-related trauma, and it is hoped that writing this article will help toward that effort. If your organization would like to collaborate with the Sensory-Enhanced Yoga Institute on research, please contact firstname.lastname@example.org.
Copyright © 2019, Sensory-Enhanced Yoga® Institute. All Rights Reserved.
Readers should assume that everything that they see or read on this site is copyrighted by the Sensory-Enhanced Yoga ® Institute and/or others unless otherwise noted. You may download material from the site for non-commercial, personal use as long as you also retain all copyright and other proprietary notices contained on the materials. You may not use, modify, distribute or transmit the contents of the site for public or commercial purposes without the written permission of the Sensory-Enhanced Yoga® Institute.