R E S E A R C H

A R T I C L E

Weight Status and Gross Motor Skill in Kindergarten Children Dawn Roberts, PT, PhD; Diana Veneri, PT, EdD; Robert Decker, PhD; Mary Gannotti, PT, PhD Department of Physical Therapy, Springfield College (Dr Roberts), Springfield, Massachusetts; Departments of Rehabilitation Sciences (Drs Veneri and Gannotti) and Mathematics (Dr Decker), University of Hartford, West Hartford, Connecticut.

Purpose: Childhood obesity rates are increasing globally. Physical activity is one behavioral variable that influences weight status. Participation in physical activity requires basic gross motor proficiency in early childhood. The purpose of this study was to examine the relationship between gross motor skill level and weight status in a large national representative sample of kindergarten-aged children. Methods: Body mass index percentile ranking was calculated for 4650 children from the Early Childhood Longitudinal Study-Birth Cohort. Children were classified into underweight, healthy, overweight, or obese categories according to the Centers for Disease Control and Prevention criteria. The Early Screening Inventory Revised was used to evaluate gross motor skill level. Results: Children with obesity displayed lower gross motor skill levels compared with peers of healthy weight. Largest differences were seen in locomotor and balance skills. Conclusions: Clinicians should consider adjusting gross motor expectations for locomotor or stability tasks in young children with obesity. (Pediatr Phys Ther 2012;24:353–360) Key words: child, female, human, locomotor activities, male, motor skills, obesity INTRODUCTION AND PURPOSE More than 30% of children are classified as obese or overweight in the United States, with similar rates in other developed countries around the world.1 Overall gross motor skills of young children with obesity have been described as delayed or less proficient compared with peers of healthy weight.2 Yet, the association of obesity with delay in acquisition or decreased proficiency of specific gross motor skills has not been clearly described.2 Physical therapists working with children of kindergarten age will benefit from a better understanding of the influence of weight status on specific gross motor items when interpreting standardized assessments or planning treatments. School-aged children classified as overweight or obese demonstrate lower gross motor proficiency compared with peers of healthy weight.2-10 Nervik et al2 recently reported

0898-5669/110/2404-0353 Pediatric Physical Therapy C 2012 Wolters Kluwer Health | Lippincott Williams & Copyright  Wilkins and Section on Pediatrics of the American Physical Therapy Association

Correspondence: Dawn Roberts, PT, PhD, Springfield College, 263 Alden Street, Springfield, MA 01109 ([email protected]). The authors declare no conflicts of interest. DOI: 10.1097/PEP.0b013e3182680f19

Pediatric Physical Therapy

that preschool-aged children classified as overweight or obese scored lower on a standardized gross motor scale compared with peers who were not overweight, though the number of subjects in higher body mass index (BMI) categories in that study were small. Graf et al8 also found that first-grade children in obese categories had lower overall gross motor skill levels compared with peers who were of normal weight or underweight. Several studies have examined the influence of weight status on gross motor skill categories including locomotor skills,6,9,10 object manipulation,7,9 and dynamic body coordination skills.3,4 Most studies have found a negative relationship between weight and locomotor skills in both boys and girls.7,8,10 Similarly, several studies report lower dynamic body coordination in children who are obese.4,11,12 Specific gross motor skill delays were not reported in these studies. Obesity affects the speed and kinematics of gait and static balance in children,13,14 which may influence skill acquisition. The effect of weight status on object manipulation is less clear with many studies showing no association between weight status and object manipulation skills,6,9 and others reporting that children with obesity score lower in this category compared with children of healthy weight.3 Childhood obesity and gross motor skill development are a result of both intrinsic (child) and extrinsic (environmental) factors. The World Health Organization’s Weight Status and Gross Motor Skills 353

Copyright © 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins and the Section on Pediatrics of the American Physical Therapy Association. Unauthorized reproduction of this article is prohibited.

