Relationship between BMI, physical fitness, and motor skills in youth ...

International Journal of Obesity (2006) 30, 861–867 & 2006 Nature Publishing Group All rights reserved 0307-0565/06 $30.00 www.nature.com/ijo

ORIGINAL ARTICLE Relationship between BMI, physical fitness, and motor skills in youth with mild intellectual disabilities GC Frey1 and B Chow2 1

Department of Kinesiology, Indiana University, Bloomington, IN, USA and 2Department of Physical Education, Hong Kong Baptist University, Kowloon Tong, Hong Kong

Introduction: The negative impact of obesity on physical fitness and motor abilities has been documented in youth of various ages; however, this issue has not been explored in youth with mild intellectual disabilities (ID). Youth with ID are considered more overweight, less physically fit, and less motor proficient than peers without ID, so it is important to determine if these variables are associated in this population. The purpose of this study was to examine the relationship between body mass index (BMI), physical fitness, and motor skills in a large sample of youth with mild ID. Method: A systematic, stratified sampling method was used to select 444 youth with mild ID, aged 6–18 years, from eight special education schools in Hong Kong. Physical fitness was assessed using items from the national Hong Kong assessment: 6(ages 6–8 years) or 9- (ages 9–18 years) minute run, sit-up, isometric push-up, sit and reach, and sum of skinfold. Functional motor skills were assessed in 244 youth from the fitness sample using the Test of Gross Motor Development-II. Subjects were categorized into normal or overweight/obese BMI groups based on international cutoff points. Results: Approximately 20% of the sample was classified as overweight/obese (average BMI normal ¼ 17.4772.69; overweight/ obese ¼ 24.7874.05). ANCOVA controlling for age and gender revealed group differences in the run (P ¼ 0.001) and push-ups (P ¼ 0.05), but not in the motor or other fitness variables. After controlling for age and gender, BMI was correlated with the run (r ¼ 0.27, P o 0.001) and push ups (r ¼ 0.18, P ¼ 0.008). Age and gender were entered as the first block in hierarchical regression and accounted for most of the variance in all dependent variables, except sit and reach. The inclusion of BMI in the second block added to the model for run and push-ups only (DR2 run ¼ 0.04, push-ups ¼ 0.03, Po0.001). Conclusion: Overweight/obesity is minimally associated with aerobic fitness and muscular strength in youth with mild ID. BMI did not impact other fitness measures (sit-up, sit and reach) or motor skills in the sample. The undesirable level of overweight/ obesity in this sample requires increased attention and immediate intervention. International Journal of Obesity (2006) 30, 861–867. doi:10.1038/sj.ijo.0803196; published online 10 January 2006 Keywords: mental retardation; gross motor skills; children; adolescents

Introduction The exponential increase in obesity among children and adults across many industrialized countries has been the topic of ongoing research;1 however, there is relatively little data on this topic in people with intellectual disabilities (ID). Although only a few large-scale studies have been conducted, obesity rates are reportedly higher among adults with ID compared to the general population.2–3 Less is known about youth with ID and available data are contradictory, with reports of similar4 or higher5 obesity rates

Correspondence: Dr GC Frey, 1025 E 7th St./HPER 179, Indiana University, Bloomington, IN 47408, USA. E-mail: [email protected] Received 22 March 2005; revised 24 August 2005; accepted 23 October 2005; published online 10 January 2006

compared to peers without ID. There have been considerable efforts to understand the rise in obesity among youth without ID, but it is unclear if this information can be generalized to those with ID since it is well documented that this group experiences certain deficiencies, particularly with regard to motor ability and physical fitness.6–7 There is evidence that a high body mass index (BMI) negatively affects motor performance and physical fitness in youth without ID.8–10 Youth with ID perform poorly on motor and fitness assessments,6 but the influence of BMI on this outcome has not been clarified. Some researchers have observed that BMI and aerobic fitness are correlated in youth with ID,11 while others have not,12 and the relationship between BMI and motor performance in these individuals has not been explored. It is often speculated that this population is at greater risk for obesity than peers due to high rates of sedentary behavior. Sedentary behavior in

