Mindful energy balance exercise protocol as an adjunctive intervention for pediatric tourette syndrome: A randomized controlled trial

Participants

Sample size was calculated using G*Power 3.1 based on a pilot study (n = 20) conducted by our team, which observed a large effect size (Cohen’s d = 0.80) for tic reduction. With a significance level of α = 0.05 and statistical power (1 − β) = 0.80, the required sample size was estimated to be 52 participants per group. To account for an anticipated attrition rate of 15%, the final recruitment target was set at 135 participants.

From January 2021 to December 2023, 135 children aged 4–12 years who met the DSM-5 diagnostic criteria for Tourette Syndrome were recruited from three tertiary hospitals in Xinxiang, Henan Province. The age range of 4–12 years was chosen because it corresponds to the typical onset period of TS and represents a critical developmental window for behavioral interventions. To ensure developmental appropriateness, the MEBE program was adapted for different age groups. For younger children (4–6 years), sessions emphasized play-based mindful movement and simplified breathing exercises, supported by visual and manual aids. For older children (7–12 years), the program incorporated structured attention training and body awareness practices to build on their greater cognitive and attentional capacities.

Diagnoses were confirmed by trained child psychiatrists using the Kiddie Schedule for Affective Disorders and Schizophrenia—Present and Lifetime Version (K-SADS-PL)¹⁴. Comorbid conditions were assessed using validated instruments, including the Conners’ Parent Rating Scale for ADHD¹⁵, the Children’s Yale–Brown Obsessive-Compulsive Scale (CY-BOCS)¹⁶ for OCD, and the Screen for Child Anxiety Related Emotional Disorders (SCARED)¹⁷ for anxiety. Comorbid conditions were systematically evaluated at baseline by trained clinicians using these standardized instruments. Comorbidity status was reported based on instrument-specific clinical thresholds and clinician judgment. Severe psychiatric conditions, such as psychosis or acute suicidality, were exclusionary.

Inclusion criteria required participants to have a minimum of one year of tics and clinically significant functional impairment. Exclusion criteria included unstable psychiatric treatment or changes in psychoactive medication within the past three months. Per protocol, participants could initiate or standardize guideline-concordant pharmacotherapy at baseline, provided there had been no unstable treatment in the prior three months. Additional exclusions were a history of major neurological disorders, suicidal ideation, psychosis, or other severe psychiatric symptoms. No adverse events occurred during the study period.

Participants were randomly assigned (1:1) to the experimental group (EG) or control group (CG) using computer-generated, stratified block randomization (block size = 4), with stratification based on age and baseline tic severity. Allocation concealment was achieved using sequentially numbered, opaque, sealed envelopes. Outcome assessors and statistical analysts were blinded to group assignments.

Of the 135 randomized participants, 67 were assigned to the EG and 68 to the CG. Fourteen participants did not complete the trial due to family relocation (n = 5), scheduling conflicts (n = 4), withdrawal of consent (n = 3), or unexplained loss to follow-up (n = 2). In total, 121 participants (EG = 61; CG = 60) completed the intervention and evaluation protocol. Primary analyses used a modified intention-to-treat (mITT) approach restricted to completers (n = 121), with all randomized participants described in the CONSORT flow. Clinical assessments were conducted at four time points: baseline (T0), mid-intervention (T1, Week 6), post-intervention (T2, Week 12), and follow-up (T3, Week 24). All evaluations were performed face-to-face by trained assessors who were blinded to group allocation and not involved in intervention delivery. Baseline comparisons revealed no significant differences in demographic or clinical characteristics between groups.

Outcome assessments utilized validated instruments described in the Measures section. The primary outcome was tic severity, assessed with the Yale Global Tic Severity Scale (YGTSS)^18,19. Secondary outcomes included dispositional mindfulness (Child and Adolescent Mindfulness Measure, CAMM)^20,21, anxiety symptoms (Spence Children’s Anxiety Scale, SCAS)^22–24, and quality of life (Pediatric Quality of Life Inventory—Parent Proxy Report, PedsQL-P)^25–27. Anxiety comorbidity was screened at baseline with the SCARED, and longitudinal anxiety outcomes were assessed from T0 to T3 using the SCAS. A CONSORT flow diagram is provided as Fig. 1 to illustrate participant enrollment, allocation, follow-up, and analysis.

