Obesity Research & Clinical Practice (2012) 6, e167—e173
ORIGINAL ARTICLE
Relationship between body mass index and body fat in children—–Age and gender differences Biljana Srdi´ c a,∗, Borislav Obradovi´ c b, Goran Dimitri´ c b, Edita Stoki´ c c, ca Siniˇ sa S. Babovi´ a
Department of Anatomy, Faculty of Medicine, Novi Sad, Serbia Faculty of Sport and Physical Education, Novi Sad, Serbia c Department of Endocrinology, Institute of Internal Disease, Clinical Centre Vojvodina, Novi Sad, Serbia b
Received 17 July 2010 ; received in revised form 28 April 2011; accepted 22 August 2011
KEYWORDS Children; Obesity; Overweight; BMI; Adiposity
∗
Summary When defining obesity body mass index (BMI) has been used as the main criterion. However it indicates only the nutritional status, whereas body fat demonstrates the real body composition picture. This study aimed at analyzing the relationship between nutritional status and adiposity in the population of 2284 Serbian children (1217 boys and 1067 girls). According to BMI subjects were divided into underweight, normal-weight, overweight and obese, and %BF values (based on skinfold thickness measurements) were analyzed with regard to BMI-category, age and gender. Girls showed stronger correlation between BMI and %BF comparing to boys (r = 0.834 vs. 0.577). Differences in %BF between underweight, normal weight and overweight children from different age groups were more obvious in girls, whereas in boys younger than 8 years overlapping in %BF values between different BMI-categories was registered. In normal weight children we found age-related oscillations in %BF values: 8- and 9-year-old boys had lower %BF comparing to 7-yearold boys, which was followed by %BF increasement in 10- and 11-year-old ones; in girls %BF values gradually increased with aging, with significant jumps in 9-, 10and 11-year-old ones. Thus, adiposity rebound may appear somehow later in boys. In overweight and obese children of both genders %BF continually increased with aging, whereas in underweight children %BF values remained unchanged. Our results pointed to age- and gender-dependent variations of %BF in normal weight and overweight children. We also indicated inconsistency between %BF and BMI especially in boys, and the need for definition of references for %BF. © 2011 Asian Oceanian Association for the Study of Obesity. Published by Elsevier Ltd. All rights reserved.
Corresponding author at: Department of Anatomy, Faculty of Medicine, Hajduk Veljkova 3, 21000 Novi Sad, Serbia. Tel.: +381 21 6615 775; fax: +381 21 6615 751. c). E-mail address:
[email protected] (B. Srdi´
Introduction Body mass index (BMI) is the recommended index to define overweight and obesity. BMI is a useful
1871-403X/$ — see front matter © 2011 Asian Oceanian Association for the Study of Obesity. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.orcp.2011.08.153
e168 indicator of nutritional status, which correlates highly with body fat mass and risk of metabolic and cardiovascular complications associated with obesity [1,2]. Recommended BMI cut-off points for children are based on percentile curves that also represent age-related variations of BMI. However, BMI also reflects fat-free mass, mainly muscular and bone mass, resulting in numerous variations of these body components within the same body mass. Thus, BMI cannot be considered an accurate indicator of the body fat mass [3,4]. Current concept of obesity implies an increase of the total body fat mass, which can be determined only through body composition analysis. From the point of the future cardiometabolic risk, body fat mass evaluation is more important than BMI. Proportion of body fat in total body mass shows high variability mainly as a function of age, gender, maturity and race, and, probably, does not follow BMI changes, especially in children. It is showed that BMI changes during growth reflect changes in adipocyte size and number [5]. BMI increases from birth to 1 year, then gradually declines and reaches a minimum at 5—6 years, and then begins to increase throughout the remainder of childhood. The point at which the BMI reaches a minimum is the start of the adiposity rebound, which corresponds to a rise of adipocyte number. However, several studies suggested that BMI changes may not accurately reflect changes in adiposity [6,7]. The aim of this study was to analyze age- and gender-related relationship between BMI and body fat mass in population of Serbian children aged 4—11 years.
