The mounting evidence associating physical activity (PA) and various dimensions of health (5,50) coupled with increasing prevalence of childhood overweight and obesity (17,20) have led to a greater interest in measuring children's activity levels. In addition to measuring the number of minutes children devote to different types of activities, it is often desirable to be able to estimate rates of energy expenditure (EE; or energy cost) in those activities. This permits an overall estimate of daily EE and hence allows comparison with, and recommendations for, energy intake. It also allows for estimation of how much time children spend in sedentary, moderate, and vigorous PA (MVPA), as these constructs are defined relative to rates of EE, for example, MVPA ≥ 3 METs (26).
Unfortunately, there are relatively few data on the energy cost of everyday activities in children and adolescents. As a result, adult data are often used as surrogates (1). However, evidence suggests that when performing the same or similar activities, in particular walking and running, adults have lower rates of EE when expressed as oxygen consumption (V˙O2) or caloric cost per unit body mass (2,33,37,46). This increased energy cost is likely to be due to size- and physiology-related mechanisms, such as immature motor patterns and increased breathing rates (36). Disparities may also be due to children living in an adult-sized world. For example, children need to reach higher and further to manipulate everyday objects that are proportionally larger and heavier for children compared to adults.
The most desirable solution would be to use a compendium compiled from EE measured in children. However, such a compendium will always be a work in progress because of the methodological difficulties associated with such measurements, differences across age groups, and the wide and evolving range of activities that need to be examined. Although there are still gaps in child-measured data, we need strategies for assigning energy costs on the basis of adult values.
Existing techniques used to assign energy costs tochildren.
Various strategies have been used to estimate rates of EEin children on the basis of adult values. These include the following:
- Using Metabolic Units (METs). Many compendia express rates of EE in METs, which are multiples of resting metabolic rate (RMR), either measured or estimated (8). For adults, RMR (i.e., 1 MET) is usually taken to be 3.5 mL O2·kg−1·min−1 or 4.2 kJ·kg−1·h−1 (=1 kcal·kg−1·h−1). Mass-specific child RMR is considerably higher and varies with sex, age, body mass, pubertal status, and body composition (13,23,38). It has been suggested that using METs may assist in equalizing differences in adult/child energy costs (13). For example, a value of 2 METs for a given activity may then equate to 7 mL·kg−1·min−1 for adults but 9 mL·kg−1·min−1 for children, given the difference in RMR.
- The FAO/WHO/UNU Method (12). The FAO/WHO/UNU (12) recommendations involve multiplying either measured or estimated child RMR by factors for different types of activities. Going to school and other light activities are classified as 1.5 METs, moderate activity = 2.2 METs, and high activity = 6.0 METs.
- The Method of Torun (47). The Torun (47) recommendation involves using prescribed MET values (for children aged 1-15 yr) or multiplying adult MET values by an age-correction factor, depending on the activity being performed. Sedentary activities are classified as follows: lying down = 1.1 METs, sitting = 1.2 METs, and standing = 1.4 METs. Nonwalking light activities are assigned 2.0 METs for 1.5-6.9 yr olds or 2.2 METs for 6.0-14.9 yr olds. Walking at normal pace on level ground OR moderate activities are assigned 2.2 METs for 1.5-6.9 yr olds or 2.9 METs for 6.0-14.9 yr olds. Assigning MET costs for heavier activities involves multiplying adult MET values by age-correction factors of 0.50 for 1.5-6.9 yr olds, 0.65 for 6.0-12.9 yr olds, and 0.80 for 13.0-14.9 yr olds.
- The Method of Sallis et al. (37). In contrast, the Sallis et al. (37) recommendation involves applying an age-correction factor to adult V˙O2 data. The correction factors are on the basis of five large (n > 100) treadmill studies and decrease with age from 1.37 at age 5 yr to 1.03 at age 17 yr.
