Effects of physical exercises on balance in children with down syndrome: a systematic review and meta-analysis

Abstract

Objective

This study aims to systematically explore the effects of different physical exercises on balance ability in children with Down syndrome and to clarify the differences in the effects of various physical exercises on promoting the development of balance ability in children with Down syndrome.

Methods

The database search period was from July 1, 2024, to August 31, 2024. PubMed, Embase, Scopus, Web of Science, and Engineering Village databases were searched for all published studies involving randomized controlled trials from database establishment to August 31, 2024. The screening procedure complied with the stated inclusion and exclusion criteria and adhered strictly to the established inclusion and exclusion criteria. Using the Cochrane risk of bias assessment tool, the methodological quality of the chosen trials was determined. RevMan5.4.1 was used for data analysis, whereas Stata 16.0 was used to assess publication bias.

Results

This systematic review and meta-analysis incorporated eight randomized controlled trials comprising 260 participants. Meta-analysis showed that physical exercise significantly improved the dynamic balance (SMD = 1.45, 95%CI [1.14, 1.76], p < 0.00001) and total balance (SMD = 1.44, 95%CI [1.04, 1.85], p < 0.00001) in children with Down syndrome.

Conclusion

Physical exercise can improve balance in children with Down syndrome. Among the various exercise modalities, treadmill and core stability training appear to be among the more promising interventions, while Pilates and isokinetic training may also offer beneficial effects.

Registration

The study protocol has been duly registered with PROSPERO (CRD42024557710).

Supplementary Information

The online version contains supplementary material available at 10.1186/s13102-025-01222-2.

Introduction

Down syndrome (DS), also known as trisomy 21, is a genetic disorder caused by the presence of an extra copy of chromosome 21 [1, 2]. Its global prevalence is estimated at 1/800 to 1/1000 live births, though rates vary depending on access to prenatal screening and differences in health systems [3–5]. Trisomy 21 typically arises from nondisjunction during meiosis, a process strongly associated with increased maternal age [6–9].

Children born with Down syndrome typically develop their motor development at a slower pace than those without this condition [10]. As a result, they tend to acquire basic motor milestones—such as sitting, crawling, and walking—at a later age than typically developing children. Research suggests that this delay is mainly due to low muscle tone and joint laxity [11]. Most children with DS do eventually learn to walk, although typically at a later age [12, 13]. They manifest poor coordination and bad results in running and jumping-related activities. In addition, poor health may slow down the child’s capacity to exercise. For instance, vision problems also worsen the balance and coordination in this group of children [7, 14]. Consequently, children with DS often struggle to acquire new motor skills and to maintain postural balance and control.

Due to low motor capacity and inadequate coordination, children with DS require more significant support from their caregivers while conducting daily tasks, such as dressing, washing, and feeding [15]. If their balance is impaired, they are more likely to have a fall, which puts them at a higher risk for injury [13]. Diminished motor competence may adversely affect children’s social engagement, as such limitations often preclude sustained peer synchronization and developmentally appropriate participation in collaborative play activities [12]. It appears that the therapeutic benefit of physical exercise in children with DS becomes evident when the exercise is vigorous and the duration is sufficient [16, 17]. Physical exercise widens the range of activity and enhances balance [11, 12]. Physical exercise increases the adrenaline level in children with DS, and it can significantly reduce levels of depression and anger [18]. Some studies have shown that physical exercise can promote cognitive development in children with DS [14, 15, 19]. Exercise is closely related to cognitive development, and physical exercise can indirectly promote improving learning ability [20]. These studies highlight the importance of designing and implementing physical exercise for children with DS.

While most studies support the benefits of physical exercise, there is still debate about which exercise regimen is significantly effective [17, 21, 22]. Moreover, there’s no consensus on the intensity and duration of exercise that works best [16]. In addition, further research is still needed to ensure the durability of the intervention effect. Therefore, future studies are needed to more precisely measure intervention’s type, intensity, and duration and account for individual differences and long-term follow-up.

