Secondary Logo

Journal Logo

Exploring the Relationship between Adiposity and Fitness in Young Children

FAIRCHILD, TIMOTHY JOHN; KLAKK, HEIDI; HEIDEMANN, MALENE SØBORG; ANDERSEN, LARS BO; WEDDERKOPP, NIELS

Medicine & Science in Sports & Exercise: September 2016 - Volume 48 - Issue 9 - p 1708–1714
doi: 10.1249/MSS.0000000000000958
EPIDEMIOLOGY
Free

Purpose High levels of cardiorespiratory fitness (CRF) may attenuate the association between the excessive adiposity and the risks of cardiovascular and metabolic disease. The purpose of this study was to stratify children according to their body mass index (BMI) and adiposity (body fat percentage [BF%]) and to compare levels of CRF across subgroups.

Methods This prospective cohort study comprises a cross-sectional and longitudinal analyses of data collected at baseline (n = 641) and 2 yr later (n = 579) on children (7.4–11.6 yr) attending public school in Denmark. Levels of CRF were measured using the Andersen test, whereas BF% was measured by dual-energy x-ray absorptiometry.

Results There were 560 children (87.4%) classified as normal weight according to BMI at baseline, of which 46 (7.4%) were identified as having excessive BF%. These children had significantly lower CRF (mean [95% confidence interval]: −63.1 m [−100.2 to −25.9]) than children with normal BMI and normal BF%, and the effect of BF% on CRF was significantly worse in boys than girls. Overweight children with high BF% had significantly lower prospective (2 yr) CRF levels (−34.4 m [−58.0 to −10.7]) than children with normal BMI and BF%. However, children who improved their BMI and/or BF% classification during the 2-yr period achieved CRF levels (8.9 m [−30.2 to 47.9]), which were comparable with children with normal BMI and BF% at both measurement time points.

Conclusion The CRF levels in children are affected by BMI and BF%, although BF% appears to play a greater role. This association between BF% and CRF is sex dependent, with CRF levels in boys being affected to a greater extent by BF%. Children identified as “normal weight” by BMI but presenting with excessive BF% had significantly lower CRF than “normal weight” children with low BF%.

1School of Psychology and Exercise Science, Murdoch University, Murdoch, AUSTRALIA; 2Centre of Research in Childhood Health, Department of Sports Science and Clinical Biomechanics, University of Southern Denmark, Odense, DENMARK; 3School of Physiotherapy, University College Lillebaelt, Odense, DENMARK; 4Hans Christian Andersen, Children’s Hospital, Odense University Hospital, Odense, DENMARK; 5Department of Sport Medicine, Norwegian School of Sport Sciences, Oslo, NORWAY; and 6Orthopedic Department, Institute of Regional Health Services Research, University of Southern Denmark, Hospital of Middelfart, Middelfart, DENMARK

Address for correspondence: Timothy Fairchild, Ph.D., School of Psychology and Exercise Science, Murdoch University, 90 South Street, Murdoch WA 6018, Australia; E-mail: t.fairchild@murdoch.edu.au.

Submitted for publication September 2015.

Accepted for publication April 2016.

Global estimates using objectively measured physical activity data in children (4–11 yr old) indicate they perform between 22 min (95% confidence interval [CI] = 19.9–24.1) and 45 min (95% CI = 39.6–50.4) of moderate-to-vigorous physical activity per day (13), which falls well below the recommended 60 min of daily moderate-to-vigorous physical activity (34). The low levels of physical activity in children correspond with a high prevalence of overweight and obesity (6,20,23) as well as low levels of cardiorespiratory fitness (CRF) (23,24). This is alarming considering both low CRF and high adiposity are associated with increased cardiovascular and metabolic disease risk (4,8,9,11,15,16,18,27).

