Skip to main content

Archived Comments for: The problem of obesity among adolescents in Hong Kong: a comparison using various diagnostic criteria

Back to article

  1. Weight for height and BMI references for overweight and obesity in adolescents

    Sophie Leung, Affiliate, Centre of Nutritional Studies, Chinese University of Hong Kong , Dr. KH Mak, Student Health Service, Hong Kong

    6 August 2008

    Ko et al’s paper on obesity in adolescents [1] reported that measuring the prevalence of obesity in their studied adolescents using the Hong Kong growth standard charts was markedly different from those using the International IOTF and the Chinese reference COTF. The authors raised the concern that the difference might be due to the out-datedness of the Hong Kong reference. We would like to point out a few pitfalls in the way these various criteria were adopted that might have misled the discussion.

    In the section on methodology when referring to IOTF, the authors said ‘Overweight and obesity were defined as BMI for age > 25 and > 30…’ The fact should be: ‘Overweight and obesity in children aged 2 to 18 years were defined as the percentiles that correspond to those that pass through BMI 25 and 30 kg/m2 respectively at age 18 years.’ Similarly for COTF, they said: ‘Overweight and obesity were defined as BMI for age >24 and 28...’ The correct way to say should have been: ‘For children 7 - 18 years, overweight and obesity were defined as BMI that correspond to BMI 24 and BMI 28 kg/m2 respectively at age 18 years and these were close to P85 and P90 [2].’ It is essential to recognize that, childhood obesity, unlike adult obesity is not defined by a simple number in BMI but by a variable BMI according to age simply because weight and height of children normally increase with age. This may just be a matter of clarity in the text since similar errors were not found in the tables. Anyhow, it is not clear from either the text or table whether the authors had included obesity in the calculation of the prevalence of overweight. This is important when it comes to the comparison with other references.

    Years of growth data collection should not be confused with years of publication. The authors have correctly pointed out that the China study COTF was performed in 2000 and published in 2004, and the CDC data on adolescents collected between 1963 and 1994. However, the authors wrongly stated that the international IOTF data were based on an international growth study in 2000. The fact should be that the IOTF was published in 2000. Growth data for IOTF were retrieved from six previous studies performed between 1967 and 1993, one of which is from Hong Kong: this is the same set of data that the authors referred to as HKGS 1993 which also provides the weight for height reference. Why should IOTF accept data collected more than two or three decades ago and propose it to be used internationally? The reason is simple and clear. Theoretically, the best time to set the standard is the time when the problem of obesity has just started to emerge. In practice, this is not always possible. Therefore, the time nearest to the best available was generally accepted and this time varies with different populations. Among the six set of growth data in IOTF, the HKGS is one of the most recent ones and it is close to the time when childhood obesity was noted to appear. And it is also worth to note that the HKGS BMI reference is not that different from the IOTF.

    The authors attempted to compare the prevalence of obesity of the studied adolescents using the HKGS weight for height reference and the other international BMI references. Since weight for height and BMI are two different parameters, a more suitable way of doing so should be to convert the BMI equivalent of the 120 % median weight for height before comparison. That is, from the median height (ht) for each age from 11 to 18 years find the corresponding median weight (wt) [3] and then multiply by 120% to get the cutoff. The BMI equivalents (wt/ht2) would be found to lie between 19.9-23.1 kg/m2 which is close to the P85 of the HKGS BMI reference [4] and not so much different from the BMI cutoff for overweight using the IOTF or COTF.

