1 SUBMITTED 1 SEP 21 1 REVISION REQ. 27 OCT 21; REVISION RECD. 24 NOV 21 2 ACCEPTED 28 DEC 21 3 ONLINE-FIRST: JAN 2022 4 DOI: https://doi.org/10.18295/squmj.1.2022.008 5 6 Patterns of Movement Performance Among Japanese Children and Effects 7 of Parenting Practices 8 Latent class analysis 9 Zhu Zhu,1,2 Cunyoen Kim,3 Dandan Jiao,1 Xiang Li,1 Ammara Ajmal,1 10 Munenori Matsumoto,1 Yuko Sawada,4 Toshiyuki Kasai,5 Taeko 11 Watanabe,6 Etsuko Tomisaki,7 Emiko Tanaka,8 Sumio Ito,9 12 Rika Okumura,9 *Tokie Anme10 13 14 1School of Comprehensive Human Science, University of Tsukuba, Tsukuba, Japan; 2Faculty 15 of Preschool and Special Education, Xuzhou Kindergarten Teachers College, Xuzhou, China; 16 3School of Education Science, Leshan Normal University, Leshan, China; 4Faculty of Health 17 Medicine, Morinomiya University of Medical Sciences, Osaka, Japan; 5Faculty of 18 Foundational Academics, Miyagi University, Miyagi, Japan; 6Faculty of Nursing, Shukutoku 19 University, Chiba, Japan; 7Faculty of Nursing and Medical Care, Keio University, Tokyo, 20 Japan; 8Faculty of Nursing, Musashino University, Tokyo, Japan; 9Department of Public 21 Welfare, Tobishima, Aichi, Japan; 10Faculty of Medicine, University of Tsukuba, Tsukuba, 22 Japan. 23 *Corresponding Author’s e-mail: anmet@md.tsukuba.ac.jp 24 25 Abstract 26 Objectives: The study aimed to examine the long-term effects of parenting practice during 27 preschool years on children’s movement performance in primary school. Methods: This 28 study involved a three-year longitudinal study including 225 children aged 3–6 years old. 29 Parents reported baseline parenting practice and evaluated children’s movement performance 30 three years later. Latent class analysis was used to explore latent classes of movement 31 performance. A post hoc test was used to identify the characteristics of different patterns. 32 mailto:anmet@md.tsukuba.ac.jp 2 Finally, adjusted multinomial logistic regression models were used to test the influence of 33 parenting practice on identified patterns of movement performance. Results: Children in this 34 study were grouped into three movement performance pattens, labelled as ‘least difficulties’ 35 (58.2%, n = 131), ‘low back pain’ (30.2%, n = 68), and ‘most difficulties’ (11.6%, n = 26). 36 After controlling for age, gender, having siblings or not, family structure, BMI SDS, sleep 37 condition and dietary habits, we found that if parents played games with children frequently, 38 the children would have a 0.287 times lower probability of being in the ‘low back pain’ class, 39 95%CI [0.105, 0.783], and if parents take children to meet peers of a similar age frequently, 40 the children would have a 0.339 times lower probability of being in ‘most difficulties’ class, 41 95%CI [0.139, 0.825]. Conclusions: Primary healthcare providers should pay careful 42 attention to children with movement difficulties. The study provides longitudinal evidence to 43 support the applicability of positive parenting practice in early childhood to prevent 44 children’s movement difficulties. 45 Keywords: Movement performance; Parenting practice; Latent class analysis; Child; 46 Longitudinal study; Japan. 47 48 Advances in Knowledge 49  The study originally used person-centred method to explore three patterns of children’s 50 movement performance in a Japanese community context. 51  This study confirmed the long-term effects of parenting practice during preschool years 52 on children’s movement performance when they enter primary school. We indicated that 53 playing games with children frequently contributed to preventing children from 54 developing low back pain, while taking children to meet peers of a similar age helped in 55 preventing children’s movement difficulties during their school age. 56 57 Application to Patient Care 58  Primary healthcare providers should pay special attention to children with movement 59 difficulties. The study provides longitudinal evidence to support the applicability of 60 positive parenting practice in early childhood to prevent children’s movement difficulties. 61 62 Introduction 63 Movement performance is defined as the competence or skills related to motor coordination, 64 muscle strength and balance, which are shown in self-care, sport, and other daily activities.1 65 3 School aged children need to possess motor skills, coordination, and body control in order to 66 complete daily activities.2 Movement difficulties in childhood may reduce a child’s 67 participation in daily activities and even impact their quality of life in adulthood.3,4 The 68 prevalence of movement difficulties has been rising worldwide recently.5 In Oman (N = 97; 69 Mage = 12.9, SD = 1.6), 55% of the total sample developed low grip strength and around 45% 70 were scored low in flexibility and sit-up tests.6 National reports in Japan also show a decline 71 in school-aged children’s movement performance, particularly among boys, which is at a 72 historically low level.7 However, there is no gold standard to measure children’s movement 73 performance in existing research.8 Therefore, person-oriented cluster analysis might be a 74 possible method to identify the characteristics of movement performance of children in a 75 community. 