Measurement and Definitions of Obesity In Childhood and Adolescence: A field guide for the uninitiated

1475-2891-6-32 1475-2891 Review Measurement and Definitions of Obesity In Childhood and Adolescence: A field guide for the uninitiated Sweeting N Helen helen@sphsu.mrc.ac.uk

MRC Social and Public Health Sciences Unit, 4, Lilybank Gardens, Glasgow, G12 8RZ, UK

Nutrition Journal 1475-2891 2007 6 1 32 http://www.nutritionj.com/content/6/1/32 17963490 10.1186/1475-2891-6-32 15 3 2007 26 10 2007 26 10 2007 2007 Sweeting; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

This paper aims to guide readers embarking on the complex literature in respect of childhood and adolescent obesity. It opens with a discussion of definitions of 'obesity' based on overall fat levels and the significance of fat distribution. This is followed by simple descriptions of the various techniques used to measure fat, including density-based, scanning, bioelectrical impedance and anthropometric methods. The paper then turns to 'overweight' and the measurement of weight in relation to height, particularly via body mass index (BMI). While it is a relatively simple measure and a valuable tool, BMI has several disadvantages, which are described. These include a lack of consensus on which values should be used to define 'overweight' or 'obese', with the result that the literature contains a confusing multiplicity of child and adolescent obesity rates.

Introduction

At one level, research on child and adolescent obesity rates is easy to understand. Based on recent studies, the BBC News website includes statements such as 'obesity affects 12% of under-11s' (14^thDecember, 2006), and 'Levels of obesity in children aged two to 10 years rose from 9.9% to 13.4% between 1995 and 2004, according to the Health Survey for England.' (25^thJanuary, 2007). However, for the researcher who wishes to gain a clearer understanding of how obesity is measured, or to delve into these figures in more detail, the literature may prove quite challenging.

One difficulty is that descriptions of the measurement of obesity are littered with acronyms, an understanding of which is often assumed. But for the uninitiated, what do ADP, DXA or BIA stand for? And perhaps even more puzzling, how is it possible to accurately measure levels of fat within a living individual? A second source of puzzlement is apparently conflicting numbers. For example, given that obesity levels are lower in the UK than the US 1, why would one paper report rates for US 12–19 year olds in 1999–2002 of 16%, but another report rates for English 11–15 year olds in 2000 of 18% 23?

This paper is written by one who was puzzled, and is now less so, to help others embarking upon this literature. It begins with a discussion of the significance of overall fat levels and fat distribution, followed by a description of methods to measure fat. The second half of the paper describes measurements of overweight rather than fat. It focuses particularly on body mass index, since this is by far the most common indicator, exploring how it has been used to define obesity, and how accurately it identifies the fattest children. From this point forwards, terms commonly used in this literature are in bold the first time they appear.

Obesity, fat levels and fat distribution

Obesity should be defined as excess body fat or adipose tissue; it is this, not weight which is associated with the comorbid conditions 4. This being the case, the next question is what level of fat should be defined as 'obese'. Studies of children and adolescents which have examined the relationship between percentage of body fat calculated from skinfold measurements and indicators of biomedical status such as blood pressure and blood lipids, have suggested 30% fat in females and 20–25% in males 56. There is also evidence of ethnic differences, for example, South Asian people appear to be sensitive to the metabolic consequences of obesity at lower levels than white people 7.

This is further complicated by findings that it is central (also described as intra abdominal – IA, or visceral) fat which is more pathogenic 48. Adults with large waist circumferences have excess morbidity, including back pain, diabetes and CVD risk factors 9, and although less clear, there is some evidence of health risks associated with excess abdominal fat in children 1011. There is also evidence that the excess fat in obese children and adolescents is likely to accumulate in the abdominal regions 1213.

Overall levels, as well as the distribution of fat, differ according to both sex and ethnicity. The android (male, or 'apple shaped') fat pattern is represented by relatively greater amounts in the upper body, the gynoid (female, or 'pear') pattern by greater amounts in the hip and thigh areas 14. Female lower body fat is less metabolically active than that in the abdominal region, and is programmed to become mobilized during pregnancy and lactation. In relation to the greater pathogenicity of abdominal fat, it is interesting that mortality rates are higher among females with android fat patterning 8. Sex differences in fat levels have generally been considered to become manifest during puberty 15. Thus, in samples followed up through adolescence, levels of fat are higher among females, and of fat-free mass among males 1617181920. However, more recent studies of pre-pubertal children, some as young as 3 years old, in the US, UK, Germany, Italy and China, have also found higher percentages of body fat and evidence of the gynoid pattern among females 2122232425.

Percentage body fat also appears to be lower in black, perhaps particularly black African children (and adults) compared with Caucasians. In other words, for any given body mass, black African children have higher fat-free and lower fat mass. Levels of abdominal fat also tend to be lower among Black Africans. There is, in addition, some evidence that these differences are more pronounced among females than males 10172126. In contrast, many Asian races, and possibly also Hispanics and Chinese, carry a higher percentage fat mass than Caucasians 42526.