    The way to define obesity should be seen from a historical perspective. Life span in adults is the main concern for insurance companies. For a long time, it was noticed that adult adiposity is related to life span. Therefore life insurance companies had used BMI values of adults to calculate the premium. Subsequently more and more research has documented various health risks related to specific BMI values which affected medical insurance. As a result, different degrees of obesity were therefore described: overweight, obesity and super-obesity. Nobody has ever used weight for height to describe obesity in adulthood! On the other hand, the use of weight for height as a reference to describe nutritional status in infancy and childhood has a long historical background. In the case of children, under-nutrition of children has always been a major concern, particularly in the developing countries. As early as 1966, WHO [5] [6] recommended the following cutoffs to identify children who are at nutritional risk:

    Underweight means weight for age < -2SD of mean or <80 % of median

    Wasting means weight for height < -2SD of mean or <80 % of median

    Stunting means height < -2SD of mean or < 90 % of median

    Overweight means weight for age > 2SD of mean or >120 % of median

    Obesity means weight for height > 2SD of mean or >120 % of median

    That means for a long time, ‘overweight’ has been a term of weight for age and ‘obesity’ a term of weight for height in children. At that time, the NCHS reference was proposed to be used as the international reference, with the understanding that the weight for height reference was available only for those below 10 years of age. And, there was no grading of childhood obesity in this weight for height definition.

    On the other hand, in some western countries, particularly USA, the problem of over-nutrition in childhood and adolescence had become a serious issue in the sixties. This problem was measured by BMI and skin-fold thicknesses apart from weight for height. With the rapid increase in the problem it was then felt necessary to divide it to two categories, with the less severe one named overweight matching to that used for adults. It is therefore important to understand these terms- obesity or overweight in their context.

    Hong Kong has changed rapidly from a developing to a developed economic status over the last few decades. This change has affected the nutritional concern for the local infants and children. In the sixties, most children belonged to the first generation of the mainland Chinese immigrants and they were reported to be undernourished [7]. In the eighties, there was confusion over whether the Child Health Care workers were faced with under- or over- nutrition in children. Many Chinese parents continued to consider their children to be thin and not eating enough. Yet weight and height of many children were noted by clinicians to be far above the local growth reference which was constructed in 1963 [8]. An in- depth scientific study was therefore performed between 1984 and 1990 to look into children’s diet and health. It was then confirmed that infants and children were already overfed with animal protein and fat. Childhood obesity (using the NCHS weight for height reference) and hyperlipidemia had emerged [9][10][11]. With this background, a territory-wide growth survey was performed in 1993, with data on various parameters including weight, length/height, sitting height, weight for height, BMI and skin-fold thicknesses for ages from birth to 18 years and puberty staging for children older than six years. Weight was very much skewed to the heavy side. Many children diagnosed as ‘overweight’ (120 % median weight for age) might have received unnecessary medical intervention simply because they were also tall for their age. Therefore weight for height would be a better reflection of the need of intervention. So a cutoff using 120 % of the median was published in 1996 to identify obesity [12]. For those below 10 years, the Hong Kong reference fell closely with that of the NCHS reference. Obese children identified clinically were found to match very well to these objective tools. In the last decade, a lot of education and intervention for these obese children were performed by hospitals, schools and related professionals. In particular the Hong Kong Student Health Service has taken up the responsibility of monitoring and managing this newly emerged disorder among school children. The prevalence of obesity in primary and secondary schools were monitored using 120 % median weight for height. In fact, if 120 % median BMI reference was used instead, the prevalence of obesity identified would be similar. However this latter method of defining obesity was not popular.

    The Hong Kong BMI reference was published at about the same time as the weight for height reference. At the time, there was no consensus as to the percentile cutoff. If P85 were used as a cutoff for over-nutrition, then by definition all the top 15 % of all ages in the year of data collection would have fallen into this category. This is absurd because clinically, not that many infants and preschool children were abnormal and the prevalence of obesity should not be the same through all age groups. However if it was just used as a reference for the future years to come then using percentile as a cutoff can be acceptable.

    With the incorporation of the Hong Kong BMI references into the construction of the IOTF in 2000, a cutoff was set with an online program made available for easy calculation. We have to accept that the use of BMI has gained more popularity over weight for height for the school aged group. Compared to using weight for height, using BMI is much easier to process any calculation or comparison between groups. Early this year, the Student Health Service has taken the lead in Hong Kong to match along with the more recent and more popular meaning of overweight and renamed the definition of >120 % median weight for height from ‘obesity’ to ‘overweight including obesity’ in childhood. In order to further divide it into two categories, the Hong Kong growth standard (HKGS) BMI reference was adopted: P85 as cutoff for overweight including obesity and P95 for obesity alone. It was found that the prevalence of overweight including obesity in the primary and secondary school children assessed in this way was similar to that using 120 % weight for height.

