Effects of Mindfulness-Based Interventions on Promoting Athletic Performance and Related Factors among Athletes: A Systematic Review and Meta-Analysis of Randomized Controlled Trial

n recent years, mindfulness-based interventions (MBIs) have been widely applied in competition sports with respect to athletic performance and mental health promotion, whereas evidence of randomized controlled trials (RCTs) has not been well summarized. Therefore, this study aimed to systematically review and meta-analyze the existing evidence on the effects of MBIs on improving athletic performance, mindfulness level, mindfulness-related psychological components (e.g., acceptance, self-compassion, ow), and mental health (e.g., burnout, stress, psychological well- being) among athletes. Following the PRISMA guidelines, a literature search was implemented on ve electronic databases (Web of Science, PubMed, Scopus, ProQuest, and ScienceDirect) and relevant review papers. The article selection, risk of bias assessment, and data extraction were performed by two investigators independently. The standardized mean difference (SMD) was calculated to evaluate the effects of interventions using the random effect model. Among the 1897 original hits, thirty-two eligible RCT studies were included in the systematic review, of which seven were involved in the meta-analysis. The results showed that MBIs were effective in promoting athletes' athletic performances (by narrative synthesis), mindfulness-level (n = 3; SMD = 0.50, 95% CI = [0.17, 0.83]; I 2 = 45%,p= 0.16), and mindfulness-related psychological components (n

Among the 1897 original hits, thirty-two eligible RCT studies were included in the systematic review, of which seven were involved in the meta-analysis. The results showed that MBIs were effective in promoting athletes' athletic performances (by narrative synthesis), mindfulness-level (n = 3; SMD = 0.50, 95% CI = [0.17, 0.83]; I 2 = 45%,p= 0.16), and mindfulness-related psychological components (n = 5; SMD = 0.81, 95% CI = [0.53, 1.10], I 2 = 77%,p=0.001), while no signi cant intervention effects were found on the mental health of athletes (n = 4; SMD = 0.03, 95% CI = [ 0.35, 0.29], I 2 = 89%,p< 0.001). Our ndings preliminarily support the potential effectiveness of MBIs, whereas more high-quality RCTs were needed in the future. Keywords: mindfulness; athletic performance; athletes; randomized controlled trial; psychological components; mental health 1. Introduction In competitive sports, pursuing optimal athletic performance is the ultimate goal for athletes [1]. During past decades, sport psychologists have developed diverse psy- chological training approaches, endeavoring to help athletes promote their competitive performance [2]. As a traditional stream of psychological training, cognitive behavioral therapies (CBTs) have been widely applied and shown effectiveness in the promotion of athletic performance among athletes [3]. CBT emphasizes the self-control of internal states through the training of psychological skills, with a view of improving motor perfor- mance [3,4]. However, an increasing group of evidence has shown that negative effects might be caused by controlling, eliminating, or suppressing athletes' internal states [5]. Some studies have also indicated that goal-oriented behaviors might be replaced by exces- sive cognitive activities [5]. Int. J. Environ. Res. Public Health2023,20, 2038.

