Secondary Logo

Journal Logo


Prevention and Control of Obesity in Preschool Children: Importance of Normative Standards

Uauy, Ricardo*,‡; Rojas, Juanita; Corvalan, Camila§; Lera, Lydia*; Kain, Juliana*

Author Information
Journal of Pediatric Gastroenterology and Nutrition: December 2006 - Volume 43 - Issue - p S26-S37
doi: 10.1097/01.mpg.0000255848.84658.52
  • Free


The combined objectives of promoting child growth while preventing obesity are often not recognized as intertwined components of good nutrition; rather, they are often considered to be opposite poles of the malnutrition spectrum. Providing complementary foods to young children with the worthy objective of preventing malnutrition without considering the need to avoid obesity in stunted children may in fact do potential harm. Nutrition programs addressed at preventing malnutrition may have built-in mechanisms that can easily promote positive energy balance and thus increase risk for obesity. This is particularly relevant when feeding underweight stunted children who may be of normal or even excessive weight for stature. Thus, it is of the utmost importance to define what is normal weight and height and to apply normative standards to assess growth and to establish energy intake recommendations that are consistent with good nutrition and health during childhood and beyond (1).

This manuscript examines recent advances in defining norms to assess the growth of young children, the latest energy intake recommendations for children released by Food and Agriculture Organization/World Health Organization/United Nations University (FAO/WHO/UNU), the need to include obesity prevention in the design and implementation of nutrition programs for children with a special emphasis in Latin America, and propose recommendations to prevent malnutrition from energy deficit and excess in countries undergoing rapid epidemiological changes.


Reference Growth Standards to Define Normal Weight, Height and Body Mass Index

To assess normal growth we need an adequate reference of central tendency and variability in anthropometric indices. Because good health and nutrition are defined by the capacity to support normal growth, the reference/standards used to assess growth are fundamental for both clinical practice and to establish public health recommendations. Thus, the definition of “normal” growth is of paramount importance to secure the normal health and nutrition of populations. Normative gender- and age-specific data of weight, height and body mass index (BMI) are needed to define who is undernourished, normal, and obese. Anthropometry is only an indirect marker of body composition (lean body mass and fat stores), therefore, weight and length measurements serve only as proxy markers for increased or reduced adipose tissue energy stores. Weight, height and weight/height and BMI for age and sex have been used as indicators of nutritional adequacy in infancy, childhood and adolescence (2).

Present growth charts are based on what is observed growth for a normal reference population rather than recommended growth-based on health outcomes throughout the life course. The most commonly used growth charts are based on the National Center for Health Statistics ([US]NCHS), implying that children growing in an affluent society in North America should be considered the standard of healthy growth. The NCHS growth reference served as the basis for the present WHO international growth standards for infants 0 to 36 months, which was derived from children growing in an affluent rural society in the town of Yellow Springs, OH (3). The present WHO infant growth norm has major flaws: it is derived from a nonrepresentative sample of the population and the data are drawn from a group of predominantly formula-fed infants who received energy-dense complementary foods.