Procedure

MEBE Intervention Development The Mindful Energy Balance Exercise (MEBE) program was developed as a structured mindfulness-based intervention tailored to meet the developmental and clinical needs of children with Tourette Syndrome (TS). The program aimed to enhance self-regulation, improve attentional control, and increase emotional awareness through movement-integrated mindfulness practices specifically designed to address TS-related symptoms such as motor tics and vocal tics.

Developed by a multidisciplinary team of clinicians, educators, and mindfulness practitioners, the program integrates age-appropriate techniques, including guided breathing exercises, body scans, mindful movement, and interactive games. The intervention was refined through pilot testing (n = 33) and feedback from both clinicians and caregivers, which helped iteratively adjust the delivery format and content for optimal engagement.

Each session lasted approximately 20 min and was delivered by certified instructors in hospital-based training facilities. The 12-week program was structured into three progressive phases: an intensive phase (Weeks 1–3, five sessions per week), a consolidation phase (Weeks 4–9, three sessions per week), and a maintenance phase (Weeks 10–12, two sessions per week). Parents were encouraged to support their children’s practice with 10-minute daily home sessions, using illustrated manuals and video guides.

All program components were reviewed for developmental appropriateness and physical safety by pediatric specialists. Training manuals were standardized to ensure implementation fidelity, and physiological indicators (e.g., heart rate variability) were monitored during pilot testing to assess tolerability.

Session durations and home practice dosages were guided by prior research indicating that mindfulness interventions are most effective for children when individual sessions are under 30 min, and brief, consistent daily practice supports adherence and positive behavioral outcomes. The finalized MEBE protocol is designed to be scalable and adaptable for integration into pediatric behavioral health services, and unlike traditional practices such as yoga or Tai Chi, it was structured specifically to address tic regulation through simplified, breath-paced movement and urge-monitoring cues. A more detailed comparison with other contemplative practices is provided in the Discussion.

Intervention Implementation Children in the control group (CG) received standard pharmacological treatment based on national clinical guidelines for pediatric tic disorders. The primary medications prescribed were clonidine hydrochloride (0.1 mg, twice daily) for mild cases, and aripiprazole (2.5–5 mg/day) for moderate to severe tics, depending on symptom profile and tolerance²⁸. Medication profiles did not differ between groups at baseline (e.g., clonidine: 62% EG vs. 63% CG; aripiprazole: 32% vs. 31%; other agents: 6% vs. 6%; all p > 0.40). Biweekly dose adjustments were comparable across groups (p > 0.20). This indicates that medication exposure was balanced and unlikely to confound intervention effects. Medication type and dosage were determined by board-certified pediatricians who were blinded to group assignment, with adjustments made biweekly based on clinical response and side effect profiles. No medication-related adverse events were reported during the trial period. To minimize pharmacological confounds, all medication adjustments were documented biweekly, and dosage changes were included as covariates in supplementary sensitivity analyses. These analyses confirmed that the intervention effects on tic reduction and anxiety remained robust after controlling for medication dosage variability.

To prevent contamination, caregivers in the control group were provided with general health education materials that excluded any content related to mindfulness, breathing techniques, or emotion regulation. Weekly check-ins and caregiver adherence logs were used to monitor exposure; no families in the control group reported contact with mindfulness-based practices during the intervention period.

Children in the experimental group (EG) received the same pharmacological treatment alongside the 12-week MEBE program. The intervention sessions followed the three-phase structure described earlier, delivered by certified instructors in hospital-based training rooms. To enhance engagement, motivational strategies such as small rewards (e.g., stickers or point systems) were incorporated after each session.