Methods This was cross-sectional study that included 2284 children (1217 boys and 1067 girls) aged 4—11 years, randomly selected from daycare centers and schools in four cities located in the north part of Serbia. Subjects were screened by medical history and were excluded if they were taking medications or were diagnosed with syndromes or any illness known to affect body composition. The study was approved by the ethics committee, local and school authorities, and parental consent was obtained for all participants. The study was conducted in accordance with the Declaration of Helsinki. Anthropometric measurements were carried out by trained staff. Body height (BH) was measured by GPM anthropometer (Siber&Hegner, Zürich, Switzerland) with a 0.1 cm precision, and body
B. Srdi´ c et al. Table 1 The number of examined subjects in age subgroups. Age (years)
4 5 6 7 8 9 10 11 Total
Number of subjects Boys
Girls
85 185 252 204 110 135 125 121 1217
66 163 232 205 95 114 95 97 1067
weight (BW) was measured using standard balance beam scale with a 0.1 kg precision. Participants were weighed wearing light clothing and no shoes. BMI (kg/m2 ) was calculated from weight and height and nutritional state was assessed according to the CDC criteria [8]; underweight was defined as BMI below 5th percentile, being at risk for overweight was defined as having a BMI between the 85th and 94th percentile, while obesity was defined as BMI at or above the 95th percentile. Triceps- (T) and subscapular (SS) skinfold thickness were measured on the left side of the body using a GPM caliper (Siber&Hegner, Zürich, Switzerland) with 0.2 mm precision. All skinfold measures were done in triplicate and average of the three measurements was used for data analysis. Total body fat percent (%BF) based on skinfold thickness measurements was assessed using the equation given by Slaughter et al. [9]. Statistical analyses were performed using SPSS statistical program (version 11.5 for Windows; SPSS, Inc., Chicago, IL, USA). Data were expressed as mean ± standard deviation (X ± SD), median (Med), minimal and maximal value (Min—Max) and percent (%). Subjects were divided in eight age subgroups (Table 1) and differences between age- and gender groups were tested using one way analysis of variance (ANOVA) with post hoc Bonferonni’s test. Association between BMI and %BF was tested using Pearson’s correlation coefficient.
Results Table 2 displays basic anthropometric characteristics of subjects of both genders. Boys were statistically significantly higher and heavier comparing to girls. They also had significantly higher values of BMI (16.82 ± 2.68 vs. 16.54 ± 2.58, p < 0.01) and significantly lower values of %BF
Anthropometric characteristics of examined subjects. Total N = 2284 X ± SD
Age (years) Body height (kg) Body weight (cm) BMI (kg/m2 ) BF% BF (kg) BMI-percentiles 95 Total
Boys N = 1217 Med (Min—Max)
7.25 ± 2.05 6.80 126.31 ± 13.36 125.00 27.30 ± 8.95 25.30 16.69 ± 2.63 16.07 14.84 16.30 ± 6.03 4.76 ± 3.44 3.58 N (%) 126 (5.52%) 1636 (71.63%) 308 (13.48%) 214 (9.37%) 2284 (100%)
Differences between boys and girls: * p < 0.05. ** p < 0.01. † p < 0.001.
