Time to Step Up to the Plate: Adopting the WHO 2006 Growth Curves for US Infants
Article Outline
Abbreviations: CDC, Centers for Disease Control and Prevention, WHO, World Health Organization
The Centers for Disease Control and Prevention's (CDC) 2000 growth curves for infants, children, and adolescents are used by the vast majority of pediatric health care providers in the United States and in many other parts of the world as well.1 These charts, a growth reference, describe how children and adolescents in the United States actually grow across a wide range of social, ethnic, and economic conditions. But since the publication of the World Health Organization's (WHO) 2006 growth standard curves, which describe how infants should grow under ideal conditions not subject to economic restraints,2 there has been much discussion about adopting the WHO's curves for US infants and children.3, 4 At a meeting held June 28-29, 2006 at the National Center for Health Statistics in Hyattsville, Maryland, representatives from the CDC, National Institutes of Health, and American Academy of Pediatrics discussed how the WHO curves might be used by clinicians in the United States, and how these curves interface with the CDC 2000 growth curves for infants, children, and adolescents. The article by Mei et al in this issue of The Journal summarizes the data presented at that meeting and compares the impact of using the CDC 2000 curves or the WHO 2006 curves on the prevalence of shortness, underweight, and overweight in US children age 0 to 59 months.5
See related article, p 622
The WHO 2006 standard curves include charts from birth to age 2 years (longitudinal data) as well as charts for age 2 to 5 years (cross-sectional data). Details on the populations of children used for these curves from 6 different countries are available elsewhere.6 The WHO curves for birth to 2 years are based on 903 infants who where exclusively/predominantly breast-fed for 4 to 6 months and who continued breast-feeding for at least 12 months. Each infant was weighed and measured 21 times in 24 months. The median duration of any breast-feeding was 17.8 months, and complementary foods were introduced at a mean age of 5.1 months. In contrast, the CDC 2000 growth charts represent a reference of growth for the general US population between 1970 and the early 1990s, excluding very low birth weight infants (<1500 g). There are separate curves for age 0 to 36 months and for age 2 years to 20 years. For the CDC 2000 curves, data from more than 82 million US birth certificates and various cross-sectional databases, which included a total of 4697 infants age 2 to 24 months, were used. But the data points were sparse in the first few months of life; in fact, there were no data between birth and age 2 months. Thus, the curves for birth to age 2 months represent only the mathematical function used to smooth the data for the entire infant growth chart. Unlike the WHO curves, the CDC curves use data from infants that approximated the mix of feedings that infants received in the 1970s and 1980s. During this period, 1/3 of US infants were breast-fed up to age 3 months, and the other 2/3 were predominantly formula-fed.
The growth data from both the CDC 2000 and WHO 2006 curves for age 2 to 5 years are cross-sectional. Both take into account the approximately 1-cm decrease when examiners switch from length measurements to height measurements at age 2 years. The cross-sectional data from the WHO 2006 curves includes 6669 young children who were breast-fed for at least 3 months. They were from the same countries and socioeconomic groups used to create the longitudinal WHO curves for the first 2 years of life.2 The comparable portion of the CDC 2000 curves are based on cross-sectional data from 9894 young children (age <71 months) and represent all socioeconomic and racial groups in the United States.1
The importance of the article by Mei et al is that it is the first direct comparison in a US population of the prevalence of shortness, underweight, and overweight using the CDC 2000 curves and WHO 2006 curves for age 0 to 59 months.5 Overweight is defined by both high weight for height and high body mass index for age (after 24 months), and shortness is defined as low length/height for age. For their direct comparison of the curves, Mei et al use National Health and Nutrition Examination Survey data collected from 3920 US children between 1999 and 2004. It is clear from this article that the differences between the curves for the prevalence of overweight and shortness in these children are minimized if the recommended (but different) CDC and WHO cutpoints for the curves for overweight (>95th percentile for CDC 2000; >97.7th percentile for WHO 2006) and shortness (<5th percentile for CDC 2000; <2.3 percentile for WHO 2006) are used. In fact, for the prevalence of shortness, there are virtually no differences between age 0 and 59 months, and for the prevalence of overweight, there are no real differences by age 12 to 17 months.