Parents were actively involved in the program through the provision of illustrated instructional materials and participation in a digital support group. Daily home practice was strongly encouraged, and adherence was monitored via parental logs, reviewed biweekly by the study team. For participants showing reduced engagement, gamified reinforcement strategies and individualized feedback were introduced to ensure sustained participation.

Intervention fidelity was maintained through weekly instructor compliance checks and supplemental coaching as needed. Participants demonstrating high adherence were invited to optional advanced sessions, which included dynamic movement activities and parent–child mindfulness exercises. Adverse events were monitored weekly through structured parent reports and assessed by pediatric clinicians. Ongoing quality assurance was ensured through periodic expert consultations and iterative program refinement to maintain the integrity and effectiveness of the intervention. Adherence thresholds were prespecified in the protocol: ≥70% of assigned daily home practice and ≥ 90% session attendance.

Measures

Yale Global Tic Severity Scale Tic severity was measured using the Yale Global Tic Severity Scale (YGTSS), a clinician-administered tool recognized as the gold standard for assessing tic symptoms in children and adolescents. The scale includes two components: the total tic score and the impairment score.

The total tic score (range: 0–50) evaluates motor and phonic tics across five dimensions: number, frequency, intensity, complexity, and interference. Each dimension is rated on a 6-point scale (0 = none, 5 = severe), producing motor and phonic subscale scores of 0 to 25. The Impairment Score (range: 0–50) is independently rated and captures functional disruptions across emotional, social, familial, and academic domains. The global severity score (0–100) is calculated by summing the total tic and impairment scores, with higher scores reflecting more severe symptoms and greater impairment.

In this study, the Chinese version of the YGTSS showed acceptable internal consistency, with McDonald’s ω values as follows: motor tic number = 0.84, frequency = 0.85, intensity = 0.86, complexity = 0.81, interference = 0.83; phonic tic number = 0.85, frequency = 0.86, intensity = 0.87, complexity = 0.82, interference = 0.84; Total Tic Score = 0.86.

Child and Adolescent Mindfulness Measure Mindfulness was assessed using the Chinese version of the child and adolescent mindfulness measure (CAMM), a 10-item unidimensional self-report scale originally developed by Greco et al. and adapted for Chinese school-aged populations. Each item is rated on a 5-point Likert scale ranging from 0 (never) to 4 (always), yielding a total score between 0 and 40, with higher scores indicating greater dispositional mindfulness. In the present study, the CAMM demonstrated good internal consistency across repeated assessments in both groups (McDonald’s ω = 0.81).

Spence Children’s Anxiety Scale Anxiety symptoms were assessed using the Spence Children’s Anxiety Scale (SCAS), a 38-item self-report measure. We used a validated Chinese version with five domains (separation anxiety, social anxiety, generalized anxiety, panic/agoraphobia, and somatic symptoms), reflecting the factor structure supported in Chinese pediatric samples. Detailed item-domain mapping is provided in the Supplementary Methods (Table S1). Each item is rated on a 4-point Likert scale from 0 (never) to 3 (always), resulting in a total score range of 0–114, with higher scores indicating more severe anxiety symptoms. Internal consistency for the SCAS in this sample was excellent, with McDonald’s ω values as follows: separation anxiety = 0.88, social anxiety = 0.89, generalized anxiety = 0.90, panic/agoraphobia = 0.91, somatic symptoms = 0.87, and total score = 0.90.

To ensure developmental appropriateness, administration procedures were adapted according to age. For children aged 4–7 years, all questionnaire items (CAMM and SCAS) were administered orally by trained assessors using simplified wording and pictorial Likert anchors, such as a five-face scale. Comprehension was checked item by item, and neutral re-prompts were provided as needed. Responses were recorded verbatim without interpretation. Assessors administering oral versions were trained to standardized scripts and remained blinded to group assignment to minimize measurement bias. For participants aged 8 years and above, questionnaires were self-administered under assessor supervision.