This study aims to systematically evaluate the effects of different types of physical exercise on balance in children with DS, with particular attention to exercise modality, intensity, and duration. The findings are intended to inform clinical practice by supporting the development of tailored and evidence-based intervention programs, and to identify directions for future research.

Methods

This study adheres to the PRISMA statement guidelines for systematic reviews and meta-analyses in its research methodologies selection and application [23]. The study protocol has been duly registered with PROSPERO (CRD42024557710).

Literature search

The database search period was from July 1, 2024, to August 31, 2024. PubMed, Embase, Scopus, Web of Science, and Engineering Village databases were searched for all published studies involving randomized controlled trials from database establishment to August 31, 2024. The search strategy incorporated a mix of keywords and Medical Subject Headings (MeSH) terms such as “Balance,” “Down syndrome,” “children.” Boolean operators “AND” and “OR” were used to amalgamate subject terms and free-text words (Table 1).

Table 1.

Search strategy

Databases	Search terms
PubMed	(“Child“[MeSH Terms] OR “children“[MeSH Terms]) AND (“down syndrome“[MeSH Terms] OR “down’s syndrome“[Title/Abstract] OR “Mongolism“[Title/Abstract] OR “trisomy 21“[Title/Abstract] OR “trisomy g“[Title/Abstract] OR ((“partial“[All Fields] OR “partials“[All Fields]) AND “trisomy 21 down syndrome“[Title/Abstract])) AND (“Balance“[Title/Abstract] OR “postural control“[Title/Abstract] OR “postural stability“[Title/Abstract] OR “Equilibrium“[Title/Abstract] OR “motor coordination“[Title/Abstract] OR “fall prevention“[Title/Abstract])
Embase	(child/exp OR child*:ti, ab, kw) AND (‘down syndrome’/exp OR ‘down syndrome’:ti, ab, kw OR ‘down’s syndrome’:ti, ab, kw OR ‘mongolism’:ti, ab, kw OR ‘trisomy 21’:ti, ab, kw OR ‘trisomy g’:ti, ab, kw OR (‘partial’:ti, ab, kw AND ‘trisomy 21 down syndrome’:ti, ab, kw)) AND (balance: ti, ab, kw OR ‘postural control’:ti, ab, kw OR ‘postural stability’:ti, ab, kw OR equilibrium: ti, ab, kw OR ‘motor coordination’:ti, ab, kw OR ‘fall prevention’:ti, ab, kw)
Scopus	(TITLE-ABS-KEY(child OR children) AND TITLE-ABS-KEY(“down syndrome” OR “down’s syndrome” OR mongolism OR “trisomy 21” OR “trisomy g” OR (partial AND “trisomy 21 down syndrome”)) AND TITLE-ABS-KEY(balance OR “postural control” OR “postural stability” OR equilibrium OR “motor coordination” OR “fall prevention”))
Web of Science	TS=(child* OR children) AND TS=(“down syndrome” OR “down’s syndrome” OR mongolism OR “trisomy 21” OR “trisomy g” OR (partial AND “trisomy 21 down syndrome”)) AND TS=(balance OR “postural control” OR “postural stability” OR equilibrium OR “motor coordination” OR “fall prevention”)
Engineering Village	((child OR children) WN KY) AND ((“down syndrome” OR “down’s syndrome” OR mongolism OR “trisomy 21”) WN KY) AND ((balance OR “postural control” OR “postural stability” OR equilibrium) WN KY)

The search outcomes underwent several rounds of pre-screening, supplemented by manual searches to enhance the automated process. References and related articles were also scrutinized to identify additional sources that met the eligibility criteria. We included only literature published in English, and grey literature was excluded from our review.

Inclusion and exclusion criteria

The selection criteria for this study were as follows: (1) Participants: children under 12 years of age with a clinical diagnosis of DS; (2) Study Design: randomized controlled trials (RCTs); (3) Intervention Type: physical exercise, body practice, or exercise intervention supplementing the control group’s treatment; (4) Outcomes: total balance and dynamic balance, assessed using balance scales such as the Bruininks-Oseretsky Test of Motor Proficiency (BOTMP), Bruininks-Oseretsky Test of Motor Proficiency Second Edition (BOT-2), PBS (Pediatric Balance Scale), and Biodex Balance System (BBS). Studies were excluded if they were non-RCT studies or duplicate publications. (5) Language: only studies published in English were included; (6) Literature Type: only peer-reviewed journal articles were included; conference abstracts, unpublished reports, and dissertations were excluded.

