THE ROLE OF SPORT PARTICIPATION AND BODY MASS INDEX IN PREDICTING MOTOR COMPETENCE OF SCHOOL-AGE CHILDREN

The purpose of this study was to investigate the association between Sport Participation (SP) and Motor Competence (MC) in school-age children, and to assess the mediation role of Body Mass Index (BMI) on this association. 707 children (332 boys, 375 girls) aged between 6to 10 years participated in the study. Motor competence was determined using the BruininksOseretsky Test of Motor Proficiency (BOT-2). Parents were asked about organized SP, categorized as 1) no participation, 2) partial participation and 3) consistent participation, and BMI was calculated based on the child’s height and weight. The results indicated significant differences for all MC components and total score for SP categories, all p < .05, with higher results for consistent participation in sports. For BMI, significant differences occurred in all components, with the exception of fine manual control and manual coordination. Our findings showed that SP was associated with MC, and BMI was not an important factor in these relationships. However, a small mediation of BMI was found, but only for obese children. In summary, we conclude that sport participation plays a relevant role in the development of motor competence in school-age children, and that association is, in general, not mediated by weight status. Palavras-chave: Motor competence. Body mass index. Sport participation. School-age children. RESUMO O objetivo deste estudo foi investigar a associação entre Participação Esportiva (PE) e Competência Motora (CM) em crianças em idade escolar e avaliar o papel da mediação do Índice de Massa Corporal (IMC) nesta associação. Foram avaliadas 707 crianças (332 meninos, 375 meninas) com idades entre 6 e 10 anos participaram do estudo. A competência motora foi determinada usando o Teste Bruininks-Oseretsky de Proficiência Motora (BOT-2). Os pais foram questionados sobre a SP organizada, categorizada como 1) sem participação, 2) participação parcial e 3) participação consistente, e o IMC foi calculado com base na altura e no peso da criança. Os resultados indicaram diferenças significativas para todos os componentes do CM e escore total para as categorias de PS, todos p <0,05, com resultados mais altos para a participação consistente nos esportes. Para o IMC, ocorreram diferenças significativas em todos os componentes, com exceção do controle manual e da coordenação manual. Nossos achados mostraram que a SP estava associada à CM, e o IMC não foi um fator importante nessas relações. No entanto, uma pequena mediação do IMC foi encontrada, mas apenas para crianças obesas. Concluímos que a participação esportiva desempenha um papel relevante no desenvolvimento da competência motora em crianças em idade escolar, e essa associação, em geral, não é mediada pelo status de peso. Palavras-chave: Competência motora. Índice de massa corporal. Participação esportiva. Crianças em idade escolar.


Introduction
Motor competence (MC) is defined as proficiency in fundamental motor skills including locomotor and object control skills 1 , and represents the degree of skilled performance in a wide range of motor tasks as well as the movement quality, coordination and control underlying a particular motor outcome 2 .Most importantly, MC has been established as one of the most significant predictors of health-related fitness 3 .In school-age children, the role of motor competence in several important health outcomes has been recognized, such as physical activity, cardiorespiratory fitness 4 and adiposity 1,3 .However, it is a well-known fact that most children show poor levels of motor competence across a range of ages, in early childhood 5 , school age 6 , and adolescence 7 .
Good health-related fitness is maintained through consistent participation in physical activity.There is compelling evidence suggesting that a physically active lifestyle in early life prevents the development of chronic diseases in adulthood 8 .Many studies have already explored the relationship between motor competence and physical activity levels, generally finding positive, significant associations between the two variables 9 .Furthermore, recent research has shown that organized interventions and participation in physical education 10 can improve motor skills.However, little is known about the influence of organized sports and the mediation of other factors on the development of motor competence in school-age children 11 .
Sport participation can be characterized as involvement in organized practical sessions specific to a sport 12 , which provides opportunities to be physically active as well as favors psychosocial development and the acquisition of life skills 13 .Several studies have reported a positive relationship between sport participation and MC 12 .However, this relationship seems weak in terms of magnitude.Moreover, the impact of potential confounding variables in this relationship have not been explored.For example, body mass index (BMI) is a good indicator of health-related fitness 14 , and most studies show that children with high BMI tend to have lower MC 15,16 .Recently, Cairney and Vedhuizen 17 demonstrated an association between sport participation and BMI, but the prediction effects of this relationship were small.
While it is known that sport participation can affect MC in children, little has been done to understand how sport participation is associated to motor competence when mediated by BMI.Most studies have looked at the effects of MC on physical activity, even though the relationship is considered cyclical or reciprocal 2 .Furthermore, no studies have examined motor competence from a general motor proficiency perspective -most use the Test of Gross Motor Development, 2nd edition (TGMD-2), an assessment of fundamental motor skills 16 .Very few studies utilized an assessment such as the Bruininks-Oserestsky Test of Motor Proficiency -2nd edition 18 (BOT-2).The BOT-2 is considered one of the most comprehensive assessments of motor proficiency, and is also one of the most widely used test of motor skills 19 .Cools et al. 20 report that the BOT-2 is a very detailed instrument that provides information on skill mastering: beneath and above skill level, while accounting for qualitative and quantitative aspects of movement behavior.
Therefore, the purpose of the present study is to investigate the relationship between Sport Participation (SP) and MC in school-age children, as well as to assess the mediation role of BMI on this relationship.More specifically, we were interested in determining what components of motor competence are associated with sport participation (components determined by the BOT-2 assessment, Fine Manual Control, Manual Coordination, Body Coordination and Strength & Agility).We also looked at those relationships in children categorized in all groups based on weight status (normal weight, overweight, obesity) when mediated by BMI.