International Classification of Functioning, Health, and Disability15 provides a broad framework for examining the reciprocal interactions between health, behavior, and the environment. Given the variability among children in factors that may influence weight status and gross motor skill development, only a large nationally representative sample of children will provide valid, generalizable information on the association between these 2 factors. Thus, the purpose of this study was to evaluate whether differences existed in specific gross motor skills among kindergarten-aged children of different weight categories in a large nationally representative sample. We hypothesized that children who were obese would have difficulty performing specific gross motor skills where stabilization or movement of body mass influenced performance as compared with peers of healthy weight status. We defined stabilization tasks as tasks requiring single or double limb stability such as balancing on one foot. Movement of body mass included tasks where body mass was moved volitionally in any plane. METHODS Study Design The study was designed as a cross-sectional descriptive study using a previously collected national data set, the Early Childhood Longitudinal Study–Birth Cohort (ECLSB). This data set was collected by the US Department of Education, National Center for Education Statistics and is a nationally representative sample of 14 000 children born in the year 2001. Oversampling of children from specific subgroups (eg, American Indian and Alaska Native infants, low birth weight, and twins) allows for describing the range of developmental experiences of children.16 Cognitive, social, emotional, and physical development of the children was evaluated at multiple time points across multiple settings, and demographic and social information was collected on families and caregivers. We used cross-sectional data from the kindergarten wave of children, which included 10 700 children (5450 males), with a mean age of 5 years, 5 months, and a range of 4 years, 8 months, to 6 years, 2 months. Subjects Children with reported physical, medical, or developmental disability at birth, preschool, or kindergarten, or children who had a reported mobility problem, were receiving physical, occupational, speech, or special education therapy were eliminated from the analysis (n = 3000). Children who had missing height or weight information (n = 3050) were also eliminated from the analysis. No differences in socioeconomic status (SES) or racial distribution were found between children with and without height and weight data. Twins made up less than 10% of the sample (n = 300) and 50% of twin pairs had only one twin included in the analysis. Assessment of the distribution of twins among weight categories revealed a distribution similar to the overall sample. 354

Roberts et al

Data Collection Tools Age, gender, height, and weight available from the ECLS-B database were used to calculate BMI and ageadjusted BMI percentile ranking by gender (BMI-a). Standardized procedures adopted from the National Health and Nutrition Examination Survey were used to collect height and weight and are described in details elsewhere.16,17 Ageadjusted BMI percentiles rankings of underweight, healthy weight, overweight, and obese were determined from the Centers for Disease Control and Prevention (CDC) BMI-for-age growth charts using gender, age, height, and weight. The spreadsheet is available online.18,19 A composite score for SES was calculated by the ECLSB project staff using the following variables16 : education of the mother, education of the father, occupation of the mother, occupation of the father, and household income. Occupation was recoded to reflect the average of the 1989 General Social Survey20 prestige score of the occupation. The variables were imputed in a sequential order and separately by type of household (female single parent, male single parent, and both parents present). Once the components of the SES variable were imputed, their corresponding Z score or normalized value was computed. As described, the SES composite is the average of up to 5 measures, each of which was standardized to have a mean of 0 and a standard deviation of 1, providing a continuous measure of SES.16 The distribution of SES was examined and a categorical variable of SES was created. Five levels of SES were identified, from lowest to highest.16 For this analysis, the categorical variable of SES was used to describe participants. The ECLS-B used a combination of individual gross motor items from the Early Screening Inventory Revised,21 the Early Childhood Longitudinal Study Kindergarten Cohort of 1998-1999,16 the Bruininks-Oseretsky Test of Motor Proficeincy,22 and the Movement Assessment Battery for Children23 to assess motor abilities. Validity of these tests in children of kindergarten age has been reported to be high.24-26 Skill items assessed included jumping, balancing and hopping on 1 foot, walking backwards, skipping, and catching a beanbag. Table 1 displays each gross motor item and criteria measured in the ECLS-B. A gross motor composite score was created by adding up the scores on the 5 pass/fail items (pass = 1, fail = 0) with a maximal score of 5. Data Analysis Descriptive statistics were calculated to find mean, median, and standard deviation of demographic and descriptive variables. Analysis of variance was used to identify differences among groups of children by weight category and gross motor skills. Cross-tabulations were used to calculate odds ratios and evaluate odds for failure on pass/fail items. Correlation analysis tested for associations of BMI-a and SES with the gross motor composite score. Data from boys and girls were analyzed separately, as gender is a covariate of gross motor skills.27 SPSS 18 (Chicago, IL) was Pediatric Physical Therapy