The assignment techniques yield different energy cost values for individual activities. There has been little research comparing these methods, and there is currently no consensus on which energy cost assignment technique is best. Most raw data sets are too small to make definitive decisions. Spadano et al. (39) compared the age adjustments recommended by Torun (47), the FAO/WHO/UNU (12), and the adult METs from the adult compendium compiled by Ainsworth et al. (1) to measure METs in a sample of seventeen 12-yr-old girls while sitting, standing, and walking at three intensities. It was concluded that adult METs provided the best overall approximation of the energy cost across the five activities. A large study by Harrell et al. (13) compared children's energy costs measured in 18 activities and converted to METs using measured child RMR, with estimations from the compendium-derived adult METs (1). The use of child RMR resulted in a more accurate estimation of the energy cost of activities than the standard adult RMR. The authors therefore recommended the use of MET values derived from published adult compendia combined with age- and/or and pubertal status-adjusted RMR predictions when child-measured values are not available.
This review will add to the work undertaken by Spadano et al. (39) and Harrell et al. (13) by collating and combining all existing energy cost data collected on children. The aims of the study are to
- a) review current data of energy costs of everyday activities in children, and
- b) compare four existing techniques for assigning energy costs to children.
An extensive literature search was conducted for studies that report the energy cost of children and adolescents performing everyday activities. The literature search involved searching both online and CD-ROM bibliographic databases (MEDLINE, Sport Discus, Educational Resources Information Centre, Current Contents, Academic Search Elite, Australian Digital Theses Program, Digital Dissertations) using the following keywords: energy cost, energy expenditure, caloric expenditure, oxygen consumption, aerobic demands, and MET. The following modifiers were used in conjunction with the keywords: child/children, adolescent/adolescence, youth, boy, and girl. As many studies use criterion energy cost measures to validate accelerometry, pedometry, and HR measures, the keywords, namely, accelerometry/accelerometer, pedometer/pedometry, and heart rate, were also used with the following modifiers: valid/validity/validation. The reference lists of all published studies that were obtained through the computer searches were also examined and cross-referenced. All studies that involved measuring EE in healthy 6.0- to 17.9-yr-old children and adolescents using criterion methods (i.e., oxygen consumption or CO2 breath tests) were included. Only those studies that were published in English were used (n = 54).
As studies report rates of EE in different metrics, these were initially standardized to milliliters of oxygen per kilogram per minute. Studies that reported data in calories or joules were converted to V˙O2 using a standard energy equivalent for oxygen uptake: 21 kJ·L−1 O2 (27). Child METs for each of the activities were calculated by dividing mean oxygen consumption by mean predicted child-specific RMR for each sample. The Schofield equations (38) were used to estimate RMR. These equations have been validated against indirect calorimetry (34,52). Although some studies measured the RMR of their subjects, details were rarely provided. The measurement of RMR in children is extremely difficult, and studies vary in their methodologies (38). Therefore, standard RMR equations (38) were used to predict METs across all studies to reduce methodological variability.
The Schofield equations require body mass as an input, which was not always reported in the studies located. If body weight was not reported, mean age-specific values were calculated using a regression equation on the basis of the masses of 70216 Australian children aged between 5 and 18 yr measured between 1937 and 2000: male mass (kilograms) = 0.104 (year of measurement) + 4.02 (age) − 210.5, r = 0.81, SEE = 8.1 kg; female mass (kilograms) = 0.058 (year of measurement) + 3.7 (age) − 115.9, r = 0.82, SEE = 7.7 kg [on the basis of the data from Ref. (25)]. Where studies combined male and female values, weighted means were used applying the RMR and estimated mass formulae for both sexes. Less than 1% of the total number of studies did not report body mass.
Creation of a comparative adult-child data set.