Int. J. Environ. Res. Public Health2023,20, 2038 2 of 16 To address the limitations of traditional CBTs, mindfulness-based interventions (MBIs) have raised increasing concerns, which have shown a great potential for promoting ath- letes' competitive performances as an alternative of traditional psychological training approaches [6]. The term of mindfulness is derived from Eastern Zen [7], which is de ned as non-judgmental, conscious, purposeful focus, understanding, and acceptance of the things around one in the present moment [8]. In contrast to the traditional behavioral therapies, mindfulness emphasized that individuals should be aware of and experience the internal and external thoughts and feelings without any judgments [9]. In the sports domain, mindfulness has been used to cultivate athletes' abilities to accept and realize, to inspire athletes to effectively control their attention and improve their body perception ability [8–10]. Mindfulness underlines the process from perceiving to experiencing the constant changes of current things and one's own emotions without the inhibition and control of undesirable responses in order to remove the emotional state of pervious worries, avoid performance failure under pressure, and adjust the mental state of athletes [11]. Mindfulness provides a potential not only to improve performance but also to decrease the risks of mental health problems [11]. With increasing understandings of the mechanisms of mindfulness with different outcomes, a series of mindfulness training approaches have been developed, such as Mindfulness-Acceptance-Commitment (MAC) training, Mindfulness-Based Stress Reduc- tion (MBSR) training, and Mindfulness-based Cognitive Therapy (MBCT) [12–15]. Mind- fulness has shown to be positively associated with a wide range of indicators among athletes, including greater athletic performance, better mindfulness, and related psycho- logical components (e.g., acceptance, ow, and psychological exibility) and lower risks of mental health problems (e.g., stress, anxiety, depression, and burnout) [9,11–15]. Two previous review and meta-analysis studies compiling evidence from diverse types of studies (e.g., RCT, case study, and observational) provided preliminary support for the effectiveness of mindfulness-based interventions (MBIs) on improving athletes' sports performance [9,12,16]. Furthermore, previous review papers have also examined the ef- fects of MBIs on the mindfulness-related psychological components (i.e., ow) and mental health among speci c athlete

previous review and meta-analysis studies compiling evidence from diverse types of studies (e.g., RCT, case study, and observational) provided preliminary support for the effectiveness of mindfulness-based interventions (MBIs) on improving athletes' sports performance [9,12,16]. Furthermore, previous review papers have also examined the ef- fects of MBIs on the mindfulness-related psychological components (i.e., ow) and mental health among speci c athlete samples (e.g., elite athletes and long-distance runners) [17–19]. However, none of previous review studies focused only on the evidence from RCTs (the rigorous study design), which may be due to the limited numbers of studies. In addition, the above review studies did not provide a comprehensive overview of the effectiveness of MBIs on a wide range of outcomes (i.e., performance, psychological components, and mental health) targeting all types of athletes (e.g., elite, collegiate, and retired). Therefore, the present study aimed to systemically review and meta-analyze existing RCT evidence on the effects of MBIs in improving athletic performance, mindfulness, and mindfulness-related psychological components and mental health among athletes. It is ex- pected that the research findings may contribute to future mindfulness research, interventions, and policymaking, with the aim of promoting sport performance and health among athletes. 2. Materials and Methods 2.1. Search Strategy and Study Selection The current systematic review and meta-analysis were conducted following the Pre- ferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines (Supplementary Materials); the review protocol has been prospectively registered on PROS- PERO (CRD42022299940). Based on a prede ned literature search strategy, the following ve electronic databases were searched using Boolean logic's multi- led search format with no limit to the date range: Web of Science, PubMed, Scopus, ProQuest, and ScienceDirect. In addition, the reference lists of relevant review papers were also checked. We combined three groups of keywords in the search: (1) mindfulness; (2) sport* OR athlete* OR player* OR athletic* OR exercise* OR training; and (3) performance OR achievement. The literature search was limited to human participants and the publication language was limited to English.

three groups of keywords in the search: (1) mindfulness; (2) sport* OR athlete* OR player* OR athletic* OR exercise* OR training; and (3) performance OR achievement. The literature search was limited to human participants and the publication language was limited to English.