What may be of concern to some, however, are the differences in the prevalence of underweight using the CDC 2000 and WHO 2006 curves using the recommended cutpoints of <5th percentile for CDC 2000 and <2.3 percentile WHO 2006. Mei et al describe the prevalence of underweight using 2 different definitions of underweight: low weight for age and low weight for height. The prevalence of underweight is notably lower when using the WHO 2006 curves with a cutpoint of <2.3 percentile. Thus, in Mei et al's Figure 3, using low weight for age as the definition for underweight, the prevalence of underweight at age 12 to 17 months would be approximately 11% when using the CDC 2000 curves (cutpoint, <5th percentile), but only about 2% when using the WHO 2006 curve (cutpoint, <2.3 percentile). These differences are significant, but it is not tenable that 11% of toddlers in the United States are underweight. This large difference probably reflects a problem in the generation of the CDC 2000 curves, which overestimate underweight. As matter of fact, the percentage of infants who were underweight was increased with adoption of the 2000 CDC curves compared with the previous 1977 National Center for Health Statistics growth charts, an observation that has largely gone unnoticed.1 Perhaps an upside of using the WHO 2006 curves, in which fewer children are categorized as underweight, would be a decrease in the number of children referred to pediatric endocrinologists for growth hormone therapy and a decrease in the number of mothers overfeeding their children who have been labeled as “underweight.” Whether all of the children identified as underweight on the CDC 2000 curves but not on the WHO 2006 curves are truly at a healthy weight remains to be determined. If one accepts the WHO curves as a growth standard for infants and children who are predominantly breast-fed early in life and whose diet and environment are optimal, then the likelihood that infants no longer categorized as underweight are actually unhealthy is low. Clearly, more research is needed to evaluate the effects of the WHO 2006 curves if and when they are widely adopted in the United States. In addition, such groups as the American Academy of Pediatrics should provide appropriate instruction for health care providers who will use the WHO 2006 curves, especially regarding the significance of the new cutpoints. For example, would the present definition of obesity (>95th percentile on the CDC 2000 curves) be changed to the 98th percentile (97.7th percentile) if the WHO curves were used for age 0 to 59 months? After 59 months, would the definition of obesity revert to the 95th percentile on the CDC 2000 curves, which would be used after age 5 years?
Pediatricians currently use one CDC 2000 curve to evaluate growth from age 0 to 36 months and a second CDC 2000 curve to evaluate growth from age 2 to 20 years. If pediatricians were to use the 2 WHO 2006 curves for age 0 to 24 months and age 25 to 59 months, then they would still have to use the CDC 2000 curves for age 5 to 20 years. Thus, 3 growth curves would be needed for each patient. One might conclude that for predominantly breast-fed infants, the WHO 2006 curves for age 0 to 24 months, based on longitudinal data, are the best choice, because fewer exclusively breast-fed infants would be categorized as underweight. But what about their use in predominantly formula-fed infants? If one accepts the WHO 2006 curves as a growth standard for all infants, then there would then be less concern for the exclusively formula-fed infant at age 5 months with complementary food introduced by 4 months who is <5th percentile on the CDC 2000 curves but still in the normal range on the WHO 2006 curves.
Beyond age 24 months, the WHO 2006 curves offer little advantage, because they are also based on cross-sectional data, and offer a major disadvantage in that yet a third curve would have to be used after age 59 months, as noted earlier. On a practical level, the WHO 2006 curves are only available using weight in kilograms and height in centimeters, and they have fewer percentile lines (2.3, 15th, 50th, 85th, and 97.7th percentiles) compared with the CDC 2000 curves. This would make the WHO 2006 curves more difficult to use in office practice with US parents and more difficult to compare with the CDC 2000 curves. It is imperative that the WHO 2006 curves for age 0 to 24 months be made available in pounds and inches and with the 2.3, 10th, 25th, 50th, 75th, 90th, and 97.7th percentile lines, to facilitate their acceptance and use in the United States. This not only will allow pediatric health care providers to more readily make their own comparisons between the two growth curves for their patients, but also will improve the ability of interested researchers to assess the impact of switching from the CDC 2000 curves to the WHO 2006 curves in the US population as a whole.
The primary concern regarding direct comparison of the 2 sets of curves is the difference in the prevalence of underweight. However, for the vast majority of children who are not close to the 2.3 percentile for weight, as well as those at the opposite extreme of the curve for weight at the 97.7th percentile, using the WHO 2006 curves for infants age 0 to 24 months, as demonstrated by Mei et al, would not make any significant difference. Pediatricians would need to more closely evaluate those infants approaching the 2.3 percentile on the WHO 2006 curves, just as they do for infants approaching the 5th percentile on the CDC 2000 curves. It is certainly appropriate to begin using the WHO 2006 growth standard to follow the growth of US infants between birth and 24 months using the recommended cutpoints of <2.3 percentile and >97.7th percentile for abnormal growth. I strongly urge our colleagues at the CDC and National Center for Health Statistics to make the WHO 2006 curves available in a format that can be readily used in pediatric practice. The American Academy of Pediatrics also must advocate for the use of these curves and get involved with the education process required for users of these new growth curves. It is time for all of us to step up to the plate!
References
- 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;109:45–60
- . WHO Child Growth Standards: Height for Age, Weight for Age, Weight for Length, Weight for Height and Body Mass Index for Age: Methods and Development. Geneva: World Health Organization; 2006;
- . Childhood growth charts. Pediatrics. 2002;109:141–142
- . Comparison of the WHO child growth standards and the CDC 2000 growth charts. J Nutr. 2007;137:144–148
- . Comparison of the prevalence of shortness, underweight, and overweigh among US children age 0 to 59 months using the CDC 2000 and WHO 2006 growth charts. J Pediatr. 2008;153:622–628
- . Enrollment and baseline characteristics in the WHO Multicentre Growth Reference Study. Acta Pediatr. 2006;450(Suppl):7–15
PII: S0022-3476(08)00671-9
doi:10.1016/j.jpeds.2008.07.055
© 2008 Published by Elsevier Inc.
Refers to article:
- Comparison of the Prevalence of Shortness, Underweight, and Overweight among US Children Aged 0 to 59 Months by Using the CDC 2000 and the WHO 2006 Growth Charts , 11 July 2008