Participants
A total random sample of 707 children (332 boys, 375 girls) aged 6-to 10 years and one of their parents participated in the study.Mean age was 8.21±1.25 for girls and 8.24±1.18for boys.All children were recruited from public schools in a large urban area in southern Brazil (Maringá -Paraná State).This sample was selected based on the following steps: At the time of the study, there were a total of 16.335 children between 6 and 10 years of age enrolled in the 49 public schools that make up the school district in Maringá.According to the sample size calculations, the minimum sample size for a 95% confidence interval was of 579 children.To account for attrition, we handed 869 consent forms in 12 schools randomly chosen in all geographical areas of the city (north, south, east, and west).We received 750 forms back, and were able to test 725 children with the BOT-2 test (25 children missed school on the testing days).Of those, it was only possible to schedule appointments for the other assessments with 707 families, and that was our final sample.
The sample included a wide range of socioeconomic classifications based on family income and parental education.Data were collected in two phases.In Phase 1, researchers assessed motor competence (BOT-2) and body composition (BMI) at the child's school.In Phase 2, parents were visited at home according to their availability and administered the Sport Participation (SP) questionnaire.The University Institutional Research Board approved all procedures.Written informed consent from parents as well as verbal assent from the child were provided prior to data collection.

Measures
Sport Participation (SP).Information about organized Sport Participation (SP) was obtained through a questionnaire answered by children's parents during an interview.In this questionnaire, the parents or guardians answered whether children did or did not participate in any organized sport (e.g., sports, dance and martial arts).Organized sports involved regular classes, training, or competition (structured or formal), as well as had a coach, instructor, or teacher associated with the practice.Children were divided into three groups: 1) "no participation": not actively involved in an organized sports club during the last two years; 2) "partial participation": has participated in organized sport activities for less than two years; 3) "consistent participation": consistent engagement in sport activities for at least 1 h/week during the last two years (adapted from 12 ).
Body Mass Index (BMI).After removing shoes and heavy clothing, children's height and weight were measured in the facilities of the school.A portable stadiometer was used to measure standing height with the value recorded to the nearest centimeter.Body mass was measured using an electronic calibrated scale, with weight recorded to the nearest tenth of a kilogram.BMI was calculated as weight in kilograms divided by the square of the height in meters (expressed as kg/m 2 ).BMI was classified according to Sisvan guidelines: 21 (a) underweight (less than the 3th percentile), (b) healthy weight (3th percentile to less than the 85th percentile), (c) overweight (85th to less than the 97th percentile), obese (97th to less than the 99th percentile), and (e) high obese (equal to or greater than the 99th percentile).No children in our sample was considered underweight, and we combined the two higher categories (obese and high obese) into one category (obesity) due to the small number of children being categorized as "high obese".
Motor Competence (MC).All children were administered the Bruininks-Oseretsky Test of Motor Proficiency, 2 nd edition (BOT-2) 18 .The BOT-2 is a standardized test that measures motor proficiency in individuals ages 4-to 21 years.It assesses both fine-and gross motor skill performance using 53 items divided in four motor composite areas including fine manual control, manual coordination, body coordination, and strength and agility.The composite scores are derived from eight subscales.The BOT-2 was administered on a one-toone basis in a location free from distraction for the child and took about 40-to 60 minutes to complete.For each test item, the assessor introduced the task and then observed the child's performance to determine a raw score based on specific rating criteria.For the purposes of this study, norms for each gender were used.Here, we used standard scores and descriptive categories based on standard scores (well-above average [70 or greater], above average [60-69], average [41-59], below average [31-40], well-below average [30 or less]).