BMI, fitness, motor skills and intellectual disabilities GC Frey and B Chow

862 youth with ID may be influenced by motor and physical fitness deficits since movement competence is important to activity participation.9 People with ID account for approximately 3% of the population worldwide and obesity will exacerbate the limitations already experienced as part of the disability, consequently hindering opportunities for maximal integration into society.13 Thus, there is a need to better understand factors that contribute to the health and well-being of these citizens. The purpose of this study was to examine (a) the incidence of obesity in a cross-sectional sample of children with mild ID, aged 6–18 years and (b) the relationship between physical fitness, motor performance and BMI in this population. It was hypothesized that there would be a strong relationship between the variables of interest.

unreliable and unrepresentative results. Owing to time constraints, approximately half of those involved in fitness testing were selected for motor testing, using the same gender and grade level stratified sampling method previously described. Parents were contacted directly by the school and participation was contingent on provision of a signed informed consent document, previously approved by the university research committee.

Methods

Procedures Fitness testing was conducted within the first 3 months after the start of the school term and motor testing was conducted the following 3 months. Height and weight measures for the calculation of BMI were obtained from school records, which had been updated at the beginning of the school year. BMI classifications of normal, overweight, and obese were determined using international standards.16

Subjects A total of 736 youth with mild ID (477 males, 259 females), aged 6–18 years, from eight special education schools in Hong Kong participated. The majority of Hong Kong youth with disabilities attend segregated schools according to learning ability and diagnosis. Schools for youth with ID are organized by IQ classifications of mild (IQ ¼ 50–70), moderate (IQ ¼ 25– 49) or severe (IQ ¼ o25) ID14 and placement is primarily based on IQ. Focus was placed on those with mild ID, which accounts for approximately 95% of Hong Kong citizens with this condition.15 Although IQ scores were not available, verification that the sample represented youth with mild ID was based on (a) strong adherence to school placement according to IQ and (b) teacher confirmation. A stratified random sample method was used to select the eight research sites from 17 schools that admit the target population. The final sample was distributed among the major districts of Hong Kong Island (N ¼ 1), Kowloon Peninsula (N ¼ 4), and New Territories (N ¼ 3). The two former districts are classified as urban, while the latter is rural and represents the largest geographic region, with the largest population. Approximately 1420 students with mild ID were served in these facilities. A total of 736 students were selected according to gender and grade level using a stratified, systematic odd number sampling process (e.g., 1, 3, 5, etc.). Participants with Down’s syndrome (n ¼ 63), William’s syndrome (n ¼ 12), identified co-occurring conditions, specifically autism (n ¼ 144) and a variety of other physical disorders, such as congenital heart defects and cerebral palsy (n ¼ 60), were excluded from the analysis for two reasons. First, individuals with certain genetic conditions should be considered as a separate sample due to documented differences in physiological variables compared to those with ID and no diagnosis.7 Second, co-occurring conditions could impact the ability to participate in testing (e.g., sensory sensitivity associated with autism, tests not validated for those with physical disabilities) leading to