Int. J. Environ. Res. Public Health2023,20, 2038 3 of 16 Following the Population, Intervention, Comparison, Outcome, and Study design (PICOS) principles, the study selection criteria included: Population:any types of athletes without limits to the demographics and sports levels (e.g., collegiate, recreational/amateur, professional, elite, retired, or handicapped) were included. Intervention:any types of mindfulness-based interventions (MBI) without limits to the training types and components (e.g., traditional Mindfulness-Acceptance-Commitment intervention, Mindfulness-Based Stress Reduction intervention, mindfulness-based yoga, or mindfulness-based cognitive therapy), frequency, and duration were included. Comparison:any control conditions without mindfulness components (e.g., non- treatment control, waiting-list control, and active control using CBT) were eligible. Outcome:objectively or subjectively measured athletic performance was the primary outcome and secondary outcomes included the mindfulness level, mindfulness-related psy- chological components (e.g., acceptance, self-compassion, psychological flexibility, flow, and ruminative response) and mental health indicators (e.g., anxiety, stress, and depression). Study design:due to the large number of studies found in the literature search after protocol submission, we decided to limit our analysis to randomized controlled trials (RCTs) as this type of study provided the highest standard of evidence. All the RCT designs were included, while quasi-experimental, observational, and case studies were excluded. In addition, book chapters, editorial letters, commentaries, conference proceedings, qualitative articles, and review articles were not eligible for inclusion. All the initially searched articles were exported into the Mendeley to remove the dupli- cations. Two investigators (W.Y. and F.J.J.) independently screened the titles and abstracts of all the remaining articles, where all clearly irrelevant articles were deleted. Afterwards, the full texts of all the identified articles were checked for eligibility by two investigators independently according to the aforementioned inclusion criteria. During the review process, any disagreements were addressed by consensus or by involving a third investigator (SML). 2.2. Data Extraction A standardized form was used to extract data from the full texts of eligible articles by one investigator (WY). A second investigator (FJJ) checked the accuracy of the data extrac- tion. All the information including authors, publication year, study design, participants, sample size, mean age, intervention duration, frequency, measures, and main ndings were extracted to the Excel sheet for

Extraction A standardized form was used to extract data from the full texts of eligible articles by one investigator (WY). A second investigator (FJJ) checked the accuracy of the data extrac- tion. All the information including authors, publication year, study design, participants, sample size, mean age, intervention duration, frequency, measures, and main ndings were extracted to the Excel sheet for the following narrative synthesis and meta-analysis. 2.3. Risk of Bias Assessment The study quality of included studies was evaluated using the Cochrane risk-of- bias tool in Review Manager software (RevMan, version 5.4, Cochrane Collaboration, Oxford, UK). Seven aspects of biases, including random sequence generation (selection bias), concealment of allocation sequence (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective outcome reporting (reporting bias), and others, were assessed. The overall quality of the study was categorized as low risk (“low risk” in all aspects), unclear risk (at least one aspect with “some concerns of biases”), and high risk (at least one aspect with “high risk” or several aspects with “some concerns”). The study quality was evaluated by two investigators (W.Y. and F.J.J.) independently and any discrepancies were resolved by consensus or con rmation of a third investigator (SML). 2.4. Statistical Analysis The meta-analyses were implemented using RevMan 5.4 (Cochrane Collaboration, Oxford, UK) if at least three studies presented the same exposure with adequate information for the effect size estimates [20–22]. The standardized mean difference (SMD) with 95% con dence intervals (CI) was extracted from the pre- and post-intervention. For studies that reported other effect sizes (e.g., odds ratio and Cohen'sf 2 ), the SMDs were calculated using a conversion spreadsheet [22,23]. Pooled effect sizes were estimated using a random

Int. J. Environ. Res. Public Health2023,20, 2038 4 of 16 effects model and were presented in forest plots. The heterogeneity was assessed using the I 2 , with I 2 values of 25%, 50%, and 75% indicating low, medium, and high heterogeneity, respectively [24]. The publication bias was evaluated using funnel plots and Egger's regression tests [25,26]. The subgroup analyses were implemented on study quality and demographics in case the number of included samples is no less than 10 [22,27,28]. The statistical signi cance level was set asp< 0.05 for effect sizes estimation andp< 0.10 for heterogeneity and publication bias assessments [27,28]. 3. Results 3.1. Study Characteristics Figure the databases search and identi ed an additional ve articles from other sources. After the duplicates were removed, 1897 articles were screened by two investigators at the title and abstract level and the full texts of 66 potential articles were assessed. Finally, we included 32 articles after removing ineligible articles.Int. J. Environ. Res. Public Health 2023, 20, x FOR PEER REVIEW 5 of 18 Figure 1. PRISMA flow chart was used to identify of the include studies. The feature coding results show that the studies included in this review came from a wide range of countries and territories: seven studies reported data from Iran; five from the USA; three each from Switzerland, Brazil, and Sweden; two each from Portugal, India, Australia, China, and Germany; and one from Norway. Figure 1.PRISMA ow chart was used to identify of the include studies.