76 77 Movement performance is determined by complex interactions between biological 78 development and social environment.9 Differences are always expected for the movement 79 performance of children in terms of age, gender, body size and lifestyles.10-13 Home rearing 80 environment is one of the most important social environments, in which parenting practice 81 affects children directly.14 Parenting practice refers to the observable behaviours that parents 82 use to socialise their children in daily activities.15 A cross-sectional study demonstrated that 83 maternal permissive parenting was gender-specifically associated with better PA performance 84 in children experiencing authoritative parenting.16 However, results were not consistent with 85 the findings of Bradley et al. that indicated high parental monitoring was associated with 86 poorer PA performance for boys experiencing later puberty but increased PA performance in 87 boys experiencing early puberty using longitudinal data.17 Furthermore, only sixteen of the 30 88 quantitative studies in an integrative review showed significant positive associations between 89 supportive PA parenting and children’s physical performance.18 The majority of studies to 90 date, have focused on the intensity and frequency of PA instead of using health conditions or 91 function status as the outcomes. Limited studies have explored the relationships between 92 parenting practice and movement performance. 93 94 To fill gaps in existing research, the present three-year longitudinal study examined the 95 influence of specific parenting practices for preschool children on pattens of movement 96 performance while school aged. To avoid bias of variable-centred methods, we aimed to 97 investigate (1) the patterns of children’s movement performance based on person-oriented 98 4 cluster analysis and (2) the effects of daily parenting practice on children during the 99 preschool period. We hypothesised that (1) patterns of children’s movement performance 100 could be identified using different characteristics in a typical community and (2) more 101 positive stimulations in parenting contribute to preventing children from developing 102 movement difficulties. 103 104 Methods 105 Study design and participants 106 Our three-year longitudinal research study was part of a cohort study named ‘Community 107 Empowerment and Care for well-being and healthy longevity’ (CEC), involving all residents 108 in T village, a typical suburban community of Japan with a population of almost 5,000 from 109 1991. The inclusion criteria were as follows: (1) being aged 3–6 years old, (2) living in T 110 village, and (3) having at least one parent living together. The exclusion criteria were as 111 follows: (1) having a disability, serious disease, or developmental delay and (2) not living in 112 T village for the next three years. In the baseline survey, 289 parents with children aged 3–6 113 years provided the information on demographics and parenting practice. After 3 years, 114 children’s movement performance was evaluated by parents. As 27 families dropped out of 115 the project and 37 were excluded due to incomplete evaluation of movement performance, 116 the final sample size was 225. All research procedures were reviewed and approved by the 117 institutional review board and ethics committee of [blinded for review]. All participants 118 provided written consent before participation. 119 120 Measures 121 Parenting practice 122 Parenting practice was measured using the Index of Child Care Environment (ICCE), which 123 has been used in Japanese child cohort study for over 20. 19,20 ICCE is Japanese questionnaire 124 edition of the globally-used scale called the Home Observation for Measurement of the 125 Environment (HOME) and shows high reliability (α = 0.891).21 126 127 The ICCE is a self-reported questionnaire for parents, consisting of 13 items regarding 128 parenting practice, which are used independently in the present study. Questions for the 13 129 parenting practices are as follows: (1) How often do you play games with your child? (2) 130 How often do you go shopping with your child? (3) How often do you read to your child? (4) 131 5 How often do you sing songs with your child? (5) How often do you go to the park with your 132 child? (6) How often do you and your child meet with friends or relatives with children of a 133 similar age? (7) How often do you talk with your spouse about child care? (8) How often 134 does your spouse or other caregiver help you with the child? (9) How often do you and your 135 spouse eat meals together with the child? (10) What do you do if your child spills milk on 136 purpose? (11) How many times did you spank your child last week? (12) Do you have 137 anyone else that helps you with daily home-rearing? (13) Do you have anyone to consult with 138 about child care? Items 1–9 were measured using five-point Likert scale (1 = rarely, 2 = 1–3 139 times per month, 3 = 1–2 times per week, 4 = 3–4 times per week, 5 = almost every day). As 140 the responses were not normally distributed, binary-category classification was used in the 141 analysis based on ICCE manual (Unfavourable group = the bottom 25% of the total sample, 142 favourable group = the rest). Item 10 had five options (1= hit the child, 2 = scold the child, 3 143 = discipline in another way, 4 = determine how to prevent it in the future, 5 = in other ways). 