Copyright © 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins and the Section on Pediatrics of the American Physical Therapy Association. Unauthorized reproduction of this article is prohibited.

used for all data analysis. An α level of .05 was established to judge statistical significance.

weight and children who were classified as overweight jumped further and hopped longer than children in the obese category (Table 3, P < .05). Girls in the underweight category hopped fewer times on the left foot than girls in the healthy weight category (Table 3, P < .05). Girls in the underweight and obese categories balanced for less time on their right foot compared with girls of healthy weight (Table 3, P < .05). Though statistically significant, this difference was less than 0.6 seconds and likely does not denote a clinically meaningful difference. Boys and girls with obesity had lower overall gross motor composite scores than children who were of healthy weight and overweight (Table 3, P < .05). There was no association of gross motor composite score with BMI-a (r = −0.06, P < .000). Odds ratios between healthy weight and obese groups showed that boys with obesity were 1.6 times less likely (95% confidence interval [CI], 1.3-2.0; P < .001) and girls with obesity were 2.2 times less likely (95% CI, 1.8-2.8; P < .001) to pass the left foot hopping test. Similar findings were seen on the right foot as boys with obesity were 1.6

RESULTS Of the original sample, 7700 children met inclusion criteria. Of these, 3050 had missing BMI data and were removed from the sample resulting in 4650 children (2150 males) used in subsequent analyses. Mean age was 5 years, 3 months (SD = 4 months) with a range of 4 years, 8 months, to 6 years, 1 month. There was no significant difference in age between weight categories (P = 0.26). See Table 2 for subject characteristics and breakdown of race, age, and gender by weight status. The mean BMI-a percentile ranking for the sample of children was the 65th percentile (SD = 28) with a range between the 0 and 100th percentiles. Gross Motor Findings by Gender Individual gross motor skills assessed by gender, for both girls and boys, revealed that children of healthy

TABLE 1 Gross Motor Skills from ECLS-B Test

Criteria

Scoring

Jumping Balance on 1 foot (right and left tested)

Distance jumped from standing start Balance for 10 seconds in 1 of 3 trials

Hop on 1 foot (right and left tested)

Hop 5 times in 1 of 3 trials

Skipping Walk backwards

Skip 8 steps Walk backwards on line 6 steps

Bean bag catch

Tossed bean bag 5 times

Distance in centimeters Pass/fail Maximum number of seconds balanced in 1 of 3 trials Pass/fail Maximum number of hops in 1 of 3 trials Pass/fail 0 = no steps on line 1 = >2 steps off line 2 = 1-2 steps off line 3 = all steps on line Number caught

TABLE 2 Participant Characteristics by Weight Status

Age Gender, % Boys Girls BMI, %ile rank (SD) Race, % White, non-Hispanic African American Hispanic Asian or Pacific Islander American Indian or Alaska Native More than 1 race Socioeconomic status, % Lowest socioeconomic status Middle to low socioeconomic status Middle socioeconomic status Middle to higher socioeconomic status High socioeconomic status

Pediatric Physical Therapy

Overall, N = 4650 (100%)

Underweight, n = 150 (3%)

Healthy Weight, n = 3050 (66%)

Overweight, n = 750 (16%)

Obese, n = 700 (15%)

5 y, 3 mo

5 y, 3 mo

5 y, 3 mo

5 yr, 3 mo

5 y, 3 mo

50 50 65.0 (28)

50 50 2.1 (1.6)

50 50 53.7 (22.1)

50 50 90.2 (2.8)

50 50 97.9 (1.5)

41 13 20 14 4 8

42 15 12 25 0 6

44 12 18 15 3 8

37 14 23 13 4 9

32 15 28 9 8 8

17 18 18 20 27

13 20 14 12 41

15 17 18 21 29

19 18 19 20 24

26 19 21 18 16

Weight Status and Gross Motor Skills 355

Copyright © 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins and the Section on Pediatrics of the American Physical Therapy Association. Unauthorized reproduction of this article is prohibited.