Although the association between low CRF and adiposity is well recognized (21,25–27,29), the magnitude of this association remains equivocal. This is due in part to the differences in the techniques used to measure adiposity, which range from the direct measurement of body fat using dual-energy x-ray absorptiometry (DXA) to the adoption of a combination of anthropometric techniques (i.e., height and weight; waist circumference; skinfolds). Of these techniques, body mass index (BMI) or BMI z-score are most commonly used as the outcome measure (17). This is despite the growing number of studies reporting substantial variance in the adiposity of children within the given BMI categories (32), with a particular concern being the concealment of excessive adiposity in children categorized as “normal” by BMI (the so-called thin–fat phenotype). This is a concern because total body fat percentage (BF%) has been identified as a stronger predictor of composite and single cardiovascular risk factors than either BMI or waist circumference in children (18).

The purpose of the present study therefore was to stratify children according to BMI and DXA-derived adiposity and to identify differences in the children’s CRF. We hypothesized that children identified as being of “normal weight” (BMI) and low BF% would have the highest CRF, whereas those with high BF%—irrespective of their BMI—would have the lowest CRF. Children were then tested 2 yr later to explore the effect of an increase (considered detrimental; increasing adiposity or shifting into the overweight/obese BMI category) or decrease (considered beneficial; decreasing adiposity or shifting into the normal weight BMI category) in weight status on their CRF. We hypothesized that children who demonstrated a beneficial shift in their weight status would demonstrate similar CRF to those children who were constantly “normal weight” and “low adiposity,” but that these children would have significantly higher CRF than those who maintained a high adiposity at each time point.

Back to Top | Article Outline

METHODS

Study design

This prospective cohort study is nested in the Childhood Health, Activity, and Motor Performance School Study in Denmark (CHAMPS study-DK [31]). The study herein comprises two discrete analyses of data collected from children attending public school in the municipality of Svendborg, Denmark. The first analysis is a cross-sectional analysis of data collected during testing in 2008 (T1). This analysis was conducted to determine the association of CRF with weight status and adiposity (four categorical levels). The second analysis comprised two separate longitudinal models to determine whether baseline (T1) weight status or adiposity effected children’s CRF 2 yr later (T2). Only children with complete data sets were included in the longitudinal analysis.

Back to Top | Article Outline

Participants

Children from the CHAMPS study-DK who were in second to fourth grade (7.4–11.6 yr) at baseline were enrolled in this study, which has previously been described in detail (30). This subsample was chosen because these children had whole-body DXA (GE Lunar Prodigy; GE Medical Systems, Madison, WI) scans during T1 and T2, which allowed the direct measure of total BF%. All children and parents from the 10 participating schools received information about the study through school meetings and additional written information, and all examinations took place at The Hans Christian Andersen Children’s Hospital (Odense, Denmark). Participation in the study was voluntary, and all parents provided their written informed consent. Permission to conduct the CHAMPS study-DK was granted by the Regional Scientific Ethical Committee of Southern Denmark (ID S-20080047).

Back to Top | Article Outline

Fitness assessment

Cardiorespiratory fitness was measured using the Andersen test (2). The test was conducted indoor on one-half of a handball court (wood flooring) with 20 m running lanes marked by cones. Participants were required to run from one line to the other, where they had to touch the floor behind the line with one hand, turn around, and run back. At 15 s, the test leader blew a whistle, and the participants stop as quickly as possible and rest for 15 s. This procedure was repeated for 10 min. The test leader announced the end of each resting period by counting backward from 3 to 0. The laps and the distance covered by each child were counted by research staff with groups of 6–10 children running at the same time. The total distance measured in meters was the test result. This test has previously been shown to demonstrate good test–retest performance (988 ± 77 and 989 ± 87 m, r2 = 0.86, CV = 3%, n = 31) and concurrent validity when compared with V˙O2max testing (r2 = 0.85) in this age-group (6–9 yr old [1]).

Back to Top | Article Outline

Anthropometrics and adiposity

Weight was measured to the nearest 0.1 kg on an electronic scale (Tanita BWB-800S; Tanita Corporation, Tokyo, Japan) wearing light clothes. Height was measured to the nearest 0.5 cm using a portable stadiometer (SECA 214; Seca Corporation, Hanover, MD). Both anthropometrics were conducted barefoot, and the BMI was subsequently calculated (kg·m−2) and defined according to the International Obesity Task Force criteria (10).

