Abstract
Objective:
This study aims to perform a systematic review and meta-analysis of the effect of MBIs (Mindfulness, Tai Chi, Yoga, and Qigong) on symptoms and executive function (EF) in ADHD.
Method:
PubMed, Web of Science, the Cochrane Library, PsycINFO, CINAHL, Embase, and CNKI databases were searched to collect randomized controlled trials (RCTs) on the effects of MBIs on symptoms and EF in ADHD. Data extraction and methodological quality evaluation were conducted by two researchers, and a meta-analysis was conducted by Stata SE.
Results:
The pooled meta-analyses of MBIs revealed a positive and small effect on inattention (g = −0.26), hyperactivity/impulsivity (g = −0.19), and EF (g = −0.35).
Conclusion:
Results suggest that MBIs have a significant improvement relative to the control condition. Although some results show that symptoms are affected by age, interventions, and total time of moderators, while EF is not affected by age and measurement, it needs to be supported by more research evidence. (J. of Att. Dis. XXXX; XX(X) XX-XX)
Introduction
ADHD is one of the most common neurobehavioral disorders in childhood (Wolraich et al., 2019), and can be a lifelong condition (Thapar et al., 2017), which is characterized by developmentally inappropriate levels of inattention or hyperactivity and impulsivity (Biederman & Faraone, 2005; Faraone & Larsson, 2019). In 2015, a meta-analysis of 175 studies estimated the worldwide prevalence of ADHD in subjects aged ≤18 years to be 7.1% (Thomas et al., 2015). Individuals with ADHD can raise the risk of educational and occupational failure, other psychiatric disorders, criminality, accidents, addictions, and social disability in their lifetime (Faraone et al., 2015).
Executive function (EF) refers to a set of top-down mental processes involved in controlling thoughts and goal-oriented behaviors, including inhibition, cognitive flexibility, and working memory (Diamond, 2013). Previous studies have shown that ADHD is associated with impaired EF (Craig et al., 2016; Doyle, 2006). A meta-analysis of 83 studies reported that inhibitory control and working memory are common impaired EF domains in ADHD (Willcutt et al., 2005). In addition, EF deficits associated with ADHD persist from childhood to adulthood (Shuai et al., 2021) and have been related to impairments in academic and social function (Tamm & Nakonezny, 2015). EF are highly correlated with the prefrontal cortex (Yuan & Raz, 2014). Currently, ADHD individuals have structural and functional brain abnormalities in the frontal cortex, cingulate cortex, basal ganglia, thalamus, and primary motor cortex (McLeod et al., 2014; Qiu et al., 2011; Zang et al., 2007). Further, dysfunction of the neurotransmitter catecholamines has also been found in people with ADHD (Prince, 2008). The etiology of ADHD can be explained by the above neurophysiological disorders.
Until now, the first-line pharmacologic treatment for ADHD has been stimulants, which can enhance dopamine and norepinephrine by inhibiting dopamine transporters and NE transporters (Cortese, 2020), but they may also have adverse effects such as low appetite, insomnia, stomachaches, headaches, and cognitive rigidity (Barkley et al., 1990; Lee et al., 2011; Levy, 2013). Harmless and effective alternative or complementary therapies are urgently needed. Yoga, Tai Chi, Qigong, and Mindfulness as mediation-based mind-body interventions (MBIs) have been widely used to improve general health (Huston & McFarlane, 2016; Perry-Parrish et al., 2016; Sivaramakrishnan et al., 2019; Zou et al., 2018). Breathing technologies, controlled physical movements, full body stretching, and a mediation component are the common characteristics of these therapies (Vancampfort et al., 2021). A study investigated the prevalence rate of interventions in the United States in the past 20 years. It collected data from three different time periods: 1995, 2005, and 2015. The result shows that the prevalence rates among 2,262 community adults are 20%, 17%, and 18%, respectively (Bhattacharyya et al., 2020). Studies have confirmed that MBIs can reduce chronic pain (Majeed et al., 2018), enhance cognitive function (Wu et al., 2019), and relieve depression symptoms (Bo et al., 2017). Studies suggest that MBIs seem particularly suited to address the deficits associated with ADHD (Evans et al., 2018). An RCT study confirmed that 6 months of mindfulness training can improve the symptoms, EF, and learning problems of children with ADHD compared with waiting conditions (Valero et al., 2022). Another RCT study found that, compared with waiting conditions, mindfulness-based cognitive therapy can improve the symptoms, anxiety, and depression of ADHD college students (Gu et al., 2018). The study reviewed the effects of a mind-body intervention on ADHD, which showed positive effects on symptoms and EF, and the intervention was low-cost, easy to carry out, and had no adverse side effects (Herbert & Esparham, 2017).