Methods to measure fat

Table 1 describes the main techniques which have been adopted to measure fat in human subjects. They are categorized as density-based (hydrodensitometry; air displacement plethysmography), scanning (computerized tomography; magnetic resonance imaging; dual-energy x-ray absorptiometry), bioelectrical impedance and anthropometric (skinfold; waist circumference; waist-hip ratio) methods, according to the general principle on which they are based. More complex and detailed descriptions are available 272829. As Table 1 demonstrates, the majority of these methods are complex and limited to research settings.

Table 1

Methods to measure fat

General principle

Method

Acronym

Methodology

Method-specific principle

Further comments

Density-based methods

If the density (weight per unit volume) of a human body is known, then the relative proportions of fat and fat-free mass can be estimated using an equation such as those of Siri [72] or Lohman [73]. While mass can be easily determined by weighing, volume measurements are more difficult [74].

Hydrodensitometry (underwater weighing)

UWW

Weighs the subject while submerged in a large tank (having exhaled maximally) and also outside the tank [29].

Based on Archimedes' principle (buoyancy law) that if the density of an object exceeds that of water, it will sink. Given two people of equivalent weight outside the tank, the one with more fat, which is less dense than water, will weight less in water than the one with more fat-free tissue (such as bone and muscle) which is more dense than water [75]. (In fact, it is unnecessary to actually weight the subject underwater, since their volume can also be assessed via the amount of water displaced when they are submerged.)

Often described as 'the gold standard', but time-consuming and requires the subject to submerge themselves, so particularly unsuitable for certain populations, such as children, and limited to research settings [76].

Air Displacement Plethysmography

ADP

Measures the volume of air the subject displaces inside an enclosed chamber.

Given the subject's volume and weight, their density can be calculated.

Early plethysmographs were complex, inconvenient and required temperature-controlled surroundings. A simple, quick automated plethysmograph [77] has been available since mid 1990s, but is still limited to research settings [74].

Scanning methods

Can assess not just overall fat mass, but also its regional distribution.

Computerised Tomography; Magnetic Resonance Imaging

CT; MRI

CT – a series of x-rays pass through the body at different angles. MRI – uses a strong magnetic field and a radio wave antenna which sends signals to the body and then receives them back. These are used to produce internal images.

Both allow for the creation of cross-sectional high-resolution internal images.

Expensive, involve radiation exposure (CT) and limited to research settings [27-29].

Dual-Energy X-ray Absorptiometry

DEXA or DXA

A series of transverse scans, via low energy x-ray beams, progress inch-by-inch across the body and are collected by an external detector.

The beams are differentially absorbed by the various different tissues (fat, bone, etc) in the body.

Can be used to calculate fat and fat-free mass, and both total and regional body composition in subjects over a wide range of ages and body sizes. Relatively low radiation dose. Validated against UWW and comparison with animal carcasses in the pediatric weight range. Use limited to research settings [27, 28, 46, 78, 79].

Bioelectrical impedance methods

Electric currents pass more easily through body fluids in muscle and blood, but encounter resistance ('biolectical impedance') when they pass thorugh fat, since it contains little water.

Bioelectrical Impedance Analysis

BIA

Conductors are attached to the subject's body, and a low, safe, current is sent through. Electrodes are generally placed at wrist and ankle; an increasingly commonly used analyser requires subjects to stand on it in bare feet and hold a handgrip in each hand. Foot-to-foot BIA measures the impedance of the lower body and only requires the subject to stand on pad electrodes.

The resistance between the conductors provides a measure of body fat.

Although less accurate than more sophisticated measurements, some current analysers are relatively inexpensive, portable, simple and quick, meaning BIA can now be used in the field and with large samples [20, 80].

Anthropometric methods

Direct measurements of various body parameters.

Skinfold measurements

Subcutaneous (but not internal) fat is measured by firmly grasping a fold of skin with callipers and raising it, with no muscle included. Single site measurements, e.g. triceps skinfolds [16] are simplest. An alternative is to add skinfolds from a variety of sites, generally representing both peripheral and trunk areas [17].

Subcutaneous fat may be taken as an indicator of total fat. Fat distribution can also be determined via the ratio of trunk to peripheral skinfolds [81]. It is also possible to calculate total body fat via equations: Slaughter's equations predict percent body fat from the sum of triceps plus subscapular, or triceps plus calf in children and young people [82]; more recent equations by Dezenberg use triceps skinfolds plus body weight, sex and ethnicity [83].

Cheap and fairly simple, but the need to partially undress may put some subjects off, leading to bias. Also difficult to measure reproducibly, particularly if the subject is fat [84].

Waist circumference

Ideally measured using a flexible plastic tape with a sprung handle to ensure reproducible levels of tension [29]. Since a potential source of error is incorrectly positioning the tape, the measurement site is generally specified by reference to specific anatomic landmarks [85].

WC reflects total and abdominal fat levels, and as an indicator of adiposity is not greatly influenced by height [86].

WC centiles for children have been developed in a number of countries [86-89]. It has also recently been suggested that the ratio of waist to height could be used as a rapid screening tool [90].