Activities with comparable adult MET values listed in the adult compendium (1) were included in the data set, except continuous running and walking at various speeds that formed separate data sets. Cycling was excluded because of complexities such as speed, type of bicycle, gears, and cycling surface, all of which affect EE. To calculate a sample-weighted mean child MET score across all studies, the Monte Carlo data simulation procedure was used. This procedure involves using a random generator to create "pseudo data" in a normal distribution, matching the original data set in summary characteristics where only means and SD were available (45). The simulation procedure was applied iteratively until the mean of the generated data was within 0.5% and the SD was within 2.5% of those of the original data set. Subsequently, four existing assignment techniques were used for comparison: adult METs, FAO/WHO/UNU (12), Torun (47), and a modified version of the Sallis et al. (37) technique. The adult METS were derived from the adult compendium (1). The mean ages of the children from each energy cost study were then used to assign MET estimations on the basis of the age-specific recommendations of the FAO/WHO/UNU (12), Torun (47), and Sallis et al. (37). Some subjective decisions were made to assign the FAO/WHO/UNU and Torun energy costs, for example, is "hanging out the washing" considered "standing-little or no movement" or "walking at normal pace OR moderate activity?" To calculate METs using the Sallis et al. technique, age-correction factors were applied to adult METs, not V˙O2 values. However, it is acknowledged that Sallis et al. (37) did not investigate the MET technique as a means of assigning energy costs to children. Therefore, by converting V˙O2 values to METs, the Sallis technique referred to in subsequent analyses is not an exact application of the recommended technique and will be referred to hereafter as the modified Sallis technique.
Creation of a comparative adult-child data set for continuous walking and running.
For the continuous locomotor activities of running and walking, energy cost measurements had been made at a range of intensities. Grade running and walking were not included because of lack of data. To combine these data, a Monte Carlo procedure was used at each intensity. As the FAO/WHO/UNU (12) and Torun (47) techniques are not designed to assign energy costs to running and walking at different speeds (each speed is likely to be assigned the same MET value),only adult METs were compared to predicted child METs for running and walking. The running and walking pseudo data sets were converted into METs by dividing by Schofield-predicted RMR (38).
Paired t-tests were performed to determine whether there were any significant differences between the estimated child METs, calculated from the energy cost review on the basis of measured V˙O2 values, and the four energy cost assignment techniques. The assignment techniques were also compared using Bland-Altman analyses and intraclass coefficients (ICC). Multiple regression was used to create predictive equations for the MET cost of walking and running on the basis of age and speed.
Energy cost data in children and adolescents.
Data from 54 studies were combined with walking and running forming separate data sets. The resultant data set (excluding walking and running) contained 5592 data points encompassing 51 activities. Table 1 summarizes the data obtained via the literature review and lists the adult compendium codes, compendium descriptions, FAO/WHO/UNU classifications, and Torun classifications used to assign MET classifications for each of the 51 activities. The majority of studies measured EE by indirect calorimetry using respiration chambers, Douglas bags, metabolic carts, or portable oxygen consumption systems. Two studies (16,30) measured energy costs using CO2 breath tests. The mean ± SD predicted RMR across all studies was 4.59 ± 1.07 mL O2·kg−1·min−1.
Comparing existing techniques to assign energy costs to children.
The 51 nonlocomotor activities, weighted mean METs calculated from the pseudo data set, weighted mean ages, and METs assigned by the four techniques are displayed in Table 2. Table 3 compares the biases, SEM, upper and lower limits of agreement (LoA), and ICC for the METs assignment techniques. Figure 1 shows the Bland-Altman plot comparing the estimated child METs, on the basis of measured V˙O2 values, and adult METs. Paired t-tests found significant differences between the measured child METs and both the FAO/WHO/UNU METs (t = 4.96, P < 0.0001) and the Torun METs (t = 2.78, P = 0.008).
Comparing existing techniques to assign running and walking energy costs to children.
