| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Supplements |
1 From The Rowett Research Institute, Aberdeen, Scotland, and the Division of Nutrition and Physical Activity, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONAN INDEX TO MEASURE...DEVELOPING A REFERENCE...DEFINING CUTOFF POINTS FOR...USE OF ANCILLARY MEASURES...SUMMARY CONCLUSIONS FROM THE...REFERENCES |
|---|
Key Words: Children • adolescents • obesity • international • standards • BMI • body mass index • skinfold thickness
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONAN INDEX TO MEASURE...DEVELOPING A REFERENCE...DEFINING CUTOFF POINTS FOR...USE OF ANCILLARY MEASURES...SUMMARY CONCLUSIONS FROM THE...REFERENCES |
|---|
|
AN INDEX TO MEASURE OVERWEIGHT |
|---|
|
TOPABSTRACTINTRODUCTIONAN INDEX TO MEASURE...DEVELOPING A REFERENCE...DEFINING CUTOFF POINTS FOR...USE OF ANCILLARY MEASURES...SUMMARY CONCLUSIONS FROM THE...REFERENCES |
|---|
BMI
The measurements used to derive BMI as an assessment of adiposity in children and adolescents are reliable and nonintrusive; furthermore, BMI has been validated against measures of body density. Many countries have published BMI-for-age charts for their populations, and some have also defined cutoff points on these charts to define overweight and obesity. However, the cutoff points vary from country to country and the rationale for the choice of cutoffs is rarely provided. As a result, different BMI-for-age values define overweight and obesity in different populations.
Participants in the workshop debated the validity of BMI as a measure of adiposity in children and adolescents. An ideal index will predict early morbidity or mortality from chronic diseases. However, because obesity in children is rarely associated with morbidity or mortality, the likelihood that obesity will persist into adulthood may have to be used as such an indicator. For example, the risk of developing adult obesity (BMI >28) in children aged >9 y who are obese [defined as BMI above the 95th percentile of weight in the second National Health and Nutrition Examination Survey (NHANES II)] is up to 80% at age 35 y (1).
Limitations of BMI
Although weight was considered one indicator of fatness, children of the same weight but at different stages of height can have widely different levels of adiposity. BMI, however, adjusts for height by using the square power. Although the adjustment for height does not completely eliminate the stature effect, it partially adjusts weight for stature. One effect of this correlation between BMI and height is that taller populations will appear to have a higher prevalence of obesity (2). Use of BMI in a clinical setting requires additional measures to confirm that a high BMI reflects excess body fat and not height.
Other weight-to-height indexes
Other options for a weight-to-height index include a series of weight-to-height ratios such as the ponderal index (weight:height3) or the Benn Index (weight:heightp). These ratios were discussed in detail by Franklin (2). If height is restricted to whole-power indexes, height squared is the best compromise because it is already used widely and has been validated against measures of body density. If a fractional index is to be used, then researchers must decide whether such an index would be acceptable on a global basis.
Ethnic differences in BMI
One study in whites and Asians showed differences of 2–3 BMI units in adults with the same body fat composition (3). In another study of BMI and body fat in young adults (4), differences were appreciable by ethnic and racial groups. Because the methods were standardized the differences were considered biological. However, there are still no cross-cultural data available in children that compare the validity of BMI.
Use of BMI for adolescents
In boys, skinfold thicknesses (SFTs) vary during the adolescent growth spurt. Extreme SFT values generally begin to decline 
6–12 mo before peak height velocity occurs, suggesting that fat is redistributed during growth. One study found that obese adolescents, regardless of how they were classified as obese, tended to be more advanced in skeletal maturity than nonobese adolescents (5). Physical maturity, which may vary by ethnicity, may confound analysis of BMI in adolescents of different cultures. However, the overall efficiency of BMI in predicting percentage body fat as measured by triceps SFT or densitometry in adolescents of different ethnic origin was >0.8 (5), suggesting that BMI could be useful in international epidemiologic studies.
Skinfold thickness and circumference measures
The workshop participants debated whether SFT measurements provide an advantage over BMI or whether the difficulty in measuring SFTs outweighs the additional information it provides. SFTs or circumferences may indicate fat distribution and may help capture ethnic and sex differences in fat distribution. In addition, SFTs may help identify patients who are overfat and not merely overweight. However, the validity of SFT measurements in different populations has not been carefully explored. Furthermore, no evidence suggests that SFT measurements predict hyperinsulinemia, hypertension, or other illnesses in children better than does BMI. Results from several data sets suggest that childhood BMI remains stable into adulthood better then does SFT. This observation may reflect measurement error in SFT or changes in fat distribution.