(3.50—11.39) (96.60—173.40) (13.40—78.90) (12.21—38.85) (4.15—48.89) (1.06—35.79)
Girls N = 1067
X ± SD
Med (Min—Max)
X ± SD
7.29 ± 2.09 126.84 ± 13.18* 27.75 ± 9.05* 16.82 ± 2.68** 15.61 ± 6.38† 4.63 ± 3.57*
6.82 125.60 26.00 16.22 13.87 3.42
7.21 ± 2.01 125.71 ± 13.54 26.80 ± 8.81 16.54 ± 2.58 17.09 ± 5.50 4.90 ± 3.28
N (%) 78 (6.41%) 832 (68.36%) 169 (13.89%) 138 (11.34%) 1217 (100%)
(3.50—11.39) (97.10—173.40) (13.90—73.20) (12.31—33.67) (4.15—48.89) (1.22—35.79)
Med (Min—Max) 6.79 (3.51—11.38) 124.10 (96.60—166.10) 24.60 (13.40—78.90) 15.96 (12.36—38.85) 16.00 (6.63—38.20) 3.81 (1.06—24.54) N (%) 48 (4.50%) 804 (75.35%) 139 (13.03%) 76 (7.12%) 1067 (100%)
Relationship between body mass index and body fat in children—–Age and gender differences
Table 2
e169
Post hoc comparison with Bonferonni adjusted ␣ level of 0.05. Boys: a versus 7; b versus 4, 5, 6 and 7; c versus 8; Girls: a versus 4,5,6 and 7; b versus 8; c versus 9; d versus 4; e versus 5; f versus 6; g versus 7.
11.90 ± 1.78 12.04 ± 2.55 0.068 17.52 ± 4.95a,b 18.82 ± 5.13a,b,c 0.000 27.16 ± 4.51d,e,f 30.85 ± 4.78 d,e,f,g 0.000 11.40 ± 2.53 16.86 ± 4.38a 28.86 ± 4.74d,e,f,g 10.06 ± 3.58 15.56 ± 3.60 25.32 ± 4.99d,e 12.50 ± 1.03 14.17 ± 2.67 20.33 ± 4.79 11.94 ± 2.03 13.92 ± 2.51 18.16 ± 3.85
10.80 ± 1.58 10.94 ± 2.77 14.72 ± 3.06 15.06 ± 3.38 22.70 ± 3.99d 23.36 ± 5.04d,e
0.031 0.000 0.044 9.17 ± 2.32 14.64 ± 4.41 28.72 ± 7.77b,c 12.25 ± 5.17 14.05 ± 3.96 27.55 ± 7.47b,c 8.11 ± 0.30 12.90 ± 3.96a 24.54 ± 5.58b 9.69 ± 1.34 13.35 ± 3.49 22.61 ± 6.07b 15.27 ± 7.21 15.45 ± 6.57 17.82 ± 6.76 14.46 ± 3.93 14.50 ± 5.23 15.12 ± 5.19 14.23 ± 4.16 13.83 ± 4.30 15.08 ± 4.81
11 10 9 8 7 6 5 4
11.78 ± 2.57 13.90 ± 4.08 11.96 ± 2.34
Boys 85 Girls 85
BMIAge (years) percentiles
This research was carried out with an aim of analyzing the relation between BMI and %BF in children aged 4—11 years. Main results reveal age-related variations in %BF within the same BMI-category, with highest discrepancy between BMI and %BF in boys. BMI cannot represent fatness in boys under 8
Body fat percent in different BMI- and age-categories.
Discussion
Table 3
(15.61 ± 6.38 vs. 17.09 ± 5.50, p < 0.001). Underweight was observed in 5.52% examined children, with higher prevalence in boys (6.41% vs. 4.50%). The percentage of overweight (BMI values between 85th and 95th percentile) was 13.48%, while the contribution of obesity was 9.37%. The proportion of overweight and obesity was higher in boys (13.89 vs. 13.03, and 11.34 vs. 7.12, respectively). Table 3 presents %BF values in different age groups within three BMI-categories: underweight (BMI < 5th percentile), normal weight (BMI = 5th—85th percentile) and overweight and obesity (BMI > 85th percentile). These %BF variations are shown as interquartile ranges, outliers, and extreme cases in Fig. 1. In underweight children of both sexes %BF values did not significantly differ between age groups. Differences in %BF between underweight, normal weight and overweight children from different age groups were more obvious in girls, whereas in boys we registered overlapping in %BF values between different BMI categories for age groups younger than 8 years. In younger age groups (4—8 years) normal weight boys did not differ significantly in %BF. %BF values started to decrease in 8-year-old boys, and were significantly lower in 9-year-old boys comparing to 7-year-old counterparts. This was followed by increasement of %BF in 10- and 11-year-old ones. In normal weight girls %BF values gradually increased with aging, with significant jumps in 9-, 10- and 11-year-old girls. In overweight and obese children %BF continually increased with aging, with significant rises in 8-, and 10-year-old boys and in 6-, 7- and 9-year-old girls. The weakest correlation between BMI and %BF was established in underweight girls (Table 4). Even negative correlation was established in underweight boys. In boys, the strongest correlation was registered in overweight subjects, while the best correlation in girls was established in normal weight individuals. Girls showed higher correlation between %BF and BMI (r = 0.834, p < 0.01) comparing to the boys (r = 0.577, p < 0.01).