Studies were excluded if they: (a) were not randomized controlled trials; (b) were duplicate publications; (c) lacked sufficient outcome data on balance; or (d) did not clearly describe the intervention protocol or assessment instruments.

Study selection

The two independent reviewers independently screened the titles and abstracts of all retrieved records based on the predefined inclusion and exclusion criteria. Full-text articles were obtained for studies that appeared to meet the eligibility criteria or where eligibility was unclear. Although the reviewers were not blinded to the study authors or journals, efforts were made to minimize selection bias through independent and standardized screening. Discrepancies between two independent reviewers in the implementation process were resolved with the involvement of the corresponding author.

Data extraction

The two reviewers independently conducted the data extraction process using a standardized data extraction form. The following details were meticulously collected: (1) fundamental study information, which encompassed the author’s name, year of publication, country of origin, trial design, and participant characteristics; (2) experimental characteristics, including the type of exercise, duration, frequency, and intervention cycle; (3) outcome measures.

Risk of bias assessment

The risk of bias was evaluated using a standardized instrument from the Cochrane Collaboration that included a randomized method of allocation, maintenance of confidentiality, implementation of blinding, and possible outcomes, including loss of data from the study and selective reporting of research results. Each item was graded as high, low, or unclear risk of bias [24, 25]. In terms of bias risk, the possible thumbs down could affect the results because of the incompleteness of the random sequence generation or allocation concealment, which results in selection bias. If allocation concealment is insufficient, it can affect the reliability of the results. Impartiality of the research can suffer from the lack of blinding to either related or unrelated parties during intervention implementation or outcome evaluation, which can result in treatment implementation or output assessment. Missing data may have affected the outcome of the study if they felt it might have affected the intervention outcome. Selective reporting of certain studies, the factual part on which the review is based, can lead to publication bias and, therefore, not present a complete picture of future studies of relative fields. Funding sources or guidelines might affect the integrity of the review results. There was a risk of bias evaluation by two independent researchers, and the discrepancy was resolved with the involvement of the corresponding author.

Qualitative analysis

The two independent researchers performed the methodological quality assessment of all research papers in line with the scale for randomized clinical trials from the Physiotherapy Evidence Database (PEDro) [26]. The PEDro reformulated the scale designed to dream up the quality of any treatment or intervention study based on randomized studies, blinding, attrition, characteristics of study design, and statistics. The PEDro scale scoring standard aims to evaluate all ten items separately by assigning “yes” as 1 and the case “no” or “unclear” as 0, respectively. The maximum score of ten items will be 10 for all the 11 criteria. Low scoring can suggest injury from the risk of bias, which is determined from the information drawn from the abstract background. Less than 5 out of 10 item marking can be considered high, or over 5 out of 10 grade shipping can be considered low [27]. If there are gaps or uncertainties in the details within the article, then we will count it as a “risk of bias” in clarity and ask the corresponding author for more details on the writing part. If the corresponding author does not receive a clear answer within 10 working days, an article will be rated 0.

Data analysis and synthesis

Unqualified data were obtained, and the analysis was performed using RevMan 5.4.1, with standardized mean difference (SMD) and 95% confidence intervals (CIs) calculated for continuous variables. SMD was selected as the effect size due to variability in outcome measurement tools across the included studies, allowing for comparison on a common scale. The heterogeneity of the data was judged with the help of a p-value and I² statistic. A high heterogeneity was suggested when p < 0.01 and I² > 50%; random-effects model data were introduced. Conversely, low heterogeneity, indicated by a p-value > 0.01 and an I² < 50%, warranted using a fixed-effects model for the meta-analysis. The potential for publication bias was assessed using Egger’s regression test and funnel plots, conducted in Stata 16.0 software.