Waist-hip ratio

WHR

A larger WHR in adults indicates relatively larger amounts of abdominal fat and has been used to describe body fat distribution. However it is influenced by several other bodily factors and there is some evidence that it is a poorer measure of body fat distribution in children [91].

Infrequently used in studies of children and adolescents.

When fat cannot be measured – weight in relation to height

In contrast to obesity, which is excess levels of fat, overweight is excess weight in relation to height, and can be easily measured using only a set of scales and a stadiometer. Indeed, it may not even require this; some studies have used self-report weights and heights. However, comparisons with measured values show that although correlations are high, weight tends to be under-reported, particularly among females and the overweight, while height may be over-reported. There is a lack of consensus on the impact that this has on measures of weight in relation to height, some authors suggesting that self-report values are adequate 30, others advocating caution 3132.

Measures of weight in relation to height

Although body weight, particularly at very high levels, tends to be associated with adiposity, weight alone is an insufficient measure of obesity, because it is correlated with height 33. A number of measures of weight in relation to height have been devised. The simplest is weight for height. In 1977 a World Health Organization (WHO) bulletin noted that in undernourished populations, 80% median weight for height (which corresponds to approximately -2.0 standard deviations) was suitable for classifying malnourished children. Following this principle, it was suggested that 120% (or +2.0 standard deviations) could be used in populations where over-nutrition was a problem 34. Although adopted in some recent studies 3536, such measures tend not to be used currently. Body Mass Index (BMI), defined as weight (kg)/height squared (m²) is the most frequently used measure of weight in relation to height, but there are others. These include Rohrer's Ponderal Index (termed either Rohrer's Index – RI, or the Ponderal Index – PI), defined as weight/height³. This has been compared with BMI in respect of its ability to predict percentage body fat in children and adolescents, and its long-term associations with adult obesity 3738. Although it may perform as well or better in some respects than BMI, it is much less used with children and adolescents, although it remains popular with neonatologists 39. Finally, there is Benn's Index, defined as weight/height^pwhere the power p is chosen so the index is independent of height. While this may be 'the ideal index' (39, p.289), the fact that p is neither constant, nor necessarily a whole number, means the calculations are very complicated, and such indices are rarely used 4041.

Body Mass Index

As noted above, BMI is the most frequently used measure of weight in relation to height. It has been described as 'the backbone of the obesity classification system and surveillance statistics ... an immensely valuable tool' (4, p.141). However, a number of authors have detailed its disadvantages 42434445.

The first of these is that BMI varies between males and females and according to age and level of maturity. Thus, while male and female BMIs tend to be similar in childhood, they are higher among females in adolescence. In respect of age, BMI increases from birth to around one year, then declines to around age six, then increases through the remainder of childhood and adolescence. The point at which BMI reaches its lowest level and begins to increase is termed 'adiposity rebound', with earlier adiposity rebound being associated with increased risk of subsequent overweight 46. Such variations mean that among children and adolescents the significance of any particular BMI is more difficult to determine than within adult populations.

A second, and related limitation of BMI, is that it reflects both fat and fat-free components of body weight. However, as described earlier, populations differ in respect of both percentage fat mass and fat distribution, and in the relation between body composition and morbidity. This means, again, that the significance of any particular BMI will vary. Thus, among children with the same BMI, fat measurements are higher for whites than for blacks 47. Further, recent studies have suggested that increases in overall BMIs have been accompanied by larger increases in the percentage as fat mass and concomitant decreases in fat free mass (attributed to reduced activity levels). Importantly, this suggests that recent increases in adiposity are even greater than those suggested by increases in BMI 132348.

A third disadvantage of BMI is that since one of its components is height, the index also varies according to height, and this association in turn varies according to sex and age. Its relationship to height means that BMI is also affected by relative leg length.

A further disadvantage is that since BMI does not measure fat directly, there is no consensus about which cut-off to use in order to define obesity in children and adolescents. Among adults, a WHO Expert Committee on Physical Status agreed, in 1993, to classify BMIs as follows: 25–29.9 as 'overweight' (or 'pre-obese'); 30–34.9 as 'obese class I'; 35–39.9 as 'obese class II'; and over 40 as 'obese class III'. This classification was based on the risk of comorbidities, rising across the four categories from 'increased', to 'moderate', 'severe' and 'very severe' 4950. Unfortunately, much less is known about levels of risk associated with specific BMI levels in children and adolescents. This means that statistical approaches have often been used. These involve working out the distribution for a particular population and rather arbitrarily choosing particular values – often the 85^thor 95^thcentiles, which distinguish those with the highest BMIs from the rest of the population 33. Since the distribution of BMI varies according to sex, age and ethnicity, sex- and age- (and, sometimes, in the US, race-) specific centile curves are calculated.

BMI and overweight/obesity – where to draw the line?

Clearly, if the aim is to track levels of obesity over time, or to compare populations, the BMI centile values defining obesity must be fixed. The question is not only which centile they should be fixed at, but also which population (and so which point in time) should be used as the basis for the calculations?