At present, the existing meta-analysis focused on a single mindfulness intervention, and studies have confirmed that mindfulness has a positive impact on ADHD symptoms (Cairncross & Miller, 2020; Poissant et al., 2020; Xue et al., 2019). From the comprehensive perspective of Tai Chi, Qiong, yoga, and mindfulness, there is still a lack of corresponding meta-analysis. One study (Chimiklis et al., 2018) analyzed the impact of yoga, mindfulness, and meditation on ADHD, and the other (Zuo et al., 2020) analyzed the impact of Tai Chi and yoga on ADHD. However, there were few studies included, and the design and methodology of the studies included were not robust, which affected the robustness of the research findings and reduced their generalizability. Based on the above shortcomings, this study uses systematic review and meta-analysis to evaluate the effects of MBIs on symptoms and EF in ADHD, to provide evidence for clinical practice. This study hypothesized that MBIs had a positive effect on ADHD symptoms and EF.
Methods
This study was conducted following the preferred reporting items for Systematic Reviews and Meta-analyses statement (Liberati et al., 2009). The protocol for this study was registered with PROSPERO (CRD42022353747).
Eligibility Criteria
The inclusion criteria were: (1) subjects had a diagnosis of ADHD according to the International Classification of Diseases (ICD) edition or Diagnostic and Statistical Manual of Mental Disorders (DSM), and there was no age limit on the subjects; (2) the interventions included mindfulness, Qigong, Tai Chi, and yoga, or supplemented with usual treatment such as medication; (3) the control conditions were manual treatment, medication, sitting, reading, psychoeducation, and waiting list control; and (4) the primary outcome was ADHD symptoms, including inattention and hyperactivity/impulsivity. The evaluators could be self-reported, parents, teachers, and researchers; the secondary outcome was EF, which consisted of inhibition control, cognitive flexibility, and working memory. Assessment can be done through questionnaires or behavioral tasks. (5) the study design was a randomized controlled study. The exclusion criteria were: (1) non-randomized controlled trial; (2) non-peer-reviewed articles; (3) due to language restrictions, we excluded non-Chinese and non-English studies; (4) meeting summary; and (5) outcome data cannot be converted to mean and standard deviation.
Literature Search
Data sources used in this systematic review and meta-analysis included PubMed, Web of Science, the Cochrane Library, PsycINFO, CINAHL, Embase, and China National Knowledge Infrastructure (CNKI). The retrieval time limit was from the inception of each database to July 31, 2022. There was no language restriction for this search. The key phrases and Medical Subject Heading (MeSH) terms we used were as follows: Mind-Body Therapies, Mind-body practice, Mind-body exercise, traditional exercise, Taiji, Qigong, Chi kung, Yoga, Mindfulness, Meditation, Mindfulness training, Attention Deficit Disorders with Hyperactivity, Attention Deficit/Hyperactivity Disorder, Hyperkinetic Syndrom, Impulsivity, Hyperactivity, Attention Deficit, ADHD, ADD, Randomized Controlled Trials as Topic, Randomized Controlled Trial. Additionally, the reference lists of the retrieved literature were tracked, and the relevant meta-analysis literature was read to comprehensively collect the relevant literature.
