Abstract
Purpose
Trauma-adapted yoga (TAY) is a yoga adaptation developed for trauma-exposed populations that emphasizes autonomy and targets nervous system regulation. This study evaluates TAY’s impact on distress and psychiatric symptoms over the course of a psychiatric hospitalization, building on the results of a prior feasibility study that suggested TAY reduced participants’ anxiety and emotional pain immediately following group sessions.
Methods
This is a retrospective study. Adolescents voluntarily joined either a yoga or non-yoga (control) group. TAY was offered twice weekly alongside standard care. The Youth Outcome Questionnaire 2.0 was administered at admission and discharge to measure patient distress and assess change over the course of hospitalization. Participants also completed ratings of anxiety, emotional pain, and physical pain pre-and-post-yoga sessions.
Results
Over 27 weeks, 129 adolescents participated with 112 opting into the yoga group. Participants were predominantly female, white, and non-Hispanic. Major Depressive Disorder was the most common discharge diagnosis; 71% of participants reported trauma histories. Yoga participation significantly reduced Y-OQ scores whereas controls had no significant reductions over the course of hospitalization.
Conclusion
Findings show TAY practice during psychiatric hospitalization may reduce overall distress and imply a need for future inpatient standard programming and research.
Plain Language Summary
Adolescents admitted to psychiatric units can struggle with a variety of mental health concerns including depression, anxiety, and post-traumatic stress disorder (PTSD). The current standard for psychiatric treatment consists of therapy and medication. However, there is an evolving interest in alternative modalities that support patients holistically. One increasingly popular alternative is yoga, a practice research suggests improves physical activity as well as mental and emotional well-being. Yoga is unique when used as an adjunct to medicine as its adaptability promotes accessibility for adolescents of all backgrounds and abilities. This study looked at an alternative form of yoga known as trauma-adapted yoga (TAY), designed specifically for people who have experienced trauma. Unlike traditional yoga, TAY avoids things like instructors touching participants to correct poses or language that can sound controlling; instead, it promotes safety, participant choice, and body awareness. Over 27 weeks, 129 adolescents aged 12–18 admitted to an inpatient psychiatric unit were offered the choice to participate in yoga in addition to their usual care. TAY sessions were led by trained clinicians twice a week and patients provided feedback on how they felt before and after yoga. Most chose to take part in yoga and said they felt less emotional pain, as well as less distress and anxiety. Participation in TAY demonstrated significant reductions in overall distress, anxiety, and emotional pain over the course of hospitalization. The findings suggest that participating in TAY has potential to lessen adolescent patients' distress and anxiety while hospitalized on a psychiatric unit. This study is one of few, but its findings and growing interest in holistic methods of care encourage more research to be done on the effectiveness of TAY as a treatment resource within psychiatry.
Keywords
Within the clinical setting, yoga has gained substantial interest as an adjunct to standard treatment of care (Macy et al., 2018). Yoga is recognized as a form of complementary and alternative medicine (CAM), offering accessible symptom relief to patients who may require alternatives to standard traditional medicine (Wemrell et al., 2020). Yoga is the union of physical postures (asanas), and controlled breathing (pranayama), which are used to promote mindfulness and grounding. Rivest-Gadbois and Boudrias’s (2019) narrative review found that the practice of yoga may improve interoception in the general population, regulate the sympathetic nervous system while increasing parasympathetic activity, and decrease pain as well as the anxiety and distress associated with it.
Research suggests yoga is an effective adjunct to treatment for psychiatric disorders, with specific studies demonstrating a reduction in depression and anxiety in adults admitted to the inpatient psychiatric unit when compared to treatment as usual (Kumar et al., 2019). Recent studies suggest this benefit extends to the child and adolescent population (James-Palmer et al., 2020). The integrative review by Kerekes et al. (2024) evaluated 16 articles from more than 1,000 children from ages six to eighteen in school-based and psychiatric care settings, and identified yoga’s positive effects on psychiatric symptoms, cognitive functioning, relaxation, and well-being. Their review determined that yoga practice provided participants with an improved capacity to relax and reduce stress while improving self-esteem, resilience, and overall quality of life.