Physical fitness assessment. Test items were selected in accordance with those used to establish territory-wide fitness norms for all Hong Kong youth.17 Participants completed five activities as follows: (a) 1-min sit-up (muscular endurance); (b) isometric push-up (muscular strength); (c) sit and reach (flexibility); (d) 6-min (ages 6–8 years) or 9-min (ages 9–18 years) run/walk test (cardiovascular), and (e) triceps and calf skinfold (sum of skinfold) measures. Testing was conducted on an outdoor covered playground in six schools and on an indoor playground in two schools. The 1-min sit-up is valid and reliable for youth with and without ID.18–19 Both feet were held stationary and arms placed in a crossed position over the chest. Movement completion occurred and was counted when the participant was able to sit-up and touch elbows to thigh. Isometric pushup protocol involved timing students according to their ability to maintain ‘up’ push-up (straight leg) position in a stationary manner. This item is considered valid and reliable in youth with and without ID.20 The back-saver sit and reach test is also considered to be valid and reliable for youth with and without ID.18–20 The test was conducted with students sitting on the floor in bare feet, one leg extended with the foot pushed against a standard sit and reach box. The other leg was flexed and foot flat on the floor 2–3 inches next to the extended leg. One hand was placed on top of the other, palms down, while arms extended and participants leaned forward from the waist as far as possible and held the position for 1 s. The distance reached in centimeters was recorded. This task was performed twice with each leg extended. The 6/9-min run/walk is considered a valid and reliable field test of cardiovascular fitness in youth.8,18,21–22 The course was typically set-up on a basketball court at the school and measured 15 25 m. Students were paced by test administrators and the number of laps completed in the specified time was recorded and converted to total meters. Those aged X9 and p8 years completed the 9-min or 6-min test, respectively. Triceps and calf skinfold thicknesses are a

International Journal of Obesity


863 standard indices of body fat in youth.23 This item was measured using Harpenden calipers and the final value was averaged over two trials to obtain sum of skinfold thickness. Seven individuals with fitness testing certification from the Hong Kong Physical Fitness Association and at least 20 h of practical experience administered the tests. Supplemental instruction was provided to all test administrators to prepare them for assessing children with ID (e.g., communication issues, ensuring task understanding). A school physical education instructor was present at all sessions to assist with and facilitate the procedures. Only one class of 20–30 students was administered the test at a time and activity stations were used to assess small groups of 5–8 participants. Several days before actual data collection, participants were familiarized with testing procedures and practice trials were administered in physical education classes to enhance task understanding and completion. Demonstrations and verbal cues, and verbal encouragement were provided during assessment and all subjects completed the entire test.

Data analysis ANCOVA was used to examine differences between those classified as overweight, obese, and normal BMI, with age and gender as covariates. Pearson correlations controlling for age and gender were calculated to examine relationships between variables and interpreted as recommended by Cohen28 (small/weak ¼ 0.10–0.29; medium/moderate ¼ 0.30–0.49; large/strong ¼ 0.50–1.00). Hierarchical linear regression analyses, with age and gender entered as a block followed by BMI, were used to evaluate the relative influence of these factors on motor performance and physical fitness. The run data are presented in total meters for descriptive purposes, but a relative measure of meters per min was used in the analysis. Data were analyzed using Statistical Package for the Social Sciences software version 10.0. Means and standard deviations are reported and significance was set at Po0.05.

Results Motor assessment. Fundamental motor skills were assessed using the Test of Gross Motor Development-II, which is a valid and reliable, norm and criterion referenced assessment of locomotor and object control movement patterns.24 Although the instrument is not specifically designed for youth over age 10 years, the purpose of this study was to obtain a measure of motor proficiency in this population, not to make normative comparisons. In addition, items from the TGMD have been previously used to assess motor performance in adolescents without ID.8,25–26 There are 12 skills (six locomotor ¼ gallop, hop, horizontal jump, leap, run, slide; six object control ¼ strike stationary ball, stationary dribble, catch, kick, overhand throw, underhand roll) and participants are scored on the ability to meet performance criteria on each subtest item (range ¼ 3–5 criteria per item). Subjects are allowed two test trials and each trial was scored separately for a maximal 48 points on each subtest. Since the standard scores are validated only for youth ages 3–10 years, raw scores were used for analysis. Six test administrators received 6 h of formal training, which included a thorough explanation of procedures and scoring, watching training videotapes, practice trials using videos, and practical training in a school setting. Since these individuals had relatively less experience with motor testing, compared to those involved in the physical fitness testing, interobserver reliability was assessed in the field practice session. Each test administrator provided an independent score for six different subjects and an intraclass, two-way mixed model revealed acceptable reliability for all motor items (R ¼ 0.88).27 Subjects were tested in groups of 20, and one test administrator assessed five subjects according to stated guidelines. Prior to performing the tasks, a verbal explanation and physical demonstration were provided. Each subject was allowed a practice trial to ensure task understanding and an additional demonstration was provided if a lack of comprehension was evident.24