B. Srdi´ c et al.
ANOVA Sig.
e170
Relationship between body mass index and body fat in children—–Age and gender differences
e171
Figure 1 Age-related distribution of %BF in different BMI-categories.
Table 4
Correlation between BMI and %BF.
BMI-percentiles
Boys
85 Total
−0.397 0.265** 0.629** 0.577**
* **
Girls **
0.351* 0.755** 0.704** 0.834**
Correlation is significant at the 0.05 level (2-tailed). Correlation is significant at the 0.01 level (2-tailed).
years. According to %BF values, adiposity rebound may appear somehow later in boys. BMI is recommended as a suitable indicator of adiposity, which is quite simple to calculate, and cut-off values for children are adjusted to the age and gender [10]. However, numerous studies indicated some inconsistencies in BMI and body fat mass; thus risk of failures in diagnosing obesity by the use of BMI in both, adults, and children [11—15]. BMI does not discriminate between fat, muscle and skeletal mass; their ratio can vary without any significant change of the total body mass. Childhood is mainly characterized by constant changes of body composition, predominantly encompassing increase of the bone mineral mass and decrease of the total body water. Widhalm et al. showed that BMI explains 73% and 63% of the variance of body fat in boys and girls younger than 10 years, respectively, while it has lower predictive value in older children [16]. According to our results, BMI was more selective in prediction of %BF in girls, whereas in boys the differences in %BF among underweight, normal weight and overweight individuals were not distinct until the 8th year of age. At the same time, correlation between %BF and BMI
was stronger in girls, which corresponds to results reported by Schaefer et al. [11], Morimoto et al. [15] and Yoshinaga et al. [17]. Numerous studies emphasized that the accuracy of BMI as an indicator of fatness is strongly determined by the degree of fatness, thereby BMI proved better in children with higher body mass [11,18,19], as in the boys’ population in our study. In girls, however, the best correlation was established among normal weight individual, and accuracy of BMI decreased with more extreme weight values. The onset of adiposity rebound, which implies the increase of number of adipocytes, is usually determined by monitoring the changes of BMI values, and mostly occurs between 3rd and 7th year of age [20]. Our results revealed that %BF significantly changes with aging in normal weight and overweight children, but %BF changes did not follow the expected curve of BMI-changes. The significant %BF increase corresponds with the period of adiposity rebound in normal weight girls (fat mass significantly increased after the age of 7 years), while the same phenomenon in normal weight boys occurs somewhat later (drop started at 8th year with substantial increase at the age of 10 years). At the same time, continual increase of %BF is recorded in overweight and obese children, with several pronounced jumps that appeared earlier in girls. These changes could reflect the changes in body components ratio within the same nutritional level. Maynard et al. suggested that increase in BMI in children is mainly due to increases in fatfree mass, particularly in boys [21]. Contrary to our results, research of Schaefer et al. revealed constant increase of body fat mass in boys after 6th
e172 year of age, reaching the maximum value at the age of 13 [11]. Their results for girls’ population correspond with our findings. According to same authors, decrease in body fat, which was more pronounced in boys, occurs somewhat later during puberty, and is associated with the increased excretion of sexand growth hormone and their lipolytic effect [22]. In a view of potential risk of obesity development, measuring of the fat mass should be the most important parameter, and BMI only an indirect indicator of the nutritional status [23—25]. The main problem in such assessments is the lack of cut-off values for body fat mass, which should be specific and adjusted for sex, age or ethnic background. Some studies offered useful suggestions on such referent values that should present the basis for future international consensus [12,26,27]. Furthermore, there are numerous methods for assessment of body composition and results thereof diverge to a great extent [28]. All this impedes comparing of the results from different studies, calling for establishing of referent method for body composition analysis. The main shortcoming