ORIGINAL ARTICLE

Exercise modality and metabolic efficiency in children

Bob G. F. Verweij & Lee Stoner & Sarah P. Shultz

Received: 24 February 2013 /Revised: 22 April 2013 /Accepted: 23 April 2013 /Published online: 5 May 2013# Springer-Verlag Berlin Heidelberg 2013

Abstract Current exercise prescription guidelines for chil-dren recommend at least 60 min of moderate-to-vigorousphysical activity every day. However, little is known aboutthe efficacy of different cardiorespiratory exercise modali-ties prescribed to the pediatric cohort. Sixteen healthy chil-dren (8–12 years) completed 5-min trials of treadmillwalking, cycling, and elliptical training in a randomizedorder. The treadmill walking speed was determined frommeasurements collected during a self-selected walking trial.The workloads for treadmill walking, cycling, and ellipticaltraining were matched (40.3 W). Mechanical efficiency(ME%), perceived exertion (RPE), oxygen uptake, metabolicequivalents, and net energy expenditure were measured.ME% in walking was significantly higher than in cycling(P=0.001) and elliptical training (P<0.001), and cyclingwas significantly higher than elliptical training (P=0.003).RPE in walking was significantly lower than in ellipticaltraining (P=0.006) but not from cycling (P=0.314), and cy-cling resulted in significantly lower RPE than elliptical train-ing (P=0.021). Conclusion: Treadmill walking appears to bethe most efficacious exercise prescription for otherwisehealthy children; however, longitudinal studies need to beimplemented to investigate the long-term benefits of eachexercise modality.

Keywords Physical activity . Pediatrics . Bioenergetics .

Physical exertion

Introduction

A number of lifestyle factors [19, 26, 38] are leading togrowing rates of physical inactivity among children [20].

This is predisposing children to impaired cognitive devel-opment [16], motor skill development [32], musculoskeletalhealth [30], and cardiovascular health [30]. These impair-ments are leading to chronic noncommunicable diseases(NCDs), including obesity [5], metabolic syndrome [4],cardiovascular diseases [37], and subsequent decreased lifeexpectancy [37]. Moreover, the advancement of these NCDsis placing a burden not only on the individual but also onsociety at large. For example, the direct annual costs ofchildhood obesity in the USA are estimated at about $14.3billion [2, 34]. These NCDs have a high likelihood ofpersisting into adulthood [27], placing a continued burdenon the individual and on society.

Appropriate exercise prescription during childhood hasbeen associated with improved lipid profiles and insulinlevels in obese children [8, 12] and improved lipid profiles,blood pressure, and adiposity in normal-weight children[30]. Increasing physical activity has also been shown todecrease the risk of cardiovascular disease and increase lifeexpectancy [37]. The World Health Organization, in recog-nition of declining physical activity levels and the subse-quent rise in NCDs, published the Global Recommendationson Physical Activity and Health [36]. The WHO recom-mends that children aged 6–17 years participate in at least60 min of moderate-to-vigorous physical activity every dayand perform vigorous exercise, muscle-strengthening andbone-strengthening exercise, on at least 3 days each week[36]. These recommendations are supported by the BritishAssociation of Sport and Exercise Sciences [24], the Centersfor Disease Control and Prevention [3], and the NationalAssociation for Sport and Physical Education [22].

Although professional guidelines prescribe substantialaerobic activity, little is known about the efficacy of differ-ent cardiorespiratory exercise modalities. Because walkingis a common form of physical activity, there has beensignificant research focus on the energetics of walking [21,28]. However, some children, specifically those carryingadditional mass, may find a weight-bearing task, such aswalking, difficult [28]. Stationary cycling is considered non-weight-bearing, and aerobic exercise interventions, based on

B. G. F. VerweijDepartment of Medicine, Utrecht University, Utrecht, Netherlands

L. Stoner : S. P. Shultz (*)School of Sport and Exercise, Massey University, PO Box 756,Wellington 6140, New Zealande-mail: S.P.Shultz@massey.ac.nz