Fat mass and BF% was measured by DXA (GE Lunar Prodigy; GE Medical Systems) only in the children in second to fourth grade. Participants were instructed to lie still in a supine position wearing underwear, a thin T-shirt, and stockings, and a blanket was provided for the duration of the scan. All scans were performed by the same two operators and analyzed by one using the ENCORE software (version 12.3, Prodigy; Lunar Corp., Madison, WI). The total body composition was calculated after exclusion of the head. We adopted percentage body fat measures of 25% and 30% as the cut points for the categorization of adiposity in boys and girls, respectively, which was based on the findings of Williams et al. (33). The DXA machine was calibrated each day in accordance with the standardized procedures.

Back to Top | Article Outline

Pubertal stage

The Tanner pubertal stages self-assessment questionnaire was used to determine pubertal status (30). Boys and girls were presented with five pictures of Tanner staging for pubic hair development, and children were asked to indicate which stage best referred to their own pubertal stage with explanatory text in Danish supporting the self-assessment. The procedure took place in a private setting with sufficient time to allow for the self-assessment. The accuracy of this technique has previously been shown to be sufficient in a similar cohort and setting (28).

Back to Top | Article Outline

Data analysis

The cross-sectional analysis and estimate of prospective fitness was performed on stratified data comprising “overweight or obese” (OW/OB) or “normal weight” (NW) classification according to BMI, and “higher%BF” (BF% greater than predefined cut points) or “lower%BF” (BF% less than predefined cut points) classification based on DXA measurement of total body fat. The children were then stratified into categories consisting of NW + lower%BF, NW + higher%BF, OW/OB + lower%BF, and OW/OB + higher%BF. For inclusion in the cross-sectional analysis, BMI, adiposity, age, sex, and CRF test score during T1 was required. For inclusion in the prospective fitness assessment, BMI, adiposity, age, sex, and CRF test score at T1 were required, and CRF test score at T2 (outcome measure) was required. Children with complete data sets (T1: BMI, adiposity, age, sex, and CRF; T2: BMI, adiposity, age, sex, and CRF) were included in the longitudinal analysis and classified as i) constant NW + lower%BF, ii) constant NW + higher%BF, iii) constant OW/OB + higher%BF, iv) improving BMI or %BF, and v) worsening BMI or %BF. The classifications of “constant,” “improving,” and “worsening” in the longitudinal data set refer to categorical changes in BMI or adiposity.

The association of adiposity and BMI with fitness was examined using multilevel mixed-effects analyses using the xtmixed procedure in STATA, with school class and school modeled as random effects to comply with the cluster structure of the school-based design (30). Each (cross-sectional and longitudinal) model was also adjusted for sex, age, pubertal status, testing time point, and fitness as indicated, and the beta coefficients were calculated. Where an interaction on sex was identified, a separate analysis was performed. Sex-based differences in CRF within each category of interest were further explored using the marginsplot command in STATA, and contrasts were computed (fixed portion: distance covered in meters; Figs. 1A, 1B, and 2). All analyses were conducted using NW + lower%BF as the reference group, and children with missing data were excluded from relevant analysis. Additional post hoc analyses were conducted to further explore the associations between the independent variables BMI z-scores (SD) and the BF% with CRF, using the multilevel mixed-effects analyses (adjusted for age, sex, and puberty). The interaction term of sex and the independent variable were included when significant. Data are presented as mean and SD or 95% CI. All analyses were completed in STATA version 12.1 (StataCorp, College Station, TX) with α = 0.05 (two-sided).

FIGURE 1

FIGURE 1

Back to Top | Article Outline

RESULTS

Descriptive data for all children are presented in Table 1. Between T1 and T2, the children demonstrated an increase in their BMI (mean [95% CI]; boys: 0.98 kg·m−2 [0.86–1.09]; girls: 1.10 kg·m−2 [0.98–1.22]), BF% (mean [95% CI]; boys: 1.92% [1.47–2.37]; girls: 1.58% [1.12–2.04]) and CRF (mean [95% CI]; boys: 64.1 m [54.6–73.5]; girls: 70.5 m [61.8–79.2]), with no significant between sex differences observed in any of these outcome measures.