After excluding subjects with incomplete or missing data (n ¼ 13), the final sample sizes for fitness and motor testing were 444 (264 males, 180 females) and 244 (134 males, 110 females), respectively. BMI and sum of skinfold measures were highly correlated (r ¼ 0.80), so the latter was not entered into the regression equation and will not be addressed in the Results or Discussion. Few subjects were classified as obese, so the overweight (n ¼ 57, 13%) and obese (n ¼ 31, 97%) categories were combined for further analysis, resulting in approximately 20% sample identified as overweight/obese. This was equally represented across both the fitness (n ¼ 88, 20%) and motor test samples (n ¼ 49, 20%). Descriptive information according to BMI category is presented in Table 1. The only differences between BMI categories was in the 6/9-min run (Po0.001, Z2p ¼ 0.04) and push-up (P ¼ 0.05, Z2p ¼ 0.01) (Table 1). BMI was weakly correlated with the run and push-up, but none of the motor variables (Table 2). Of the variables studied, age and gender accounted for the most variance in the 6/9-min run (15%), push-up (11%), object control (21%), locomotor control (10%), and sit-up (18%), but not sit and reach (1%). BMI significantly contributed an additional 4 and 3% to the variance of both the 6/9-min run and push-up, respectively, but not the other variables (Table 3).

Discussion A total of 31 subjects (97% of sample) were classified as obese, which is lower than the 14% reported in the general Hong Kong youth population, aged 6–18 years.29 There are no available estimates of both overweight and obesity in Hong Kong youth, so it is difficult to adequately interpret the 20% overweight/obesity rate International Journal of Obesity


864 Table 1

Descriptive characteristics

Age Height (cm) Weight (kg) BMI (kg/m2) Locomotor Object control Run (m)* 6-min 9-min Sit-up (number) Push up (min)* Sit and reach (cm)

Normal (n ¼ 356)

Overweight/obese (n ¼ 88)

Total (n ¼ 444)

12.3373.05 146.83716.63 38.75712.35 17.4672.69 31.16710.88 n ¼ 195 29.30711.19 n ¼ 195

11.6773.53 146.90717.26 55.46719.52 24.7774.05 29.51710.34 n ¼ 49 27.65710.74 n ¼ 49

12.2473.15 146.84716.74 42.06715.54 18.9174.19 30.83710.77 n ¼ 244 28.97711.10 n ¼ 244

599.787114.36 n ¼ 45 984.397204.37 n ¼ 292 18.87711.62 n ¼ 334 33.54739.47 n ¼ 335 21.9279.50 n ¼ 335

577.087126.98 n ¼ 12 891.427179.42 n ¼ 65 16.05711.95 n ¼ 79 23.42725.24 n ¼ 79 22.07710.19 n ¼ 80

595.007116.32 n ¼ 57 967.467203.03 n ¼ 357 18.33711.72 n ¼ 413 31.61737.36 n ¼ 414 21.9479.62 n ¼ 415

*Significant difference between BMI groups at Po0.05.

Table 2

Correlations between BMI, motor skills, and fitness variables controlling for age and gender

Locomotor Object control Run Sit-up Push-up Sit and reach

BMI

Locomotor

0.01 P ¼ 0.89 0.02 P ¼ 0.78 0.27* P ¼ 0.008 0.05 P ¼ 0.45 0.18* P ¼ 0.007 0.01 P ¼ 0.89

0.67* Po0.001 0.36* Po0.001 0.33* Po0.001 0.22* P ¼ 0.001 0.08 P ¼ 0.24

Object control

Run

Sit-up

Push-up

0.25* Po0.001 0.21* P ¼ 0.002 0.19* P ¼ 0.006 0.18* P ¼ 0.007

0.32* Po0.001 0.29* Po0.001 0.11 P ¼ 0.11

0.19* Po0.007 0.21* Po0.002

0.14* P ¼ 0.04

*Significant correlation at Po0.05.