144 Item 11 had five different options (1 = never, 2 = 1-2 times, 3 = 3-4 times, 4 = 5-6 times, 5 = 145 almost every day). For items 10 and 11, responses were categorised into two groups 146 (unfavourable = spank children and favourable = no spank). For items 12 and 13, responses 147 were originally measured in a binary manner (i.e., yes or no), in which the answer ‘yes’ was 148 evaluated as favourable and ‘no’ was evaluated as unfavourable. 149 150 Movement performance 151 Movement performance of children in the present study was investigated using a nine-item 152 parent-reported movement performance questionnaire, which have been used by community 153 government in large scale population-based surveys of the general population in Japan for 154 over 20 years.22 Parents were required to compare their children’s coordination with other 155 children of the same age based on their daily observations after the community government 156 explained evaluation points in detail. The nine items included the following: (1) Does your 157 child always appear energetic before and after school? (Keep active) (2) Are there any 158 difficulties for your child to keep running? (Keep running) (3) Does your child have 159 difficulties maintaining correct sitting posture? (Good sitting posture) (4) Does your child 160 have any arm pain? (Arm strength) (5) Does your child have any lower low back pains? (Low 161 back strength) (6) Does your child have any leg pain? (Leg strength) (7) Are there any 162 difficulties for your child in moving agilely to avoid obstacles? (Agility) (8) Dose your child 163 have any difficulties balancing? (Balance) (9) Does your child have any difficulties moving 164 6 their body flexibly? (Flexibility). Participants could respond to each item with ‘no’ (without 165 any difficulties) or ‘yes’ (having some difficulties). 166 167 Covariates 168 Demographics, children’s sleep condition, and their dietary habits were considered covariates 169 in the analysis models. Demographics included children’s age, gender, BMI (standardised 170 BMI, BMI SDS, was used in the analysis), having siblings or not, and family structure (e.g., 171 nuclear family type and extended family type). Children’s sleep condition was reported by 172 parents as ‘sufficient’ or ‘not sufficient’. Dietary habits were also reported by parents as ‘no 173 fussy eating’ or ‘having fussy eating behaviours’. 174 175 Statistical analysis 176 First, we used descriptive statistics to confirm demographics, baseline condition of parenting 177 practice and follow-up year’s movement performance. Second, latent class analysis (LCA) 178 was used to explore patterns of movement performance.23 Third, A post hoc test for the chi-179 square test (Bonferroni) and ANOVA analysis (LSD and S-N-K) was used to clarify 180 differences in demographics among the patterns of movement performance and identify the 181 characteristics of the patterns. Finally, adjusted multinominal logistic regression analysis was 182 applied to confirm the associations between parenting practice and movement performance 183 patterns. 184 185 All statistical analyses were performed using SPSS (Version 26.0; SPSS Inc., Chicago, IL) 186 and Mplus (Version 8.0; Muthén and Muthén, Los Angeles, CA, USA). 187 Results 188 Table 1 shows descriptive statistic results of demographic background. A total of 225 children 189 (Age: M = 4.13, SD = 0.87; BMI SDS: M = 0.12, SD = 0.98) was even distributed in gender 190 and family structure (boys: n = 119, 52.9%; girls: n = 106, 47.1%; Nuclear family: n = 107, 191 47.6%; Extended family: n = 118, 52.4%), while 83.6% children (n = 188) had siblings. 192 85.8% children (n = 193) had sufficient sleep while 68.9% children (n = 155) had fussy 193 eating behaviours. 194 195 Table 2 shows baseline parenting practice conditions and follow-up year’s movement 196 7 performance of children. In baseline year, the item with most negative evaluations was ‘How 197 many times did you spank your child last week?’, in which 37.8% parents (n = 85) reported 198 they had spanked their child in the last week. The item with least negative evaluations was 199 ‘Do you have anyone else help you in daily home-rearing?’, in which only 2.2% parents (n = 200 5) reported they took care of children without any help from others. As for the movement 201 performance of children three years’ later, our study showed that more than half of the 202 children were reported to have some difficulties on (1) maintaining right sitting posture (n = 203 139, 61.8%), (2) arm strength (n = 127, 56.4%), (3) agility (n = 114, 50.7%), and (4) 204 flexibility (n = 163, 72.4%). 