Pediatric Quality of Life Inventory—Parent Proxy Report (PedsQL-P) Daily functioning was evaluated using the Pediatric Quality of Life Inventory—Parent Proxy Report (PedsQL-P), developed by Varni et al. and validated for use in Chinese pediatric samples by Wu et al. This 23-item scale assesses four dimensions: physical, emotional, social, and school functioning. Items are rated on a 5-point Likert scale (0 = never, 4 = almost always), reverse-coded, and transformed into a 0–100 scale, with higher scores indicating better quality of life. To minimize bias, the inventory was completed independently by parents without researcher guidance, and data collection staff remained blinded to group assignment. Internal consistency for the subdomains in this study was excellent: physical = 0.92, emotional = 0.89, social = 0.90, school = 0.88, and total score = 0.94 (McDonald’s ω).

Baseline characteristics

At baseline (T0), no statistically significant differences were observed between the experimental group (EG) and the control group (CG) in demographic or clinical characteristics, including age, gender, illness duration, lifestyle habits, screen time, socioeconomic status, or parental education (p > 0.05), confirming the equivalence of the two groups at baseline (Table 1).

The sex distribution was approximately 1:1 overall (61 boys, 60 girls), with a balanced distribution across groups (EG: 31 boys, 30 girls; CG: 30 boys, 30 girls). Baseline YGTSS severity did not differ significantly by sex. A detailed breakdown of baseline comorbid conditions, including ADHD, OCD, anxiety disorders, depression, sleep disturbances, and other developmental or psychiatric conditions, is provided in Supplementary Table S2. No significant imbalances were observed between groups at baseline. Sensitivity analyses comparing mITT results with complete-case sensitivity analyses yielded consistent inferences (Supplementary Table S3).

Tic severity

Tic severity, measured using the YGTSS, was the primary clinical outcome. At baseline (T0), no significant differences were observed between the EG and CG across any YGTSS domains, confirming baseline equivalence between the groups (Table 2).

Throughout the intervention, the EG showed significantly greater reductions in motor tics, phonic tics, total tic severity, functional impairment, and global severity scores compared to the CG. Between-group comparisons of change scores revealed consistent differences across all five domains (p < 0.05). Although effect sizes were small to moderate (e.g., motor tic score d ≈ 0.43), such changes are clinically meaningful in pediatric populations, given the short session length and the developmental limitations of younger children (Table 3).

Repeated-measures ANOVAs confirmed significant main effects of time, group, and time × group interactions for all YGTSS domains (p < 0.001), indicating meaningful differences in tic symptom trajectories between groups (Table 4). These results support the efficacy of the MEBE intervention in reducing tic symptoms and related functional impairments.

Mindfulness

Mindfulness, assessed using CAMM as a secondary psychosocial outcome, significantly improved following the intervention. No significant group differences were observed at baseline. From mid-intervention (T1) to follow-up (T3), CAMM scores were significantly higher in the EG than in the CG. Between-group comparisons at each time point showed moderate to large effect sizes, indicating a consistent and meaningful increase in mindfulness. Importantly, higher CAMM scores may reflect enhanced self-regulation, which could in turn facilitate tic suppression and reduce anxiety symptoms (Table 5).

A two-way repeated-measures ANOVA revealed a significant time × group interaction (F(3,357) = 42.17, p < 0.001, η² = 0.262), indicating differential changes in mindfulness scores between the EG and CG over time (Table 6).

Anxiety

Anxiety symptoms were assessed using the SCAS across five domains. At baseline (T0), no significant group differences were observed in any subscale or in the total anxiety score, confirming initial equivalence (Table 7).

Following the intervention, the EG showed significantly greater reductions in separation anxiety, social anxiety, and generalized anxiety compared to the CG, with between-group differences in change scores reaching statistical significance (p < 0.001; Table 8).

Consistent with this pattern, somatic/physical anxiety symptoms showed smaller group differences. The effects on the SCAS panic and somatic domains were not significant (p > 0.05). Baseline scores for panic and somatic symptoms were lower than those for separation and social anxiety (Table 7), and the curriculum focused more on attentional control and somatic anchoring rather than interoceptive exposure. These factors may have limited the detectable between-group changes. The total SCAS score also showed a significantly greater reduction in the EG compared to the CG, indicating a broader decline in overall anxiety levels attributable to the intervention. These improvements were further supported by significant time × group interaction effects observed in repeated-measures ANOVAs, with moderate effect sizes (η² = 0.261–0.286; Table 9).