Screening and Data Extraction
According to the above retrieval strategy, the first author (ZP) imports the obtained literature into EndNote X9 software and establishes a document library. Firstly, two researchers (XL and WJ) removed duplicate literature. Second, based on the inclusion and exclusion criteria of the literature, two researchers (XL and WJ) read the title and abstract for preliminary screening and excluded literature that did not meet the criteria. Thirdly, two researchers (XL and WJ) downloaded the full text for secondary screening, determined the literature to be included, and extracted the data. Data extraction includes (1) basic information included in the study (i.e., first author, publication time, country, and study design); (2) the basic characteristics of the subjects (i.e., age, sample size, diagnostic criteria, and ADHD type); (3) specific details of the intervention; (4) outcome measures and data (mean and standard deviation); (5) risk of bias and key elements of literature quality evaluation. Two researchers (XL and WJ) extracted data before cross-checking. They were resolved through negotiation with the first author (ZP) when discrepancies were encountered.
Risk of Bias Assessment
The Physiotherapy Evidence Database Scale (Cashin & McAuley, 2020) was used to evaluate the methodological quality of the included literature. There were 11 items (eligibility criteria, randomization, allocation concealment, baseline similarity, blinding of subjects, blinding of therapists, blinding of evaluators, dropout rate, intention-to-treat-analysis, between-group-comparison, and point measures). Items 2 to 11 were scoring items, 1 point was given if they meet the standard, and 0 points were given if they do not meet or are unclear. The total score on this scale was 10. Referring to previous studies, the quality level was divided into three categories: high quality ≥6, moderate quality = 4 to 5, and low quality ≤3 (Fang et al., 2019). The literature quality score was independently conducted by two researchers (XL and WJ). If there were any discrepancies in allocation concealment and blinding, the first author (ZP) discussed them together until a consensus was reached.
Statistical Analysis
The data were input into Stata SE (version 14.0) software for meta-analysis. The data included the sample size of the two groups (experimental group and control group), and the mean and standard deviation of the combination of pre-test and post-test in the two groups, separately. In the included study outcomes: (1) if there is no standard deviation, it was converted into standard deviation by the corresponding formula; (2) in the evaluation of symptoms in children, the parent as evaluator was the first choice, and the teacher was the second choice; In the evaluation of symptoms in adults the first choice was an observer or investigator evaluation, and the second choice was self-evaluation; (3) if there were multiple dependent variables in an EF task, the outcome of the more executive demanding condition was included (e.g., reaction time in the flanker task); and (4) we tended to use intent to treat (ITT) data, if ITT data was presented in the included studies. Due to the small sample size, Hedges’ g was used to measure the effect size regarding ADHD symptoms and EF. The magnitude of Hedges’ g values was interpreted as 0.8 for large effect, 0.5 for medium effect, and ≤0.2 for small effect (Hedges, 1981).
Between-trial heterogeneity was evaluated using the Q statistic and the I2 statistic. The Q statistic indicates whether heterogeneity exists, and the I2 statistic quantifies the magnitude of heterogeneity. The I2 values defined the proportion of heterogeneity across the studies as small (≤25%), medium (50%), and large ( ≥75%; Higgins & Thompson, 2002). Considering the heterogeneity of sample characteristics, intervention measures, and evaluation methods of outcomes, we used a random effect model for pooled analysis (Vysniauske et al., 2020). The summary results of each study were presented in the form of forest plot. Furthermore, we used planned subgroup analysis to explore the effects of a moderator (i.e., age, interventions, and total time) on symptoms and EF. We verified the reliability and stability of our meta-analysis results by sensitivity analysis (removing each study one by one). In addition, publication bias was examined by funnel plot and Egger’s linear regression method.
Results
Screening Results
The flow of records over the review is summarized in Figure 1. We retrieved a total of 1,543 reports or articles that might be eligible. A total of 342 duplicate works of literature were excluded. Based on the title and abstract, we excluded 1,132 articles. In the remaining 69 works, 50 were excluded because they were non-randomized controlled studies, lack of required data, non-Chinese and non-English works, and meeting abstracts. Finally, 19 RCTs (Amiri et al., 2022; Bachmann et al., 2018; Behbahani et al., 2018; Chen & Cheng, 2016; Converse et al., 2020; Fritz & O’Connor, 2022; Gu et al., 2018; Hepark et al., 2019; Hoxhaj et al., 2018; Janssen et al., 2019; Kiani et al., 2017; Lo et al., 2020; Mitchell et al., 2017; Moritz et al., 2021; Rezaei et al., 2018; Robe & Dobrean, 2022; Schoenberg et al., 2014; Siebelink et al., 2022; Valero et al., 2022) met the inclusion conditions and were synthesized for effectiveness.