Int. J. Environ. Res. Public Health2023,20, 2038 5 of 16 The feature coding results show that the studies included in this review came from a wide range of countries and territories: seven studies reported data from Iran; ve from the USA; three each from Switzerland, Brazil, and Sweden; two each from Portugal, India, Australia, China, and Germany; and one from Norway. The characteristics of studies included in this review are presented in Table. All these 32 studies were RCT studies. A total sample of 1788 athletes from a variety of sports were included, with an average age of 23.67 years. The athletes were involved in basketball, football, shooting, volleyball, wushu, cycling, track and eld, and other sports. There was also a range of athletic experience from amateur to elite international athletes, with most studies including athletes competing at university level or higher. All the included trails were published after 2009, ranging from 2010 to 2022. The average intervention period was 6.3 weeks and the intervention training period was different in different programs. The intervention programs mainly include MAC, MBCT, MBSR, brief mindfulness training, and other mindfulness-based intervention types. For the risk of bias assessment, 71% of the included studies had a low risk of bias, while six studies (29%) were rated as a high risk of bias. Most studies showed a high risk of bias with respect to the allocation concealment (22%), blinding of participants and personnel (29%), incomplete outcome data (29%), and selective reporting (19%) (Figure).Int. J. Environ. Res. Public Health 2023, 20, x FOR PEER REVIEW 6 of 18 The characteristics of studies included in this review are presented in Table 1. All these 32 studies were RCT studies. A total sample of 1788 athletes from a variety of sports were included, with an average age of 23.67 years. The athletes were involved in basket- ball, football, shooting, volleyball, wushu, cycling, track and field, and other sports. There was also a range of athletic experience from amateur to elite international athletes, with most studies including athletes competing at university level or higher. All the included trails were published

of sports were included, with an average age of 23.67 years. The athletes were involved in basket- ball, football, shooting, volleyball, wushu, cycling, track and field, and other sports. There was also a range of athletic experience from amateur to elite international athletes, with most studies including athletes competing at university level or higher. All the included trails were published after 2009, ranging from 2010 to 2022. The average intervention pe- riod was 6.3 weeks and the intervention training period was different in different pro- grams. The intervention programs mainly include MAC, MBCT, MBSR, brief mindfulness training, and other mindfulness-based intervention types. For the risk of bias assessment, 71% of the included studies had a low risk of bias, while six studies (29%) were rated as a high risk of bias. Most studies showed a high risk of bias with respect to the allocation concealment (22%), blinding of participants and per- sonnel (29%), incomplete outcome data (29%), and selective reporting (19%) (Figure 2). Figure 2. Risk of bias assessment for the included studies. Figure 2.Risk of bias assessment for the included studies.