Daily functioning

Daily functioning was assessed using the PedsQL-P, covering physical, emotional, social, and school domains. At baseline, both groups had comparable levels of functioning (Table 10).

Over time, the EG demonstrated significantly greater improvements than the CG in the School Function (p = 0.026) and Total (p = 0.005) domains, with larger gains observed in the EG across these areas. Physical Function showed a trend toward significance (p = 0.056), and Social Function also showed a trend-level difference (p = 0.071). Emotional functioning showed minimal between-group differences and did not reach statistical significance (p = 0.538).

Repeated-measures ANOVA confirmed significant time × group interaction effects in physical, social, and school functioning (p < 0.001), with the largest effects seen in school functioning (η² = 0.181) and physical functioning (η² = 0.172). No significant interaction was observed for emotional functioning (Table 11).

These findings suggest that integrating the mindfulness-based intervention with standard pharmacological treatment led to significant improvements in various domains of daily functioning in children with Tourette syndrome. While between-group change scores for Physical and Social Function approached significance at the endpoint (Table 12), the time × group interactions were robustly significant for these domains (Table 11), indicating differential trajectories favoring the experimental group across the study period. Additionally, adherence to the intervention was high: 86% of participants in the experimental group completed at least 70% of the assigned daily home practice (per parental logs), and 92% completed all 12 weeks of in-person sessions. Adherence thresholds (≥ 70% of assigned home practice; ≥90% session attendance) were prespecified in the study protocol. These results indicate that the observed clinical improvements are closely tied to consistent engagement with the MEBE program.

Key findings and theoretical implications

This study demonstrated that integrating the MEBE program with pharmacological treatment yielded significant improvements in mindfulness, tic severity, anxiety symptoms, and quality of life among children with TS. These results align with and extend the growing body of evidence supporting developmentally adapted mindfulness-based interventions for neurodevelopmental disorders such as TS^29–31.

Beyond clinical outcomes, the observed improvements can be interpreted in light of neurobiological mechanisms. Breathwork may reduce tic severity by enhancing parasympathetic activity and stabilizing CSTC function, while body awareness supports detection of premonitory urges. For tic symptoms, the mechanism likely involves strengthening prefrontal–motor pathways to suppress involuntary movements. For anxiety symptoms, the mechanism is distinct—breath control and somatic anchoring likely disrupt maladaptive cognitive–affective cycles, reducing separation and social anxiety. Thus, MEBE exerts symptom-specific effects through partially different neural pathways. This mechanistic mapping supports the hypothesis that MEBE operates through both autonomic regulation and top–down cognitive control, offering a neurodevelopmentally plausible framework.

Children in the experimental group exhibited significantly greater reductions in both motor and vocal tics compared to the control group, with particularly pronounced improvements in motor symptoms. This pattern may reflect the intervention’s emphasis on movement-based attention and sensory grounding, which have been shown to facilitate emotion regulation and attenuate premonitory urges^32,33. These findings are consistent with the hypothesis that embodied practices enhance voluntary tic suppression by strengthening top-down regulatory processes. While the observed effect size for motor tic reduction (Cohen’s d = 0.43) is modest, it remains clinically meaningful given the brevity of sessions and younger participant age. For comparison, meta-analyses of HRT/CBT in pediatric TS report larger effect sizes (d = 0.6–0.8), but these approaches typically require longer, more intensive protocols delivered by highly trained therapists. The present findings suggest that MEBE, as a low-intensity, scalable adjunct, can provide meaningful tic reduction even with a smaller effect size. To contextualize discrepancies with our prior pilot trial, the previously reported large effect size (Cohen’s d = 0.80) was calculated as a within-group pre–post change in the experimental arm on the YGTSS Total Tic score. By contrast, the present RCT reports between-group effect sizes on change scores (EG vs. CG), which are inherently more conservative. Additional factors—such as greater sample heterogeneity across sites, the pragmatic lower-intensity intervention schedule, and concomitant medication adjustments (recorded and controlled in sensitivity analyses)—likely further attenuated between-group effects while preserving clinical relevance.