TABLE 1

TABLE 1

Back to Top | Article Outline

Cross-sectional analysis of fitness with weight status and adiposity

Data from 322 girls and 319 boys at T1 were included in the cross-sectional analysis (Table 2). Sex and age were identified as significant variables in the regression model, with boys running farther (74.9 m; P < 0.001) than girls and older children running farther (24.1 m; P < 0.001) than the younger children. Puberty status was not a significant variable in the model (P = 0.957). When children from each category were stratified by sex, the associations followed similar trends for girls and boys (Fig. 1A). However, a higher BMI and a higher BF% were significantly more detrimental on CRF in boys than girls (OW/OB + higher%BF: −55.8 m, P = 0.009).

TABLE 2

TABLE 2

Back to Top | Article Outline

Effect of weight status and adiposity at baseline on prospective fitness

Data from 579 children (boys, n = 290; girls, n = 289; age = 11.3 ± 0.8 yr) were used in the analysis (Table 3). Children increased the distance they ran in the CRF test between T1 (930.2 ± 101.8 m) and T2 (997.1 ± 100.3 m). Sex, age, and CRF (run distance) at T1 were identified as significant variables affecting the prediction of CRF at T2 (all P < 0.001), whereas puberty status was not significant (P = 0.677). Boys ran farther (34.3 m) than girls, whereas older children ran farther (14.6 m·yr−1) than the younger children in the CRF test at T2. After adjusting for sex, age, and CRF at T1, children categorized as OW/OB + higher%BF at T1 ran significantly less distance (34.4 m; P = 0.004) during the CRF test at T2 than individuals categorized as NW + lower%BF at baseline. When children were stratified by sex (Fig. 1B), the associations followed similar trends for girls and boys. However, BF% had a significantly greater detrimental effect on CRF in boys than in girls categorized as NW + higher%BF (−65.1 m, P = 0.002), but this was not observed in the other categories (OW/OB + higher%BF: −32.3 m, P = 0.071; OW/OB + lower%BF: 6.6 m, P = 0.882).

TABLE 3

TABLE 3

Back to Top | Article Outline

Effect of longitudinal weight status and adiposity on prospective fitness

Data from 502 children (boys, n = 254; girls, n = 248; age = 11.3 ± 0.9 yr) were used in this analysis (Table 4). As expected, children increased the distance they ran during the CRF at T2 (1003.0 ± 100.1 m) when compared with T1 (934.5 ± 100.9 m). Children maintaining a normal BMI and BF% (constant NW + lower%BF; Table 4) ran significantly farther than those identified as being constantly NW + higher%BF (−52.4 m; P = 0.015) and constantly OW/OB + higher%BF (−56.0 m; P < 0.001). There was an interaction of sex in this association with boys classified as constantly NW + higher%BF and constantly OW/OB + higher%BF having their CRF affected significantly more (constantly NW + higher%BF: −69.4 m, P = 0.014; constantly OW/OB + higher%BF: −45.4 m, P = 0.021) than girls (Fig. 2). Puberty status was not a significant variable in the model (P = 0.602).

TABLE 4

TABLE 4

FIGURE 2

FIGURE 2

Back to Top | Article Outline

Post hoc analysis

The cross-sectional relationship between CRF and BMI was explored further using the BMI z-scores, which showed that for every 1-point increase in the child’s BMI z-score, the child ran 32.7 m less (95% CI = −38.9 to −26.6), but there was no significant (P = 0.305) interaction with sex. When the association between CRF and BF% was calculated, every 1% increase in BF was associated with the child running 5.6 m less (95% CI = −6.5 to −4.7), and a significant (P = 0.003) interaction between sex and BF% on CRF was identified, with higher BF% affecting boys to a greater extent (mean [95% CI] = −1.9 m [−3.2 to −0.7]).