Within the US, reference growth charts based on nationally representative surveys (the National Health Examination Surveys – NHES, later called the National Health and Nutrition Examination Surveys – NHANES) have been produced since 1977 51, and in 1991, race-specific and population-based BMI centiles, covering ages 6–74 were generated. These are often described as the MDD definitions, based on the initial letters of the authors' surnames 52. An expert committee recommended their use for children and adolescents, with the 95^thBMI centile for age and sex (or BMI 30 kg/m²) as cut-offs for 'overweight', and the 85^thcentile as 'at risk of overweight' for screening purposes. The fact that the committee decided not to use the term 'obese' (which they associated with excess fat rather than weight) 53 has lead to some confusion 33. The most recent US charts were produced by the Centers for Disease Control in 2000 (hence termed 'CDC 2000'). The inclusion of more recent data would have shifted the weight and BMI curves upwards, resulting in a high proportion of fatter children being characterized as 'normal'. In order to avoid this, data obtained since 1988 from those aged over 6 years were excluded 54. These charts also tend to be used within Canada and Australia.

Within the UK, charts ('UK90') have been produced, based on data from several surveys, conducted 1978–90 and including around 30,000 subjects. Although the authors suggested the 98^thcentile, which at age 20 is 29.0 kg/m²(thus close to the adult definition of 30 kg/m²) as 'a reasonable definition of child obesity' (55, p.28), the 95^thcentile is more commonly adopted for epidemiological purposes. Similar BMI-for-age sex-specific reference charts have been developed in several other European countries including France, Germany and Denmark 565758.

Against this background, the International Obesity Task Force (IOTF), a global network of expertise, concluded that the definition of obesity in children and adolescents should be consistent with that for adults, and that, ideally, it should be based on a reference representative of the world's population 59. Subsequently, data from 6 countries, collected 1963–93, was pooled, and in 2000, centile curves were published that pass through the points of 25 kg/m²and 30 kg/m²(reflecting WHO recommended definitions of adult overweight and obese) at age 18 60. In fact, the available reference data do not adequately represent the world's population. Further, as noted earlier, ethnic differences in body composition and the percentage of body fat associated with adverse health consequences mean that a single international definition of obesity may not be appropriate 44. However, it has been suggested that despite these limitations, prevalence studies should present results based on the IOTF cut-offs as well as national definitions, since this would allow for comparison across populations 61.

The confusing multiplicity of child and adolescent obesity rates seen in the literature results from the use of different definitions; a number of studies have demonstrated this by applying several definitions to the same population. For example, one US study found rates of 13%, 11% and 8% when applying the MDD 'overweight', CDC 2000 'overweight' and IOTF 'obesity' criteria to (1998–1994) data in respect of 6–8 year old males 62. Similarly, a UK study of 4–11 year olds (1994 data) found male and female obesity rates of 1.7% and 2.6% when applying the IOTF definition, but of 2.5% and 2.2% when applying the UK90 definition. Thus, while the IOTF definition suggested higher rates among females, the reverse was suggested by the UK90 definition 63.

Crucially, given that BMI measures excess weight rather than fat, studies have also examined the accuracy with which it identifies the fattest children and adolescents. This translates into its sensitivity (how good it is at identifying the truly obese) and specificity (how good it is at identifying the non-obese). Ideally, this requires a definition of 'true' obesity, however, as described earlier, this is lacking. Studies have therefore adopted a variety of methods. These include correlating BMI with percentage body fat (measured in various ways); comparing BMI-defined obesity (various different cutoffs) with obesity defined in terms of percentage body fat (also various different cutoffs, generally around 30% for females and 20–25% for males – see earlier and 56); comparing BMI-defined obesity with percent body fat levels deemed 'high' (typically 85^thor 95^thpercentile); and finally, comparing cardiovascular risk factors among those defined via BMI as obese and non-obese 64656667. The consensus from such studies is that BMI is a reasonable measure of adiposity, although the relationship differs not only according to age, sex and ethnicity, but also degree of fatness, being better at high levels. This means that it is of more use in epidemiological studies, but is relatively poor at predicting fat mass in any individual child. Further, current BMI cutoffs have relatively high specificity, but lower sensitivity. This means that non-obese children are unlikely to be wrongly labeled, however obese children may be missed (for example, see 22686970).

Conclusion

Increasing rates – regardless of definition – have highlighted the importance of the measurement of child and adolescent obesity on a population level. However 'obesity' is a slippery concept. Even the WHO definitions of 'overweight' and 'obesity' adopted for use with adults are based on BMI levels associated with raised risks of comorbidities. However, 25 kg/m²and 30 kg/m²are also easily remembered, neat and tidy numbers. Other definitions would have been possible. In fact, lower cut-offs have subsequently been suggested for Asians, given their tendency to both carry a higher percentage fat and be more sensitive to its metabolic consequences 71.

Decisions in respect of children are even more complex. As with adults, the relationships between BMI and levels of fat, and between levels of fat and associated morbidities, differ between ethnic groups. In addition, BMI levels, and their relationship with levels of fat vary according to sex, age and maturity. This has resulted in attempts to define obesity on the basis of percentage body fat, waist circumference and, most frequently, BMI, which reflects weight in relation to height. The consequence of this is that different studies may define obesity in different ways. None could be said to be 'correct' 62. It has been argued that efforts to assess body fat and develop population standards against which individuals can be compared should be increased 4. However, even this would require decisions as to cutoffs.