The group noted that circumferences (eg, waist or hip) may reflect morbidity in adults, but the relation between visceral fat and morbidity in children and adolescents has not yet been clarified.
|
DEVELOPING A REFERENCE POPULATION |
|---|
|
TOPABSTRACTINTRODUCTIONAN INDEX TO MEASURE...DEVELOPING A REFERENCE...DEFINING CUTOFF POINTS FOR...USE OF ANCILLARY MEASURES...SUMMARY CONCLUSIONS FROM THE...REFERENCES |
|---|
Reasonable criteria for combining data are: 1) the sample is representative of the country, 2) the data collection process included measures of reliability and validity and data were collected within a specified period, 3) the origin of the sample is specified, 4) the population is adequately described, and 5) the sample is large (n 
10000). Note that the representativeness of data may be less important than other considerations because many countries in Africa and Asia lack nationally representative samples.
The group concluded that data conforming to the criteria would be used to produce growth charts for the different populations. Differences between them would be explored. Of particular interest is how superimposable those data sets are, how they differ at the extremes, and whether population-specific differences exist in the use of BMI.
|
DEFINING CUTOFF POINTS FOR OVERWEIGHT AND OBESITY |
|---|
|
TOPABSTRACTINTRODUCTIONAN INDEX TO MEASURE...DEVELOPING A REFERENCE...DEFINING CUTOFF POINTS FOR...USE OF ANCILLARY MEASURES...SUMMARY CONCLUSIONS FROM THE...REFERENCES |
|---|
A less satisfactory approach is to define obesity statistically by a cutoff such as a percentile-for-age relative to a reference population. For example, the United States has adopted the 85th and 95th percentiles of NHANES II to define overweight and obesity in children (6). Similarly, the 97.7th percentile of a reference population of British children is sometimes used in the United Kingdom to define obesity (7). A consequence of this approach is that the prevalence of overweight and obesity is set to a fixed proportion for all ages of both sexes. Thus, 15% of the NHANES II population is overweight (>85th percentile) and 5% is obese (>95th percentile). Because populations may change with time, a further problem arises if new standards are based on current data because use of current data to establish standards eliminates the capacity to show secular trends. Therefore, percentiles should be based on a reference population that does not change with time.
A third, prescriptive, approach defines a healthy population and then defines normal growth by the distribution of that population. This approach has been adopted by a World Health Organization working group in its definition of a reference population aged 0–5 y (8). Breast-fed children were considered the physiologic prescriptive population because the pattern of their growth seems to be the same now as it was 100 y ago.
The problem with the prescriptive approach is that characteristics of the reference population may change with time. In addition, long-term follow-up studies that aid in defining healthy BMIs for children have not been performed. Nevertheless, the prescriptive approach was not considered to be unworkable but to require further discussion.
Setting cutoff points
Children tend to retain the same BMI percentile ranking as they become older. Therefore, another alternative would be to use the percentiles in late adolescence identified by the adult BMI values used to establish overweight and obesity (25 and 30, respectively). A BMI of 25 corresponds to about the 80th percentile of the National Center for Health Statistics (NCHS) standards at age 18 y in both males and females. Therefore, the 80th percentile would be used to define overweight in children and adolescents of all ages. Interestingly, a BMI of 30 corresponds with the 95th percentile of the NCHS standards for 18-y-old males. Therefore, the 95th percentile could be used throughout childhood and adolescence to identify obesity. This method makes the definition of overweight and obesity for children and adolescents consistent with that for adults.
The development of a reference population is critical to this approach. However, once a reference population is defined, the percentile values at different ages corresponding to BMIs of 25 and 30 at age 20 y would serve as the absolute reference cutoff points throughout childhood and adolescence. For example, if a BMI of 30 identifies the 95th percentile for a 20-y-old, the 95th percentile would be used to identify obesity in adolescents and children. This approach could be tested on available but pooled data sets.