205 206 Table 3 shows the model fit information for five LCA models with two to six latent classes. 207 Akaike information criterion (AIC), Bayesian information criterion (BIC) and sample-208 adjusted Bayesian information criterion (aBIC) in three-class model decreased sharply than 209 two-class model and the decline scope was the biggest among all the models (AIC = -210 71.126, BIC = -36.965, aBIC = -68.657). Entropy in three-class model was the highest in 211 all the models (0.935). The smallest sample size of the latent class is just over 25 (n = 26). 212 And the three-class model was significantly better than two-class model (p＜0.01). Based on 213 model selection recommendations for LCA model, we considered three-class model as the 214 best identified class. 215 216 Table 4 presents the results of the chi-square test and one-way ANOVA analysis, showing 217 demographics and movement performance characteristics of the three latent patterns. There 218 was no significant difference between the demographics of the three movement performance 219 patterns (p＞0.1). All the nine items, except flexibility, showed significant differences among 220 three movement performance patterns (p＜0.05). The results of post hoc test indicated the 221 number of responses of movement with difficulties in class 3 was significantly greater than 222 that in class 1 among all the nine items, except flexibility (p＜0.05). No significant difference 223 between class 2 and class 1 was found in the following categories: keep active, keep running, 224 arm strength, agility, and flexibility. No significant difference between class 2 and class 3 was 225 shown in the following categories: good sitting posture, arm strength, leg strength, and 226 balanced (p＞0.05). The number of responses indicating having low back pain in class 2 was 227 significantly greater than that in class 1, but less than that in class 3 (p＜0.05). Class 1 was 228 labelled as having the least difficulties (LD), class 2 was labelled as having low back pain 229 8 (LBP), and class 3 was labelled as having the most difficulties (MD). Figure 1 shows the item 230 probability of movement performance without difficulties in LD, LBP, and MD classes. The 231 LD class contained 58.2% (n = 131) of the sample and had high probabilities of movement 232 performance without difficulties. The LBP class contained 11.6% (n = 26) of the sample, and 233 all samples showed low back pain in the group. The MD class contained 30.2% (n =68) of the 234 sample and had low probabilities of movement performance without difficulties. 235 236 Table 5 show the associations between parenting practice and children’s movement 237 performance. In the multinomial logistic regression models, each parenting practice was 238 considered as independent variable respectively, while age, gender, having siblings or not, 239 family structure, BMI SDS, sleep condition and dietary habits were included in the models as 240 covariates. The LD class was used as the reference class to show the effect of positive 241 parenting practice on preventing movement difficulties. Model 1 indicated that if parents 242 played games with children frequently, the children would have a 0.287 times lower 243 probability of being in the LBP class, 95%CI [0.105, 0.783]. Model 2 indicated that if parents 244 take their children to meet peers of a similar age frequently, the children would have a 0.339 245 times lower probability of being in the MD class, 95%CI [0.139, 0.825]. 246 247 Discussion 248 To the best of the authors’ knowledge, this study is the first in Japan to examine the long-term 249 effects of parenting practice in children’s preschool period on their movement performance 250 outcomes when they are school age. We originally explored three patterns of children’s 251 movement performance and identified their characteristics in a sample of children from a 252 suburban area in central Japan. Based on our longitudinal results, we indicated that more 253 positive stimulations in parenting practice, such as playing games with children frequently 254 and frequently taking children to meet peers of a similar age, contribute to preventing 255 children from developing movement difficulties three years later. 256 257 Several studies have used person-oriented method to explore pattens of movement 258 performance, however, got inconsistent results. Jaakkola et al. investigated PA, sedentary 259 time, perceived competence, motor competence, cardiorespiratory fitness, and muscular 260 fitness in a Finnish elementary school student sample (N = 491; Mage = 11.27, SD = 0.32) and 261 labelled three movement profiles as ‘at-risk’ (37.7%, n = 185), ‘intermediate’ (49.3%, n = 262 9 242), and ‘desirable’ (13.0%, n = 64).24 Four movement profiles, which were ‘poor movers’ 263 (27.9%, n = 129), ‘average movers’ (38.4%, n = 177), ‘skilled movers’ (18.9%, n = 87), and 264 ‘expert movers’ (14.8%, n = 68), were identified when the performance of leap, throw-catch, 265 jump, push-up, sit-up tests were focused on.25 Our study explored three patterns of children’s 266 movement performance and originally identified the characteristics associated with different 267 types of movement difficulties. The biggest cluster, LD (58.2%, n = 131), received 268 significantly higher probability of ‘no difficulties’ than the MD cluster (30.2%, n = 68) for all 269 nine items except flexibility. All samples in the LBP cluster (11.6%, n = 26) reported having 270 pains in their low back, which was significantly different from the other two clusters. 