The literature in this area can be confusing. While that seems unlikely to change in the near future, this guide may increase the understanding of those embarking upon it.

Acknowledgements

The author would like to thank Professors Sally Macintyre and Patrick West and Dr. Seeromanie Harding for comments on an earlier draft. The author is funded by the UK Medical Research Council. This paper was funded in part by the European Men's Health Development Foundation.

Obesity: the scale of the problem - fact sheet Association for the Study of Obesity (ASO) http://www.aso.org.uk National prevalence of obesity: Prevalence of obesity in the United States Baskin ML Ard J Franklin F Allison DB Obesity Reviews 2005 6 1 5 7 10.1111/j.1467-789X.2005.00165.x 15655032 Health Survey for England 2004. Updating of trend tables to include childhood obesity data Information Centre for Health and Social Care http://www.ic.nhs.uk Beyond body mass index Prentice A Jebb SA Obesity Reviews 2001 2 141 147 10.1046/j.1467-789x.2001.00031.x 12120099 Body fatness and risk for elevated blood pressure, total cholesterol, and serum lipoprotein ratios in children and adolescents Williams DP Going SB Lohman T Harsha DW Srinivasan S Webber LS Berenson G American Journal of Public Health 1992 82 358 363 1694353 1536350 Defining obesity in children by biological endpoint rather than population distribution Dwyer T Blizzard CL International Journal of Obesity 1996 20 5 472 480 8696427 Early evidence of ethnic differences in cardiovascular risk: cross sectional comparison of British South Asian and white children Whincup P Gilg J Papacosta O Seymour C Miller G Alberti K Cook D British Medical Journal 2002 324 1 6 1121932 11777781 10.1136/bmj.324.7338.635 Human obesity: an evolutionary approach to understanding our bulging waistline Lev-Ran A Diabetes/Metabolism Research Reviews 2001 17 5 347 362 10.1002/dmrr.230 Impairment of health and quality of life in people with large waist circumference Lean MEJ Han TS Seidell J Lancet 1998 351 853 856 10.1016/S0140-6736(97)10004-6 9525361 Relation between visceral fat and disease risk in children and adolescents Goran MI Gower B American Journal of Clinical Nutrition 1999 70 (suppl) 149S 156S Waist circumference and cardiovascular risk factors in prepubertal children Maffeis C Pietrobelli A Grezzani A Provera S Tato L Obesity Research 2001 9 3 179 187 11323443 Fat distribution in obese and nonobese children and adolescents Moreno LA Fleta J Mur L Sarria A Bueno M Journal of Pediatric Gastroenterology & Nutrition 1998 27 2 176 180 10.1097/00005176-199808000-00009 Central overweight and obesity in British youth aged 11-16 years: cross sectional surveys of waist circumference McCarthy HD Ellis SM Cole TJ BMJ 2003 326 7390 624 151972 12649234 10.1136/bmj.326.7390.624 Gender-specific aspects of obesity Legato MJ International Journal of Fertility & Women's Medicine 1997 42 3 184 197 Obesity in children and young people: a crisis in public health Lobstein T Baur LA Uauy R Obesity Reviews 2004 5 (Suppl. 1) 4 85 10.1111/j.1467-789X.2004.00133.x 15096099 Body composition of African American and white children: A 2-year follow-up of the BAROC study Bray GA DeLany JP Harsha DW Volaufova J Champagne CM Obesity Research 2001 9 10 605 621 11595777 Longitudinal analysis of changes in indices of obesity from age 8 years to age 18 years Dai SF Labarthe DR Grunbaum JA Harrist RB Mueller WH American Journal of Epidemiology 2002 156 8 720 729 10.1093/aje/kwf109 12370160 Age- and maturity-related changes in body composition during adolescence into adulthood: The Fels Longitudinal Study Guo SS Chumlea WC Roche AF Siervogel RM International Journal of Obesity 1997 21 12 1167 1175 10.1038/sj.ijo.0800531 9426385 Do changes in body mass index percentile reflect changes in body composition in children? Data from the Fels Longitudinal Study Demerath EW Schubert CM Maynard LM Sun SS Chumlea WC Pickoff A Czerwinski SA Towne B Siervogel RM Pediatrics 2006 117 3 e487 95 10.1542/peds.2005-0572 16510627 Body fat reference curves for children McCarthy HD Cole TJ Fry T Jebb SA Prentice AM International Journal of Obesity 2006 30 4 598 602 10.1038/sj.ijo.0803232 16570089 Physical activity related energy expenditure and fat mass in young children Goran MI Hunter G Nagy TR Johnson R International Journal of Obesity 1997 21 3 171 178 10.1038/sj.ijo.0800383 9080254 Sensitivity and specificity of body mass index and skinfold thicknesses in detecting excess adiposity in children aged 8-12 years Bedogni G Iughetti L Ferrari M Malavolti M Poli M Bernasconi S Battistini N Annals of Human Biology 2003 30 2 132 139 10.1080/0301446021000033409 12637189 Body composition of healthy 7- and 8-year-old children and a comparison with the 'reference