Results

Search results

The literature search and screening process is illustrated in the flowchart presented in Fig. 1. The initial database search yielded a total of 1,059 studies, which was reduced to 566 after the removal of duplicates. A comprehensive screening of titles, abstracts, and full texts was conducted, and studies that failed to meet the inclusion criteria were excluded. Consequently, eight articles were ultimately included in the meta-analysis.

Fig. 1.

PRISMA flowchart of included and excluded studies

Characteristics of the included studies

The characteristics of the included studies are shown in Table 2. Eight RCTs involving 260 DS patients aged 4–13-year-old were included in this review. The included studies were performed in four countries, namely Saudi Arabia [28, 29], Egypt [30–33], India [34], and Turkey [35]. Participants were recruited from schools, communities, or hospitals. All the studies listed inclusion and exclusion criteria for a participant’s diagnosis of DS. The intervention included standard physical therapy plus physical exercise, including weight-bearing exercises, Pilates, treadmill training, and core stability training. The control group received the same physical therapy as the experimental group, although the therapy protocol varied slightly among the studies. The frequency of these interventions was 1 to 3 times per week, lasting 30 to 90 min. The duration of the intervention ranged from 6 to 24 weeks. The included studies described the specific content of the intervention. Outcome measures used in the meta-analysis included total balance and dynamic balance. Similar studies were evaluated using a variety of measurement tools, including BBS, PBS, BOTMP, BOT-2 and Berg Balance Scale. Of the studies included in the meta-analysis, six studies used BBS for evaluation [28–33], two studies used BOTMP [30, 34], one study used BOT-2 [31], one study used PBS [35], and one study used Berg Balance Scale [28].

Table 2.

Characteristics of the inclusion studies

ID	Study	Country	Sample size (E1/E2/C)	Gender (F/M)	Age (years)	Intervention arm	Time (min)	Frequency (times/week)	Cycle	Control arm	Instruments
1	Abdel rahman, 2010 [30]	Egypt	13/13	EG:8/5 CG: 7/6	EG:4.56 ± 0.44 CG:3.92 ± 1.16	Weight Bearing exercises + PT	EG:30 + 30 CG:60	—	6weeks	Physical therapy	1.BOTMP 2.BBS
2	AL-Nemr, 2024 [31]	Egypt	20/20	EG: 12/8 CG:14/6	EG:8.86 ± 0.58 CG:9.16 ± 0.48	Pilates exercises + PT	EG:45 + 15 CG:90	3	12weeks	Physical therapy	1.BOT-2 2.BBS
3	Alsakhawi, 2019 [28]	Saudi Arabia	15/15/15	—	4.59 ± 0.53	Treadmill exercise program/Core stability exercise training + PT	EG1:30 + 30 EG2:30 + 30 CG:60	3	8weeks	Physical therapy	1.Berg Balance Scale 2.BBS
4	Azab, 2022 [29]	Saudi Arabia	16/15	EG: 7/9 CG:4/11	EG:9.19 ± 0.75 CG:8.60 ± 0.98	Trampoline-Based SSC exercises + PT	EG:45 + 15 CG:45	2	12weeks	Physical therapy	BBS
5	Eid, 2017 [32]	Egypt	15/16	EG: 7/8 CG:7/9	EG:10.26 ± 0.79 CG:10.05 ± 0.68	Isokinetic training + PT	EG:45 + 15 CG:60	3	12weeks	Physical therapy	BBS
6	Eid, 2015 [33]	Egypt	15/15	EG: 7/8 CG:6/9	EG:8.93 ± 0.7 CG:9.26 ± 0.79	Whole-Body vibration training + PT	EG:60+(5–10) CG:60	3	24weeks	Physical therapy	BBS
7	Gupta, 2011 [34]	India	12/11	EG: 4/8 CG:5/6	EG:13.17 ± 0.84 CG:12.60 ± 1.26	Progressive resistive exercises for lower limbs and balance training	—	3	6weeks	Regular activities	BOTMP
8	Kaya, 2023 [35]	Turkey	17/17	EG: 13/4 CG:6/11	EG:10.12 ± 3.30 CG:8.18 ± 2.74	Hippotherapy program + PT	EG:30 + 30 CG:30	3 (HG:1)	6weeks	Physical therapy	PBS