271 Previous studies highlighted the prevalence of low back pain in school-aged children was 272 24% in a British sample (N = 1376) while it was 22% in an American sample (N = 1241) and 273 51% in a Danish sample (N = 1395. This suggests low back pain is an important and 274 relatively common problem in school children.26 Our results are consistent with the existing 275 research and additionally suggested low back pains should also be given attention in Japan. 276 277 Many previous studies have indicated parent-related factors, such as parents’ attitude towards 278 children’s PA, parents’ exercise habits, and parenting practice, are associated with children’s 279 daily physical activities, and therefore, influence children’s motor competence and physical 280 performance.27 A systematic review indicated supporting children to do PA or enrol in PA 281 classes, doing PA together significantly contributed to improving children’s physical 282 performance.28 Davison originally created the Activity Support Scale (ACTS) to measure 283 parental support for children's PA and confirmed that providing children with the chance or 284 places to be active, and playing sports with children is beneficial for children to improve their 285 physical activity levels.29 In addition, previous studies also highlighted the important role of 286 peer interactions on children’s motor performance.30 One systematic review reported positive 287 influence of peers’ support on PA and health outcomes.31 Our results are consistent with 288 previous studies and further clarified long-term effects of parenting practice during preschool 289 years on children’s movement performance when they entered primary school. We indicated 290 that playing games with preschool children frequently contributes to preventing children from 291 developing back pain three years later; while taking children to meet peers of a similar age is 292 beneficial to the prevention of children’s movement difficulties when reaching school age. 293 294 On the other hand, children’s age, gender, BMI, sleep condition, and dietary habits were not 295 10 significantly associated with children’s movement performance in the current study, which 296 are not consistent with existing research. Boys performed better in ‘walking’, while girls 297 performed better in ‘ball control’, and no gender difference were observed in ‘running’ and 298 ‘kicking’ in a meta-analysis for Japanese preschool children.32 Cardio-respiratory fitness 299 (CRF) and flexibility decreased with increasing age in a sample of 4,903 European children 300 aged 6–11 years.33 Sleep duration did not have a consistent significant effect on physical 301 fitness while fruit and vegetable intake positively related to physical performance with small 302 effects.34 Inconsistent results suggested influence factors and their effects of movement 303 performance are complex and different across culture. 304 305 Several limitations should be considered when interpreting our results and designing future 306 studies. First, children’s movement performance was only measured by parent-reported 307 questionnaires in the present study. Objective tests should be performed to verify the 308 consistency of the results in the future. Second, although we have controlled several 309 covariates, more related factors, such as SES and baseline movement performance, should 310 also be included in the final analysis model. Finally, the sample size was small because of the 311 loss of follow-up. 312 Conclusions 313 In conclusion, children in this study were grouped into three movement performance pattens 314 labelled ‘least difficulties (LD)’, ‘low back pain (LBP)’, and ‘most difficulties (MD)’, based 315 on a person-oriented perspective and cluster analysis. The LD group was characterised as 316 having highest probability of having no difficulties for all items, while the MD group was 317 characterised as having lowest probability of having no difficulties. The LBP group was 318 characterised by having all samples in the group develop low back pain. More positive 319 stimulations in parenting practice during preschool years, such as frequently playing games 320 with children and taking children to meet peers of a similar age, contributed to preventing 321 children’s movement difficulties when they entered primary school. Children with movement 322 difficulties should be carefully monitored by healthcare providers. Parents’ support is 323 beneficial for children to prevent developing movement difficulties. Nevertheless, there is 324 still a great need for more diverse samples and sufficient sample sizes to confirm the results 325 across cultures. 