Study selection. We retrieved 83 articles through the database search. After repeated literature elimination, title reading and abstract screening to exclude non-RCTs, rigorous experimental design and unsuccessful research data, 16 articles were ultimately included, including 14 Chinese articles and 2 English articles. There were a total of 1296 experimental participants.
Characteristics of the exercise interventions
The literature inclusion criteria and exclusion criteria generated a total of 16 studies. We recorded the study characteristics of the author, the year of publication of the article, the type of study, the frequency and duration of intervention.
All included studies reflect the effects of roller skating on children and adolescents in terms of their body shape, physiological function, physical fitness and adaptability(Table 1). The most frequently used evaluation indicators, balance ability and explosive force, are used to measure changes in neuromuscular control of movement after participating in roller skating; secondary evaluation indicators include speed, endurance, sensitivity, flexibility, vital capacity, and social adaptation.
Table 1. Four aspects of physical health consisting of 9 Variables
Aspects
|
Variables
|
Body Shape
|
Body Height
|
Physical Fitness
|
Speed, Lower limb Strength, Endurance, Sensitivity, Balance, Flexibility
|
Physiological Function
|
Vital Capacity
|
Adaptive Capacity
|
Social Adaptive Capacity
|
In the 16 articles included in this study, a total of 1296 participants with mixed genders, age range of 4 to 21 years and experiment range from 8 weeks to 36 weeks. There were large differences in the intervention programmes among the studies. The duration of a single intervention ranged from 15 to 120 minutes, and most concentrated on 30 to 60 minutes. The frequency of interventions ranged from 1 to 7 times per week, and the intervention cycles also varied greatly. The shortest cycle was 8 weeks, and the longest cycle was 36 weeks. Table 2 lists the basic characteristics of each study.
Table 2. Summary of the observed data from the included studies n=1296
NO.
|
Study
|
Age
mean±SD
|
Quantity
EG/CG
|
Intervention
period
|
Exercise intensity
|
Intervention time
|
Exercise frequency
|
Variables
|
ID1
|
Chen Siping [13]
|
15-16
|
43/43
|
20 wk
|
——
|
4 min
|
3 times/week
|
①↑, ②↑, ③↑, ④↑, ⑤→, ⑥→
|
ID2
|
Chen Siping [27]
|
15-16
|
43/43
|
20 wk
|
——
|
4 min
|
3 times/week
|
⑨↑
|
ID3
|
Fan Peng[14]
|
19.21±0.37 19.34±0.37
|
F:48/50
|
36 wk
|
Medium/high intensity
|
90 min
|
3 times/week
|
①↑
|
ID4
|
Han Qi [15]
|
19.15±1.6
|
20
|
16 wk
|
Medium/high intensity
|
90 min
|
4 times/week
|
①↑
|
ID5
|
Hu Jiayi[16]
|
20.1±0.6
|
20/20
10/10
|
12 m
|
——
|
90 min
|
1 times/week
|
④↑, ①↑, ③→, ⑧↑
|
ID6
|
Jing Xiaowei [17]
|
20-22
|
23/40
|
18 wk
|
——
|
90 min
|
1 times/week
|
⑤↑, ④↑, ②→, ①↑
⑧, ⑨↑
|
ID7
|
Kong Xiangzhen
[24]
|
11.25±0.64
11.35±0.67
|
20/20
|
16 wk
|
medium intensity
|
50 min
|
1 times/week
|
⑤↑, ④↑, ②↑, ⑧↑
|
ID8
|
Liu Lexing [19]
|
4—6
|
25/27
|
20 wk
|
——
|
——
|
——
|
⑦↑, ⑤↑, ②, ④↑, ①↑, ⑨↑
|
ID9
|
Liu Zhongxi [20]
|
4—6
|
20/20
|
12 wk
|
——
|
——
|
——
|
①↑
|
ID10
|
Qi Hongchun [21]
|
19. 21±0. 37
19. 34±0. 37
|
50/50
|
36 wk
|
medium intensity
|
90 min
|
1 times/week
|
⑥↑, ②↑, ④↑, ①↑, ③↑
|
ID11
|
SunChao [26]
|
11±2.3
|
12/12
|
18 wk
|
High intensity
|
40 min
|
3 times/week
|
⑦↑, ⑧↑
|
ID12
|
Wang Hai [22]
|
20±0.3
|
F:40/40
|
15 wk
|
medium intensity
|
90 min
|
1 times/week
|
①↑
|
ID13
|
You Yang [25]
|
19-20
|
271/280,
131/140
|
16 wk
|
medium intensity
|
90 min
|
2 times/week
|
⑦↑, ⑥↑, ②↑, ③↑, ⑤↑, ⑧↑
|
ID14
|
zhang Jian[28]
|
21.32±0.78
20.88±0.72
|
42/37
|
12 wk
|
——
|
90 min
|
3 times/week
|
①↑, ⑦↑, ⑨↑
|
ID15
|
Zhou You[23]
|
5-6
|
20/20,
20/20
|
24 wk
|
——
|
---
|
|
⑦↑, ⑤↑, ④↑, ②→, ①↑
|
ID16
|
LidaZare Dizajdizi
[18]
|
14-16
|
F:10/10
/10/10
|
8 wk
|
——
|
105 min
|
——
|
①↑,③→, ⑨↑
|
Abbreviations: F=female; EG=experimental group; CG=control group; ①Balance ability, ②Flexibility, ③Sensitivity, ④Lower limb strength, ⑤Speed quality, ⑥endurance, ⑦Body Height, ⑧Vital capacity, ⑨Social Adaptabilit
Methodological quality and risk assessment of bias
Using the Cochrane risk bias assessment tool for randomized controlled trials, the evaluation results of the included 16 studies showed that there were 1 low-risk study, 14 medium-risk studies, and 1 high-risk study(12). In Fig 2, "+" means compliance, "-" means failure, and "?" means that the text has not been described in detail. Figure 3 is a statistical chart of the proportion of each item of methodological assessment.