Paired t-tests revealed significant differences between child and adult MET values for running (t = −11.35, P < 0.0001) and walking (t = 8.004, P < 0.0001). The best prediction equations for METs during running and walking use both age (in years) and speed (v, in meters per second). The running MET prediction equation was on the basis of 1974 data points: 0.27 age + 1.91 speed (meters per second) + 0.46; r = 0.61, SEE = 1.38 METs. The walking MET prediction equation was on the basis of 1187 data points: 0.07 age − 1.21 speed (meters per second) + 1.65 speed2 (meters per second) + 1.72; r = 0.65, SEE = 1.0 MET. The studies used to collect the data for the running and walking analyses are available by request to the authors.
The Bland-Altman analyses revealed that using adult METs would have been the most accurate technique to assign energy costs to children in the 51 selected nonlocomotor activities if measured values had not been available. Overall, published adult METs were slightly lower than the estimated child METs calculated using measured V˙O2 data (bias = −0.03 METs). The adult METs technique produced the lowest bias and the tightest limits of agreement compared to the other techniques. Energy costs were also underestimated using the Torun METs and FAO/WHO/UNU METs techniques. Although the METs calculated using the modified Sallis correction factors overestimated the child METs, the values were not significantly different statistically. It is important to note that these comparisons were on the basis of only 51 activities, so the results should be interpreted with caution.
Using adult values to predict child energy costs of continuous running and walking has been shown to produce significant biases (37). Previous research suggests that the V˙O2 of walking and running increases with speed but decreases with age (2,37). However, within this data set, the MET cost of walking and running (as opposed to the V˙O2) was found to generally increase with speed and increase with age. This finding is consistent with Torun's findings that the MET cost of walking and moving around increases with age from preschool years to adolescence (47). Hence, Torun's recommendation for assigning METs to heavy activities involves applying correction factors that get progressively closer to 1.0 with increasing age, this is, 0.65 for 6.0 to 12.9 yr olds and 0.80 for 13.0 to 14.9 yr olds. These findings suggest that although both RMR and oxygen cost of locomotion decrease with age, the declines do not occur at a proportional rate. It may be that the oxygen cost of locomotor activities remains elevated compared to changes in RMR because of the inefficiency in gait in children (9,35,36). Alternatively, the oxygen cost of running and walking may remain relatively stable, in comparison to other activities, as children may be relatively more efficient at running and walking than at other tasks (such as household chores), as running and walking are more practiced activities. In addition, the environmental factors related to a child living in an adult's world (e.g., table heights, size of equipment, etc.) are not as influential during running and walking as children are usually able to self-select their gait pattern without having to maneuver around or use adult-sized paraphernalia. Hence, changes in the oxygen cost of running and walking may be less evident than in other activities. As the MET cost of walking and running is significantly influenced by age, adult MET values should not be used to assign energy costs to children for these activities.
The analysis of everyday nonlocomotor activities found that eight activities resulted in mean differences of ≥1.5 METs. The energy costs of rollerblading (+6.0 METs), step aerobics (+2.3), lacrosse (+1.6), and cricket (+1.5) were overestimated using adult METs, whereas bowling (−1.8 METs), soccer (−1.7), dusting the table (−1.7), and climbing stairs (−1.5) were underestimated. As the walking and running analyses revealed inaccuracies when using adult METs to assign costs to children, these sporting and household chore activities that involve interspersed periods of walking and running at various intensities may be more poorly estimated using adult METs. Another likely explanation for the increased mean error in these activities is the variation in speed and intensity possible within many of these activities. In particular, the rollerblading METs in the adult compendium may be on the basis of a faster speed than the child study (i.e., 1.4-3.6 m·s−1). It is important to note that complications arising from individual variations in intensity, and so on, are not limited to energy cost assignments in children. Researchers working with adult data are also conscious of the limitations of using compendia-derived EE. Ainsworth et al. (1) warn that: "It (the compendium) does not take into account individual differences when estimating the energy cost of physical activity in individuals."