of this study is its crosssectional design, and the obtained differences with respect to body fat mass among different age groups cannot be interpreted with full accuracy. On the other hand, certain limitations of the use of anthropometric method in assessing the body fat mass are to be considered. It is based on indirect assessment of body density only on the basis of subcutaneous adipose tissue thickness [28]. Furthermore, besides the standardized measurement procedure, individual diversities in the ratio of subcutaneous fatty tissue and total fat tissue must not be neglected [29]. Finally, one must keep in mind that relation of BMI and body fat mass is determined by numerous factors, such as sexual maturity, race, gender and fat tissue distribution pattern [7,30]. The obtained results strongly suggest some discrepancies between BMI and body fat, particularly in boys, as well as irregular dynamics of age-related change of body fat mass in relation to BMI, requiring an additional evaluation in longitudinal studies.
Acknowledgements We would like to thank all children who participated in the study, their parents and the staff of daycare centers and schools where our study was performed. Part of this study was financially supported by Provincial Secretary for Science and Technological Development (114-451-00606/200702, 2005—2008).
B. Srdi´ c et al.
References [1] Geiß HC, Parhofer KG, Schwandt P. Parameters of childhood obesity and their relationship to cardiovascular risk factors in healthy prepubescent children. Int J Obes 2001;25:830—7. [2] Lindsay RS, Hanson RL, Roumain J, Ravussin E, Knowler WC, Tataranni PA. Body mass index as a measure of adiposity in children and young adults: relationship to adiposity by dual energy X-ray absorptiometry and to cardiovascular risk factors. J Clin Endocrinol Metab 2001;86: 4061—7. [3] Warner JT, Cowan FJ, Dunstan FD, Gregory JW. The validity of body mass index for the assessment of adiposity in children with disease states. Ann Hum Biol 1997;24: 209—15. [4] Pietrobelli A, Faith MS, Allison DB, Gallagher D, Chiumello G, Heymsfield SB. Body mass index as a measure of adiposity among children and adolescents: a validation study. J Pediatr 1998;132:204—10. [5] Rolland-Cachera MF, Deheeger M, Bellisle F. The adiposity rebound and adult obesity. In: Madeiros-Neto G, Halpern A, Bouchard C, editors. Progress in obesity research: 9. Proceedings of the 9th International Congress on Obesity. London: John Libbey Eurotext Ltd; 2003. p. 456—62. [6] Demerath EW, Schubert CM, Maynard M, Sun SS, Chumlea WC, Pickoff A, et al. Do changes in body mass index percentile reflect changes in body composition in children? Data from the Fels Longitudinal Study. Pediatrics 2006;117:e487—95. [7] Daniels SR, Khoury PR, Morrison JA. The utility of body mass index as a measure of body fatness in children and adolescents: differences by race and gender. Pediatrics 1997;99(6):804—7. [8] Ogden CL, Kuczmarski RJ, Flegal KM, Mei Z, Guo S, Wei R, et al. Centers for Disease Control and Prevention 2000 Growth Charts for the United States: Improvements to the 1977 National Center for Health Statistics Version. Pediatrics 2002;109(1):45—60. [9] Slaughter MH, Lohman TG, Boileau RA, Horswill CA, Stillman RJ, Van Loan MD, et al. Skinfold equations for estimation of body fatness in children and youth. Hum Biol 1988;60:709—23. [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] Schaefer F, Georgi M, Wühl E, Schärer K. Body mass and percentage fat mass in healthy German schoolchildren and adolescents. Int J Obes 1998;22:461—9. [12] Tyrrell VJ, Richards GE, Hofman P, Gillies GF, Robinson E, Cutfield WS. Obesity in Auckland school children: a comparison of the body mass index and percentage body fat as the diagnostic criterion. Int J Obes 2001;25:164—9. [13] Wells JCK, Fewtrell MS, Williams JE, Haroun D, Lawson MS, Cole TJ. Body composition in normal weight, overweight and obese children: matched case-control analyses of total and regional tissue masses, and body composition trends in relation to relative weight. Int J Obes 2006;30:1506—13. [14] Tennefors C, Forsum E. Assessment of body fatness in young children using the skinfold technique and BMI vs. body water dilution. Eur J Clin Nutr 2004;58:541—7. [15] Morimoto A, Nishimura R, Sano H, Matsudaira T, Miyashita Y, Shirasawa T, et al. Gender differences in the relationship between percent body fat (%BF) and body mass index (BMI) in Japanese children. Diabetes Res Clin Pract 2007;78:123—5.