The WHO/UNU growth standards released in May 2006 correspond to a representative sample of disease-free individuals in 6 sites around the globe. The new Multicountry Growth Reference Standards (MGRS) were developed based on the growth of infants who were predominantly breast-fed for 4 to 6 months and appropriate complementary foods after weaning (4). In fact, infants fed according to present WHO recommendations and living under conditions that favor the achievement of genetic growth potential, grow less rapidly in weight than the present WHO/NCHS reference, particularly after 4 to 6 months. Present distributions of normal weight-for-age and weight-for length are skewed toward higher values, relative to those observed in predominantly breast-fed infants; this may be a contributory factor to the increase in childhood obesity because normal children are being considered underweight and are thus prescribed additional energy. These drawbacks led a WHO Expert Committee in 1995 to support the development of a new growth reference. The multicountry (Brazil, Norway, India, Ghana, the US and Oman) growth reference study, mentioned above, is specifically designed for this purpose; data collection began in 1997 and was completed by 2003. Analysis of the data has been completed and new reference standards were released in May 2006. The study undertaken in diverse geographical areas included population-based samples of infants and children whose mothers were nonsmokers of middle to high income so that environmental conditions were not restrictive and caregivers followed the established WHO feeding recommendations. The research design combined a longitudinal study from birth to 24 months of age of 300 newborns per country, with a cross-sectional study of 1400 children ages 18 to 71 months per site. More than 13,000 healthy infants and children were involved in the study. The new international growth reference provides a scientifically reliable descriptor of physiological growth and a powerful tool for advocacy in support of good health and nutrition. Another objective is to support, based on actual evidence, the concept that human growth during the first years of life is similar across groups of children of different ethnic/genetic backgrounds (5). There are genetic differences in linear growth, but these appear to be of a similar nature across population groups. Adverse environmental factors by themselves or those that interact with genetic makeup explain differences in growth observed in low income groups. The corollary for this concept is that existing differences in growth are predominantly environmentally derived and can be subject to improvement. Prevalence estimates of undernutrition and obesity in children will clearly be affected to the extent that the new reference differs from the WHO norms being used. Most important, this reference sets the growth of the breast-fed infant as the normative standard. Optimal infant feeding, as presently defined by WHO, is exclusive breast-feeding for the first 6 months of life, followed by continued breast-feeding with adequate complementary foods for up to 2 years and beyond. The issue of whether this standard should be applied to all children 0 to 5 y requires ample discussion, yet for most the breast-fed infant represents the gold standard for infant growth. The long-term consequences of this pattern of growth remain to be fully determined, however. Existing national and international standards have defined normal growth based on the weight and length gain observed in apparently healthy children. This has led in practice to support the notion that “bigger is better.” This is a reasonable proposal if the objective is to enhance survival in infancy and early childhood in areas where malnutrition and infection in synergy claim the lives of infants and young children. However, it is certainly not the case in countries where deaths of young children are rare and the concern has shifted to the prevention of obesity and related burden of chronic disease (6). To make appropriate comparisons across countries and to monitor trends in childhood overweight prevalence it is crucial to have appropriate growth standards and to agree on a clear definition of childhood overweight and obesity. The former is available, at least for children 0 to 5 y, but the latter remains a problem.

Reference Standards to Define Overweight and Obesity in Children

There is a lack of consistency in the use of the terms overweight and obesity in children. The WHO Expert Committee on Physical Anthropometry provided in 1995 a definition for overweight and recommended that the term obesity not be used for children (3). More recently, the NCHS proposed to classify children in the upper end of the reference distribution as either at risk of overweight or overweight, again avoiding the use of the term obesity (7). However, in 1999, the International Obesity Task Force (IOTF) Expert Committee used the terms overweight and obesity to provide their recommendations for assessment of children from 2 to 18 y to be consistent with adult definitions (8). All of the recommendations take into account 2 levels of excess weight, but use of different definitions may lead to confusion in interpreting results, monitoring trends or comparing prevalence within and across countries. This is further complicated by the need to define the appropriate anthropometric indices and cutoff points that best predict long-term adverse health outcomes. The use of weight-for-height z scores to assess the nutritional status of children is still a common practice in pediatrics; recently BMI has been incorporated, after 2 y old, to achieve concordance with adult assessment. In 1995, the WHO Expert Committee recommended the use of weight-for-height >2 z score to define overweight in children under 10 y old (9). Body mass index is being progressively adopted as a valid measurement of obesity in childhood, adolescence and adulthood. Body mass index changes substantially with age and sex and thus recommendations are age- and sex-specific. The 2000 Centers for Disease Control (CDC) growth charts provide BMI-for-age curves for the US population. Based on the NCHS recommendations, children with BMI ≥95th percentile are classified as overweight, whereas children with BMI between the 85th and the 95th percentile are classified as at risk of overweight (10). The use of BMI curves based on projected adult cutoff points has been recommended by the IOTF. These cutoffs are based on age- and sex-specific percentiles for children that project to the adult cutoffs of 25 kg/m2 for overweight, and 30 kg/m2 for obesity (8). The IOTF approach has been considered to have a low sensitivity to detect excess adiposity; in fact, IOTF obesity cutoff projects to the 97th to 99th percentile values of the WHO standard. This may prove helpful in terms of specificity. Another limitation of the IOTF approach is that it does not provide a way to assess the full distribution of values, so BMIs in excess of the obesity cutoff cannot be quantified. When z scores are used as part of the evaluation, the magnitude of the excess can be established with greater precision. At the population level, overweight and obesity prevalence vary depending on the definition used. We have demonstrated, using a large data set for school children across 13 y of monitoring, that obesity prevalence estimates determined using the IOTF cutoff points are significantly lower than those based on the WHO or the NCHS-CDC 2000 references (11). Despite differences in absolute values, trends are similar independent of the standard used. At the individual level, the use of different criteria has an impact on the assessment of the nutritional status both for underweight and overweight children. This is exemplified in Table 1, which uses 3 hypothetical children of equal height but different weights at each age between 2 and 5 y.