BOTMP = Bruininks-Oseretsky Test of Motor Proficiency; BOT-2 = Bruininks-Oseretsky Test of Motor Proficiency Second Edition; BBS = Biodex Balance System; PBS = Pediatric Balance Scale; PT = Physical therapy; HG = Hippotherapy program; “-” = not reported. EG: experimental group; CG: control group

Risk of bias

All the eight studies included in this review were RCTs. However, one study had an inadequate description of the randomization method [34]. The remaining studies provide comprehensive details about random allocation and allocation concealment methods. one study did not report whether blinding was performed [30], while the other studies adequately implemented blinding. All eight studies provided complete outcome data, as shown in Fig. 2, suggesting that the overall quality of the included literature was high.

Fig. 2.

Risk of bias assessment for included studies

Quality assessment

The methodological quality of the 8 included studies was assessed using the PEDro scale. The PEDro scores of the eight studies were ≥ 6 points, indicating good quality. However, in the PEDro quality assessment, one study did not describe whether participants and therapists were blinded. Details of the records are shown in Table 3.

Table 3.

Assessment of quality of study design using PEDro

ID	Study	Eligibility and source	Random allocation	Concealed allocation	Groups similar at baseline	Participants blinding	Therapist blinding	Assessor blinding	< 15% dropouts	Intension-to- treat analysis	Between-group difference reported	Point estimate and variability reported	Total
1	Abdel rahman 2010	Y	Y	N	Y	N	N	N	Y	Y	Y	Y	6/10
2	AL-Nemr 2024	Y	Y	Y	Y	N	Y	Y	Y	Y	Y	Y	9/10
3	Alsakhawi 2019	Y	Y	Y	Y	N	Y	Y	Y	Y	Y	Y	9/10
4	Azab 2022	Y	Y	Y	Y	N	Y	Y	Y	Y	Y	Y	9/10
5	Eid 2017	Y	Y	Y	Y	N	Y	Y	Y	Y	Y	Y	9/10
6	Eid 2015	Y	Y	Y	Y	N	Y	Y	Y	Y	Y	Y	9/10
7	Gupta 2011	Y	Y	Y	Y	N	Y	Y	Y	Y	Y	Y	9/10
8	Kaya 2023	Y	Y	Y	Y	N	Y	Y	Y	Y	Y	Y	9/10

Meta-analysis

Effect of physical exercise on total balance in children with DS

Five studies (n = 168) evaluated the effect of physical exercise on total balance in children with DS [28, 30, 31, 34, 35]. Figure 3 showed high heterogeneity (I²= 82%, p < 0.0001), indicating significant differences between studies. The pooled standardized mean difference (SMD = 1.28, 95%CI [0.94,1.62], p < 0.0001) indicated a significant difference between the experimental and control groups. Sensitivity analysis identified two studies that may have been the main sources of heterogeneity [34, 35]. The meta-analysis showed a significant reduction in heterogeneity (I²= 47%, p = 0.13; SMD = 1.44, 95%CI [1.04, 1.85], p < 0.0001) after excluding these two studies.

Fig. 3.

Forest plot of the effect of physical activity on total balance. Note: Values to the right favor the physical exercise (experimental) group

In the study of Alsakhawi & Elshafey (2019), the effect size of experimental group 2 was 2.12 (SMD = 2.12, 95%CI [1.20, 3.04], p < 0.0001), experimental group 1 was (SMD = 1.98, 95%CI [1.08, 2.87], p < 0.0001), indicating that core stability and treadmill training improved balance significantly. In addition, Pilates and weight-bearing exercise also improved balance significantly. The comparison of effect sizes among different exercise regimens indicated that core stability and treadmill training seem to be the significantly effective interventions, and Pilates and weight-bearing exercise were also effective options.