Data analysis
Chi-square analysis was used to investigate the association between frequencies for each group of motor competence.To determine the relationship among numerical variables, Pearson correlations were used.
To look at differences on MC based on SP groups and BMI groups, a 3-way Analysis of Variance (ANOVA) was conducted (SP -no participation, partial participation, consistent participation) (BMI -healthy weight, overweight, obesity), using BOT-2 standard scores as the dependent variable for the BOT-2 subtests, and total motor composite for total MC scores.Tukey post-hoc analyses were conducted if necessary.
General Structural equation modeling (GSEM) was used to assess: a) the impact of SP on MC, b) the to analyze the impact of BMI on MC and c) the possible mediation effects of BMI on relationship between SP and MC.The results were presented in three different components: direct association (SP-MC) -straight association coefficient between two variables; indirect association -association between two variables that is mediated by a third variable (SP-BMI-MC) and total association -the sum of the direct and indirect association between variables.GSEM expressed all measures of relationship as standardized coefficients (r).

Results
The distribution of the children and their BMI and some selected demographic variables was shown in Table 1.

General motor classification, SP and BMI
According to total BOT-2 scores, 75% of children were classified as average and above (n = 494 for average; n = 36 above average, n = 2 well-above average), and 25% scored below average (n = 165 below average; n = 10 well-below average).For BMI, 52.6% (n = 372) of the children were considered within healthy weight, 16.7% (n = 118) were considered overweight, and 30.7 % (n = 217) were identified as obese.

Correlations
Pearson correlation analysis determined inverse and low relationships between most components of the MC and BMI, with the exception of Manual Coordination.Correlation coefficients for the MC components and SP were positive and moderate, ranging between .41 and .48,with .62 for the BOT-2 total score and SP.Table 2 shows all correlation coefficients.

Group differences
Table 3 shows means and standard deviations for MC standard scores and results of the ANOVA for BMI and SP.Results indicated significant differences for all MC components and total score for SP, all p < .05.For BMI, significant differences occurred in all components, except for Fine Manual Control and Manual Coordination.All post-hoc differences are listed on Table 3.The interaction between SP and BMI was not significant.Mediation SP and BMI were negatively related to each other (r = -.11[95% CI= -.18 to -.03]; p = .00).In the mediation analysis, the association of SP with all components of motor competence remained significant (p < .01),independent of BMI mediation (correlation coefficients slightly varied).On the other hand, there was a negative association of BMI with MC and the components Body Coordination and Strength and Agility (regardless of SP) (Table 4).In the final GSEM analysis, the association of SP with MC (also mediation effect attributed to BMI) was stratified into three groups based on weight status (and adjusted by age and sex) in order to understand the actual influence of BMI on these relationships.Among healthy weight and overweight school-age children, the mediation of BMI and SP was not significant, but was significant for the obese group (r = -.15 [95% CI = -.28 to -.02]; p = .02).In all GSEM, the positive association of SP and MC occurred despite of BMI group status.However, BMI and MC were not significantly associated (except in the component Strength and Agility for the overweight group, r = -.23 [95% CI= -.47 to -.00]; p = .04)(Table 5).