Eur J Pediatr (2013) 172:1191–1196DOI 10.1007/s00431-013-2025-4

stationary cycling, have reported improved fitness levels inprepubertal children [33] and adolescents [14]. However,these studies did not examine the acute metabolic costsassociated with stationary cycling. While elliptical trainersare not often prescribed for children, elliptical training hasproduced lower rates of perceived exertion (RPEs) whencompared to treadmill running and stationary cycling inadult populations [1, 10]. Further research is required todetermine whether this exercise modality can be efficacious-ly prescribed in children. An exercise modality will mostlikely be more effective if the activity can be aerobicallysustained by the child. Therefore, the purpose of this studyis to compare the metabolic efficiency of three cardiorespi-ratory exercise modalities with varying degrees of weightbearing (treadmill, cycling, and elliptical training). The nullhypothesis is that these three exercise modalities will evokea comparable metabolic workload.

Materials and methods

Participants

Sixteen healthy children (8–12 years) were recruited fromschools in the central Wellington area. Data were not in-cluded from one participant, who did not complete the entiretesting session. Prior to participation, parents completed aself-reported health history and physical activity question-naire. Participants were excluded if they had had a history oflower limb injury within 6 months, previous orthopedicsurgery, diabetes, and moderate to severe asthma. Partici-pants and their parent or legal guardian gave informedwritten assent and consent, respectively, in accordance withthe Massey University Human Ethics Committee. Table 1summarizes the physical characteristics of the participants.

Experimental design

Speed and cadence for each exercise modality was deter-mined from measurements collected during self-selectedwalking trials. A 6-m walkway was established using twopairs of timing gates spaced 1 m wide. Participants were

asked to walk normally through the gates, without anyreferences made to speed, in order to collect the most naturalwalking speed possible. Self-selected speed was thenconverted from meters per second to kilometers per hourfor the purpose of establishing the treadmill speed. Cadencewas calculated as steps per minute. The cycling and ellipti-cal training were set at a constant load, which was calculatedbased on the participant’s mechanical work duringoverground walking. Because the workload on the ellipticaltraining could not be set precisely, the workload was esti-mated to the nearest 5 W; dynamic resistance was employedto maintain the set workload regardless of changes in ca-dence. Participants were asked to maintain a cadence duringthe cycling trial, which was calculated as 50 % of theoverground walking cadence. During familiarization, partic-ipants were allowed to select a comfortable cadence for theelliptical trainer; they were asked to maintain this cadencefor the duration of the trial. Participants completed thetreadmill walking, cycling, and elliptical training in a ran-domized order. All trials lasted 5 min, with a 5-min washoutperiod between exercise modalities.

Anthropometric measurements

Weight was assessed to the nearest 0.05 kg using an electricscale (A&D Instruments, Adelaide, Australia). Height wasassessed to the nearest 0.1 cm with a stadiometer (Surgicaland Medical Products, Seven Hills, Australia). Body MassIndex (BMI; in kilograms per square meter) was calculatedusing the assessed height and weight. Waist circumferencewas measured to the nearest 0.1 cm with an anthropometricmeasuring tape according to internationally standardizedprotocol [29]. Waist-to-height ratio was calculated as waistcircumference (in centimeters)/height (in centimeters).

Energetic measurements

All participants were instructed to have fasted for 4 h priorto the testing session. Breath analysis was performed with ametabolic cart (Sensormedics Corporation, Yorba Linda,CA) via a mask placed securely on the face with a headstrap. While the mask may result in underestimation ofmetabolic equivalents (METs) compared to the canopymethod, the mask has been shown to be better tolerated bychildren [18]. The metabolic cart was calibrated for air flowand gas analysis before each participant. Participants laid ina supine position for 10–20 min prior to assessing restingmetabolic rate [18]. Resting metabolic rate was collected for15 min, during which time participants continued to restquietly in a supine position while watching an animatedmovie [13].