Nutritional status of hypothetical boys 2–5 y old using 4 different criteria

Table 1 shows weight-for-height and BMI-for-age for NCHS-CDC 2000; both measures show excellent agreement in the identification of overweight and obesity (4/4 and 3/4, respectively). The WHO 1978 scale compared with the BMI-for-age criterion of NCHS-CDC 2000 have an excellent concordance for overweight (4/4), but no agreement for obesity; the WHO 1978 criteria are systematically more tolerant. When comparing weight-for-height criteria, these agree in every case for the diagnosis of overweight, but in only 1 case when the diagnosis is obesity. Again, the WHO 1978 criteria are systematically more tolerant. There was scant agreement in the classification of overweight and obesity between BMI-for-age and the IOTF criteria–based classification, and the latter has a systematically lower sensitivity. In the case of undernutrition for both sexes and all ages, NCHS-CDC 2000 weight-for-height and BMI-for-age have excellent concordance and a higher sensitivity than the WHO 1978 criteria. This implies that more children would be diagnosed as underweight using either of the CDC indices as compared to WHO 1978 weight-for-height z score. These differences in estimates of both underweight and overweight have clear implications in terms of the preventive interventions to which children may be subjected to improve their nutritional status. Overall, these results support the idea that at this age, weight-for-height and BMI-for-age perform equally well as measures of overweight and highlight the relevance of establishing an adequate reference population and cutoff points based on long-term health outcomes rather than populations that shift according to stage of epidemiological transition that they face (12,13). Given the need for a continuous indicator across age groups into adulthood we consider BMI to be the best measure of overweight and obesity. The need for specific cutoff values based on normative population or projected BMI into adulthood has not been fully agreed upon. The merit of the IOTF criteria in providing child BMI values projecting to adult BMI of 25 and 30 for overweight and obesity is offset by not having a continuous measure of the range. This is especially relevant for children who are above the obesity cutoff point; the IOTF criteria cannot discriminate whether a child is 3 or 5 SD above the median BMI for age. Acceptable international criteria to evaluate overweight and obesity in children remain to be defined; for now the best practice is to record BMI data for age and present median and SD values by gender (14).


Optimal nutrition should be defined based on growth in weight and length associated with the lowest risk of adverse health outcomes during infancy, but they also should consider the burden of chronic disease in later life. The redefinition of normal growth for children 0 to 5 y by WHO/UNU is a step in the right direction because these growth standards are based on longitudinal data from infants fed according to the internationally agreed-upon norms, exclusively breast-fed for the first 6 months of life and with added complementary foods after that. However, the long-term effect of this dietary prescription remains to be fully established (15). The prevention of obesity will indeed be supported by having a standard that is based on breast milk feeding, which is known to decrease the risk of childhood obesity. In addition, there is an added dimension to this policy because the prescription for energy intake of normal children has also been redefined by the FAO/WHO/UNU report published in late 2004.