Discussion
The purpose of the present study was to investigate the association between SP and MC in school-age children, and the potential mediation of BMI on this relationship.Our findings indicated that SP was associated with MC, and in general, the impact of SP on MC was not mediated by BMI values.However, when splitting children into groups, a small mediation of BMI occurs for obese children.These results reinforce the notion that sport participation is crucial for the development of motor competence 2 .In addition, it shows that BMI is not relevant for these relationships, except in extreme cases such as the obese group.
First and foremost, it is important to highlight the notion that MC needs instruction to be developed and improved 22 .Motor skill development, which leads to motor competence, needs instruction, practice, and experience.Our results show that consistent participation in organized activities leads to better scores in MC (components and total) more so than partial participation, which showed better results than no participation at all.Along the same line, work by Robinson et al. 5 found that children who are directed by specialists to learn motor skills display greater increases in MC than children who engage in free play.Robinson et al. 5 also notes that the instructional approach used to teach motor skills along with basic learning principles and the amount and context of experiences influence the stability of MC.Thus, it is important to foster continued learning and development of MC through practice and participation in developmentally appropriate activities that demand more advanced movement patterns and higher levels of performance in a variety of movement contexts 22 .
Our findings confirm a trend showing that SP has an important association with MC, and extended these results by using an assessment of general motor proficiency.For example, Vandorpe et al. 12 found that children who consistently practiced sports in a club environment displayed better coordination levels than children who only partially participated or did not participate in a club environment at all.Fransen et al. 23 found that children with a relatively high MC are more involved in sports than children with a lower MC.Both studies evaluated gross motor coordination via a product-oriented assessment (Körperkoordinationstest für Kinder-KTK).One of the main difficulties of this type of measure is that skill outcomes do not necessarily highlight the developmental process (i.e., qualitative movement pattern development) that results in the achieved score.Other studies have utilized the BOT-2, but many used only the Short Form of the assessment 9,24 .To the best of our knowledge, this is the only study to date that used the BOT-2 complete form to investigate motor competence and its associations with both sport participation and BMI.
Here, we established the direction of the relationship between the variables (as in sport participation affecting motor competence), but it is important to highlight the fact that motor competence and physical activity are related in a dynamic and synergistic way 1 .Stodden's model 1 suggests that physical activity (as in participation in activities, including sports) in early childhood will initially promote the development of MC as basic motor patterns are developed through a variety of exploratory movement experiences.Seeing sport as a special context for the acquisition of motor skills in young children may help promoting long-term adherence to physical activity 25 .This relationship becomes more cyclical in middle and late childhood, when a high level of motor competence is expected for children to engage in physical activities.This progression fosters continued participation in a variety of physical activities as children enjoy success and are motivated to continue to improve motor competence 1 .
While our results emphasize the significant association between SP and MC, it is important to note that BMI showed significant differences for body coordination, strength and agility, and total MC (the higher the weight, the lower the motor competence of children).Even though these differences were not as strong as SP, we can infer that children with higher BMIs tend to have lower motor competence.In general, research shows negative low to moderate correlations between weight status and MC 24,26 , with coefficients usually ranging from 0.05 to 0.49 15 .
It is important to mention that in the final GSEM analysis, weight did not mediate the relationship between SP and MC in children with healthy weight and overweight.Mediation was only significant for the obese group, although small.This implies that only in extreme cases BMI seems to be a factor for the development of motor competence.D'Hondt and colleagues 27 have also demonstrated that childhood obesity is associated with lower scores on the Movement Assessment Battery for Children (MABC).In their study, no differences between children with healthy weight and overweight children were found, against their expectations.This suggests the existence of a certain weight cut-off from which movement difficulties appear 27 .However, it is possible to change this with a focused program for overweight and obese children.D'Hondt et al. 28 showed that children in these categories are able to improve their motor competence after participating in a regular physical activity program, and can partially catch up to their healthy weight peers.
The limitations of the study should be recognized.Initially, we investigated specific patterns between gender and age -an initial look at the data showed no significant differences between gender and age and all variables (MC, SP, and BMI).However, we acknowledge that it is possible that the fact that we adjusted scores to gender and age with the BOT-2, and the calculation of BMI was also adjusted to these two factors, may have influenced the results.While we accounted for participation in sports or lack of, we did not measure engagement and enjoyment in sport participation.We also only tested children up to age 10, and it is possible that SP decreases after this age 29 .Another factor to be considered is that the questionnaire used to assess SP was not validated, as well as the BOT-2 test, which is currently not validated to use with the Brazilian population.However, it is important to note that this assessment has been frequently used in studies in Brazil 30,31,32 .This study, as many others, is cross-sectional.However, the results do not outweigh the fact that our large sample size and clear results provide several answers that can move our understanding of associations among the three study variables.

Conclusion
In terms of practical applications, our findings highlight that is fundamental to promote community-based and school-based programs to enhance motor competence of all children.
Teachers and coaches need to offer the children a challenging, non-competitive environment with appropriate equipment (e.g.monkey bars, tires, ropes, etc.) and need to build structured skill learning/enhancement routines into their classes.Obviously, weight status should not be overlooked.But our results emphasize that is necessary for children to participate in physical/motor activities where the instruction of skills is prioritized.In summary, it is possible to conclude that sport participation plays a relevant role in the development of motor competence in school-age children, and that association is generally not mediated by weight status.

Table 1 .
Demographic characteristics Source: Authors

Table 3 .
Motor competence according to BMI classification and sport participation in schoolage children 2ote: a Healthy weight versus overweight.bHealthyweightversusobesity.1Consistentparticipationversuspartial participation.2Consistentparticipation versus no participation ., 3 Partial participation versus no participation Source: Authors

Table 4 .
Relationship between sport participation and motor competence mediated by BMI (n = 707) Source: Authors