PRISMA flow chart depicting study search and selection process.
Study Characteristics
The 19 RCTs were all peer-reviewed journals, with 18 studies published in English and 1 study in Chinese, and the time range was from 2014 to 2022. A total of 1,018 subjects participated in the 19 eligible studies, of which 9 studies were children/adolescents, and 10 studies were adults. Concerning MBIs, 15 studies were mindfulness (Amiri et al., 2022; Bachmann et al., 2018; Behbahani et al., 2018; Gu et al., 2018; Hepark et al., 2019; Hoxhaj et al., 2018; Janssen et al., 2019; Kiani et al., 2017; Lo et al., 2020; Mitchell et al., 2017; Moritz et al., 2021; Robe & Dobrean, 2022; Schoenberg et al., 2014; Siebelink et al., 2022; Valero et al., 2022), 2 studies were Tai Chi (Chen & Cheng, 2016; Converse et al., 2020), and 2 studies were yoga (Fritz & O’Connor, 2022; Rezaei et al., 2018). Regarding the intervention total time, one study was a one-time intervention of 10 minutes (Robe & Dobrean, 2022), one study was not reported in detail (Hepark et al., 2019), and the other studies were all 480 minutes or more. Of the 19 included studies, 14 studies reported the outcome of inattention; 15 studies reported outcome measures of hyperactivity/impulsivity; 10 studies reported EF (questionnaires or behavioral tasks). See Table 1 for specific research characteristics.
Summary of Characteristics of Included Studies.
Note. NR = not reported; IG = intervention group; CG = control group; CPRS = Conners’ Parent Rating Scale; SNAP-IV = Swanson, Nolan, and Pelham-IV rating scales; ADHD-I = Inattention; ADHD-H = Hyperactivity/Impulsivity; ADHD-C = Combined; MAP = Mindful awareness practices; MPT = Mindful parenting training; MBCT = Mindfulness-based cognitive therapy; MMT = Mindfulness meditation training; FBMI = Family-based mindfulness intervention; DBT-ST = Dialectical Behavior Therapy Skill Training; MBI = Mindfulness-based intervention; TAU = Treatment as usual; CAARS-O = Conners Adult ADHD Rating Scale-Observer rating; CAARS-S = Conners Adult ADHD Rating Scale-Self-rating; CAARS-INV = Conners Adult ADHD Rating Scale-Investigator rating; CTRS = Conners’ Teacher Rating Scale; CPRS = Conners’ Parent Rating Scale; CASS-C = Current ADHD Symptoms Scale-Clinician; ASRS = ADHD Self Rated Scale; CBCL = Child Behavior Checklist; BRIEF-A = Behavior Rating Inventory of Executive Function-Adult; BRIEF-P = Behavior Rating Inventory of Executive Function-Parent; SWAN = The Strengths and Weaknesses of ADHD Symptoms and Normal Behaviors Rating Scale; CPT = Continuous performance test; SCWT = Stroop Color-Word Interference Test; Conners-3 = Conners-third edition.
Risk of Bias Assessment
The risk of bias assessment of all included studies is presented in Figure 2. All included studies fulfilled at least four criteria, with an average score of 5.6, in which eight studies were high quality (≥6), and 11 studies were moderate quality (4–5). All studies were in line with random allocation, similar at baseline, between-group comparison, and point measures. A 4 studies used allocation concealment, 7 applied assessor blinding, 14 met dropout rate ≤15%, and 5 followed intent-to-treat analysis. However, none of the studies succeeded in blinding participants and therapists due to the challenges of performing double-blind procedures in non-pharmacological studies (Liang et al., 2021; Table 2).

Forest plot of the meta-analysis of inattention.
Methodological Quality Assessment for Included Studies.
Note. Yes = 1; No or unclear = 0.