child' Ruxton CH Reilly JJ Kirk TR International Journal of Obesity 1999 23 1276 1281 10.1038/sj.ijo.0801067 10643684 Gender differences in fat mass of 5-7-year old children Mast M Kortzinger I Konig E Muller MJ International Journal of Obesity 1998 22 9 878 884 10.1038/sj.ijo.0800675 9756246 Age- and sex-specific body composition of Chinese children Li S Zhang M Yang S Okada T Iwata F Harada K Acta Paediatrica 2005 94 8 1139 1142 10.1080/08035250510028588 16188862 Relation of BMI to fat and fat-free mass among children and adolescents Freedman DS Wang J Maynard LM Thornton JC Mei Z Pierson Jr RN Dietz WH Horlick M International Journal of Obesity 2005 29 1 1 8 10.1038/sj.ijo.0802735 15278104 Selected body composition methods can be used in field studies Ellis KJ Journal of Nutrition 2001 131 1589s 1595s 11340123 Pediatric body composition in clinical studies: which methods in which situations? Pietrobelli A Peroni DG Faith MS Acta Diabetologica 2003 40 Suppl 1 S270 3 10.1007/s00592-003-0084-0 14618491 Body composition measurements: From the past to the future Pietrobelli A Tato L Acta Paediatrica, International Journal of Paediatrics, Supplement 2005 94 448 8 13 10.1080/08035320510035221 Accuracy of teen and parental reports of obesity and body mass index Goodman E Hinden BR Khandelwal S Pediatrics 2000 106 1 Pt 1 52 58 10.1542/peds.106.1.52 10878149 Reliability and validity of self-reported height and weight among high school students. Brener ND McManus T Galuska DA Lowry R Wechsler H Journal of Adolescent Health 2003 32 4 281 287 10.1016/S1054-139X(02)00708-5 12667732 Factors associated with errors in self-reports of stature, weight and body mass index in Minnesota adolescents Himes JH Hannan P Wall M Neumark-Sztainer D Annals of Epidemiology 2004 15 272 278 10.1016/j.annepidem.2004.08.010 Overweight prevalence among youth in the United States: Why so many different numbers? Troiano RP Flegal KM International Journal of Obesity 1999 23, Suppl 2 S22 S27 10.1038/sj.ijo.0800855 The presentation and use of height and weight data for comparing the nutritional status of groups of children under the age of 10 years. Waterlow JC Buzina R Keller W lane JM Nichman MZ Tanner JM Bulletin of the World Health Organization 1977 55 4 489 498 304391 Nutritional status of Kuwaiti elementary school children aged 6-10 years: comparison with the NCHS/CDC reference population Al-Isa AN Moussa MAA International Journal of Food Sciences and Nutrition 2000 51 4 221 228 10.1080/09637480050077103 Prevalence and trends of obesity among school children in Taiwan - The Taipei Children Heart Study Chu NF International Journal of Obesity 2001 25 2 170 176 10.1038/sj.ijo.0801486 11410816 Validity of body mass index compared with other body-composition screening indexes for the assessment of body fatness in children and adolescents Mei ZG Grummer-Strawn LM Pietrobelli A Goulding A Goran MI Dietz WH American Journal of Clinical Nutrition 2002 75 6 978 985 12036802 Use of weight-for-height indices in children to predict adult overweight: the Bogalusa Heart Study Valdez R Greenlund KJ Wattigney WA Bao W Berenson GS International Journal of Obesity & Related Metabolic Disorders: Journal of the International Association for the Study of Obesity 1996 20 8 715 721 Birthweight for length: ponderal index, body mass index or Benn index? Cole TJ Henson GL Tremble JM Colley NV Annals of Human Biology 1997 24 4 289 298 10.1080/03014469700005032 9239434 Weight-for-height in children aged 4-12 years. A new index compared to the normalised body mass index. Chinn S Rona R Gulliford MC Hammond J European Journal of Clinical Nutrition 1992 46 489 500 1623853 Defining childhood obesity: the relative body mass index (BMI) Poskitt EM Acta Paediatrica 1995 84 961 963 7488831 Three limitations of the body mass index Garn SM Leonard WR Hawthorne VM American Journal of Clinical Nutrition 1986 44 996 997 3788846 Childhood obesity in Europe: a growing concern Livingstone MB Public Health Nutrition 2001 4 1A 109 116 11255500 Epidemiology of childhood obesity - Methodological aspects and guidelines: What is new? Wang Y International Journal of Obesity 2004 28 SUPPL. 