During the exercise bouts, breath analysis was performedduring the entire 5-min trial. Data were analyzed over a

Table 1 Study population physical characteristics

Characteristic Boys (N=9) Girls (N=7)

Age (years) 9.97±1.01 9.73±0.77

Height (m) 1.42±0.09 1.40±0.05

Weight (kg) 32.5±7.93 34.96±6.43

Body Mass Index (kg/m2) 16.01±1.85 17.70±2.41

Waist circumference (cm) 57.77±6.39 61.80±9.99

Waist-to-height ratio 0.41±0.03 0.44±0.06

1192 Eur J Pediatr (2013) 172:1191–1196

period of 1 min, beginning at 3.5 min, and consideredsteady-state. Steady-state data were used for further statisti-cal analysis. METs were calculated as VO2 (in milliliters perminute per kilogram)/3.5. The rate of energy expenditure(EE) was calculated using the abbreviated Weir [35] formu-la: EE=[(1.106×RQ)+3.941]×VO2, where RQ representsthe respiratory quotient, estimated by the measured respira-tory exchange ratio (RER) and VO2 represents the corre-sponding oxygen uptake.

Net energy expenditure was calculated as the differencebetween EE at rest and during the exercise modality. Themechanical work rate (W) of each exercise modality wasconverted to kilocalories per minute and metabolic efficien-cy was calculated as mechanical efficiency (ME%)=netEE/work rate.

Rating of perceived exertion

Rating of perceived exertion was assessed after 1 min ofsteady-state oxygen consumption measurement, using theEston-Parfitt perceived exertion scale [6]. This is a curvilin-ear scale with verbal anchors and images of a person indifferent stages of exertion, corresponding with a 0–10 scorefor perceived exertion on the x-axis. This scale is believed toprovide a more reliable RPE assessment in children [7].

Statistical analysis

A one-way analysis of variance (ANOVA) with repeatedmeasures on exercise modality assessed potential differ-ences in workload. Independent t tests examined sex differ-ences in all anthropometric characteristics, which couldhave influenced metabolic efficiency and prevented apooling of participants. Separate one-way repeated mea-sures ANOVA were used to examine differences in VO2,work rate, RER, RPE, and metabolic efficiency betweenexercise modalities. Prior to running all ANOVA tests, datawere checked for sphericity using Mauchly’s test. Wheresphericity violations were noted, Huynh–Feldt correctionswere used to modify the degrees of freedom employed in thesubsequent statistical analyses. Where statistical signifi-cance was noted, Bonferroni pairwise comparisons weremade to determine specifically where differences existed.All data were analyzed using the Statistical Package forSocial Sciences (SPSS Inc, Chicago, IL) 20.0 for Windows,with a critical α level set at 0.05 for all analyses.

Results

There were no significant differences in workload betweenexercise modalities; therefore, workload was not considereda covariate in subsequent analysis. Similarly, there were no

differences in boys and girls for height, weight, BMI, waistcircumference, or waist-to-height ratio. Because waist-to-height ratio is considered predictive of adiposity in children[31], sex was not considered a confounding factor for bodycomposition, and therefore exercise performance, in thisstudy. All participants were pooled for subsequent analyses,regardless of sex.

Table 2 characterizes the effects of exercise mode onwork rate, VO2, RER, RPE, and ME%. There was nosignificant interaction between mode of exercise and workrate (P=0.932).

VO2

There was a significant interaction between mode of exer-cise and VO2 (P<0.001). Walking produced significantlylower VO2 than cycling (P<0.001) and elliptical training(P<0.001). VO2 in cycling was significantly lower thanelliptical training (P=0.007). The differences in VO2 perkilogram were of the same level of significance as they werewhen not accounted for body weight.