Meta-Analysis of Effects of MBIs on ADHD Inattention
Fourteen studies (Amiri et al., 2022; Bachmann et al., 2018; Behbahani et al., 2018; Converse et al., 2020; Gu et al., 2018; Hepark et al., 2019; Hoxhaj et al., 2018; Janssen et al., 2019; Lo et al., 2020; Mitchell et al., 2017; Moritz et al., 2021; Schoenberg et al., 2014; Siebelink et al., 2022; Valero et al., 2022) concentrating on inattention were integrated into the meta-analysis. As shown in the forest plot (Figure 2), there was a positive and small effect of MBIs (g = −0.26, 95% CI [−0.40 to −0.12], p < .001) relative to the control condition, and with better homogeneity (I2 = 0.0%, p = .514). Three subgroup analyses were conducted. Full statistics are presented in Table 3. In the age group, subgroup analysis showed a positive and small-to-medium size effect of MBIs for trials that tested against an adults’ group (g = −0.36, 95% CI [−0.54 to −0.18], p < .001). However, MBIs intervention did not reach significance in a group of children/adolescents (g = −0.12, 95% CI [−0.34 to 0.10], p = .293). In the intervention group, mindfulness studies had a significant and small effect (g = −0.28, 95% CI [−0.42 to −0.14], p < .001). Meanwhile, the effect of the Tai Chi intervention did not achieve significance (g = 0.53, 95% CI [−0.48 to 1.54], p = .306). In the total time group, subgroup analysis showed that there was a significant and small-size effect of MBIs in the more than 1,000-minute group (g = −0.35, 95% CI [−0.57 to −0.12], p = .002). However, the effect of MBIs intervention did not reach significance in the less than 1,000-minute group (g = −0.14, 95% CI [−0.34 to 0.06], p = .166).
Meta-Analytic Findings in Subgroup Analyses.
Meta-Analysis of Effects of MBIs on ADHD Hyperactivity/Impulsivity
Fifteen studies (Amiri et al., 2022; Bachmann et al., 2018; Behbahani et al., 2018; Chen & Cheng, 2016; Converse et al., 2020; Gu et al., 2018; Hepark et al., 2019; Hoxhaj et al., 2018; Janssen et al., 2019; Lo et al., 2020; Mitchell et al., 2017; Moritz et al., 2021; Schoenberg et al., 2014; Siebelink et al., 2022; Valero et al., 2022) assessed hyperactivity/impulsivity as the outcome. The results showed that the MBIs had a significant and small effect on hyperactivity/impulsivity (g = −0.19, 95% CI [−0.33 to −0.06], p < .001) compared to the control condition (Figure 3), and with better homogeneity (I2 = 0.0%, p = .858). Three subgroup analyses were conducted. Full statistics are presented in Table 3. In the age group, subgroup analysis resulted in a positive and small effect of MBIs for trials that tested against a children/adolescents’ group (g = −0.25, 95% CI [−0.46 to −0.04], p = .018). A MBIs intervention, however, did not reach significance in the adults’ group (g = −0.15, 95% CI [−0.33 to 0.03], p = .097). In the intervention group, mindfulness-based interventions had a significant and small effect (g = −0.18, 95% CI [−0.32 to −0.04], p = .011). Meanwhile, the effect of the Tai Chi intervention did not achieve significance (g = −0.37, 95% CI [−0.92 to 0.18], p = .192). In the total time group, subgroup analysis showed that there was a significant and small effect of MBIs in a less than 1,000-minute group (g = −0.28, 95% CI [−0.47 to −0.09], p = .004). However, the effect of MBIs intervention did not reach significance in the more than 1,000-minute group (g = −0.11, 95% CI [−0.33 to 0.11], p = .324).

Forest plot of the meta-analysis of hyperactivity/impulsivity.