Relationship between body mass index and body fat in children—–Age and gender differences [16] Widhalm K, Schönegger K, Huemer C, Auterith A. Does the BMI reflect body fat in obese children and adolescents? A study using the TOBEC method. Int J Obes 2001;25:279—85. [17] Yoshinaga M, Shimago A, Noikura Y, Kinou S, Ohara T, Miyta K. Body fat percentage in girls increased steadily with age and percentile rank of body mass index. Pediatr Int 2002;44:149—52. [18] Freedman DS, Wang J, Maynard LM, Thornton JC, Mei Z, Pierson RN, et al. Relation of BMI to fat and fat-free mass among children and adolescents. Int J Obes 2005;29:1—8. [19] Mast M, Langnase K, Labitzke K, Bruse U, Preuss U, Muller MJ. Use of BMI as a measure of overweight and obesity in a field study on 5—7 year old children. Eur J Nutr 2002;41:61—7. [20] Cole TJ. Children grow and horses race: is the adiposity rebound a critical period for later obesity? BMC Pediatrics 2004;4:6. [21] Maynard LM, Wisemandle MA, Roche AF, Chumlea WC, Guo SS, Siervogel RM. Childhood body composition in relation to body mass index. Pediatrics 2001;107:344—50. [22] Loomba-Albrecht LA, Styne DM. Effect of puberty on body composition. Curr Opin Endocrinol Diabetes Obes 2009;16:10—5. [23] Dietz WH. Periods of risk in childhood for the development of adult obesity — what we need to learn? J Nutr 1997;127:1884—6.
e173
[24] Davison KK, Susman EJ, Birch LL. Percent body fat at age 5 predicts earlier pubertal development among girls at age 9. Pediatrics 2003;111(4):815—21. [25] Ribeiro J, Santos P, Duarte J, Mota J. Association between overweight and early sexual maturation in Portuguese boys and girls. Ann Hum Biol 2006;33(1): 55—63. [26] Ellis KJ. Measuring body fatness in children and young adults: comparison of bioelectrical impedance analysis, total body electrical conductivity, and dual-energy X-ray absorptiometry. Int J Obes Relat Metab Disord 1996;20:866—73. [27] Goulding A, Gold E, Cannan R, Taylor RW, Williams S, LewisBarned NJ. DEXA supports the use of BMI as a measure of fatness in young girls. Int J Obes Relat Metab Disord 1996;20:1014—21. [28] Eisenmann JC, Heelan KA, Welk GJ. Assessing body composition among 3- to 8-year-old children: anthropometry, BIA and DXA. Obes Res 2004;12(10): 1633—40. [29] Wells JCK. A critique of the expression of pediatric body composition data. Arch Dis Child 2001;85:67—72. [30] Kagawa M, Kerr D, Uchida H, Binns CW. Differences in the relationship between BMI and percentage body fat between Japanese and Australian-Caucasian young men. Br J Nutr 2006;95(5):1002—7.
Available online at www.sciencedirect.com