Table 3 The contribution of BMI to the regression model, after accounting for age and gender

Run Locomotor Object control Sit up Push up Sit and reach

b

t

R2

DR2

DR2 P-value

0.21 0.03 0.00 0.07 0.18 0.04

4.30 0.50 0.07 1.49 3.67 0.69

0.18 0.10 0.21 0.19 0.13 0.01

0.04 0.00 0.00 0.00 0.03 0.00

0.001* 0.63 0.93 0.14 0.001* 0.49

*Significant added variance at Po0.05.

observed in the current sample within a cultural context. Regardless, a 20% overweight/obesity is cause for concern, particularly since the desired rate for youth is 5%.30 To our knowledge, there are only two other studies addressing obesity in large, cross-sectional samples of youth with ID. Murphy et al4 observed no difference in age-corrected International Journal of Obesity

BMI between US youth with and without ID, but prevalence was not reported. Conversely, Takeuchi5 observed a higher incidence rate among Japanese youth with ID compared to peers (B12%), but actual BMI group comparisons were not provided. These studies were conducted before development of the international cutoffs for youth, but provide evidence to suggest that culture may influence findings in this population. For example, Hong Kong utilizes a segregated school system, while North American countries implement school integration. Research on living environments indicates that adults with ID in more controlled settings (i.e., segregated) are less obese than those in less controlled settings (i.e., integrated),31 thus school placement may also affect body composition. Therefore, it is recommended that obesity data on youth with ID be considered according to disabilityrelated cultural norms (e.g., services, education, community placement).


865 BMI had a small, negative influence on aerobic performance and muscular strength in youth with mild ID, when age and gender were controlled. Overweight/obese youth performed worse on the 6/9-min run than those with normal BMI values, which has been consistently demonstrated in youth without ID across ages.8–10,32–34 Data regarding this topic in youth with ID are somewhat varied with evidence of a strong association11 or no association between these factors.12 Of the variables examined in the current study, age and gender accounted for most of the variance in aerobic performance, contrary to previous studies on youth with ID.11,35 Although there were differences between the BMI groups regarding these variables, the importance of this result is questionable in the light of small effect sizes, and particularly since the addition of BMI to the regression model contributed little to the variance observed. Based on these data, the influence of BMI on selected fitness variables in youth with mild ID should not be overstated. Correlations between BMI and aerobic fitness were similar to those reported by Graf et al8 (r ¼ 0.20), but lower than Manios et al33 (0.53), Lloyd et al34 (0.49) in youth without ID. Differences in the strength of the relationship between these two variables may be because of test selection. Fernhall et al11 found a moderate, inverse relationship between BMI and VO2 peak as measured on a treadmill, but the correlations between BMI and field tests (e.g., 20 and 16-m shuttle run, 600 yard run) were not significant. We used the same field test as Graf et al8 and demonstrated comparable correlations, but it is difficult to interpret findings across studies because of large differences in factors such as population age and sampling procedures. However, it appears that the influence of BMI on certain fitness items may vary according to test item and this topic is worthy of further inquiry. Youth with ID perform poorly on other fitness measures (e.g., muscular strength, flexibility) compared to peers without ID,36–38 but there is limited information regarding the association between anthropometric factors and physical fitness in this group. Fernhall and Pitetti39 found that isokinetic leg strength was inversely correlated with BMI in youth with ID, and other studies on youth without ID have demonstrated weak to moderate correlations between BMI and muscular strength and endurance.10,34 It was reasonable to predict that obesity would negatively affect isometric push-up performance due to the difficulty in supporting excess weight while in a plank position. However, the lack of difference between groups on the sit-up task was surprising because obese youth without ID perform worse than normal weight peers on this item.40 The current findings can be explained by the poor sit-up performance of the whole sample, which appears to be typical for youth with ID,6–7,38 as well as Hong Kong youth without ID in general.41 Hong et al41 compared physical fitness in youth without ID from Hong Kong and mainland China, aged 9–18 years. Hong Kong youth completed significantly fewer sit-ups in 1 min compared to Chinese peers, with 14.9