Effect of physical exercise on dynamic balance in children with down syndrome

Six studies (n = 203) evaluated the effect of physical exercise on dynamic balance in children with DS. The Meta analysis showed low heterogeneity between studies (I²= 26%, p = 0.23), with pooled standardized mean differences (SMD = 1.45, 95%CI [1.14,1.76], p < 0.0001) indicating that the physical exercise intervention significantly improved dynamic balance in children with DS (see Fig. 4).

Fig. 4.

Forest plot of the effects of physical activity on dynamic balance. Note: Values to the right favor the physical exercise (experimental) group

In the study of Alsakhavi & Elshafey (2019), the effect sizes of experimental group 1 and experimental group 2 were (SMD = 2.22, 95%CI [1.28, 3.15], p < 0.00001) and (SMD = 1.95, 95%CI [1.06, 2.84], p < 0.0001), respectively, indicating that treadmill and core stability training significantly improved dynamic balance. The effect sizes of Pilates and isokinetic exercise training were (SMD = 1.83, 95%CI [1.08, 2.58], p < 0.0001) and (SMD = 1.05, 95%CI [0.29, 1.80], p = 0.007), respectively, indicating that Pilates and isokinetic exercise were effective in improving dynamic balance. In addition, the effect sizes were higher for weight-bearing exercise (SMD = 1.20, 95%CI [0.35, 2.05], p = 0.005), whole-body vibration training (SMD = 1.32, 95%CI [0.52, 2.12], p = 0.001), and trampoline exercise (SMD = 0.93, 95%CI [0.19, 1.68], p = 0.01) indicating that their improvements in dynamic balance were also significant. Thus, the comparison of effect sizes among different exercise regimens indicated that treadmill and core stability training were the significantly effective interventions, Pilates, isokinetic exercise, weight-bearing exercise, whole-body vibration training and trampoline exercise were also effective options.

Publication bias and the analysis of sensitivity

Publication bias was assessed for both outcome domains using Egger’s and Begg’s tests.

(1) Total balance: Egger’s test results (pr >|z| = 0.0666 > 0.05) and Begg’s test results (pr >|z| = 0.0894 > 0.05) suggest no statistically significant publication bias. However, the marginal result of Egger’s test (pr >|z| = 0.0666) may indicate potential small-study effects and should be interpreted with caution (see Fig. 5).

Fig. 5.

Funnel plot of the effect of physical activity on total balance

(2) Dynamic balance: Egger’s test results (pr >|z| = 0.0541 > 0.05) and Begg’s test results (pr >|z| = 0.0715 > 0.05) also did not reveal significant publication bias. Although funnel plots appear symmetrical (see Fig. 6), the limited number of included studies reduces the statistical power of these tests.

Fig. 6.

Funnel plot of the effect of physical activity on dynamic balance

In summary, while both Egger’s and Begg’s tests yielded non-significant results (p > 0.05), the small sample size may limit their ability to detect true publication bias. Therefore, these findings should be interpreted with caution, in line with recommendations for small meta-analyses [36].

Discussion

This systematic review provides an important scientific basis for the in-depth understanding of the complex relationship between physical activity and balance ability in children with DS. This meta-analysis suggests that physical exercise can significantly improve global balance, and dynamic balance in children with DS. This finding is supported by several studies included in the review, and additional implications make it possible to tentatively conclude that physical exercise remarkably contributes to the children’s health and developmental well-being with DS.

Several studies included in the meta-analysis demonstrated that children with DS had a markedly higher capacity for balance after participating in motor therapy [12, 16, 17]. In addition to concentric walking on the treadmill, which contributes extensively to balance, core stability training also emphasizes working on dynamic balance [28]. Therefore, they were very good at controlling motion. This observation aligns with earlier findings and adds to the evidence that motor therapy can be beneficial for improving motor function in children with DS [37]. The treadmill training significantly improved the balance ability among the children with DS by enhancing the lower limb’s strength level and gait coordination [11, 38, 39]. At the same time, training of the core stability also enhances the muscle groups of the trunk, which will result in an improved capability of postural control and coordination [13, 16, 17]. These findings are clinically relevant and indicate the potential of physical activity to support motor skills and balance development in children with DS.