METs

There was a significant interaction between mode of exer-cise and METs (P<0.001). Walking produced significantlyless METs than elliptical training (P<0.001) and cycling(P<0.001). METs in cycling were significantly lower thanthose in elliptical training (P=0.005).

RER

There was a significant interaction between mode of exer-cise and RER (P<0.001). RER in walking was significantlylower than that in cycling (P<0.001) and elliptical training(P<0.001). Cycling did not significantly differ from ellipti-cal training (P=1.0).

Net EE

There was a significant interaction between mode of exer-cise and net EE (P<0.001). Walking produced significantlyless net EE than elliptical training (P<0.001) and cycling(P<0.001). Net EE in cycling was significantly lower thanthat in elliptical training (P=0.006).

ME%

There was a significant interaction between mode of exer-cise and ME% (P<0.001) (Fig. 1). ME% in walking wassignificantly higher than that in cycling (P=0.001) andelliptical training (P<0.001). ME% in cycling was signifi-cantly higher than that in elliptical training (P=0.003).

Eur J Pediatr (2013) 172:1191–1196 1193

RPE

There was a significant interaction between mode of exer-cise and RPE (P=0.001) (Fig. 2). RPE in walking wassignificantly lower than that in elliptical training (P=0.006)but not from cycling (P=0.314). Cycling resulted in signifi-cantly lower RPE than elliptical training (P=0.021).

Discussion

This was the first study to determine the differences inchildren’s metabolic efficiency in different exercise modal-ities at the same workload. Treadmill walking produced thelowest energy expenditure, and subsequent highest metabol-ic efficiency, of the three modalities tested. Additionally,perceived exertion during treadmill walking was significant-ly lower than that during elliptical training.

Walking is the mode of exercise that is performed themost frequently in daily life; the higher skill level mayexplain the greater metabolic efficiency compared to sta-tionary cycling and elliptical training in children. To the bestof our knowledge, the mechanical efficiency of ellipticaltraining has not been previously reported in children or

adults. However, a previous research has shown that oxygenuptake increases when the upper and lower limbs are active.Elliptical training is the exercise modality that best incorpo-rates both the upper and lower extremities; this could ex-plain the resulting increased energy expenditure whencompared to walking and stationary cycling.

Because walking had a higher mechanical efficiency thanthe other modalities, it would be considered less difficult toperform. Similarly, elliptical training produced the highestenergy expenditure, and was perceived as the most difficult.Interestingly, there is no significant difference in RPE be-tween treadmill walking and stationary cycling. This couldindicate that a child considers walking and cycling to berequiring a similar effort, even though cycling requiresgreater net energy expenditure. Walking is considered aninverted pendulum, where energy is recovered during gait[25]. Cycling, however, does not have a similar recoverysystem, thus requiring greater energy expenditure. The sim-ilar rates of perceived exertion could indicate that the non-weight-bearing characteristics of cycling would decreasejoint loading and improve the overall experience at a givenworkload. The higher energy expenditure and similar RPEssuggest that the non-weight-bearing cycling could be an

Table 2 Effects of exercise modality on respiratory and metabolic variables

Walking Cycling Elliptical training p values Pairwise

W vs C W vs E C vs E

Work rate (W) 40.30 (13.44) 40.33 (13.42) 40.33 (13.16) 0.932

Respiratory exchange ratio 0.81 (0.05) 0.90 (0.05) 0.92 (0.09) <0.001 <0.001 <0.001 1

Rating of perceived exertion 2.17 (1.12) 3.17 (1.30) 4.22 (0.87) 0.001 0.314 0.006 0.021

Oxygen uptake (l min−1) 0.64 (0.19) 0.96 (0.23) 1.15 (0.33) <0.001 <0.001 <0.001 0.007

Oxygen uptake (ml kg−1 min−1) 19.37 (3.96) 29.16 (6.01) 35.30 (10.64) <0.001 <0.001 <0.001 0.005