When the longitudinal relationship between CRF and BMI z-scores was explored, the child’s BMI z-score during T1 did not affect the child’s CRF at T2 (mean [95% CI] = −7.2 m [−15.5 to 1.2]), although a significant interaction with sex was observed with boys being affected to a greater extent (mean [95% CI] = −13.5 m [−25.0 to −2.0]). When the longitudinal association between CRF and BF% was calculated, every 1% increase in BF% at T1 was associated with the child running 2.1 m less (95% CI = [−3.1 to −1.0]) at T2. A significant (P = 0.003) interaction between sex and BF% on CRF was identified, with higher BF% at T1 affecting the CRF at T2 in boys to a greater extent (mean [95% CI] = −2.1 m [−3.5 to −0.7]).

Back to Top | Article Outline

DISCUSSION

The main findings of the present study were i) a higher than expected proportion (7.2%) of children who were normal weight by BMI had a high BF% (NW + higher%BF); ii) children classified as NW + higher%BF demonstrated a significantly lower CRF than children classified as NW + lower%BF; iii) the detrimental effects of high BF% or OW/OB classification on CRF were similar and appeared to be additive with OW/OB + higher%BF children having the lowest CRF; iv) children OW/OB + higher%BF at T1 had significantly lower CRF at T2, even when adjusted for baseline fitness (CRF at T1); v) children who improved either their BMI or BF% between T1 and T2 no longer had a significantly lower CRF than children maintaining a constant NW + lower%BF phenotype; and vi) high levels of BF% were more detrimental on CRF in boys than in girls.

A longstanding concern with adoption of BMI as a diagnostic tool for the classification of obesity—which is defined as abnormal or excessive fat accumulation that may impair health (34)—is the concealment of individuals with excessive adiposity and increased cardiometabolic risk factors within the “normal” category, the so-called thin–fat phenotype, which includes individuals with “metabolic obesity” (14). This phenotype, characterized by a greater fat mass at any given BMI level, is believed to be more prevalent in some ethnicities (e.g., South Asians vs Europeans) (12) and appears already in early childhood (32,35). In the present study, 7.2% of children were identified as being NW + higher%BF (thin–fat phenotype); this is despite the children in this study being primarily Caucasian. This prevalence of the thin–fat phenotype is comparable with the 10.8% of children identified as being OW/OB + higher%BF, and when considered another way, 40% of children who were considered to have a high BF% were considered normal weight by BMI.

Evidence from studies in adults (5,22) and children (4,7) suggests that higher levels of CRF attenuate the health risks associated with obesity. In the current study, the cross-sectional analysis revealed significantly lower CRF in children categorized as being OW/OB (OW/OB + lower%BF) or having high BF% (NW + higher%BF) when compared with children categorized as NW + lower%BF. However, the combination of high BF% with OW/OB was most detrimental, with children identified as OW/OB + higher%BF performing worst in the CRF test. This detrimental effect of BF% on CRF was significantly greater in boys than in girls. To provide some context, the CRF of NW + higher%BF or OW/OB + lower%BF children was similar to the CRF of NW + lower%BF children who were 2 yr younger, whereas OW/OB + higher%BF children had a CRF similar to NW + lower%BF children who were 3.5 yr younger. Considering BF% has previously been shown to be a stronger predictor of composite and individual CVD risk factors in this population than BMI or waist circumference (18) and that high CRF may attenuate this risk, the poor CRF in children with high BF% is concerning, particularly in children classified as NW + higher%BF because typical screening measures (i.e., BMI classification) are unlikely to identify these children as being at risk.

To the authors’ knowledge, no studies have previously examined prospective CRF based on DXA-derived adiposity and weight status. In the current study, 579 children had complete data to conduct this analysis (Table 3). Children classified as OW/OB + higher%BF at baseline (T1) had significantly lower CRF 2 yr later (during T2) than children classified as NW + lower%BF at baseline (T1) even when adjusted for baseline CRF (T1). This difference was not observed in other categories. When stratified by sex, the detrimental effect of BF% on CRF was more profound in boys than girls.