Current literature suggests that most adolescents admitted to inpatient psychiatric hospitals have experienced a traumatic event, and up to 29% of patients are formally diagnosed with PTSD (Havens et al., 2012). This is corroborated by a similar study that showed 75% of children admitted to psychiatric hospitals were exposed to violence. Approximately 26% of them were diagnosed with PTSD and the majority reported significant trauma symptoms (Belivanaki et al., 2017; Martinaki et al., 2021).
Macy et al. (2018) found yoga reduces PTSD symptom severity in trauma survivors of natural disasters, combat, and war. However, there is limited evidence of the effects of yoga on hospitalized adolescents as well as adolescents who identify as minorities with regards to race, ethnicity, gender, and sexuality. When looking specifically at yoga’s effects on anxiety and depression in children and adolescents, the systematic review by James-Palmer et al. (2020) identified only 27 studies with children and adolescents with various health conditions and some of the included studies lacked significant findings due to a lack of consistent and quality methodology.
Trauma Adapted Yoga (TAY) is a specialized form of yoga uniquely geared towards individuals with a history of trauma and PTSD as it encourages participants to practice bodily awareness, learn to regulate their nervous system with body-based strategies, and guides facilitators to be mindful of touching participants or using controlling language that can be triggering (Davis & Aylward, 2022; Sciarrino et al., 2017). TAY uses gentle and accessible asanas and provides support towards increased breathwork without initially focusing directly on breath, given sensitivities for certain traumatized populations. Implementing TAY has improved symptoms in similar trauma exposed populations such as in prisons, where a randomized controlled study demonstrated decreases in anxiety, hostility, and emotional distress as well as improvements in personality for yoga participants (Kerekes, 2024).
The Upstate Medical Hospital inpatient unit is designed as a Dialectical Behavior Therapy (DBT)-based unit. Due to the nature of the treatment model, patients spent a substantial portion of their time in DBT skills training, which is often in a didactic format. Maintaining treatment engagement can be a challenge. Non-compliance can contribute to treatment-interfering behaviors (TIB). The TIB can become disruptive not only to the patient involved but also to the milieu of several other adolescents in mental health crisis. TAY practice provides a body-based mindfulness practice, which is a core DBT skill and could provide a sense of mastery for adolescents whose physical abilities may have been an area of strength. For some patients, other treatments that depend on verbal or language skills may be discouraging if those are not preferred ways of learning or experiencing the world.
A previous pilot study by the research team showed improvements in self-reported anxiety and emotional pain directly following TAY sessions on an adolescent inpatient psychiatric unit (Khoury et al., 2024). These positive effects indicated that TAY had potential benefit within the adolescent inpatient unit, but findings were limited due to its status as an initial feasibility pilot study. There is also limited evidence on how TAY may influence psychiatric symptoms over the total course of an inpatient psychiatric hospitalization. This study examines how participation in TAY changes distress over the total course of inpatient psychiatric hospitalization, as measured by Youth Outcome Questionnaire 2.0 Self-Report (Y-OQ) scores. Based on the initial pilot study, we hypothesized that regular participation in a group TAY session while admitted to the adolescent inpatient psychiatric unit would reduce self-reported scores of distress as measured by the Y-OQ, as well as emotional and physical pain, and anxiety, as measured by the SUDS scale. We further hypothesized that those who participated in more TAY sessions would experience greater decreases in total Y-OQ scores, due to previous findings that yoga dose and consistency is related to positive outcomes (Bazzano et al., 2022; Davis & Aylward, 2022).
Additionally, this study assessed whether specific patient factors, such as a trauma history, length of stay, and patient demographics, affect these improvements. Considering TAY’s founding purpose to benefit participants with a history of trauma, we hypothesized that patients with a known trauma history would show significant decreases in Y-OQ scores after participating in TAY during their hospitalization.
Method
This retrospective study was conducted from February 2023 to March 2024 with an estimated sample size of 100 participants. Participants were followed for the time spent for their respective admittance to the inpatient unit.