    We believe that Ko and his workers are concerned not just with those adolescents whom they considered obese but with overweight adolescents as they have correctly pointed out that the health risk is in overweight and not just obesity. It should be appreciated that in general terms, obesity covers overweight. That’s why we say childhood obesity and not childhood overweight. Had the authors put the HKGS 120 % weight for height cutoff against the overweight and not obesity BMI references, there would not be much difference. Minor differences are inevitable because they were derived from different reference populations measured at different time.

    In conclusion, it should be appreciated that none of the available cutoffs and references for over-nutrition/obesity is perfect. But for epidemiological reasons there is a real need for definition. It is very important to use the same reference in order to track the progression of the problem and the effectiveness of the intervention program. Growth references or growth standards are just tools to measure and to improve health in the population. For the first time, the Government of the Special Administrative Region of Hong Kong committed in the 2007-08 Policy Agenda to develop a comprehensive strategy to prevent and control non-communicable diseases (NCD) in Hong Kong. Priority is accorded to the prevention of the major NCD and related risk factors, one of which is obesity. We believe that it does not exclude those who are ‘overweight.’ We hope for a concerted effort by all different sectors of the society and the world to effectively bring the problem of childhood and adolescent obesity (in the broad sense) under control.

    Sincerely,

    Dr. Sophie SF Leung

    Affiliate, Centre of Nutritional Studies

    Chinese University of Hong Kong

    Dr. KH Mak

    Student Health Service

    Department of Health

    Hong Kong Special Administrative Region

    The People’s Republic of China

    References

    1. Ko GTC, Ozaki R, Wong GWK, Kong APS, So WY, Tong PCY, Chan MHM, Ho CS, Lam CWK, Chan JCN: The problem of obesity among adolescents in Hong Kong: a comparison using various diagnostic criteria. BMC Pediatrics 2008, 8:10

    2. JI CY: Report on Childhood obesity in China (1) Body mass index reference for screening overweight and obesity in Chinese school –age children. Biomedical and Environmental Sciences 2005, 18: 390-400.

    3. Leung SSF, Tse LY, Lau JTF and Oppenheimer SJ. Weight-for-age and weight-for-height references for Hong Kong children from birth to 18 years. J. Paediatr. Child Health, 1996, 32:103-109.

    4. Leung SSF, Cole TJ, Tse LY and Lau JTF. Body mass index reference curves for Chinese children. Annals of Human Biology, 1998, 25(2):169-174.

    5. Jelliffe, DB, The assessment of the nutritional status of the community. WHO Series No. 503, WHO, Geneva 1966.

    6. Jelliffe, DB and Jelliffe EEP, Advance in international maternal and child health. Clarendon Press.1986, 6: 66-78,

    7. Field CE, Baber FM. Growing up in Hong Kong, Hong Kong University Press, 1973.

    8. Chang KSF, Lee MMC and Low WD. Standards of the height and weight of Southern Chinese children. Far East Medical Journal, 1965,1:101-109

    9. Leung SSF, Lui S. Nutritive value of Hong Kong Chinese weaning diet. Nutrition Research, 1990, 10: 707-715.

    10. Leung SSF, Davies DP. Infant feeding and growth of Chinese infants: birth to 2 years. Paediatric and Perinatal Epidemiology, 1994, 8: 301-313.

    11. Leung SSF, Ng MY, Tan BY, Lam CWK, Wang SF, Xu YC and Tsang WP. Serum cholesterol and dietary fat of two populations of southern Chinese. Asia Pacific Journal Clinical Nutrition, 1994: 3: 127-30.

    12. Leung SSF, Tse LY, Leung NK. Growth and Nutrition in Hong Kong Children. Singapore Paediatric Journal, 1996, 38:61-66.

    Competing interests

    Dr. Leung is one of the authors of the HKGS 1993

Advertisement