When children were classified according to their BF% and BMI across both measurement times (T1 and T2; n = 502), children identified as constant NW + higher%BF and constant OW/OB + higher%BF had significantly lower CRF than the constantly NW + lower%BF children (Table 4). Consistent with our previous findings (18), this detrimental effect of adiposity on CRF was worse for boys than girls (Fig. 1). Girls in these categories ran on average 52.4 m/5.3% (constant NW + higher%BF) and 56.0 m/5.7% (constant OW/OB + higher%BF) less than girls in the constant NW + lower%BF categories, whereas boys ran on average 128.8 m/12.1% (constant NW + higher%BF) and 101.4 m/9.5% (constant OW/OB + higher%BF) less than boys in the constant NW + lower%BF categories. It is noteworthy that children identified as having an improved BMI or adiposity classification between T1 and T2 achieved a CRF score, which was no longer significantly different from children who were constantly NW + lower%BF. Additionally, no sex-based differences were apparent in this improvement (P = 0.413).

The current study had several strengths and limitations, which informed the interpretation of these results. The major study strengths include that we directly measured BF% by DXA in this large cohort of children, using standardized procedures. Weight and height were measured using the same equipment, and our multilevel modeling accounted for several potential sources of confounding in the analyses. In the present study, CRF was directly assessed using a valid and reproducible intermittent running test (1,2), which showed that the general fitness of the cohort was comparable with those previously reported (mean V˙O2 peak range: 36.8–66.0 mL·min−1·kg−1) (19). A limitation of the study was despite the large study sample, some of the subgroup analyses requiring stratification by sex were limited in size, therefore increasing the risk of model overfit. Although not a limitation, it is important to note that current evidence suggests the use of continuous scores for risk factors of disease classification and prognostic prediction in children (3). However, this study dichotomized adiposity and BMI according to predetermined cut points. This decision was based on aligning the major findings of the study with the current application of these measures. With respect to the BF% cutoffs, these were chosen because they were shown to be indicative of increased risk for elevated BP and lipoprotein ratios in a study conducted in more than 3300 children and adolescents (1667 males and 1653 females [33]). Although the percentage body fat cutoffs in that study were calculated using skinfold thickness, to the authors’ knowledge, there are currently no large cohort studies that have assessed associations between cardiometabolic risk factors and BF% measured by DXA in this age-group. Finally, there were significant differences in the BMI of children in respective categories and, in particular, in the BMI of children classified as “normal weight” (i.e., NW + lower%BF vs NW + higher%BF). However, when children with the lowest BMI in the NW + lower%BF category were removed from the analysis, such that the mean BMI was no longer different, the main findings remained consistent with those reported herein (results not shown).

In summary, we show not only that the thin–fat phenotype is present (~7.2%) in this European cohort but also that these children have lower CRF levels than children considered normal weight by BMI. Adiposity is known to strongly contribute to metabolic disease risk in prepubescent children, whereas CRF is protective against metabolic disease risk when adiposity is high (27). Further, the addition of CRF as a risk factor for metabolic syndrome in children improves diagnostic criteria (3). For these reasons, the finding of low CRF in children with normal BMI but high BF% is concerning because these children are not identified as being at risk during routine diagnostic screening. Data from the longitudinal analyses revealed that an improvement in BF% or BMI classification is associated with an improvement in CRF, indicating the importance of early detection and intervention for children with high BF% or BMI. The proportion of children with low levels of CRF who maintain these low levels of CRF into adolescents and adulthood is currently unknown, and the associated clinical implications remain to be resolved.

The authors gratefully acknowledge the work of numerous research staff and PhD students who have participated in the data collection for the CHAMPS study-DK. We thank children, parents, and teachers in the schools involved in the project, and we are grateful for the cooperation with the Svendborg Project, Sport Study Sydfyn, and the municipality of Svendborg.

Funding for the study has been provided by the TRYG Foundation, the University College Lillebaelt, the University of Southern Denmark, the Nordea Foundation, the IMK Foundation, the Region of Southern Denmark, the Egmont Foundation, the A. J. Andersen Foundation, the Danish Rheumatism Association, the Østifternes Foundation, Brd. Hartmann’s Foundation, TEAM Denmark, the Danish Chiropractor Foundation, and the Nordic Institute of Chiropractic and Clinical Biomechanics. Finally, the authors acknowledge the members of the CHAMPS study-DK not listed as coauthors of this article: C. T. Rexen, E. Jespersen, N. C. Moller, and C. Franz.