Meta-Analysis of Effects of MBIs on EF
Ten studies (Converse et al., 2020; Fritz & O’Connor, 2022; Hepark et al., 2019; Janssen et al., 2019; Kiani et al., 2017; Mitchell et al., 2017; Rezaei et al., 2018; Robe & Dobrean, 2022; Schoenberg et al., 2014; Siebelink et al., 2022) were incorporated into the meta-analysis. As shown in the forest plot (Figure 4), there was a positive and small-to-medium size effect of MBIs on EF (g = −0.35, 95% CI [−0.53 to −0.17], p < .001) relative to the control condition, and the test of heterogeneity revealed a small heterogeneity across included studies (I2 = 3.5%, p = .408). Four subgroup analyses were conducted. Full statistics are presented in Table 3. In the age group, subgroup analysis revealed significant point estimates for studies with children/adolescents (g = −0.41, 95% CI [−0.81 to −0.02], p = .04) and for studies with adults (g = −0.35, 95% CI [−0.58 to −0.13], p = .002). In the intervention group, subgroup analysis showed significant point estimates for studies with mindfulness intervention (g = −0.32, 95% CI [−0.50 to −0.14], p = .001). Meanwhile, there was no statistical difference both yoga intervention (g = −0.53, 95% CI [−1.35 to 0.28], p = .199) and Tai Chi intervention (g = −1.04, 95% CI [−2.11 to 0.03], p = .058). In the total time group, subgroup analysis showed significant point estimates for studies that used less than 1,000 minutes (g = −0.48, 95% CI [−0.86 to −0.10], p = .014). The effect of MBIs intervention did not reach significance in the group with over 1,000 minutes (g = −0.20, 95% CI [−0.48 to 0.07], p = .15). In the measurement group, subgroup analysis produced significant point estimates for studies with questionnaire (g = −0.26, 95% CI [−0.52 to −0.01], p = .043) and behavioral task (g = −0.50, 95% CI [−0.78 to −0.22], p = .001).

Forest plot of the meta-analysis of EF.
Sensitivity Analysis
In terms of inattention, hyperactivity/impulsivity, and EF, the results showed that the effect size was within 95% CI of the total effect size after one-by-one removing studies, indicating that the impact on the total combined effect size is small and acceptable. Therefore, the meta-analysis results were relatively stable and reliable.
Publication Bias
Through visual inspection of the funnel plot in Figures 5 to 7, it was found that there was no obvious asymmetry on both sides. Using Egger’s linear regression, further detection results showed no publication bias (p = .496 for inattention; p = .096 for hyperactivity/impulsivity; p = .098 for EF).

Funnel plot for inattention studies.

Funnel plot for hyperactivity/impulsivity studies.

Funnel plot for EF studies.
Discussion
To our knowledge, this systematic review and meta-analysis are the first to evaluate the effects of meditation-based mind-body interventions on ADHD symptoms and EF. Nineteen RCTs were included in this study. For the included studies, symptoms were assessed with questionnaires, while the EF assessment included questionnaires or behavioral tasks. The results of the meta-analysis showed that compared with the control condition, MBIs had a small to medium effect on inattention, hyperactivity/impulsivity, and EF in ADHD individuals. There was greater homogeneity among the studies, indicating that MBIs can improve symptoms and EF in ADHD. From the perspective of movement, developing attention control and inhibition ability is a common feature of MBIs. In the process of practicing Tai Chi, it is necessary to keep your attention on movement and breathing, and at the same time make movements that require inhibition. In the process of practicing yoga, practitioners are required to master certain poses and breathing skills, while mindfulness training requires learning to control attention and focus on specific purposes or actions (Herbert & Esparham, 2017), which are helpful for improving ADHD symptoms and EF. It has been found that mindfulness-based cognitive training increases the amplitude of No Go-P3 at Pz electrode in ADHD adults (Schoenberg et al., 2014), which may be related to enhancing the brain activation of ADHD. In addition, aerobic exercise can induce the increase of serotonin, dopamine, brain-derived neurotropic factor (BDNF), and brain blood flow in ADHD (Chan et al., 2022), while Tai Chi and yoga in MBIs also have certain characteristics of aerobic exercise. The author speculates that they may have the same mechanism as aerobic exercise, but further research is needed to confirm.