Participants
Staff of the adolescent inpatient psychiatric unit invited eligible patients to participate in a yoga class utilizing principles of TAY as an adjunct to their usual mental health treatment with the option to decline or leave the group as needed. Participation was open to inpatient psychiatric patients aged 12–18. Patients with serious physical illness that would prohibit participation in any physical activity were excluded from attending yoga. Recently admitted patients on their first day in unit or patients required to be in restraint or seclusion were not eligible to participate in TAY, but were able to participate at a later date.
Throughout the inpatient stay, typical treatment consisted of daily group therapy, therapeutic medication management sessions and individual therapy dependent on individual treatment plans.
Yoga Group
The IRB at SUNY Upstate Norton School of Medicine deemed this study exempt. The team of yoga group facilitators included two nurse practitioners, two child psychiatrists, one psychologist, one nurse, and one art therapist. All facilitators were certified in TAY. The yoga group was held twice a week in the evening during daily shift change of inpatient medical staff or in times of patient need identified by staff members. Yoga was conducted in the unit’s group room and lasted approximately 50 minutes to 1 hour. Participants were encouraged to practice with modifications; one of the two facilitators would demonstrate using the modified version of the yoga movements side-by-side. Facilitators used inviting language, encouraging participants to listen to their body and move however made them comfortable. Facilitators played calming music and provided yoga mats to every participant. If participants needed to leave the room, one of the staff would check in with them and encourage them to return to the group when ready to do so. Each yoga class began with a review of guidelines: “Yoga is calm. Yoga is for you. Yoga is safe.” Participants were invited to check in with their anxiety, emotional and physical pain levels at the start and end of the yoga session using a subjective unit of distress Likert scale (SUDS). Post-yoga session, participants completed the scale again and were invited to debrief what changes, if any, they observed with the group and what their experience of yoga was like. Participants completed the Y-OQ at admission and discharge from the inpatient unit.
Control Group
Controls were largely self-selected patients who did not attend any TAY sessions during their hospital stay. Patients who chose not to participate in TAY sessions were engaged in alternative quiet activities that included art, watching television, or reading. Patients who could not attend any yoga sessions during their stay were also included in the control group.
Measures
The Youth Outcome Questionnaire 2.0 Self-Report (Y-OQ) is an outcome measure for adolescents ages 12–18 that assesses various aspects of psychosocial distress to monitor treatment progress during mental health interventions (Ridge et al., 2009). It is a 64-item measure using a five-point Likert scale (i.e., 0–4) that was normed using both community and clinical samples of adolescents. This measure yields an overall distress score and various subscales including intrapersonal distress (emotional distress), somatic distress (distress presenting physically), interpersonal relations (relationship with parents, other adults, and peers), critical items (flags raised for those requiring immediate intervention beyond standard treatment), social problems (socially-related problematic behaviors), behavioral dysfunction (unhealthy behaviors). We hypothesized that symptoms related to all of these subscales would improve with TAY and did not do sub-analysis due to sample size/power. The Y-OQ was chosen as a standard outcome measure for the adolescent inpatient unit for multiple reasons including its ability to capture an overall improvement non-specific to diagnosis or category of symptoms, but most importantly because patients can share their own perspective with a change-sensitive self-report measure. Additionally, research suggests it is a psychometrically sound measure, with studies demonstrating moderate to high internal consistency, reliability, and strong concurrent validity (Ridge et al., 2009). The Y-OQ was administered to all adolescent inpatient patients on an iPad and automatically scored. The scale range utilized for the Y-OQ in this study was −16 to 240 with a clinical cut-off of 47 and index for change was 18.
The Subjective Units of Distress Scale (SUDS) is a self-report scale that measures feelings that can include anxiety, anger, agitation, stress, or other feelings of pain using an 11-point Likert scale (Wolpe, 1969). It is used as a tool used regularly in therapy including systematic desensitization and exposure therapy. It was hypothesized that participating in yoga was a novel experience and that anxiety ratings pre-yoga session may decrease in response to participation repetition. The SUDS scale is a key component of DBT where it is used to help adolescents track their emotional distress on a scale of 0–10 to build self-awareness, identify emotional triggers, and measure effectiveness of DBT skills to manage and tolerate emotional distress (Miller et al., 2007).