The authors declare no conflict of interest.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.

Back to Top | Article Outline

REFERENCES

1. Ahler T, Bendiksen M, Krustrup P, Wedderkopp N. Aerobic fitness testing in 6- to 9-year-old children: reliability and validity of a modified Yo-Yo IR1 test and the Andersen test. Eur J Appl Physiol. 2012;112(3):871–6.
2. Andersen LB, Andersen TE, Andersen E, Anderssen SA. An intermittent running test to estimate maximal oxygen uptake: the Andersen test. J Sports Med Phys Fitness. 2008;48(4):434–7.
3. Andersen LB, Lauersen JB, Brønd JC, et al. A new approach to define and diagnose cardiometabolic disorder in children. J Diabetes Res. 2015;2015:539835.
4. Andersen LB, Sardinha LB, Froberg K, Riddoch CJ, Page AS, Anderssen SA. Fitness, fatness and clustering of cardiovascular risk factors in children from Denmark, Estonia and Portugal: the European Youth Heart Study. Int J Pediatr Obes. 2008;3(1 Suppl):58–66.
5. Barry VW, Baruth M, Beets MW, Durstine JL, Liu JH, Blair SN. Fitness vs. fatness on all-cause mortality: a meta-analysis. Prog Cardiovasc Dis. 2014;56(4):382–90.
6. Bibiloni Mdel M, Pons A, Tur JA. Prevalence of overweight and obesity in adolescents: a systematic review. ISRN Obes. 2013;2013:392747.
7. Brage S, Wedderkopp N, Ekelund U, et al. Features of the metabolic syndrome are associated with objectively measured physical activity and fitness in Danish children: the European Youth Heart Study (EYHS). Diabetes Care. 2004;27(9):2141–8.
8. Brothers J, McBride M, Paridon A, Zhang X, Paridon S. Fatness is not a factor of fitness: analysis of cardiorespiratory data from healthy children over an 8-year period. Cardiol Young. 2013;23(1):47–53.
9. Brouwer SI, Stolk RP, Liem ET, Lemmink KA, Corpeleijn E. The role of fitness in the association between fatness and cardiometabolic risk from childhood to adolescence. Pediatr Diabetes. 2013;14(1):57–65.
10. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ. 2000;320(7244):1240–3.
11. Cummings DM, Dubose KD, Imai S, Collier DN. Fitness versus fatness and insulin resistance in U.S. adolescents. J Obes. 2010:2010.
12. Deurenberg P, Deurenberg-Yap M, Guricci S. Asians are different from Caucasians and from each other in their body mass index/body fat per cent relationship. Obes Rev. 2002;3(3):141–6.
13. Ekelund U, Luan J, Sherar LB, et al. Moderate to vigorous physical activity and sedentary time and cardiometabolic risk factors in children and adolescents. JAMA. 2012;307(7):704–12.
14. Hamdy O, Porramatikul S, Al-Ozairi E. Metabolic obesity: the paradox between visceral and subcutaneous fat. Curr Diabetes Rev. 2006;2(4):367–73.
15. Hurtig-Wennlof A, Ruiz JR, Harro M, Sjostrom M. Cardiorespiratory fitness relates more strongly than physical activity to cardiovascular disease risk factors in healthy children and adolescents: the European Youth Heart Study. Eur J Cardiovasc Prev Rehabil. 2007;14(4):575–81.
16. Jago R, Drews KL, McMurray RG, et al. Fatness, fitness, and cardiometabolic risk factors among sixth-grade youth. Med Sci Sports. 2010;42(8):1502–10.
17. Jiménez-Pavón D, Kelly J, Reilly JJ. Associations between objectively measured habitual physical activity and adiposity in children and adolescents: systematic review. Int J Pediatr Obes. 2010;5(1):3–18.