This study is consistent with existing meta-analyses that have used mindfulness-based interventions (Cairncross & Miller, 2020; Poissant et al., 2020), meditation based-intervention (Zhang et al., 2018), Tai Chi, and yoga intervention (Zuo et al., 2020). It is worth noting that this study has the following advantages: (1) the analyzed studies are all RCTs, the study design has strong causality, and the results are relatively objective; (2) as for the evaluators of the questionnaire, we first choose parent evaluation or teacher evaluation, and then choose self-report. Studies have shown that the results reported by parents are more predictive than those self-reported in the field of ADHD (Barkley et al., 2002); and (3) the quality of the included RCTs in the study is moderate to high, and the methodological quality evaluation results are high, which makes up for the limitations of previous studies that did not report the quality of the included studies or the quality of the included studies was poor.
Through horizontal comparison of different therapies, it is more helpful for us to determine the value of MBIs. In the aspect of medication treatment of ADHD symptoms, one study (Castells et al., 2011) found that methylphenidate has an almost medium effect size (SMD = 0.49), and another study (Cunill et al., 2013) concluded that atomoxetine has a small to medium effect (SMD = −0.33). As far as medication for executive function in ADHD, one study (Tamminga et al., 2016) showed that methylphenidate had a small to medium effect on response inhibition (SMD = 0.40) and working memory (SMD = 0.24). In terms of neurofeedback therapy, one study (Riesco-Matías et al., 2021) found that neurofeedback therapy had a small to medium effect on inattention in children with ADHD (SMD = −0.33), while hyperactivity/impulsivity had a small effect (SMD = −0.17). The results of this study show that MBIs have a small to medium effect on inattention, hyperactivity/impulsivity, and executive function. However, because MBIs have no side effects (Herbert & Esparham, 2017) and are economical, they should receive more attention and may become a complementary or alternative therapy.
The subgroup analysis of symptoms showed that MBIs had a significant improvement effect on inattention in adults, but there was no statistical difference in hyperactivity/impulsivity. Cairncross and Miller (2020) indicated that this may be related to the developmental changes of adults. With the age of ADHD individuals, inattention would become more obvious, while hyperactivity/impulsivity tends to decrease. Therefore, the improvement effect of MBIs on inattention may be better than hyperactivity/impulsivity in adults. Conversely, MBIs promote hyperactivity/impulsivity with statistical significance, while inattention has no statistical significance. This is a very interesting finding, which may be related to the effect of pharmacology. The most effective treatment for inattention is medication (Storebø et al., 2015), which may lead to less room for improvement in MBIs. Some of the participants in this study used medications. In one study (Siebelink et al., 2022), 80% of the subjects used medications. This is also speculation, and more studies are needed to confirm the effect of MBIs on inattention in children. Furthermore, mindfulness showed a positive intervention effect on inattention and hyperactivity/impulsivity, while the effect of Tai Chi had no statistical significance. This may be due to the included studies. There was only one study on inattention and two studies on hyperactivity/impulsivity, which did not meet the requirements of a combined analysis. Therefore, it is suggested that more Tai Chi interventions could be included in future studies to confirm its effectiveness. Within this total time group, more than 1,000 minutes had a statistically significant effect on inattention, but less than 1,000 minutes did not show any statistical significance. In the subgroup of less than 1,000 minutes, there was one study (Converse et al., 2020) in which the black diamond square fell on the right side of the invalid line, which led to no statistical difference in the subgroup. The reason was that the study showed a low attendance rate, and the subjects only participated in less than half of the courses. This may have affected the effectiveness of the intervention. On the contrary, fewer than 1,000 minutes had a statistically significant intervention effect on hyperactivity/impulsivity, while more than 1,000 minutes had no statistical significance. In the subgroup of more than 1,000 minutes, it was found that the black diamond square of two studies (Hoxhaj et al., 2018; Moritz et al., 2021) fell to the right of the invalid line, which led to no statistical significance in this subgroup. Further analysis of these two studies showed that both studies used psychoeducation as the content of the control group. Despite the fact that one of them was usual treatments, the main focus was on psychoeducation (Moritz et al., 2021). In this study, it was found that mindful awareness practices and psychoeducation were beneficial to ADHD adults, and there was no significant difference between the two intervention methods (Hoxhaj et al., 2018). This study believes that psychoeducation is also helpful for ADHD hyperactivity/impulsivity, and reduces the intervention effect of MBIs to a certain extent. We should conduct a large-scale, high-quality study in the future to determine whether MBIs have a significant improvement effect over psychoeducation. In addition, we should also explore the impact of MBIs with different training time on inattention and hyperactivity/impulsivity in ADHD, which is particularly critical to provide comparative evidence of different training time directly.