Metabolic efficiency (%) 28.58 (11.67) 15.34 (3.39) 12.75 (3.96) <0.001 0.001 <0.001 0.003

METs 5.53 (1.13) 8.33 (1.72) 10.09 (3.04) <0.001 <0.001 <0.001 0.005

Net EE (kcal min−1 kg−1) 2.21 (0.92) 3.85 (1.12) 4.83 (1.72) <0.001 <0.001 <0.001 0.006

Values are mean (SD)

W: Walking; C: Cycling; E: Elliptical training

0%

10%

20%

30%

40%

Walk Cycle Elliptical

Mec

han

ical

Eff

icen

cy (%

)

Fig. 1 Effects of exercise mode on metabolic efficiency

0

1

2

3

4

5

6

Walk Cycle Elliptical

RP

E

Fig. 2 Effects of exercise mode on ratings of perceived exertion (RPE)

1194 Eur J Pediatr (2013) 172:1191–1196

effective physical activity for children, specifically in thosepopulations who find it difficult to transfer body massduring locomotion [9, 28].

Walking produced the least amount of METs, but wasstill defined as a moderate activity (<6 METs). Cycling wasclassified as a vigorous activity (6–9 METs) and ellipticaltraining was classified as very vigorous (>9 METs) at thesame mechanical workload. The increased levels of METs atthe given work rate would be difficult for a child to main-tain. This is especially true for children with lower aerobiccapacity, most notably those struggling with obesity [23],chronic pulmonary disease [11], and asthma [17].

The design of the exercise machines may impact thedifferences in mechanical efficacy. Although the machinescould be adjusted, the motions of the machines were scaledto adults. The scale of the treadmill had little impact onmotions produced in normal walking cadence. However,the rigid trajectory of the pedals for both the ellipticaltrainer and stationary cycle could have produced less nat-ural movements. The altered kinematics could influencethe participant’s performance and, ultimately, the metabol-ic efficiency associated with the modality. This issupported by the contrast between our findings in childrenand those reported in adults [1, 10]. Unfortunately, jointbiomechanics were not included in this study, and thistheory cannot be substantiated.

This study was limited by the design of the exercisemodality and lack of familiarity in using the machinery.Elliptical trainers and stationary cycles are not commonlyused in homes or schools and may be novel to a child.Participants may not have been fully comfortable with theactivity, thereby altering the metabolic outputs. The level ofdifficulty could also affect the RPE associated with a par-ticular exercise, and future research designs should considerperforming these activities at similar RPE levels, rather thanworkload. Future consideration should also be given tounderstanding the role of excess postexercise oxygen con-sumption (EPOC), which may contribute substantially to theenergy expenditure resulting from exercise [15]. By notaccounting for EPOC, energy expenditure was not robustlyevaluated and could underestimate the true energy expendi-ture for each modality. The postexercise energy expenditurecould be especially important when evaluating the efficacyof an exercise modality for use in the weight management ofan obese child.

Conclusions

In summary, treadmill walking appears to be the most effi-cacious exercise prescription for a cohort of healthy childrenaged 8–12 years. As this study was a cross-sectional design,longitudinal studies need to be implemented to investigate

the long-term benefits of each exercise modality. If physicalactivity guidelines continue to recommend exercise pre-scription for weight management, then further research isalso required to determine whether the current findings canbe transferred to the obese pediatric population.

References

1. Batte AL, Darling J, Evans J, Lance LM, Olson EI, Pincivero DM(2003) Physiologic response to a prescribed rating of perceivedexertion on an elliptical fitness cross-trainer. J Sports Med PhysFitness 43(3):300–305

2. Cawley J (2010) The economics of childhood obesity. Health Aff(Millwood) 29(3):364–371

3. CDC (2008) Physical activity guidelines for Americans. U.S.Department of Health and Human Services, Washington, DC

4. Donnelly JE, Greene JL, Gibson CA, Smith BK, Washburn RA,Sullivan DK, DuBose K, Mayo MS, Schmelzle KH, Ryan JJ,Jacobsen DJ, Williams SL (2009) Physical activity across thecurriculum (PAAC): a randomized controlled trial to promotephysical activity and diminish overweight and obesity in elemen-tary school children. Prev Med 49(4):336–34