326 327 11 Conflict of Interest 328 The authors declare no conflicts of interest. 329 330 Funding 331 Japan Society for the Promotion of Science under Grants-in-Aid for Scientific Research 332 17H02604. 333 334 Acknowledgement 335 We express our deepest gratitude to all the participants and staff members of Tobishima for 336 their voluntary participation in this study. The work was also supported by Japan society for 337 the promotion of Science (JSPS) to provide the funding [Grant nos. JP21H00790 and 338 JP21K18449]. 339 340 Authors' Contribution 341 Conceptualization, Zhu Zhu and Cunyoen Kim; methodology, Zhu Zhu, Dandan Jiao, and 342 Toshiyuki Kasai; software, Yantong Zhu and Xiang Li; validation, Zhu Zhu, Dandan Jiao, 343 Ammara Ajmal, and Munenori Matsumoto; formal analysis, Zhu Zhu; investigation, Dandan 344 Jiao, Xiang Li, Ammara Ajmal, Munenori Matsumoto, Yuko Sawada, Sumio Ito, and Rika 345 Okumura; resources, Tokie Anme; data curation, Dandan, Jiao, Yuko Sawada, Taeko 346 Watanabe, Etsuko Tomisaki, Emiko Tanaka, and Tokie Anme; writing original draft 347 preparation, Zhu Zhu; writing—review and editing, Cunyoen Kim, Dandan Jiao, and Xiang 348 Li; visualization, Tokie Anme; supervision, Tokie Anme; project administration, Tokie Anme; 349 funding acquisition, Tokie Anme. All authors have read and agreed to the published version 350 of the manuscript. 351 352 References 353 1. Missiuna C, Moll S, King S, King G, Law M. A trajectory of troubles: parents' 354 impressions of the impact of developmental coordination disorder. Phys Occup Ther 355 Pediatr. 2007; 27(1):81-101. 356 2. Holm I, Tveter AT, Aulie VS, Stuge B. High intra- and inter-rater chance variation of the 357 movement assessment battery for children 2, ageband 2. Res Dev Disabil. 2013; 358 34(2):795-800. 359 12 3. Kwan MY, Cairney J, Hay JA, Faught BE. Understanding physical activity and 360 motivations for children with developmental coordination disorder: an investigation using 361 the theory of planned behavior. Res Dev Disabil. 2013; 34(11):3691-8. 362 4. Canto EG, Guillamon AR, Lopez LN. Relationship between overall physical condition, 363 211 motor coordination and perceived quality of life in Spanish adolescents. Acta Colomb 364 de 212 Psicol 2021; 24:96–106. https://doi.org/10.14718/acp.2021.24.1.9 365 5. World Health Organization. Guidelines on physical activity, sedentary behaviour and 366 sleep for children under 5 years of age. From: 367 https://apps.who.int/iris/handle/10665/311664. Accessed: Aug 2021. 368 6. Kilani H, Alyaarubi S, Zayed K, Alzakwani I, Bererhi H, Shukri R, Alrasadi K. Physical 369 Fitness Attributes, Vitamin D, Depression, and BMD in Omani's Children. Eur. Sci. J. 370 2013; 9(30): 156-173. 371 7. Japan Sports Agency. The national survey of children’s physical fitness, motor 221 372 competence, and physical activity. From: 222 373 https://www.mext.go.jp/sports/content/20191225-spt_sseisaku02-000003330_1.pdf. 374 Accessed: 223 Aug 2021. 375 8. Cools W, Martelaer KD, Samaey C, Andries C. Movement skill assessment of typically 376 developing preschool children: a review of seven movement skill assessment tools. J 377 Sports Sci Med. 2009; 8(2):154-68. 378 9. Davids K, Baker J. Genes, environment, and sport performance: why the nature-nurture 379 dualism is no longer relevant. Sports Med. 2007; 37(11):961-80. 380 10. Johnston LM, Burns YR, Brauer SG, Richardson CA. Differences in postural control and 381 movement performance during goal directed reaching in children with developmental 382 coordination disorder. Hum Mov Sci. 2002; 21(5-6):583-601. 383 11. Ikeda T, Aoyagi O. Relationships between gender difference in motor performance and 384 age, movement skills and physical fitness among 3- to 6-year-old Japanese children based 385 on effect size calculated by meta-analysis. School Health 2009; 5:9-23. doi: 386 10.20812/jash.SH-2009_031 387 12. Jaric S. Role of body size in the relation between muscle strength and movement 388 performance. Exerc. Sport Sci. Rev. 2003; 31(1): 8-12. 389 13. Tyler RPO. Health, Fitness, Lifestyle and Physical Competency of School Children in 390 Wales. Swansea University. 2018; https://doi.org/10.23889/Suthesis.50798 391 13 14. Newland LA. Family well‐being, parenting, and child well‐being: Pathways to healthy 392 adjustment. Clinical Psychologist. 2020; 19(1):3-14. 393 15. Kahraman H, Yilmaz Irmak T, Basokcu TO. Parenting Practices Scale: Its Validity and 394 Reliability for Parents of School-Aged Children. Educ. Sci.: Theory Pract. 2017; 395 17:745–769. doi:10.12738/estp.2017.3.0312 396 16. Jago R, Davison KK, Brockman R, Page AS, Thompson JL, Fox KR. Parenting styles, 397 parenting practices, and physical activity in 10- to 11-year olds. Prev Med. 2011; 398 52(1):44-7. 399 17. Bradley RH, McRitchie S, Houts RM, Nader P, O'brien M. Parenting and the decline of 400 physical activity from age 9 to 15. Int. J. Behav. Nutr. Phys. Act. 2011; 8(1): 1-10. 