Int. J. Environ. Res. Public Health2023,20, 2038 6 of 16 Table 1.Characteristics of included studies (n = 32). Study Study Design Population (n) Age (Mean SD) Duration/Frequency Measures Findings Wolch et al. (2021) [29] RCT Male recreational basketball players (n = 32) 21.22 2.01 yrs. Two sessions CSAI-2R, TMS, MAAS A 15 min mindfulness intervention appears to have some effects on participants rst basketball free-throw shot and state anxiety when performing under pressure. Shaabani et al. (2019) [30] RCT Experienced male basketball players (n = 72) 28.6 4.0 yrs. Two sessions SAS-2, DISE, DSS, PANAS, CHIME, BSCS, EDMC, TMS A brief mindfulness intervention mitigates the effects of ego depletion in a basketball free-throw task. Dehghani et al. (2018) [31] RCT The women's basketball team of both Iran and Tehran University of Medical Sciences (n = 31) 23.44 0.49 yrs.; 22.34 0.34 yrs. Eight sessions SPQ, AAQ, SCAT Mindfulness-Acceptance-Commitment-based approach is an effective intervention to increase athletic performance and reduce experiential avoidance and sports anxiety in athletes. Gross et al. (2018) [13] RCT Women's basketball players from an NCAA Division III university athletic department in the northeastern United States (n = 22) Nil 7 weeks CCAPS-62, AAQ-II, SPQ, WAI-S, DERS, MAAS, MC-C The MAC is an effective intervention for the mental health and sport performance needs of female collegiate athletes. Josefsson et al. (2019) [32] RCT Competitive elite athletes (n = 69) 20.9 4.17 yrs. 7 weeks AMQ, DERS MAC approach is a more effective intervention compared to the PST condition in reducing emotion regulation dif culties, as well as enhancing sport-relevant mindfulness skills and perceived athletic training performance in elite sport. Röthlin et al. (2016) [33] RCT Elite athletes from 23 different sports (n = 133) 23.68 6.12 yrs. Nil CHIME, CAI-T, Self-generated three-item measure Trait mindfulness is related to fewer performance worries and prevents the remaining worries from in uencing athletes' behavior, thereby helping them to perform better. John et al. (2011) [34] RCT Male elite level shooters (n = 96) 29.5 4.3 yrs. 4 weeks Measure of shooting accuracy or shooting score MMT may decrease PCS and will

CHIME, CAI-T, Self-generated three-item measure Trait mindfulness is related to fewer performance worries and prevents the remaining worries from in uencing athletes' behavior, thereby helping them to perform better. John et al. (2011) [34] RCT Male elite level shooters (n = 96) 29.5 4.3 yrs. 4 weeks Measure of shooting accuracy or shooting score MMT may decrease PCS and will enhance PS. It is concluded that 4 weeks of MMT has an effect on HPA-Axis by decreasing the level of SC as a reliable physiological marker of PCS. Kachanathu et al. (2013) [35] RCT Male elite level shooters (n = 110) 29.5 4 yrs. 4 weeks Measure of shooting accuracy MMT may decrease PCS and therefore enhance SP. Glass et al. (2019) [36] RCT University athletes (n = 52) 19.32 1.25 yrs. 6 weeks DASS-21, SWLS, FFMQ, AAQ-II, DFS-2, SAS, Mindfulness practice summary, SRF, CRF Signi cant increases in ow, trait mindfulness, satisfaction with life, and self-rated sport performance, along with reductions in worry. Röthlin et al. (2016) [33] RCT Elite, sub-elite, and recreational athletes (n = 108) Nil 5 weeks CAI-S, TOQS, TOPS, FFMQ-SF, AMQ, SEC-27, AAQ-II, EQ, DSS, ANT, TEOSQ, RSC, SCS, and BSI-18 Both PST and MBI are expected to help improve functional behavior in athletes. Mehrsafar et al. (2019) [37] RCT Elite Wushu athletes (n = 26) 25.4 2.4 yrs. 8 weeks CSAI-2R and MAAS Mindfulness-based intervention might be associated with a diminished physiological and psychological stress responses to competition. Carraça et al. (2019) [38] RCT Elite soccer athletes (n = 57) 25.79 3.3 yrs. 8 weeks AAQ-II, FFMQ, DFS-2, SCS, WBSI, BSI, Athlete's—FAIP-A MBSoccerP can be effective in enhancing elite soccer performance, self-compassion, psychological exibility, mindfulness, and ow.

A systematic review and meta-analysis of randomized controlled trials on mindfulness-based interventions for athletes.