the highest average number performed by Hong Kong youth. Hong Kong youth are viewed as some of the least active and least physically fit in the world42 and it is possible that this potential cultural trend may exacerbate the already low levels of abdominal strength and endurance frequently observed in youth with ID. The lack of association between flexibility and BMI was not surprising based on previous reports of small correlations between these variables.10,34 The current findings indicate that obesity does not influence flexibility in youth with mild ID. There is evidence to suggest that obesity hampers motor performance in youth without ID, but it is important to note that previously reported correlations between measures of body composition and motor skills have been small to moderate (range ¼ 0.10 to 0.33).8–9,43 An association between BMI and motor performance was not observed in the current study, most likely because motor scores for the entire sample placed them in the very poor performance category.24 It is well documented that youth with ID display inferior motor skills compared to peers without ID.6 The reason for these discrepancies are complex and attributed to a variety of factors such as differences in motor development, motor delays, physiologic limitations (e.g., reduced muscle activation), lack of movement opportunities, and anthropometric characteristics.6 Results from this study suggest that BMI as an anthropometric characteristic does not influence motor performance in youth with mild ID and other causes for these deficiencies should be explored. Perhaps most of interest with regard to prevention of obesity in youth with mild ID were the small to moderate correlations between motor and fitness items. There are few reports on this topic, but other researchers have reported similar relationships in youth without ID.25,44 Deficiencies in motor proficiency could predispose people to inactive behavior,26 and this may partially explain the high rates of obesity, low fitness levels, and sedentary behavior observed in adults with ID.2,3,7,45 Additional research is needed to determine if remediation of motor deficits leads to improvements in physical fitness and reduced obesity rates in this population. This study is unique because it is the first to examine obesity correlates in a large sample of youth with mild ID. The cross-sectional design precludes conclusions regarding cause and effect; but it is clear that the prevalence of overweight/obesity among Hong Kong youth with mild ID is undesirable, even though there is negligible impact on physical fitness and body composition. Data on medication were not collected and this may have influenced BMI; however, minimal differences in body weight and skinfold measures between Hong Kong youth with and without mild ID indicate that this was not an issue of significant concern.46 These findings can only be generalized to those with mild, idiopathic ID and the association between fitness, obesity, and motor abilities in individuals with genetic conditions such as Down, Prader–Willi or Bardet–Biedl International Journal of Obesity


866 syndrome, as well as those in other diagnostic categories (e.g., moderate, severe), requires further study. Much research is needed to better understand unique obesity determinants, such as physical activity behavior and medications, as well as other psychosocial and environmental factors, across this diverse population.