The modified Sallis technique was relatively successful at predicting the MET cost of the 51 measured activities. However, because the technique involves applying correction factors >1.0 to adult values, large errors occurred when the estimated child METs, on the basis of measured V˙O2 values, were lower than the published adult METs (e.g., rollerblading and step aerobics). It seems that the adjustments recommended by Torun (47) were inaccurate in heavy activities where the adult MET cost was less than the measured child MET cost, for example, swimming, basketball, and soccer. In these activities, the Torun-recommended adjustment factors of 0.65 or 0.80 widen the gap between child and adult values and subsequently underestimate the energy cost in children. The FAO/WHO/UNU recommendations were limited by a maximum MET allocation of 6.0, as many of the activities reviewed had measured energy costs >6.0 METs.
There are some limitations in this study that should be acknowledged. Although the analyses incorporated all published energy cost data available at the time the study was undertaken, the combined data set contains a limited range of activities. Predicted, rather than measured, RMR was used to calculate METs. As the RMR equations use age as a predictor variable, differences in RMR between similarly aged children may not have been accounted for. In addition, the range of subjects studied was relatively narrow. The energy cost studies typically involved normal-weight white children. Therefore, further research is required to confirm whether using adult METs to assign energy costs to children is effective across a wide range of activities and across diverse populations of children. The method of data generation should also be considered. Raw data from each study were not used in the analyses. The data generated using the Monte Carlo technique assume the data sets were normally distributed, and this may not have been the case in all studies. Although this is likely to have marginal effect on the sample-weighted mean MET scores for each activity and the walking and running regression equations, the estimations maybe less accurate than if raw data were used. Finally, as acknowledged previously, the Sallis technique was modified to report energy costs as METs rather than oxygen consumption in units of milliliters per kilogram per minute.
CONCLUSION AND RECOMMENDATIONS
In conclusion, although compendia of EE exist for adults (1), there is a lack of data on the energy cost of everyday activities in children. Nevertheless, this study provides evidence that adult METs are the best existing technique to assign energy costs to children in most activities when measured values are not available. However, as adult METs are inaccurate in assigning energy costs to children in continuous running and walking activities, predictive equations on the basis of age and speed should be used. This study provides support for the recommendations of Harrell et al. (13), that is, when research designs require calculation of gross EE expressed as joules, calories, or oxygen cost per unit body mass, adult METs should be multiplied by a child-specific RMR either measured or estimated.
Clearly, there is a need for more studies investigating the energy cost of activities in a wide range of groups, varying in age, ethnicity, and body composition, to allow for comparison. In addition, the breadth of measured activities needs to be widened. For example, there are very few data regarding sports and playground games undertaken by children. As many of these activities are also missing from adult compendia (1), it is even more imperative that energy cost data in these activities be collected. The extensive literature search undertaken in this article provides energy cost data that can serve as a basis for the development of a compendium of physical activities for children.
The authors thank Dr Jim Dollman for his advice and assistance throughout the project.
Dr Kate Ridley was supported by an Australian Postgraduate Award at the time of this project.
The results of the present study do not constitute endorsement by ACSM.
1. Ainsworth BE, Haskell WL, Whitt MC, et al. Compendium of physical activities: an update of activity codes and MET intensities. Med Sci Sports Exerc
. 2000;32(9 Suppl):S498-516.
2. Åstrand PO. Experimental Studies of Physical Working Capacity in Relation to Sex and Age
. Copenhagen (Denmark): Munksgaard; 1952. p. 171.
3. Banerjee B, Saha N. Energy intake and expenditure of Indian schoolboys. Br J Nutr
4. Banerjee B, Saha N. Energy cost of some common physical activities of Chinese schoolboys. Ann Nutr Metab
5. Bauman AE. Updating the evidence that physical activity is good for health: an epidemiological review 2000-2003. J Sci Med Sport
6. Bedale EM. Energy expenditure and food requirements of children at school. Proc R Soc Lond
7. Cox L. Comparison of Television Viewing, Arcade Game Play, and Resting Metabolic Rates in Youth
[Masters dissertation]. Chapel Hill (NC): University of North Carolina; 1999. p. 67.