Reductions in anxiety symptoms were also more pronounced in the EG, especially in domains related to separation and social anxiety. These effects may be attributable to MEBE’s integration of breath control, attention training, and somatic anchoring techniques, which may disrupt maladaptive cognitive-affective cycles frequently observed in anxious children^34–36. Interestingly, somatic and panic symptoms demonstrated smaller group differences. This likely reflects that MEBE emphasized attention training and emotional regulation rather than interoceptive exposure or body-symptom desensitization. Future protocols may consider adding specific modules targeting panic and somatic complaints to broaden clinical benefits. This profile is consistent with pediatric mindfulness trials that show larger effects on cognitive–affective anxiety than on panic/somatic complaints, and with TS-focused mindfulness studies reporting mixed or small effects on panic/somatic domains.

Enhancements in perceived quality of life—particularly in school functioning and overall quality of life—further support the intervention’s efficacy in improving daily adaptive functioning. Physical and social functioning showed trend-level improvements but did not reach statistical significance. These benefits may reflect gains in executive functioning, emotion regulation, and behavioral flexibility fostered by MEBE’s structured progression and home-based reinforcement³⁷. Gamified rewards and parental involvement likely contributed to sustained motivation and transfer of learned skills to real-world settings. Notably, emotional functioning showed minimal change and did not improve significantly, despite reductions in anxiety symptoms. One explanation may be that the PedsQL emotional subscale is less sensitive to domain-specific anxiety improvements, particularly in separation and social anxiety, which were most affected by MEBE. Alternatively, MEBE may primarily target cognitive–behavioral regulation and situational anxiety rather than broader emotional resilience. Longer interventions or supplementary modules may be required to enhance emotional quality of life outcomes. This observation aligns with prior findings that emotional resilience often follows behavioral self-regulation and may depend on sustained interpersonal support^38,39.

Importantly, the MEBE intervention presents several advantages over conventional cognitive-based therapies. Unlike verbally mediated CBT or static meditation practices, MEBE emphasizes active, body-oriented engagement, which may be more accessible for younger children with limited cognitive capacity or attentional stamina^40,41. Its phase-based delivery model, standardized instructor training, and implementation flexibility suggest strong potential for future adaptation in educational and clinical settings. Detailed session protocols and training resources are included in the Appendix I to facilitate replication.

As briefly noted in the Methods, MEBE was designed with features distinct from yoga or Tai Chi, with further conceptual and procedural distinctions elaborated below. Beyond cognitively oriented comparators, it is also important to distinguish MEBE from movement-based contemplative practices. While yoga and Tai Chi can improve motor coordination and reduce anxiety, MEBE differs in its target, structure, and dose. It is a brief, manualized protocol for pediatric TS in which mechanisms are mapped a priori to CSTC regulation and premonitory-urge monitoring through breath-paced down-regulation, interoceptive labeling, and urge–response decoupling. The movement vocabulary relies on short, low-amplitude sequences delivered in 60–90-second sets to accommodate tics and limited attentional stamina; sessions last approximately 20 min and follow a phased schedule with daily 10-minute home micro-practice to support adherence. Age-tiered adaptations (4–6 vs. 7–12 years) and caregiver copractice are embedded, and fidelity is supported by standardized scripts, checklists, and adherence logs. Conceptually, MEBE functions as a “regulation primer” that can be layered with or precede HRT/CBT, whereas yoga and Tai Chi are broader lifestyle practices that typically require longer instruction and a steeper learning curve. A component-level summary is provided in Supplementary Table S4.

Limitations and future directions

Despite promising outcomes, several limitations warrant consideration.