A concern was raised about the application and use of the cutoff points in parts of the world where undernutrition exists. A potential risk in such areas is that interventions designed for overweight persons may have adverse effects on underweight persons.
|
USE OF ANCILLARY MEASURES TO VALIDATE THE INDEX |
|---|
|
TOPABSTRACTINTRODUCTIONAN INDEX TO MEASURE...DEVELOPING A REFERENCE...DEFINING CUTOFF POINTS FOR...USE OF ANCILLARY MEASURES...SUMMARY CONCLUSIONS FROM THE...REFERENCES |
|---|
|
SUMMARY CONCLUSIONS FROM THE WORKSHOP |
|---|
|
TOPABSTRACTINTRODUCTIONAN INDEX TO MEASURE...DEVELOPING A REFERENCE...DEFINING CUTOFF POINTS FOR...USE OF ANCILLARY MEASURES...SUMMARY CONCLUSIONS FROM THE...REFERENCES |
|---|
The participants agreed that a reference population widely representative of the world's population should be established. Data from several countries should be pooled to create a provisional reference population. These countries include the United Kingdom, Singapore, the United States, Japan, France, Hong Kong, and Netherlands. The group recommended that before the data sets are pooled and additional cohorts are included, the sampling system must be specified, the data must be broadly representative of the population, and quality assurance practices must be applied to the data-checking process.
The participants also emphasized that the process of developing a reference population would have to be considered provisional because the data that now meet these criteria represent countries comprising <5% of the world population. However, the group recommended that the World Health Organization address this issue and initiate activities to establish a reference population.
A novel approach to setting the childhood percentile for overweight and obesity based on adult morbidity cutoff points was proposed. Other analyses using data sets that have information on glucose and insulin concentrations and responses to glucose tolerance tests should be used to validate these percentiles.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONAN INDEX TO MEASURE...DEVELOPING A REFERENCE...DEFINING CUTOFF POINTS FOR...USE OF ANCILLARY MEASURES...SUMMARY CONCLUSIONS FROM THE...REFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. W. Demerath, C. M. Schubert, L. M. Maynard, S. S. Sun, W. C. Chumlea, A. Pickoff, S. A. Czerwinski, B. Towne, and R. M. Siervogel Do Changes in Body Mass Index Percentile Reflect Changes in Body Composition in Children? Data From the Fels Longitudinal Study Pediatrics, March 1, 2006; 117(3): e487 - e495. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Fonseca and M. Gaspar de Matos Perception of overweight and obesity among Portuguese adolescents: an overview of associated factors Eur J Public Health, June 1, 2005; 15(3): 323 - 328. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Mandel, E. Zimlichman, F. B. Mimouni, I. Grotto, and Y. Kreiss Height-Related Changes in Body Mass Index: A Reappraisal J. Am. Coll. Nutr., February 1, 2004; 23(1): 51 - 54. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Lissau, M. D. Overpeck, W. J. Ruan, P. Due, B. E. Holstein, and M. L. Hediger Body Mass Index and Overweight in Adolescents in 13 European Countries, Israel, and the United States Arch Pediatr Adolesc Med, January 1, 2004; 158(1): 27 - 33. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Engeland, T. Bjorge, A. J. Sogaard, and A. Tverdal Body Mass Index in Adolescence in Relation to Total Mortality: 32-Year Follow-up of 227,000 Norwegian Boys and Girls Am. J. Epidemiol., March 15, 2003; 157(6): 517 - 523. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. W Taylor, I. E Jones, S. M Williams, and A. Goulding Body fat percentages measured by dual-energy X-ray absorptiometry corresponding to recently recommended body mass index cutoffs for overweight and obesity in children and adolescents aged 3-18 y Am. J. Clinical Nutrition, December 1, 2002; 76(6): 1416 - 1421. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Stoddart, K. S. Blevins, E. T. Lee, W. Wang, and P. R. Blackett Association of Acanthosis Nigricans With Hyperinsulinemia Compared With Other Selected Risk Factors for Type 2 Diabetes in Cherokee Indians: The Cherokee Diabetes Study Diabetes Care, June 1, 2002; 25(6): 1009 - 1014. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Freemark and D. Bursey The Effects of Metformin on Body Mass Index and Glucose Tolerance in Obese Adolescents With Fasting Hyperinsulinemia and a Family History of Type 2 Diabetes Pediatrics, April 1, 2001; 107(4): e55 - e55. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Maynard, W. Wisemandle, A. F. Roche, Wm. C. Chumlea, S. S. Guo, and R. M. Siervogel Childhood Body Composition in Relation to Body Mass Index Pediatrics, February 1, 2001; 107(2): 344 - 350. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