Data Analysis
Patients who attended at least one yoga session were compared to those who did not participate. Continuous variables, including age and length of stay, were analyzed using linear regression to assess mean differences between groups. Categorical variables—sex, gender, sexual orientation, race, ethnicity, trauma history, and primary diagnosis—were compared using Chi-square tests. For the primary outcome— change in Y-OQ score, computed as the difference between pre-and-post-treatment overall distress Y-OQ scores—a one sample t-test was used to determine if the change is significantly different from 0. Multivariate linear regression models were used to examine the impact of demographic and clinical characteristics, pre-treatment Y-OQ score (pre-treatment symptom severity), length of stay, and number of sessions on the observed change. To address the small and unbalanced size of the non-participant group, we also modeled the number of sessions attended as a continuous independent variable rather than using a binary participation status. For repeated measures of anxiety, emotional and physical pain collected immediately pre-and-post yoga sessions, mixed-effects linear regression models were applied to account for within-subject correlations and repeated observations over multiple yoga sessions. All analyses were conducted using STATA 18.
Results
Demographic Characteristics
Note. SD = standard deviation; ns = non-significant.
Clinical Characteristics
Note. SD = standard deviation; IQR = interquantile range; N = observation numbers; ns = non-significant.
aindicating changes in Y-OQ were significantly below 0.
Most recruited participants self-selected to participate in the yoga group (112, 86.8% of total participants) (Table 1). Most participants identified as female, white, and non-Hispanic (Table 1). Of note, chi-square analysis showed that participants identifying as racial minorities were more likely to participate than opt-out of yoga apart from patients identifying as Asian who were evenly split (χ2 (4) = 9.75 p = 0.045) (Table 1). 26 participants (20.2%) were non-cis gender (Table 1). Many participants identified as heterosexual or their sexual orientation was unknown. The most common discharge diagnosis was Major Depressive Disorder (84, 65.1%) (Table 2). 71.3% of participants indicated a trauma history and PTSD was the second most common discharge diagnosis (11.6%) (Table 2). The researchers focused on discharge diagnoses over admission diagnoses as it is typically thought to be more accurate given that one purpose of hospitalization is to clarify diagnosis.
45 yoga group sessions were conducted on the adolescent inpatient psychiatric unit with an average eight eligible patients per session. The average number of participants per session was five and the participation rate across all sessions offered was 63.4%. The median number of sessions offered per participant was three and the median number of sessions participated in was two. 86.8% of eligible inpatient adolescents participated in one or more sessions throughout the study period (Table 1).
Overall, post-treatment Y-OQ scores were significantly correlated with the pre-treatment Y-OQ scores (Pearson correlation r = 0.63, p < 0.0001, Figure 1(A)) and the correlation was similar in both participating and non-participating individuals (Figure 1(B)). The overall mean decrease of Y-OQ was 18.7 (+/−32.2, Table 2, Figure 2(A)) and was only statistically significant for the yoga-participating individuals (t(41) = 3.35, p = 0.0003) (Figure 2(B)). The length of hospitalization, the total number of yoga group sessions participated in, as well as trauma history were not associated with the change in Y-OQ scores. There was also no significant association of the pre-treatment Y-OQ score with the reported trauma history. Of note, the pre-treatment Y-OQ score was significantly and negatively associated with the decrease of Y-OQ scores (Pearson’s r = −0.4, p = 0.02), indicating that individuals with higher pre-treatment scores had greater symptom reduction. The pre-treatment Y-OQ scores were not associated with the participation status. However, among those who participated, the higher pre-treatment Y-OQ scores were marginally associated with a greater number of sessions attended (F(1, 38) = 3.1, p = 0.086). Length of stay, although significantly correlated with the total sessions participated in (r = 0.3, p = 0.003), was not associated with the pre-treatment Y-OQ scores. Inclusion of pre-treatment Y-OQ scores in the model, but not the total number of sessions or length of stay, attenuated the association between yoga participation and the decreased Y-OQ scores, rendering it non-significant. Y-OQ: Post-vs. pre-treatment. Note: Pre- and Post-Y-OQ scores were highly correlated (A), and the correlations were similar among the participating and non-participating individuals (B) Changes in Y-OQ based on TAY participation. Note. Changes of YOQ scores were significant for the participating group. Only the participating group had significantly decreased YOQ scores (one-tailed t-test t = −3.35, p = 0.001)
Those identifying as male entered the study with a significantly lower pre-treatment Y-OQ as compared to female-identifying participants. Both male-identifying heterosexual and non-heterosexual participants also had a significantly lower pre-treatment Y-OQ as compared to their female-identifying counterparts.