3 S21 S28 10.1038/sj.ijo.0802801 15543215 Trends in waist circumferences in young British children: a comparative study McCarthy HD Jarrett KV Emmett PM Rogers I International Journal of Obesity 2005 29 2 157 162 10.1038/sj.ijo.0802849 15570313 Assessing body composition among 3- to 8-year old children: anthropometry, BIA and BXA Eisenmann JC Heelan KA Welk GJ Obesity Research 2004 12 10 1633 1640 15536227 The utility of body mass index as a measure of body fatness in children and adolescents: differences by race and gender Daniels SR Khoury PR Morrison JA Pediatrics 1997 99 6 804 807 10.1542/peds.99.6.804 9164773 The contribution of fat and fat-free tissue to body mass index in contemporary children and the reference child Wells J Coward WA Cole TJ Davies P International Journal of Obesity 2002 26 1323 1328 10.1038/sj.ijo.0802077 12355328 Physical status: the use and interpretation of anthropometry WHO Geneva, World Health Organisation, Technical Report Series, 854 1995 Obesity: preventing and managing the global epidemic. Report of a WHO consultation on obesity. WHO Geneva, World Health Organisation 1998 Physical growth: National Center for Health Statistics percentiles Hamill PV Drizd TA Johnson CL Reed RB Roche AF Moore WM American Journal of Clinical Nutrition 1979 32 607 629 420153 Reference data for obesity: 85th and 95th percentiles of body mass index (wt/ht2) and triceps skinfold thickness Must A Dallal GE Dietz WH American Journal of Clinical Nutrition 1991 53 839 846 2008861 Guidelines for overweight in adolescent preventive services: recommendations from an expert committee Himes JH Dietz WH American Journal of Clinical Nutrition 1994 59 307 316 8310979 CDC Growth Charts: United States Kuczmarski RJ Ogden CL Grummer-Strawn LM Flegal KM Guo SS Wei R Mei Z Curtin LR Roche AF Johnson CL Advance data from vital and health statistics Hyattsville, Maryland, National Center for Health Statistics 2000 314 Body mass index reference curves for the UK, 1990 Cole TJ Freeman JV Preece MA Archives of Disease in Childhood 1995 73 1 25 29 7639544 Body mass index variations: centiles from birth to 87 years Rolland-Cachera MF Cole TJ Sempe M Tichet J Rossignol C Charraud A European Journal of Clinical Nutrition 1991 45 13 21 1855495 Perzentile fur den Body Mass Index fur das Kindesund Jugendalter unter Heranziehung verschiedener deutscher Stichproben Kromeyer-Hauschild K Wabitsch T Geller F Ziegler PJ Geiss LS Hesse V Hippel A Jaeger U Johnsen D Keiss W Korte W Kunze D Menner K Muller M Niemann-Pilatus A Remer T Schaefer F Wittchen HU Zabransky S Zellner K Hedebrand J Monatszeitschrift Kinderheilkunde 2001 149 807 818 10.1007/s001120170107 Body mass index of 0 to 45-y-old Danes: reference values and comparison with published European reference values Nysom K Molgaard C Hutchings B Michaelsen KF International Journal of Obesity 2001 25 2 177 184 10.1038/sj.ijo.0801515 11410817 Workshop on childhood obesity: summary of the discussion Bellizzi MC Dietz WH American Journal of Clinical Nutrition 1999 70 173S 175S Establishing a standard definition for child overweight and obesity worldwide: international survey Cole TJ Bellizzi MC Flegal KM Dietz WH British Medical Journal 2000 320 7244 1240 1243 27365 10797032 10.1136/bmj.320.7244.1240 Definitions of childhood obesity: current practice Chinn S European Journal of Clinical Nutrition 2006 10 1189 1194 10.1038/sj.ejcn.1602436 Prevalence of overweight in US children: comparison of US growth charts from the Centers for Disease Control and Prevention with other reference values for body mass index Flegal KM Ogden CL Wei R Kuczmarski RL Johnson CL American Journal of Clinical Nutrition 2001 73 6 1086 1093 11382664 International definitions of overweight and obesity for children: a lasting solution? Chinn S Rona RJ Annals of Human Biology 2002 29 3 306 313 10.1080/03014460110085340 12031139 Body composition in adolescents: measurements and metabolic aspects Rodriguez G Moreno LA Blay MG Blay VA Garagorri JM Sarria A Bueno M International Journal of Obesity & Related Metabolic Disorders: Journal of the International Association for the Study of Obesity 2004 28 Suppl 3 S54 8 10.1038/sj.ijo.0802805 Validity of BMI as a measure of obesity in Australian white Caucasian and Australian Sri Lankan children Wickramasinghe VP Cleghorn GJ Edmiston KA Murphy AJ Abbott RA Davies PS Annals of Human Biology 2005 32 1 60 71 10.1080/03014460400027805 15788355 Detection of overweight and obesity in a national sample of 6-12-y-old Swiss children: accuracy and validity of reference values for body mass index from the US Centers for Disease Control and Prevention and the International Obesity Task Force. Zimmermann MB Gubeli C Puntener C Molinari L American Journal of Clinical Nutrition 2004 79 5 838 843 15113723 Comparison of weight-for-height indices as a measure of adiposity and cardiovascular risk from childhood to young adulthood: The Bogalusa heart study Frontini MG Bao WH Elkasabany A Srinivasan SR Berenson G Journal of Clinical Epidemiology 2001 54 8 817 822 10.1016/S0895-4356(01)00343-2 11470391 Body mass index and percentage fat mass in healthy German schoolchildren and adolescents Schaefer F Georgi M Wuhl E Scharer K International Journal of Obesity & Related Metabolic Disorders: Journal of the International Association for the Study of Obesity 1998 22 5 461 469 10.1038/sj.ijo.0800608 