First, reliance on self- and parent-report, without physiological or neurocognitive indices (e.g., heart-rate variability, functional neuroimaging), limits objectivity and mechanistic inference^42,43. Although assessors were blinded, families were not, introducing expectancy bias for parent-reported outcomes (e.g., PedsQL); active-control designs are recommended. Because separation and social anxiety declined markedly, tic improvement may have been partly anxiety-mediated; mediation analyses integrating psychological and physiological markers are warranted. Despite age-appropriate administration (oral delivery, pictorial anchors, comprehension checks), self-report in 4–7-year-olds remains a limitation; adding behavioral tasks, clinician ratings, and physiological indices would strengthen validity.

Second, the broad age range (4–12 years) was chosen to capture the typical onset and developmental course of TS, yet younger children’s cognitive and attentional capacities differ substantially from those of older participants. Although age-appropriate adaptations were built into the protocol, future studies should conduct subgroup analyses or develop age-specific modules to further refine developmental fit.

Third, the three-month follow-up may be insufficient to evaluate durability in a chronic, relapsing condition such as TS; longer-term longitudinal studies (≥ 12 months) are needed to assess maintenance of gains and relapse trajectories. Future iterations will pilot a targeted somatic/panic add-on module—combining extended paced breathing, graded interoceptive exercises, and symptom-specific psychoeducation—and include longer follow-up to evaluate generalization to these domains.

Fourth, while the intervention was implemented in hospital-based environments by trained professionals, its generalizability to real-world contexts (e.g., schools, community centers, remote settings) remains uncertain. Exploring delivery through digital platforms or hybrid modalities will be essential⁴⁴ for expanding accessibility, particularly in underserved or resource-constrained regions⁴⁵.

Fifth, common comorbid conditions were present at baseline, as detailed in Supplementary Table S2. Although randomization was stratified by age and baseline tic severity, the trial was not prespecified to conduct subgroup analyses by age group or comorbidity. The sample size was also not powered to test comorbidity-by-treatment interactions. Future trials should include stratified analyses across age ranges and common comorbidities to clarify differential intervention effects. To mitigate potential confounding, we conducted exploratory covariate-adjusted analyses including baseline comorbidity indicators, and the primary inferences were unchanged. Reporting the comorbidity profile enhances interpretability and external validity. Future trials should prespecify comorbidity stratification and be adequately powered for moderator analyses across common cooccurring conditions to delineate whether MEBE exerts differential effects across clinically relevant subgroups⁴⁶.

Sixth, the sex ratio in our cohort was approximately 1:1, which differs from the male predominance reported in epidemiological studies. This likely reflects clinic-based referral patterns and the inclusion of children with functionally impairing symptoms. As such, generalizability to population-level sex distributions may be limited. Future trials should prespecify sex-stratified recruitment and be adequately powered for sex-specific moderator analyses.

Finally, ensuring fidelity of implementation across diverse settings remains a challenge. Future work should apply established dissemination and implementation science frameworks to assess adherence, contextual adaptation, and scalability. While the structured manualized protocol enhances reproducibility, the requirement for repeated hospital-based sessions may limit scalability. Future adaptations should explore briefer formats, school-based delivery, or digital platforms such as app-guided or telehealth modules to increase accessibility, especially in low-resource settings.

Clinical implications

The present findings suggest that MEBE may be particularly beneficial for children with TS who exhibit heightened anxiety, as reductions were most pronounced in separation and social anxiety domains. Accordingly, MEBE could serve not only as an adjunctive treatment but also as a preparatory intervention to strengthen self-regulation capacities prior to more cognitively demanding approaches such as HRT or CBT. From a safety perspective, no participants in this trial experienced tic exacerbations attributable to mindfulness or movement practices. This aligns with prior literature indicating that such events are rare, but ongoing monitoring remains essential in both clinical and research applications.

Competing interests

The authors declare no competing interests.

Institutional review board statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Medical Ethics Committee of Xinxiang Medical University (XYLL-2020-279-01(K), 09/05/2020). Written informed consent was obtained from all participants and/or their legal guardians for publication of identifying information and/or images in an online open-access publication.

Informed consent statement

 Informed consent was obtained from all subjects involved in the study.