References 1 Hedley AA, Ogden CL, Johnson CL, Carroll MD, Curtin LR, Flegal KM. Prevalence of overweight and obesity among US children, adolescents, and adults, 1999–2002. JAMA 2004; 291: 2847–2850. 2 Yamaki K. Body weight status among adults with intellectual disability in the community. Ment Retard 2005; 43: 1–10. 3 Hove O. Weight survey on adult persons with mental retardation living in the community. Res Dev Disabil 2004; 25: 9–17. 4 Murphy CM, Allison DB, Babbitt RL, Patterson HL. Adiposity in children: is mental retardation a critical variable? Int J Obes Relat Metab Disord 1992; 16: 633–638. 5 Takeuchi E. Incidence of obesity among school children with mental retardation in Japan. Am J Ment Retard 1994; 99: 283–288. 6 Eichstaedt CB, Lavay BW. Physical Activity for Individuals with Mental Retardation: Infancy Through Adulthood. Human Kinetics: Champaign, IL, 1992. 7 Fernhall B. Physical fitness and exercise training of individuals with mental retardation. Med Sci Sports Exerc 1993; 25: 442–450. 8 Graf C, Koch B, Kratschmann-Kandel E, Falkowski G, Christ H, Coburger S et al. Correlation between BMI, leisure habits and motor abilities in childhood (CHILT-project). Int J Obes Relat Metab Disord 2004; 28: 22–26. 9 Okely AD, Booth ML, Chey T. Relationships between body composition and fundamental movement skills among children and adolescents. Res Q Exerc Sport 2004; 75: 238–247. ¨rima¨e T. Relationships of physical activity and 10 Raudsepp L, Ju somatic characteristics with physical fitness and motor skill in prepubertal girls. Am J Hum Biol 1997; 9: 513–521. 11 Fernhall B, Pitetti KH, Vukovich MD, Stubbs N, Hensen T, Winnick JP et al. Validation of cardiovascular fitness field tests in children with mental retardation. Am J Ment Retard 1998; 102: 602–612. 12 Pitetti KH, Yarmer DA, Fernhall B. Cardiovascular fitness and body composition of youth with and without mental retardation. Adapt Phys Act Quart 2001; 18: 127–141. 13 Sutherland G, Couch MA, Iacono T. Health issues for adults with developmental disability. Res Dev Disabil 2002; 23: 422–445. 14 Lan M. The implementation of the target oriented curriculum in special schools for pupils with mental handicap in Hong Kong Paper presented at the International Special Education Congress, Manchester, UK. Retrieved March 1, 2004 from http://www. isec2000org.uk/abstracts/papers_a/au_1.htm. 2000. 15 Hong Kong Health and Welfare Bureau. Hong Kong rehabilitation programme plan (1998–99 to 2002–03). Towards a new rehabilitation era Retrieved March 1, 2004, from http://www.info.gov.hk/hwb/english/WHATSNEW/RPP/TOC.HTM 1999. 16 Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. Br Med J 2000; 320: 1240–1261. 17 Hong Kong Education and Manpower Bureau. A brief report on the surveys ’physical fitness status of Hong Kong school pupils’, 1998–2000 Retrieved 2004, from http://cd.ed.gov.hk/peweb/ kla.asp. 18 Johnson BL, Nelson JK. Practical measurements for evaluation in physical education 4th edn. Burgess Publishing Company, Minneapolis, MN, 1986.