401 18. Hutchens A, Lee RE. Parenting Practices and Children's Physical Activity: An Integrative 402 Review. J Sch Nurs. 2018;34(1):68-85. 403 19. Anme T, Tanaka E, Watanabe T, Tomisaki E, Mochizuki Y, Tokutake K. Validity and 404 reliability of the Index of Child Care Environment (ICCE). Public Health Front 2013; 2: 405 141–5 https://doi.org/10.5963/PHF0203003 406 20. Chen W, Tanaka E, Watanabe K, Tomisaki E, Watanabe T, Wu B, Anme T. The influence 407 of home-rearing environment on children’s behavioral problems 3 years’ later. 408 Psychiatry Res. 2016; 244:185–193. 409 21. Bradley RH, Caldwell BM. The HOME Inventory and family demographics. Dev. 410 Psychol. 1984; 20:315–320. 411 22. Anme T. Empowerment for Healthy Longevity–Effective Utilization as Skills for Care 412 Prevention and Health Premotion (in Japanese). Tokyo, Japan: Ishiyaku Publishers, Inc., 413 2007. P. 42-55. 414 23. Lanza ST, Cooper BR. Latent class analysis for developmental research. Child Dev 415 Perspect 2016; 10: 59–64. https://doi.org/10.1111/cdep.12163 416 24. Jaakkola T, Yli‐Piipari S, Stodden DF, Huhtiniemi M, Salin K, Seppälä S, et al. 417 Identifying childhood movement profiles and tracking physical activity and sedentary 418 time across 1 year. Transl. Sports Med. 2020;3(5):480-7. 419 25. Jaakkola T, Hakkarainen A, Grasten A, Sipinen E, Vanhala A, Huhtiniemi M, et al. 420 Identifying childhood movement profiles and comparing differences in mathematical 421 skills between clusters: A latent profile analysis. J Sports Sci. 2021;39(21): 2503-2508. 422 26. Watson KD, Papageorgiou AC, Jones GT, Taylor S, Symmons DP, Silman AJ, 423 Macfarlane, GJ. Low back pain in schoolchildren: occurrence and characteristics. Pain 424 14 2002; 97(1-2): 87-92. 425 27. Newland LA. Family well‐being, parenting, and child well‐being: Pathways to healthy 426 adjustment. Clinical Psychologist. 2020;19(1):3-14. 427 28. Sleddens EF, Kremers SP, Hughes SO, Cross MB, Thijs C, De Vries NK, et al. Physical 428 activity parenting: a systematic review of questionnaires and their associations with 429 child activity levels. Obes Rev. 2012; 13(11):1015-33. 430 29. Davison, K. Activity-related support from parents, peers and siblings and adolescents' 431 physical activity: are there gender differences? J. Phys. Activ. Health 2004; 1: 363–376. 432 30. Herrmann C, Bretz K, Khnis J, Seelig H, Keller R, Ferrari I. Connection between social 433 265 relationships and basic motor competencies in early childhood. Children 2021; 8:10. 434 266 https://doi.org/10.3390/children8010053 267 435 31. Jenkinson KA, Naughton G, Benson AC. Peer-assisted learning in school physical 268 436 education, sport and physical activity programmes: A systematic review. Phys Educ 437 Sport 269 Pedagogy 2014; 19: 253–77. https://doi.org/10.1080/17408989.2012.754004 438 32. Ikeda T, Aoyagi O. Relationships between gender difference in motor performance and 439 age, movement skills and physical fitness among 3-to 6-year-old Japanese children 440 based on effect size calculated by meta-analysis[J]. School Health, 2009, 5: 9-23. 441 33. Zaqout M, Vyncke K, Moreno LA, De Miguel-Etayo P, Lauria F, Molnar D, et al. 442 Determinant factors of physical fitness in European children. Int J Public Health. 443 2016;61(5):573-82. 444 34. Jaikaew R, Satiansukpong N. Movement performance and movement difficulties in 445 typical school-aged children. PLoS One. 2021; 16(4): e0249401. 446 15 Table 1: Demographic background in the baseline year N = 225 Variables Categories N % Age of child (years) 4.13±0.87a Gender of child Boy 119 52.9 Girl 106 47.1 Siblings Only child 37 16.4 Having siblings 188 83.6 Family structure Nuclear family 107 47.6 Extended family 118 52.4 BMI SDS of child 0.12±0.98a Sleep condition of child Fussy eating behaviour of child Sufficient Not sufficient No Yes 193 32 70 155 85.8 14.2 31.1 68.9 Note: a Mean and SD were shown for continuous variables. 16 Table 2: Parenting practice in baseline year and movement performance of children three years' later N = 225 Items Categories n % Parenting practice Play games with child Few 44 19.6 Frequently 181 80.4 Shopping with child Few 21 9.3 Frequently 204 90.7 Read books to child Few 56 24.9 Frequently 169 75.1 Sing songs with child Few 45 20.0 Frequently 179 79.6 NA 1 0.4 Take child to play outside Few 27 12.0 Frequently 197 87.6 NA 1 0.4 Take child to meet peers of similar age Few 46 20.5 Frequently 178 79.1 NA 1 0.4 Eat meals together with child Few 48 21.4 Frequently 176 78.2 NA 1 0.4 Spank child for mistakes Spank 14 6.3 Not spank 210 93.3 NA 1 0.4 Spank child last week Spank 85 37.8 Not spank 138 61.3 NA 2 0.9 Take care of child with others Few 18 8.0 Frequently 204 90.7 NA 3 1.3 Talk with spouse about child Few 50 22.3 Frequently 174 77.3 NA 1 0.4 Have helpers No 5 2.2 Yes 218 96.9 NA 2 0.9 Have someone to consult with No 7 3.1 Yes 216 96.0 NA 2 0.9 Movement performance 17 Keep active With difficulties 69 30.7 Without difficulties 156 69.3 Keep running With difficulties 62 27.6 Without difficulties 163 72.4 Good sitting posture