Meta-analysis results
The full text reading of the 16 included documents revealed that the focus of the studies on the effect of roller skating on the improvement of children and adolescents' physical health varied. The performance of the same indicator used different test methods, and some research subjects had different genders. A comprehensive analysis of the 9 indicators identified in the 16 studies involved independent analysis of each indicator to analyse the effect size heterogeneity and research publication bias. The results are as follows.
Balance ability
The meta-analysis of the balance ability of children and adolescents included 17 sets of test data from 11 studies involving 960 participants[13-23]. It included 4 sets of independent men and women, and 2 sets of dynamic and nondynamic balance ability test data[14-17,22-23]. After the heterogeneity test, I2=93%, and Q-test P<0.1, suggesting that there is strong heterogeneity between the selected studies, and random effects were selected for meta-analysis. The data from this study strongly suggest that the source of heterogeneity was inconsistent test methods. Random effects meta-analysis results showed that the experimental group's balance ability improved by 1.29, Z=11.94, P<0.01, indicating that roller skating has a positive effect on improving the balance ability of children and adolescents(Fig 4).
Lower Limb Power
The meta-analysis of the strength and quality of children and adolescents was based on seven studies and 17 sets of test data involving 502 participants[13,16-17,19-21,23-24], including 3 independent test data for men and women. After heterogeneity testing, I2=78%, and Q-test P<0.1, suggesting that there was strong heterogeneity between the selected studies, and random effects were selected for meta-analysis. There is strong evidence that the source of heterogeneity is the inconsistency of strength evaluation methods and test methods. Random effects meta-analysis results showed that Z=5.55, P<0.01, indicating a significant difference between the experimental group and the control group, which shows that roller skating can improve the strength and quality of children and adolescents(Fig 5 ).
Body Height
Meta-analysis data on the effect of roller skating on the height of children and adolescents comes from 9 sets of data involving 866 participants[16,19,23-26], including 3 sets of independent data sets for men and women[16,24,26]. The height data of the experimental group and the control group increased after exercise intervention testing. The total SMD value was 0.13, the 95% confidence interval was [-0.01, 0.27], Z=1.79, P>0.05, the data were not statistically significant(Fig 6).
Flexibility
For studies of roller skating on improvement in the flexibility of children and adolescents, we performed analysis of 6 studies and 9 groups of balance ability test data, involving 771 participants[13,17,21-25], including 6 independent test data for males and females[13,17,23]. The heterogeneity test showed I2=87%, and in the Q-test, P<0.1, suggesting that there is strong heterogeneity between the studies, and random effects were selected for meta-analysis. A total of 2 articles were left out of the sensitive analysis process, and the remaining 4 articles showed no heterogeneity (I2=0% and Q- test P>0.1); then, the combined fixed effect was used. The results showed that the flexibility of the experimental group was increased by 0.05 compared with the control group; however, the results were not statistically significant (P=0.52>0.01),indicating that the effect of roller skating on the flexibility of children and adolescents is not obvious(Fig 7).
Speed quality
A meta-analysis of the effect of roller skating on children’s and adolescents' speed and quality included data from 8 sets of data involving 810 participants in 5 studies[13,17,23-25], and three separate data sets for male and female students[17,23,25]. After exercise intervention, the speed quality of the experimental group and the control group improved (95% CI,[0.14 to 0.20], Z=11.62, P<0.01), indicating that roller skating can improve the speed ability of children and adolescents(Fig 8 ).
Sensitive quality
Meta-analysis data on the relationship between roller skating and children’s and adolescents’ sensitive qualities came from 5 studies involving 6 sets of data from 276 participants[13,16,18,19,21], including 1 group of independent data set of males and females[17]. The total SMD value was 0.64, and the 95% confidence interval was [-1.13,2.41], with Z=0.71 and P=0.48>0.05. The studies showed that the effect of roller skating on the sensitivity of children and adolescents is not significant(Fig 9).
Vital capacity
The meta-analysis data of the relationship between roller skating and the vital capacity of children and adolescents comes from 9 groups of data involving 770 participants from 5 studies[16-17,24-26], two independent data sets of 3 groups of males and females were included[16,17,26]. The heterogeneity test I2=99% and Q-test P<0.1 indicates that the included studies showed strong heterogeneity, and the random effect of the combined effect size was selected. Based on the overall study of the included studies, it is highly suspected that the heterogeneity source test methods were inconsistent. The results showed that the total SMD value was 1.92, the 95% confidence interval was [0.17, 3.67], and it was statistically significant (Z=2.15, P=0.03<0.05), suggesting that roller skating can improve the cardiopulmonary function of children and adolescents(Fig 10 ).
Social Adaptability
For meta-analysis of the social adaptability of children and adolescents affected by roller skating, the data were derived from 5 sets of information involving 298 participants in 5 studies[17,18,20,27,28]. Overall, the total SMD value was 0.68, and the 95% confidence interval was [0.32 to 1.04], and was statistically significant (Z=3.67, P=0.0002<0.05). Studies showed that roller skating can effectively improve the social adaptability of children and adolescents(Fig 11).
3.4.9 Endurance quality
For meta-analysis of the effect of roller skating on children and adolescents' endurance quality, data were from 5 studies involving 5 groups of information from 803 participants. The heterogeneity test showed I2=32%, and the Q-test P>0.1; there was no heterogeneity among the studies involving endurance quality, and the fixed effect of the combined effect amount was selected. The results were Z=3.42 and P<0.01, suggesting that roller skating can significantly improve the endurance quality of children and adolescents(Fig 12).