18. Klakk H, Grøntved A, Møller NC, Heidemann M, Andersen LB, Wedderkopp N. Prospective association of adiposity and cardiorespiratory fitness with cardiovascular risk factors in healthy children. Scand J Med Sci Sports. 2014;24:e275–82.
19. Krahenbuhl GS, Skinner JS, Kohrt WM. Developmental aspects of maximal aerobic power in children. Exerc Sport Sci Rev. 1985;13:503–38.
20. Lobstein T, Baur L, Uauy R, IASO International Obesity TaskForce. Obesity in children and young people: a crisis in public health. Obes Rev. 2004;5(1 Suppl):4–104.
21. Magnusson KT, Sveinsson T, Arngrimsson SA, Johannsson E. Predictors of fatness and physical fitness in nine-year-old Icelandic school children. Int J Pediatr Obes. 2008;3(4):217–25.
22. McAuley PA, Artero EG, Sui X, Lavie CJ, Almeida MJ, Blair SN. Fitness, fatness, and survival in adults with prediabetes. Diabetes Care. 2014;37(2):529–36.
23. Novotny R, Fialkowski MK, Li F, et al. Systematic review of prevalence of young child overweight and obesity in the United States–affiliated Pacific Region compared with the 48 contiguous states: the Children’s Healthy Living Program. Am J Public Health. 2015;105(1):e22–35.
24. Olds T, Tomkinson G, Leger L, Cazorla G. Worldwide variation in the performance of children and adolescents: an analysis of 109 studies of the 20-m shuttle run test in 37 countries. J Sports Sci. 2006;24(10):1025–38.
25. Ostojic SM, O’Neil M, Calleja J, Terrados N, Stojanovic M. Cardiorespiratory fitness and adiposity in overweight and nonoverweight 8-year-old school children. Minerva Pediatr. 2010;62(6):537–43.
26. Ostojic SM, Stojanovic MD, Stojanovic V, Maric J, Njaradi N. Correlation between fitness and fatness in 6–14-year old Serbian school children. J Health Popul Nutr. 2011;29(1):53–60.
27. Parrett AL, Valentine RJ, Arngrímsson SA, Castelli DM, Evans EM. Adiposity and aerobic fitness are associated with metabolic disease risk in children. Appl Physiol Nutr Metab. 2011;36(1):72–9.
28. Rasmussen AR, Wohlfahrt-Veje C, Tefre de Renzy-Martin K, et al. Validity of self-assessment of pubertal maturation. Pediatrics. 2015;135(1):86–93.
29. Stigman S, Rintala P, Kukkonen-Harjula K, Kujala U, Rinne M, Fogelholm M. Eight-year-old children with high cardiorespiratory fitness have lower overall and abdominal fatness. Int J Pediatr Obes. 2009;4(2):98–105.
30. Tanner JM. Growth at Adolescence: With a General Consideration of the Effects of Hereditary and Environmental Factors upon Growth and Maturation from Birth to Maturity. Blackwell Scientific Publications; 1962.
31. Wedderkopp N, Jespersen E, Franz C, et al. Study protocol. The Childhood Health, Activity, and Motor Performance School Study Denmark (The CHAMPS-study DK). BMC Pediatr. 2012;12:128.
32. Wells JC. Toward body composition reference data for infants, children, and adolescents. Adv Nutr. 2014;5(3):320S–9.
33. Williams DP, Going SB, Lohman TG, et al. Body fatness and risk for elevated blood pressure, total cholesterol, and serum lipoprotein ratios in children and adolescents. Am J Public Health. 1992;82(3):358–63.
34. World Health Organization. Global Recommendations on Physical Activity for Health. World Health Organization Geneva (Switzerland). 2010. pp. 17–21.
35. Yajnik CS, Fall CH, Coyaji KJ, et al. Neonatal anthropometry: the thin–fat Indian baby. The Pune Maternal Nutrition Study. Int J Obes Relat Metab Disord. 2003;27(2):173–80.
Keywords:

PHYSICAL ACTIVITY; OBESE; WEIGHT; EXERCISE; RUNNING

© 2016 American College of Sports Medicine