In the subgroup analysis of EF, it was found that the intervention effect of MBIs was not affected by age moderators. It was beneficial for children/adolescents and adults, reflecting the value of MBIs for all ages. In terms of the intervention group, mindfulness therapy produced results consistent with symptoms of inattention and hyperactivity/impulsivity, confirming its positive effects once again. In addition, there was no statistical difference between Tai Chi and yoga interventions. Then, because there are few studies included, including only one in the Tai Chi intervention subgroup, the results need careful explanation and more research to confirm. In the total time group, we got a similar result with the hyperactivity/impulsivity symptom, that is, the subgroup with less than 1,000-minute showed the intervention effect of EF, and the difference was significant, while the subgroup with more than 1,000-minute showed no statistical significance. In general, this is an interesting finding. In the subgroup with more than 1,000 minutes, we did not find any studies crossing the invalid line. Maybe a shorter intervention time can provide the best improvement in EF, or perhaps the subjects in the subgroup of less than 1,000 minutes have a lot of room for improvement. Restricted by inclusion into the studies, the studies only divided the intervention total time into two groups. After that, it can be divided into three subgroups: short time, medium time, and long time, which will be used to determine which short time option is best. In the measurement subgroup, both the questionnaire and the behavior task showed a significant intervention effect. The effect size of the more objective behavior task subgroup reached medium, which means that the improvement result of MBIs on EF is relatively stable, further increasing people’s confidence in MBIs to improve EF.
This study also has the following limitations: (1) In the subgroup analysis, the yoga and Tai Chi studies were fewer to be included, which reduced the robustness of the analysis results. This is related to the fewer studies about yoga and Tai Chi implemented in the RCT design. It is suggested that future studies can include yoga and Tai Chi intervention studies with different study designs, and its effectiveness can be confirmed by increasing the number of studies. (2) In the subgroup analysis of EF, we did not carry out subgroup analysis of inhibition, cognitive flexibility, and working memory due to a lack of outcome indicators. It is suggested that future research should examine the impact of MBIs on different subcomponents of EF. (3) It is difficult to implement a blind method in non-pharmacological intervention experiments. Among the RCTs included in this study, only some studies used single blinding, which is the main factor restricting the methodological evaluation. It is recommended that future studies should improve the implementation of blinding and the methodological quality of the study design. (4) In this study, there were different proportions of subjects taking medication in most of the included studies, which may have mixed effects on the intervention. However, because only one study showed that no medication was used, no subgroup analysis was performed. It is proposed that future studies investigate the differences between medication users and non-medication users. (5) This study did not search for grey literature, that is, unpublished literature. There may be deviations from the literature search. It is suggested that aspiring researchers use a comprehensive literature search strategy to avoid publication bias.
Conclusion
This study provides preliminary evidence for the treatment of ADHD by reviewing the effects of MBIs on ADHD symptoms and EF. This meta-analysis confirmed that MBIs, such as mindfulness, Tai Chi, and yoga, had small to medium intervention effects on ADHD symptoms and EF compared with the control condition. Although the results of the subgroup analysis suggest that age, interventions, and total time may be important moderators affecting MBIs intervention on the symptoms (inattention, hyperactivity/impulsivity) of ADHD, while interventions and total time may be moderators affecting EF, more research needs to be conducted to support these findings.
Footnotes
Author Contributions
Zeping Zhang, Suyong Yang, and Guangsheng Zhao developed the idea of systematic review and meta-analysis. The literature search and data analysis were conducted by Zeping Zhang. The screening and data extraction were conducted by Xiaolong Chang and Weijing Zhang. The first draft of the manuscript was written by Zeping Zhang, and all the authors made critical modifications to the manuscript. All authors read and approved the final manuscript.
Declaration of Conflicting Interest
The author(s) declared that there was no potential conflict of interest regarding the research, authorship and/or publication of this article.
Ethical Approval
The protocol for this study was registered with PROSPERO (CRD42022353747).
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