19 Pizarro DC. Reliability of the health related fitness test for mainstreamed educable and trainable mentally handicapped adolescents. Adapt Phys Act Quart 1990; 7: 240–248. 20 Johnson RE, Lavay B. Fitness testing for children with special needs: an alternative approach. J Phys Ed Rec Dance 1989; 60: 50–53. 21 Turley KR, Wilmore JH, Simons-Morton B, Williston JM, Epping JR, Dahlstrom G. The reliability and validity of the 9-minute run in third-grade children. Pediatr Exerc Sci 1994; 6: 178–187. 22 Vodak PA, Wilmore JH. Validity of the 6-min jog-walk and the 600-yard run-walk in estimating endurance capacity in boys, 9–12 years of age. Res Quart Ex Sport 1975; 46: 230–234. 23 Heyward VH, Stolarczyk LM. Applied Body Composition Assessment. Human Kinetics: Champaign, IL, 1996. 24 Ulrich DA. Test of Gross Motor Development Examiner’s Manual 2nd edn. Pro-Ed: Austin, TX, 2000. 25 Okely AD, Booth ML, Patterson JW. Relationship of cardiorespiratory endurance to fundamental movement skill proficiency among adolescents. Pediatr Exerc Sci 2001; 13: 380–391. 26 Okely AD, Booth ML, Patterson JW. Relationship of physical activity to fundamental movement skills among adolescents. Med Sci Sports Exerc 2001; 33: 1899–1904. 27 Baumgartner TA. Norm-referenced measurement: reliability. In: Safrit MJ, Wood TM (eds). Measurement Concepts in Physical Education and Exercise Science. Human Kinetics: Champaign, IL, 1989. pp 45–71. 28 Cohen J. Statistical Power Analysis for the Behavioral Sciences. Lawrence Erlbaum Associations: Hillsdale, NJ, 1988. 29 Hong Kong Commission on Youth. Youth in Hong Kong: A Statistical Profile 2002 Retrieved 2005, from http://www.info. gov.hk/coy/eng/report/index.htm. 30 US Department of Health and Human Services. Healthy People 2010: Understanding and Improving Health. US Government Printing Office: Washington, DC, 2000. 31 Rimmer JH, Braddock D, Fujiura G. Prevalence of obesity in adults with mental retardation: implications for health promotion and disease prevention. Ment Retard 1993; 31: 105–110. 32 Chatrath R, Shenoy R, Serratto M, Thoele DG. Physical fitness of urban American children. Pediatr Cardiol 2002; 23: 608–612. 33 Manios Y, Yiannakouris N, Papoutsakis C, Moschonis G, Magkos F, Skenderi K et al. Behavioral and physiological indices related to BMI in a cohort of primary schoolchildren in Greece. Am J Hum Biol 2004; 16: 639–647. 34 Lloyd LK, Bishop PA, Walker JL, Sharp KR, Richardson MT. The influence of body size and composition on the FITNESSGRAM test performance and the adjustment of FITNESSGRAM test scores of skinfold thickness in youth. Meas Phys Ed Exerc Sci 2003; 7: 205–226. 35 Fernhall B, Pitetti KH, Stubbs N, Stadler L. Validity and reliability of the 1/2 mile run-walk as an indicator of aerobic fitness in children with mental retardation. Pediatr Exerc Sci 1996; 8: 130–142. 36 Chaiwanichsiri D, Sanguanrungsirikul S, Suwannakul W. Poor physical fitness of adolescents with mental retardation at Rajanukul School, Bangkok. J Med Assoc Thai 2000; 83: 1387– 1392. 37 Gillespie M. Cardiovascular fitness of young Canadian children with and without mental retardation. Ed Train Dev Disabil 2003; 38: 296–301. 38 MacDonncha C, Watson AWS, McSweeney T, O’Donovan DJ. Reliability of Eurofit physical fitness items for adolescent males with and without mental retardation. Adapt Phys Act Quart 1999; 16: 86–95. 39 Fernhall B, Pitetti KH. Leg strength is related to endurance run performance in children and adolescents with mental retardation. Pediatr Exerc Sci 2000; 12: 324–333.


867 40 Deforche B, Lefevre J, De Bourdeaudhuij I, Hills A, Duquet W, Bouckaert J. Physical fitness and physical activity in obese and nonobese Flemish youth. Obes Res 2003; 11: 434–441. 41 Hong Y, Chan KM, Li JX. Health-related physical fitness of school children in Hong Kong and mainland China. J Comp Phys Ed Sport 1998; 20: 2–10. 42 Macfarlane DJ. Some disturbing trends in the level of habitual physical activity in Hong Kong primary school children: preliminary findings. Hong Kong J Sports Med Sports Sci 1997; 5: 42–46. 43 McKenzie TL, Sallis JF, Broyles SL, Zive MM, Nader PR, Berry CC et al. Childhood movement skills: predictors of physical activity

in Anglo American and Mexican American adolescents? Res Quart Exerc Sport 2002; 73: 238–244. 44 Reeves L, Broeder CE, Kennedy-Honeycutt L, East C. Relationship of fitness and gross motor skills for five- to six-yr-old children. Percept Motor Skills 1999; 89: 739–747. 45 Draheim CC, Williams DP, McCubbin JA. Prevalence of physical inactivity and recommended physical activity in communitybased adults with mental retardation. Ment Retard 2002; 40: 436–444. 46 Chow B, Frey GC, Cheung S-Y, Louie L. An examination of health-related physical fitness levels in Hong Kong youth with intellectual disability. J Exerc Sci Fit 2005; 3: 9–16.