Copyright © 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins and the Section on Pediatrics of the American Physical Therapy Association. Unauthorized reproduction of this article is prohibited.

356

Roberts et al

Pediatric Physical Therapy

31.9 (8.5) 31.5, 13.2-57.3 28.8 (7.4) 29.0, 11.5-49.6 9.1, (1.8) 10, 3-10 8.8 (2.2) 10, 2-10 8.8, (2.5) 10, 1-10 9.3 (1.8) 10, 2-10 4.5 (1.1) 5, 0-5 4.8 (0.7) 5, 0-5 4.7 (0.9) 5, 0-5 4.6 (1.1) 5, 0-5 3.9 (1.3) 4, 0-5 4.0 (1.2) 4, 0-5 3.6 (1.4) 4.0, 0-5 3.8 (1.3) 4.0, 0-5

9.0, (2.0) 10, 0-10 9.3, (1.8) 10, 1-10 9.0, (2.0) 10, 0-10 9.3 (1.7) 10, 0-10 4.8 (0.8) 5, 0-5 4.8 (0.7) 5, 0-5 4.7 (0.9) 5, 0-5 4.8 (0.8) 5, 0-5 4.2 (1.1) 5, 0-5 4.1 (1.1) 5, 0-5 3.7 (1.2) 4.0, 0-5 4.1 (1.1) 4.0 0-5

b Healthy

Underweight (n = 150)

32.6 (8.9) 33.4, 0-57.3 30.1 (8.2) 30.6, 0-60

values given as Mean (SD) median, range. weight differs from obese (P < .05). c Overweight differs from obese (P < .05). d Underweight differs from healthy weight (P < .05).

a All

Standing long jump, in Boysb,c Girlsb,c Balance count on right, s Boys Girlsb,d Balance count on left, s Boys Girls Hop 5 times on right Boysb,c Girlsb,c Hop 5 times on left Boysb,c Girlsb,c,d Catching beanbag 5 times Boys Girls Gross motor composite score Boysb,c Girlsb,c

Overall (N = 4650)

TABLE 3

3.7 (1.2) 4.0, 0-5 4.2 (1.1) 4.0, 0-5

4.2 (1.1) 5, 0-5 4.1 (1.2) 5, 0-5

4.7 (0.9) 5, 0-5 4.8 (0.7) 5, 0-5

4.8 (0.8) 5, 0-5 4.9 (0.6) 5, 0-5

9.1, (2.0) 10, 0-10 9.4, (1.7) 10, 0-10

9.0, (2.0) 10, 0-10 9.4 (1.6) 10, 2-10

33.0 (9.0) 33.9, 0-55.0 30.7 (8.3) 31.2, 0-60

Healthy Weight (n = 3050)

Gross Motor Skills by Gender and Weight Statusa

3.8 (1.2) 4.0, 0-5 4.0 (1.2) 4.0, 0-5

4.2 (1.1) 5, 0-5 4.2 (1.1) 5, 0-5

4.8 (0.8) 5, 0-5 4.7 (0.8) 5, 1-5

4.8 (0.8) 5, 0-5 4.8 (0.7) 5, 0-5

8.9, (2.1) 10, 2-10 9.3 (1.8) 10, 1-10

9.1, (1.9) 10, 2-10 9.3 (1.8) 10, 2-10

33.4 (8.8) 34.4, 0.5-56.5 29.8 (7.8) 30.0, 6-52

Overweight (n = 750)