With difficulties 139 61.8 Without difficulties 86 38.2 Arm strength With difficulties 127 56.4 Without difficulties 98 43.6 Low back strength With difficulties 59 26.2 Without difficulties 166 73.8 Leg strength With difficulties 54 24.0 Without difficulties 171 76.0 Agility With difficulties 114 50.7 Without difficulties 111 49.3 Balanced With difficulties 94 41.8 Without difficulties 131 58.2 Flexibility With difficulties 163 72.4 Without difficulties 62 27.6 Note: NA = No answer 447 18 Table 3: Model fit information for the LCA models Log-likelihood df G-squared AIC BIC aBIC Entropy BLRT Two-class model –1112.676 492 381.606 2263.352 2328.258 2268.043 0.827 <0.01 Three-class model –1067.113 481 278.509 2192.226 2291.293 2199.386 0.935 <0.01 Four-class model –1036.648 471 217.222 2151.296 2284.524 2160.925 0.903 <0.01 Five-class model –1018.885 462 194.023 2135.769 2303.158 2147.867 0.921 <0.01 Six-class model –1009.066 452 174.386 2136.133 2337.683 2150.700 0.935 0.122 Note: df=degrees of freedom; AIC=Akaike information criteria; BIC=Bayesian information criteria; aBIC=adjusted Bayesian information criterion; BLRT= Bootstrapped Likelihood Ratio Test 448 19 Table 4: Demographics and movement performance characteristics of three patterns 449 Variables Categories Movement performance F/c2 p Class1 Class 2 Class 3 n % n % n % Age 4.13±0.87 2.112 0.123 Gender Boy 65 54.6 14 11.8 40 33.6 1.533 0.465 Girl 66 62.3 12 11.3 28 26.4 Siblings Single child 23 62.2 4 10.8 10 27.0 0.289 0.865 Having siblings 108 57.4 22 11.7 58 30.9 Family structure Nuclear family 61 57.0 11 10.3 35 32.7 0.757 0.685 Extended family 70 59.3 15 12.7 33 28.0 BMISDS 0.12±0.98 0.389 0.678 Sleep Sufficient 19 59.4 3 9.4 10 31.3 0.175 0.916 Not sufficient 112 58.0 23 11.9 58 30.1 Fussy eating No 93 60.0 17 11.0 45 29.0 0.653 0.721 Yes 38 54.3 9 12.9 23 32.9 Keep active With difficulties 21a 30.4 5a 7.2 43b 62.3 48.721 0.000 Without difficulties 110 70.5 21 13.5 25 16.0 Keep running With difficulties 15a 24.2 4a 6.5 43b 69.4 62.315 0.000 Without difficulties 116 71.2 22 13.5 25 15.3 Good sitting posture With difficulties 66a 47.5 20b 14.4 53b 38.1 17.254 0.000 Without difficulties 65 75.6 6 7.0 15 17.4 Arm strength With difficulties 59a 46.5 16a, b 12.6 52b 40.9 18.300 0.000 Without difficulties 72 73.5 10 10.2 16 16.3 Low back strength With difficulties 12a 20.3 20b 33.9 27c 45.8 60.649 0.000 Without difficulties 119 71.7 6 3.6 41 24.7 20 Leg strength With difficulties 15a 27.8 14b 25.9 25b 46.3 30.083 0.000 Without difficulties 116 67.8 12 7.0 43 25.1 Agility With difficulties 47a 41.2 15a, b 13.2 52b 45.6 30.090 0.000 Without difficulties 84 75.7 11 9.9 16 14.4 Balanced With difficulties 42a 44.7 16b 17.0 36b 38.3 12.743 0.002 Without difficulties 89 67.9 10 7.6 32 24.4 Flexibility With difficulties 95a 58.3 18a 11.0 50a 30.7 0.175 0.916 Without difficulties 36 58.1 8 12.9 18 29.0 Note: a, b refers different groups based on the results of Post hoc test using Bonferroni method. 450 21 Table 5: Significant results of multinominal logistic regression model showing associations 451 between parenting practice and movement performance 452 Variables LBP class vs. LD class MD class vs. LD class OR 95%CI p OR 95%CI p Model 1 Play games with child 0.287 0.105 - 0.783 0.015 0.834 0.371 - 1.873 0.660 Age 0.860 0.499 - 1.480 0.585 1.389 0.965 - 1.998 0.077 Gender 1.101 0.453 - 2.674 0.833 1.491 0.804 - 2.764 0.205 Having siblings or not 0.543 0.141 - 2.092 0.375 0.847 0.362 - 1.984 0.702 Family structure 0.773 0.315 - 1.901 0.575 1.166 0.634 - 2.147 0.621 BMI SDS 0.997 0.616 - 1.612 0.989 0.957 0.695 - 1.316 0.785 Sleep condition 0.961 0.245 - 3.777 0.955 1.144 0.475 - 2.759 0.764 Fussy eating 0.914 0.349 - 2.393 0.855 0.843 0.436 - 1.631 0.613 Model 2 Take child to meet peers of a similar age 1.175 0.443 - 3.115 0.746 0.339 0.139 - 0.825 0.017 Age 0.936 0.552 - 1.586 0.806 1.401 0.973 - 2.019 0.070 Gender 1.006 0.419 - 2.413 0.990 1.399 0.745 - 2.627 0.296 Having siblings or not 0.634 0.169 - 2.378 0.499 0.837 0.357 - 1.964 0.682 Family structure 0.880 0.367 - 2.110 0.774 1.155 0.626 - 2.134 0.645 BMI SDS 1.085 0.681 - 1.728 0.732 0.941 0.685 - 1.294 0.710 Sleep condition 0.853 0.220 - 3.300 0.818 1.117 0.459 - 2.718 0.807 Fussy eating 0.921 0.356 - 2.385 0.865 0.878 0.450 - 1.712 0.703 Note: 1. Reference group: play games with child = few, encourage child to play with peers of 453 a similar age = few, gender = boy, having siblings or not = only child, family structure = 454 nuclear family, sleep = sufficient, fussy eating = no fussy eating behaviours, age, BMI SDS = 455 continuous variables. 456 2. LD = least difficulties, LBP = low back pain, MD = most difficulties 457 22 Figure 1: Item probability of movement performance without difficulties in three classes. 458 459 0.0 20.0 40.0 60.0 80.0 100.0 Keep active Keep running Keep sitting posture Arm strength Low back strength Leg strength Agility Balanced Flexible P ro b a b il it y （%） Class 1-Least difficulties (n = 131, 58.2%) Class 2-Low back pain(n = 26, 11.6%) Class 3-Most difficulties (n = 68, 30.2%)