Identification of the obese child: adequacy of the body mass index for clinical practice and epidemiology Reilly JJ Dorosty AR Emmett PM team ALSPAC International Journal of Obesity & Related Metabolic Disorders: Journal of the International Association for the Study of Obesity 2000 24 12 1623 1627 10.1038/sj.ijo.0801436 Use of BMI as a measure of overweight and obesity in a field study on 5-7 year old children Mast M Langnase K Labitzke K Bruse U Preuss U Muller MJ European Journal of Nutrition 2002 41 2 61 67 10.1007/s003940200009 12083315 The worldwide obesity epidemic James PT Leach R Kalamara E Shayeghi M Obesity Research 2001 9 Suppl 4 228S 233S 11707546 Body composition from fluid spaces and density: analysis of methods Siri WE Techniques for Measuring Body Composition Washington, DC, National Academy of Sciences / National Research Council Brozek J, Henschel A 1961 223 234 Skinfolds and body density and their relation to body fatness: a review Lohman T Human Biology 1981 53 2 181 225 A new air displacement method for the determination of human body composition Dempster P Aitkens S Medicine and Science in Sports and Exercise 1995 27 1692 1697 Body composition assessment using underwater weighing techniques Francis KT Physical Therapy 1990 70 10 657 662 2217545 Body composition measurement: a review of hydrodensitometry, anthropometry, and impedance methods Brodie MJ Moscrip V Hutcheson R Nutrition 1998 14 3 296 310 10.1016/S0899-9007(97)00474-7 9583375 Body-composition assessment via air-displacement plethysmography in adults and children: a review Fields DA Goran MI McCrory MA American Journal of Clinical Nutrition 2002 75 453 467 11864850 Visceral fat in white and African American prepubertal children Goran MI Nagy TR Treuth MS Trowbridge C Dezenberg C McGloin A Gower BA American Journal of Clinical Nutrition 1997 65 6 1703 1708 9174463 Association of body fat distribution and cardiovascular risk factors in children and adolescents Daniels SR Morrison JA Sprecher DL Khoury P Kimball TR Circulation 1999 99 4 541 545 9927401 Foot-to-foot bioelectrical impedance analysis: a valuable tool for the measurement of body composition in children Tyrrell VJ Richards GE Hofman P Gillies GF Robinson E Cutfield WS International Journal of Obesity 2001 25 273 278 10.1038/sj.ijo.0801531 11410831 Utility of different measures of body fat distribution in children and adolescents Daniels SR Khoury PR Morrison JA American Journal of Epidemiology 2000 152 12 1179 1184 10.1093/aje/152.12.1179 11130624 Skinfold equations for estimation of body fatness in children and youth Slaughter MH Lohman TG Boileau RA Horswill CA Stillman RJ Van Loan MD Bemben DA Human Biology 1988 60 5 709 723 Predicting body composition from anthropometry in pre-adolescent children Dezenberg CV Nagy TR Gower BA Johnson R Goran MI International Journal of Obesity 1999 23 253 259 10.1038/sj.ijo.0800802 10193870 Measurement and long-term health risks of child and adolescent fatness Power C Laker S Cole TJ International Journal of Obesity 1997 21 507 526 10.1038/sj.ijo.0800454 9226480 Comparisons of waist circumferences measured at 4 sites Wang J Thornton JC Bari S Williamson B Gallagher D Heymsfield SB Horlick M Kotler D Laferrere B Mayer L Pi-Sunyer FX Pierson Jr RN American Journal of Clinical Nutrition 2003 77 2 379 384 12540397 The development of waist circumference percentiles in British children aged 5.0-16.9y McCarthy AM Jarrett KV Crawley HF European Journal of Clinical Nutrition 2001 2001 902 907 10.1038/sj.ejcn.1601240 Waist circumference percentiles in nationally representative samples of African-American, European-American, and Mexican-American children and adolescents Fernandez JR Redden DT Pietrobelli A Allison DB Journal of Pediatrics 2004 145 427 430 10.1016/j.jpeds.2004.07.021 15480357 Body Mass Index, Waist Circumference, and Clustering of Cardiovascular Disease Risk Factors in a Biracial Sample of Children and Adolescents Katzmarzyk P Srinivasan S Chen W Malina RM Bouchard C Berenson G Pediatrics 2004 114 e198 e205 10.1542/peds.114.2.e198 15286257 Waist circumference percentiles for 7-to 15-year-old Australian children Eisenmann JC Acta Paediatrica 2005 94 9 1182 1185 10.1080/08035250510029352 16203670 Six reasons why the waist-to-height ratio is a rapid and effective global indicator for health risks of obesity and how its use could simplify the international public health message on obesity Ashwell M Hsieh SD International Journal of Food Sciences & Nutrition 2005 56 5 303 307 10.1080/09637480500195066 Indices of body fat distribution in Spanish children aged 4.0 to 14.9 years Moreno LA Fleta J Mur L Feja C Sarria A Bueno M Journal of Pediatric Gastroenterology & Nutrition 1997 25 2 175 181 10.1097/00005176-199708000-00008