3.4 (1.4) 4.0, 0-5 3.8 (1.3) 4.0, 0-5

4.2 (1.1) 5, 0-5 4.2 (1.0) 5, 0-5

4.5 (1.1) 5, 0-5 4.5 (1.1) 5, 0-5

4.6 (1.0) 5, 0-5 4.7 (1.0) 5, 0-5

8.8, (2.2) 10, 2-10 9.1, (2.1) 10, 2-10

8.9, (2.1) 10, 0-10 9.1 (2.0) 10, 1-10

31.0 (9.0) 31.2, 0-50.4 28.0 (8.1) 28.0, 6.4-60

Obese (n = 700)

times (95% CI, 1.3-2.1; P < .001) and girls with obesity were 2.0 times less likely (95% CI, 1.5-2.5; P < .001) to pass the hopping skill. Girls with obesity were 1.3 times less likely (95% CI, 1.1-1.7; P < .029) to be able to balance on the right foot for 10 seconds compared with peers of healthy weight. Boys with obesity were 1.3 times less likely (95% CI, 1.1-1.6; P = .005) to be able to balance on the left foot for 10 seconds. Girls with obesity were also 1.3 times less likely (95% CI, 1.1-1.6; P = .006) to be able to skip compared with peers of healthy weight. Both boys and girls with obesity were less able to walk backwards with boys 1.3 (95% CI, 1.1-1.6; P = .013) and girls 1.4 times as likely (95% CI, 1.1-1.7; P = .003) to fail this gross motor test compared with nonobese peers (Table 4). DISCUSSION This study examined the gross motor skill levels of both boys and girls in different weight categories. We hypothesized that children who were obese would display decreased proficiency with gross motor tasks involving stabilization or movement of body mass. Our hypothesis is supported by our findings as children in the obese category scored lower on gross motor skills where movement or stabilization of body mass were required such as jumping, hopping, or balancing. Children with obesity hopped fewer times and jumped shorter distances than children in both the healthy weight and overweight categories. Interestingly, underweight girls also hopped fewer times than children of healthy weight. Girls in both the underweight and obese categories balanced for fewer seconds than girls of healthy weight, but the difference is not clinically meaningful. Obesity increased the odds that both boys and girls would not pass the balance or walking backwards test,

and girls with obesity were less likely to pass the skipping item. Boys and girls in the obese category had lower overall gross motor composite scores than boys and girls of healthy weight. Our results indicate that in kindergarten children, obesity is associated with poorer gross motor performance on items where movement or stabilization of mass is required. Our findings are similar to other recent studies that have reported lower overall gross motor skill level in young children in higher weight categories.2,5 Graf et al8 reported that in a sample of 500 school-aged children in Germany, children who were obese had lower motor proficiency than peers who were of healthy weight and underweight. Children in their sample were older than those in our study with a mean age of 6.7 years. An overall gross motor score was used; thus, the influence of weight status on specific gross motor tasks in this age group cannot be compared with our results. We also looked at individual gross motor task items and the influence of weight status on specific skills. We found that that gross motor tasks involving movement of body mass (jumping, hopping, backwards walking) were more difficult for children in the obese category compared with peers of healthy weight. There was no difference in skill level between weight categories for a manipulative task not involving body mass mobilization or management (ball catching). Our results match those by Okely et al,6 who found that weight status did not influence manipulative gross motor tasks but was inversely related to locomotor skill proficiency. Similar to our findings in young children, limitations in dynamic balance control in adolescents with obesity when compared with peers of healthy weight has been reported.13 D’Hondt et al4 found that overweight and obese status was associated with decreased performance on

TABLE 4 Gross Motor Skill by Gender and Weight Status for Pass/Fail Items

Balance on right for 10 seconds (% pass) Boys Girlsa Balance on left for 10 seconds (% pass) Boysa Girls Skipping 8 steps (% pass) Boys Girlsa Hop 5 times on right (% pass) Boysa Girlsa Hop 5 times on left (% pass) Boysa Girlsa Walk backwards 6 steps on line (% pass) Boysa Girlsa a Obese