8. Dill DB. The economy of muscular exercise. Physiol Rev
9. Ebbeling CJ, Hamill J, Freedson PS, Rowland TW. An examination of efficiency during walking in children and adults. Pediatr Exerc Sci
10. Eisenmann JC, Strath SJ, Shadrick D, Rigsby P, Hirsch N, Jacobson L. Validity of uniaxial accelerometry during activities of daily living in children. Eur J Appl Physiol
11. Eston R, Rowlands A, Ingledew D. Validity of heart rate, pedometry and accelerometry for predicting the energy cost of children's activities. J Appl Physiol
12. FAO/WHO/UNU. Energy and Protein Requirements
. Geneva (Switzerland): World Health Organization; 1985. p. 206.
13. Harrell JS, McMurray RG, Baggett CD, Pennell ML, Pearce PF, Bangdiwala SI. Energy costs of physical activities in children and adolescents. Med Sci Sports Exerc
14. Heil DP. Predicting activity energy expenditure using the Actical activity monitor. Res Q Exerc Sport
15. Horswill C, Kien C, Zipf W. Energy expenditure in adolescents during low intensity, leisure activities. Med Sci Sports Exerc
16. Horswill C, Zipf W, Kien C. Measuring energy costs of leisure activity in adolescents using a CO2
breath test. Med Sci Sports Exerc
17. Jolliffe D. Extent of overweight among US children and adolescents from 1971 to 2000. Int J Obes Relat Metab Disord
18. Kjendlie PL, Ingjer F, Madsen O, Stallman RK, Stray-Gundersen J. Differences in the energy cost between children and adults during front crawl swimming. Eur J Appl Physiol
19. Klesges RC, Shelton ML, Klesges LM. Effects of television on metabolic rate: potential implications for childhood obesity. Pediatrics
20. Lobstein T, Baur L, Uauy R. Obesity in children and young people: a crisis in public health. Obes Rev
. 2004;5(Suppl 1):4-85.
21. Louie L, Eston R, Rowlands A, Tong K, Ingledew D, Fu F. Validity of heart rate, pedometry, and accelerometry for estimating the energy cost of activity in Hong Kong Chinese boys. Pediatr Exerc Sci
22. Mahadeva K, Passmore R, Woolf B. Individual variations in the metabolic cost of standardised exercises: the effects of food, age, sex and race. J Physiol
23. Malina RM, Bouchard C, Bar-Or O. Growth, Maturation and Physical Activity
. Champaign (IL): Human Kinetics; 2004. p. 728.
24. Merati G, Negrini S, Sarchi P, Mauro F, Veicsteinas A. Cardio-respiratory adjustments and cost of locomotion in school children during backpack walking (the Italian backpack study). Eur J Appl Physiol
25. Olds TS, Dollman J, Norton KI, Harten NR. A century of growth in Australian children. In: Body Composition Assessment in Children and Adolescents, Medicine and Sport Science
. Jurimae T, Hills A (Eds.). Basel (Switzerland): Karger; 2001, p. 85-103.
26. Pate RR, Pratt M, Blair SN, et al. Physical activity and public health: a recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA
27. Péronnet F, Massicotte D. Table of nonprotein respiratory quotient: an update. Can J Sport Sci
28. Pfeiffer K, Schmitz K, McMurray R, Treuth M, Murray D, Pate R. Physical activities in adolescent girls: variability in energy expenditure. Am J Prev Med
29. Poujade B, Hautier CA, Rouard A. Determinants of the energy cost of front-crawl swimming in children. Eur J Appl Physiol
30. Richards M, Davies P. Energy cost of activity assessed by indirect calorimetry and a 13
breath test. Med Sci Sports Exerc
31. Ridley K. The Multimedia Activity Recall for Children and Adolescents MARCA: Development and Validation
[dissertation]. Adelaide (Australia): University of South Australia; 2005. p. 335.