Among participants, yoga sessions were significantly associated with decreases in anxiety (chi2(1) = 25.4, p < 0.0001) and emotional pain (chi2(1) = 32.60, p < 0.0001) post-yoga session (Figure 3). The effect of yoga on self-reported physical pain was marginal and did not survive Bonferroni correction (chi2(1) = 4.32, uncorrected p = 0.04, Figure 3). Trauma history was significantly associated with higher physical pain (chi2(1) = 8.26, p = 0.0041), but not with anxiety or emotional pain. Demographic characteristics, primary diagnoses, or Y-OQ scores showed no association with anxiety or physical pain. Sex, gender, and Y-OQ scores, but not sexual orientation, were all significantly associated with emotional pain. Anxiety, physical, and emotional pain before and after TAY. Note. Dot plots of the SUDs measures of anxiety, emotional and physical pain before and after yoga sessions
Discussion
This retrospective study evaluating TAY practice on an adolescent inpatient unit demonstrates that TAY may be an effective adjunct to standard adolescent inpatient psychiatric treatment. Adolescents who participated in yoga had a statistically and clinically significant improvement of their Y-OQ scores during hospitalization. Notably, the pre-treatment Y-OQ scores accounted for the association and were negatively correlated with the symptom change, suggesting that individuals with more severe symptoms experienced more improvement following the intervention. No statistically significant improvement was observed in the non-participating group; however, the small sample size in the non-participating group limits the ability to draw definitive comparisons between the groups.
Those adolescents in the yoga group demonstrated immediate improvement in anxiety and emotional pain, and reduction in distress over their hospitalization. These results align with previous research demonstrating the benefits of yoga on stress, negative emotion, anxiety and depression symptoms, and mindfulness for youth (Kerekes et al., 2024; Weaver & Darragh, 2015). With regards to the feasibility and efficacy within inpatient settings, these results build on previous literature supporting yoga as adjunctive treatment for adults in psychiatric inpatient settings (Wu et al., 2025) and adds to the small but growing literature on yoga for adolescents specifically with psychiatric disorders in psychiatric inpatient settings (Hariprasad et al., 2013; Re et al., 2014).
Benefits to Participants
The aims of intentionally sequenced movements were to offer tools to regulate adolescents’ nervous systems, which are often impacted by trauma exposure. Participating in TAY may have provided participants with a space away from the inpatient unit and time to practice physical activity attuned to their needs in a calm group setting. Administering TAY as a twice weekly group also facilitated community and connection amongst participants. This finding resonated with Kerekes et al. (2024) and their review of yoga’s effects on adolescents and how the practice itself can serve as a source of social interaction, bonding, empathy, and socioemotional competence. Interestingly, even adolescents who participated minimally in the yoga session would report improvement in stress and anxiety at the end of the session, suggesting that the benefit of the practice may be not only in the physical poses but the relaxed and inviting atmosphere set by the facilitators or the companionship offered by the group setting.
This study did not identify any significant relationship between the number of yoga sessions participants engaged in and decreases in Y-OQ scores. Only five individuals who had Y-OQ scores participated in four sessions or more. This lack of consistency and low sample size may have contributed to the lack of significant effect of the number of sessions on Y-OQ scores. Nevertheless, our results indicated that individuals with more severe pre-treatment Y-OQ scores tended to demonstrate greater improvement following yoga, even with relatively few sessions. This highlights the potential utility of TAY for inpatient youth with severe symptoms during short-term hospital stays. The marginal association between pre-intervention symptom severity and number of sessions attended suggests an opportunity to increase participation among those who may benefit most. Accordingly, a key goal of future studies will be to expand both session participation and Y-OQ assessment, thereby improving statistical power to evaluate a potential dose–response relationship.