There is also evidence that sports such as swimming, cycling, and dancing are favorable for the improvement of balance ability of children with DS. As the report explains, swimming is a whole-body exercise that effectively improves muscle strength and coordination abilities, increasing the balance level [40]. Cycling strengthens the legs and improves coordination, which is important for improving balance [41, 42]. Dance combines rhythmical and motor exercise, a very effective tool for developing balance and coordination via dynamic swing movements [43, 44]. In addition to the above motor interventions, sensory integration training can effectively improve the balance function of children with DS by integrating tactile, vestibular, and proprioceptive inputs [45, 46]. At the same time, virtual reality technology (such as Wii Fit) has shown positive effects due to its characteristics of strong interaction and high interest [22, 47]. However, considerable heterogeneity remains in the intervention protocols, particularly in terms of intensity and duration, which were not analyzed in this study and may influence outcomes. Moreover, differences in balance assessment tools (e.g., BOT-2, PBS, BBS), may affect the comparability of results across studies and contribute to measurement bias. Although many studies suggest the benefits of physical activity, no single intervention approach can yet be identified as superior, given the variations in participant characteristics, intervention parameters, and outcome measurements.

This meta-analysis integrated various exercise intervention studies to provide a holistic perspective on the effects of physical exercise on balance in children with DS. The insights obtained are of great practical value to exercise intervention research and point the way for future research and applications. Based on the current meta-analysis, follow-up studies are warranted to explore the long-term effects of different exercise interventions. Further research should focus on other elements of intervention’s effects on children with DS, for example, methods of better inclusion of the therapy into everyday education and treatment plans. In addition, future studies should systematically evaluate the influence of exercise type, intensity, durations, and outcome measurement tools to establish more robust and generalizable evidence for targeted intervention strategies in children with DS.

Limitations

We concluded that the present study has the following limitations. First, the sample of studies included in this systematic review was relatively small; therefore, we should be cautious about generalizing the findings. Second, although the included studies reported various exercise protocols, the intensity and duration of the physical exercise interventions were not quantitatively analyzed in the meta-analysis, which limits our ability to determine optimal training parameters. Third, we used different designs, variant types, instruments, and measures in conducting the study, which may have led to inconsistencies in the measurements and affected the comparability of the results. In particular, the use of different balance assessment tools (e.g., BOT-2, PBS, BBS) may introduce measurement heterogeneity, complicating direct comparisons across studies.

Finally, our study builds on the published literature, and there is a risk of omitting unpublished trials and, therefore, a risk of publication bias. Although publication bias was not statistically significant based on Egger’s and Begg’s tests (p > 0.05), these tests are known to be underpowered in small-sample meta-analyses and should be interpreted with caution. These limitations suggest that despite the results of the study meta-analysis showing that physical activity has a positive effect on balance in children with Down syndrome (DS), the current evidence remains limited, and conclusions should be interpreted with appropriate caution.

In addition, there are no RCTs from Western countries included in this study, but this does not mean that Western countries do not contribute to the motor development of children with DS. Early intervention programs have been widely implemented in Western countries. The core goal of these programs is to provide comprehensive support for infants, including physical therapy, occupational therapy, psychotherapy, and rehabilitation [48]. Future studies that draw robust conclusions must include larger samples, more diverse participants, longer follow-up periods, and harmonized study designs and assessment tools to make the findings are robust and consistent.

Conclusion

Overall, this meta-analysis provides us with preliminary evidence of the positive effects of physical exercise on balance in children with DS. These findings have meaningful implications for improving clinical practice and educational approaches, while also highlighting key directions for future research. As we continue to explore and apply exercise interventions, we can better understand and use these interventions to improve the quality of life of children with DS.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

ZL and KY: writing– original draft, methodology, and data curation. JX: formal analysis. YM: software. YD: validation. JW: supervision, writing– review and editing. JC: conceptualization, project administration, funding acquisition, and writing– review and editing. All authors contributed to the article and approved the submitted version.

Funding

This study was funded by the National Social Science Fund of China (22BTY050) and the National Key Research and Development Program (2022YFC2705201).

Data availability

The datasets used during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.