Overall (N = 4650)

Underweight (n = 150)

Healthy Weight (n = 3050)

Overweight (n = 750)

Obese (n = 700)

76.1 82.9

79.1 83.5

77.5 83.6

74.2 82.2

70.6 79.8

75.8 82.9

79.1 83.5

77.5 83.6

74.2 82.2

70.6 79.8

38.2 57.3

32.6 48.1

38.0 59.3

39.2 56.5

35.6 50.7

88.2 92.0

83.7 92.4

89.4 93.4

90.8 91.4

81.7 86.1

85.3 89.8

83.7 86.1

86.2 92.1

89.3 89.4

78.1 80.4

41.8 47.3

51.2 44.3

43.0 49.0

41.2 47.2

36.1 40.1

differs from healthy weight, P < .05.

Pediatric Physical Therapy

Weight Status and Gross Motor Skills 357

Copyright © 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins and the Section on Pediatrics of the American Physical Therapy Association. Unauthorized reproduction of this article is prohibited.

dynamic body coordination tasks in school-aged children. Furthermore, these decreases were more pronounced in older children (10-12 years) than in younger (5-7 years). As weight status is generally stable during childhood, we can speculate that young children who are obese and display gross motor deficiencies at an early age are at risk for more pronounced delays as they age. We found that children with obesity demonstrated lower scores in tasks involving moving mass against gravity (jumping, hopping) compared with peers who were not obese. Others10,28 have also reported this. It has been suggested that mechanical movement inefficiency contributes to the motor deficiencies seen when children with obesity perform weight-bearing or limb-moving skills. Perhaps the excess inert mass impedes movement or lack of strength limits motor proficiency. Further studies exploring these hypotheses are required. We found balancing on the left foot to be more difficult for boys with obesity compared with boys of normal weight whereas balancing on the right foot was more difficult for girls with obesity compared with peers of normal weight. Though overall dynamic balance differences have been found in adolescents with obesity,13 no studies to date have examined differences in dominant compared with nondominant limb balance in children with obesity. Information on limb dominance was not available, thus we cannot determine whether it affected this finding; however, it is clear that young children who are obese seem to have more difficulty stabilizing body mass. One recent study assessing fine motor performance in children who are obese reported deficits in fine motor tasks in the sitting position when postural control demand was minimized.11 The authors suggested that children who are overweight or obese may have perceptual-motor coordination issues that influence not only gross motor but fine motor skills as well.11 We did not see deficits in the manipulative task assessed in our study (ball catching); however, we did not assess higher-level fine motor skills. The task of ball catching may not have been subtle enough to expose higher-level fine motor issues in children with obesity. Of note, although we found significant differences between children of healthy weight and children with obesity for many of our gross motor variables, few differences were found between children of healthy weight and children categorized as overweight. We found no linear correlation between BMI-a and the gross motor composite score. The CDC classifies children with a BMI percentile greater than the 85th percentile and less than the 95th percentile as overweight.19 Perhaps there is a threshold level of BMI after which gross motor skills are affected. Further exploration of the multivariate relationship between weight and gross motor skill is warranted. Many children with familial obesity tend to move to higher weight classifications with age.29,30 Future work should involve assessing gross motor skill changes over time in children who increase in weight category to better understand the influence of weight on motor skill development. 358

Roberts et al

Interestingly, we found that children in the underweight category displayed gross motor deficits in some skill areas compared with peers of healthy weight. Few studies have reported results for children who are underweight and ones that have reported on children who are underweight have had very small samples.2,8 These studies have reported no differences in gross motor skill level of children classified as underweight compared with peers of healthy weight.2,8 We found that children who were classified as underweight were not able to jump as far as peers of healthy weight. Girls classified as underweight also displayed poorer balance and hopping abilities compared with girls of healthy weight. Balance differences were very small (