32. Rieper H, Karst H, Noack R, Johnsen D. Intra- and inter-individual variations in energy expenditure of 14-15-year-old school-girls as determined by indirect calorimetry. Br J Nutr
33. Robinson S. Experimental studies of physical fitness in relation to age. Arbeitsphysiologie
34. Rodriguez G, Moreno LA, Sarria A, Fleta J, Bueno M. Resting energy expenditure in children and adolescents: agreement between calorimetry and prediction equations. Clin Nutr
35. Rowland TW, Auchinachie JA, Keenan TJ, Green GM. Physiologic responses to treadmill running in adult and prepubertal males. Int J Sports Med
36. Rowland TW, Green GM. Physiological responses to treadmill exercise in females: adult-child differences. Med Sci Sports Exerc
37. Sallis JF, Buono MJ, Freedson PS. Bias in estimating caloric expenditure from physical activity in children. Sports Med
38. Schofield W. Predicting basal metabolic rate, new standards and review of previous work. Hum Nutr Clin Nutr
39. Spadano JL, Must A, Bandini LG, Dallal GE, Dietz WH. Energy cost of physical activities in 12-y-old girls: MET values and influence of body weight. Int J Obes Relat Metab Disord
40. Spady DW. Total daily energy expenditure of healthy, free ranging children. Am J Clin Nutr
41. Spurr GB, Reina JC, Barac-Neito M. Marginal malnutrition in school-aged Colombian boys: metabolic rate and estimated daily energy expenditure. Am J Clin Nutr
42. Taylor CM, Lamb MW, Robertson ME, MacLeod G. Energy expenditure for quiet play and cycling of boys 7 to 14 years of age. J Nutr
43. Taylor CM, Pye OF, Caldwell AB. Energy expenditure of -9 to 11-year-old boys and girls (1) standing drawing and (2) dressing and undressing. J Nutr
44. Taylor CM, Pye OF, Caldwell AB, Sostman ER. Energy expenditure of boys and girls 9 to 11 years of age (1) sitting listening to the radio, (2) sitting singing, and (3) standing singing. J Nutr
45. Tomkinson GR, Léger L, Olds TS, Cazorla G. Secular trends in the performance of children and adolescents (1980-2000): an analysis of 55 studies of the 20m shuttle run test in 11 countries. Sports Med
46. Torun B. Inaccuracy of applying energy expenditure rates of adults to children. Am J Clin Nutr
47. Torun B. Energy cost of various physical activities in healthy children. In: Activity, Energy Expenditure and Energy Requirements of Infants and Children
. Schurch B, Scrimshaw N (Eds.). Lausanne (Switzerland): International Dietary Energy Consulting Group; 1990, p. 139-83.
48. Treuth MS, Butte NF, Wong WW. Effects of familial predisposition to obesity on energy expenditure in multiethnic prepubertal girls. Am J Clin Nutr
49. Treuth MS, Schmitz K, Catellier DJ, et al. Defining accelerometer thresholds for activity intensities in adolescent girls. Med Sci Sports Exerc
50. United State Department of Health and Human Services. The Surgeon General's Report on Physical Activity and Health
. Washington (DC): US Government Printing Office; 1996. p. 278.
51. Vermorel M, Vernet J, Bitar A, Fellmann N, Coudert J. Daily energy expenditure, activity patterns, and energy costs of the various activities in French 12-16-y-old adolescents in free living conditions. Eur J Clin Nutr
52. Wong W, Butte N, Hergenroeder A, Hill R, Stuff J, O'Brian Smith E. Are basal metabolic rate prediction equations appropriate for female children and adolescents? J Appl Physiol