Strengths
This study adds to the growing literature on yoga as an adjunctive treatment to psychiatric inpatient care in demonstrating the statistically significant changes found in Y-OQ scales over time for yoga participants. This study builds on the work of Re et al. (2014) who found yoga to be effective as an adjunct to adolescent inpatient psychiatric treatment and examines different measures of effect over the course of hospitalization. Current literature includes few studies involving yoga as an adjunct to treatment and the existing studies often lose significance due to small sample size, low replicability, and a lack of standardization in the measures utilized (James-Palmer et al., 2020). Our study size of 129 adolescent patients allows ample power and demonstrates the potential impact of TAY within the field of mental health treatment.
Limitations
The administration of the Y-OQ is a standard practice within the adolescent inpatient psychiatric unit and is only administered at admission and discharge. However, Y-OQ scores were not available on every patient due to staffing issues and other considerations. No significant differences were found between those individuals with Y-OQ scores and those without when it came to discharge diagnosis but missing data is a limitation. Additionally, the pre/post-Y-OQ scores may have been affected by other factors during the patient’s stay such as changes in treatment, medication administration, environment, social factors, or other factors that our study could not control for. Moreover, participation in yoga may have been a marker for capacity to participate in inpatient programming overall and not be specific to yoga.
Adolescents were offered the choice between the yoga treatment group and a quiet activity within the milieu which served as a naturalistic control, with a majority of participants choosing to practice yoga. This led to a small non-participating group which limits ability to adequately compare groups. There was no randomization or matching made between the groups. As such, some participants may have self-selected to participate because they were already aware of the benefits of yoga. In addition, our study aimed to place controls within a relaxing room or invite them to a relaxing practice for the same duration as the yoga group. This initially occurred to allow us to compare the effects of TAY versus other non-yoga relaxation methods but was soon discontinued due to uncontrollable circumstances within the unit including lack of available staff and shift-change time conflicts. Further studies should consider comparing TAY against non-yoga relaxation methods to determine its effectiveness as an optimal relaxation method.
Like existing studies on yoga as a treatment method, our study had a predominantly female-identifying population which limits the applicability of its findings. Further studies should recruit a balanced population of male-identifying and female-identifying individuals and include randomization where possible.
Implications
TAY is unique in its focus on safety and inclusivity, and its practice can be easily taught to clinicians and learned by participants as it involves a smaller set of standard yoga poses and promotes modification by the individual participating. Participants can adapt certain movements, poses, or flows to fit their environment and capabilities.
TAY may be an effective method for promoting mindfulness and well-being in other settings as well, but more research is needed. This study looked only at an adolescent inpatient psychiatric unit, but the same practice can be applied to adolescents who are outpatient psychiatric patients. Beyond the field of psychiatry and patient group of adolescents, clinicians can also bring TAY to patients of any age by the bedside or during outpatient appointments when applicable. Trauma history can be assumed to be more common in psychiatry patients, but patients in other disciplines may present with trauma histories that are not disclosed or unrelated to their current condition. Offering a body-based trauma-informed method of care can further support individual patients and their needs to offer an adjunctive source of treatment and mindfulness. Further research is needed to assess TAY effectiveness for patients in other clinical settings.
Footnotes
Acknowledgments
The authors thank our TAY-certified leaders who brought their enthusiasm to every weekly group yoga session. We would like to acknowledge the founders of TAY, Josefin Wikstrom and Nora Kerekes, who provided TAY training and ongoing support. We also would like to thank the clinical staff on the adolescent inpatient psychiatric unit for their willingness to implement and help facilitate this novel intervention. The Upstate Well Council has continuously supported the authors in funding TAY training and we could not have done this without their support. Thank you to all our participants for their participation in such a clinically significant research study.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.