5. DuBose KD, Eisenmann JC, Donnelly JE (2007) Aerobic fitnessattenuates the metabolic syndrome score in normal-weight, at-risk-for-overweight, and overweight children. Pediatrics 120(5):e1262–1268

6. Eston G, Parfitt G (2006) Percieved exertion. In: Armstrong N(ed) Paediatric exercise physiology. Elsevier, London, pp 275–297

7. Faulkner J, Parfitt G, Eston R (2008) The rating of perceivedexertion during competitive running scales with time. Psychophys-iology 45(6):977–985

8. Ferguson MA, Gutin B, Le NA, Karp W, Litaker M, HumphriesM, Okuyama T, Riggs S, Owens S (1999) Effects of exercisetraining and its cessation on components of the insulin resistancesyndrome in obese children. Int J Obes Relat Metab Disord23(8):889–895

9. Fowler EG, Knutson LM, Demuth SK, Siebert KL, Simms VD,Sugi MH, Souza RB, Karim R, Azen SP, Physical Therapy ClinicalResearch N (2010) Pediatric endurance and limb strengthening(PEDALS) for children with cerebral palsy using stationary cy-cling: a randomized controlled trial. Phys Ther 90(3):367–381

10. Glass SC, Knowlton RG, Becque MD (1992) Accuracy of RPEfrom graded exercise to establish exercise training intensity. MedSci Sports Exerc 24(11):1303–1307

11. Gosselink R, Troosters T, Decramer M (1996) Peripheral muscleweakness contributes to exercise limitation in COPD. Am J RespirCrit Care Med 153(3):976–980

12. Hardin DS, Hebert JD, Bayden T, Dehart M, Mazur L (1997)Treatment of childhood syndrome X. Pediatrics 100(2):E5

13. Jackson DM, Pace L, Speakman JR (2007) The measurement ofresting metabolic rate in preschool children. Obesity (SilverSpring) 15(8):1930–1932

14. Kelly AS, Wetzsteon RJ, Kaiser DR, Steinberger J, Bank AJ,Dengel DR (2004) Inflammation, insulin, and endothelial functionin overweight children and adolescents: the role of exercise. JPediatr 145(6):731–736

15. LaForgia J, Withers RT, Gore CJ (2006) Effects of exercise inten-sity and duration on the excess post-exercise oxygen consumption.J Sports Sci 24(12):1247–1264

16. Lambourne P, Donnelly JE (2011) The role of physical activity inpediatric Obesity. EdD Pediatr Clin N Am 58:1481–1491

Eur J Pediatr (2013) 172:1191–1196 1195

17. Lochte L, Angermann M, Larsson B (2009) Cardiorespiratoryfitness of asthmatic children and validation of predicted aerobiccapacity. Clin Respir J 3(1):42–50

18. Mellecker RR, McManus AM (2009) Measurement of restingenergy expenditure in healthy children. JPEN J Parenter EnterNutr 33(6):640–645

19. Mendoza JA, Zimmerman FJ, Christakis DA (2007) Televisionviewing, computer use, obesity, and adiposity in US preschoolchildren. Int J Behav Nutr Phys Act 4:44

20. Mitchell JA, Pate RR, Dowda M, Mattocks C, Riddoch C, NessAR, Blair SN (2012) A prospective study of sedentary behavior ina large cohort of youth. Med Sci Sports Exerc 44(6):1081–1087

21. Morgan DW, Tseh W, Caputo JL, Keefer DJ, Craig IS, Griffith KB,Akins MB, Griffith GE, Krahenbuhl GS, Martin PE (2004) Longitu-dinal stratification of gait economy in young boys and girls: thelocomotion energy and growth study. Eur J Appl Physiol 91(1):30–34

22. NASPE (2004) Physical activity for children: a statement of guide-lines for children ages 5–12. National Association for Sport andPhysical Education, Reston

23. Norman AC, Drinkard B, McDuffie JR, Ghorbani S, Yanoff LB,Yanovski JA (2005) Influence of excess adiposity on exercisefitness and performance in overweight children and adolescents.Pediatrics 115(6):e690–696

24. O'Donovan G, Blazevich AJ, Boreham C, Cooper AR, Crank H,Ekelund U, Fox KR, Gately P, Giles-Corti B, Gill JM, Hamer M,McDermott I, Murphy M, Mutrie N, Reilly JJ, Saxton JM,Stamatakis E (2010) The ABC of physical activity for health: aconsensus statement from the British Association of Sport andExercise Sciences. J Sports Sci 28(6):573–591

25. Saibene FB, Minetti AE (2003) Biomechanical and physiologicalaspects of legged locomotion in humans. Eur J Appl Physiol88:297–316

26. Salmon J, Tremblay MS, Marshall SJ, Hume C (2011) Healthrisks, correlates, and interventions to reduce sedentary behaviorin young people. Am J Prev Med 41(2):197–206

27. Serdula MK, Ivery D, Coates RJ, Freedman DS, Williamson DF,Byers T (1993) Do obese children become obese adults? A reviewof the literature. Prev Med 22(2):167–177

28. Shultz SP, Browning RC, Schutz Y, Maffeis C, Hills AP (2011)Childhood obesity and walking: guidelines and challenges. Int JPediatr Obes 6(5–6):332–341

29. Stewart A, Marfell-Jones M, Olds T, de Ridder H (2011)International standards for anthropometric assessment. ISAK,Lower Hutt

30. Strong WB, Malina RM, Blimkie CJ, Daniels SR, Dishman RK,Gutin B, Hergenroeder AC, Must A, Nixon PA, Pivarnik JM,Rowland T, Trost S, Trudeau F (2005) Evidence based physicalactivity for school-age youth. J Pediatr 146(6):732–737

31. Taylor RW, Williams SM, Grant AM, Taylor BJ, Goulding A(2011) Predictive ability of waist-to-height in relation to adiposityin children is not improved with age and sex-specific values.Obesity (Silver Spring) 19(5):1062–1068

32. Timmons BW, Proudfoot NA, MacDonald MJ, Bray SR,Cairney J (2012) The health outcomes and physical activityin preschoolers (HOPP) study: rationale and design. BMC PublHealth 12:284

33. Tolfrey K, Campbell IG, Batterham AM (1998) Exercise traininginduced alterations in prepubertal children’s lipid-lipoprotein pro-file. Med Sci Sports Exerc 30(12):1684–1692

34. Trasande L, Chatterjee S (2009) The impact of obesity on healthservice utilization and costs in childhood. Obesity (Silver Spring)17(9):1749–1754

35. Weir JB (1990) New methods for calculating metabolic rate withspecial reference to protein metabolism. 1949. Nutrition 6(3):213–221

36. WHO (2010) Global recommendations on physical activity andhealth. World Health Organisation, Geneva

37. Williams CL, Hayman LL, Daniels SR, Robinson TN, SteinbergerJ, Paridon S, Bazzarre T (2002) Cardiovascular health in child-hood: a statement for health professionals from the Committee onAtherosclerosis, Hypertension, and Obesity in the Young (AHOY)of the Council on Cardiovascular Disease in the Young, AmericanHeart Association. Circulation 106(1):143–160

38. Zimmerman FJ, Christakis DA, Meltzoff AN (2007) Televisionand DVD/video viewing in children younger than 2 years. ArchPediatr Adolesc Med 161(5):473–479

1196 Eur J Pediatr (2013) 172:1191–1196