The energy recommendations for infants and children published by FAO/WHO/UNU in 1985, based on the work of the Expert Committee Group that met in Rome in 1981 (16), were estimated from observed energy intakes of children from industrialized countries who were growing optimally according to the Harvard Growth Standard (the best available measure at the time). To this, 5% was added to support growth in conditions prevailing in developing countries where infection was more prevalent and diets may have lower digestibility. The need to consider actual expenditure rather than intake was recognized at the time, but there was insufficient data on energy expenditure of young children, except for neonates and young infants. However, the expenditure approach to estimate energy needs was used in children older than 13 y based on estimates of basal energy expenditure, plus the energy required for normal growth and defined level of physical activity (17). The dissemination of the doubly labeled water method to assess total energy expenditure during the past decades, and the development of methods to assess activity levels applicable to young children, permitted measurements of daily energy expenditure in children from different parts of the world. This allowed the 2001 FAO/WHO/UNU committee to define recommendations based on actual expenditure.

Energy recommendations for infant and child age groups published by FAO/WHO/UNU in 2004 are based on actual measurements and estimates of total daily energy expenditure either by the doubly labeled water method or estimates based on heart rate monitoring during active periods coupled with individual calibrations of oxygen consumption. The energy needs for tissue deposition related to growth in infants, children and adolescents were added to the estimate of daily energy expenditure. In the case of infants, the new recommendations were based on breast-fed infants rather than formula-fed infants; for the first year of life, the mean values for the former are 5% to 10% lower than figures for formula-fed babies. Table 2 illustrates the changes in energy recommendations for breast- and formula-fed infants during the first year of life and the differences with respect to the recommendations given in the 1985 report.

Comparison of present estimates of energy requirements of infants with those calculated in 1985 FAO/WHO/UNU report (selected ages)

The present recommendations based on expenditure are also substantially lower for children up to age 10 y in comparison with those derived on observed food intake as used in 1985 by FAO/WHO/UNU. Figure 1 compares the 2004 estimates with the 1985 estimates for boys and girls. The proposed energy requirements are 18% to 20% lower for boys and girls under 7 y, and 12% lower for boys 7 to 10 y (Table 2). From age 12 onward, the proposed requirements are 12% higher for both boys and girls, assuming moderate levels of physical activity. These modifications of the recommendations for infants and children should serve to correct the overestimations of the dietary energy needs of children from birth to 10 y. Interventions such as child feeding programs, formulated based on the 1985 recommendations, could have easily led to excess energy intake among target populations. The higher energy prescribed and the greater tolerance of the anthropometric indices with overweight and obesity could have contributed to the observed excessive weight gain among these children. This may have been a critical problem for children living in urban settings, especially in countries undergoing economic and dietary changes associated with modernization.

FIG. 1
FIG. 1:
Energy recommendations for children and adolescents based on habitual energy intake as per FAO/WHO/UNU 1985 and derived from energy expenditure FAO/WHO/UNU 2004.

The 2004 recommendations include a need for physical activity to maintain fitness and health and to reduce the risk of developing obesity and diseases associated with sedentary lifestyles. Moreover, different requirements for populations with lifestyles that involve different levels of habitual physical activity, starting at 6 y old are given. Recommendations for physical activity levels based on classification of habitual activity consistent with good long-term health are provided. The energy requirement of an individual was defined by FAO/WHO/UNU as “the level of energy intake from food that will balance energy expenditure when the individual has a body size and composition and level of physical activity consistent with long-term good health; and that will allow for the maintenance of economically necessary and socially desirable physical activity. In children and pregnant or lactating women, the energy requirement includes the energy needs associated with the deposition of tissues or the secretion of milk at rates consistent with good health” (17). For infants and young children, energy needs are equal to the sum of expenditure and energy stored as growth. Total expenditure is the sum of several components: basal metabolism, thermal effect of feeding, thermoregulation and physical activity. A lower recommended energy intake for all children <10 y, except for those that are highly physically active as prescribed by FAO/WHO/UNU 2004, will contribute to the prevention of systematic energy excess and will stress the need to improve nutrient density of diets rather than increase energy intakes. For infants the new recommendations will serve to strongly support exclusive breast-feeding because as energy needs drop, the sufficiency of breast milk energy supply is prolonged. This means that for most infants, exclusive breast-feeding will provide sufficient energy for normal growth for at least 6 months and possibly beyond this period. FAO/WHO/UNU 2004 will also serve to stress the need to enhance nutrient density of complementary foods after 6 months to meet infants' and children's micronutrient needs rather than overemphasizing the need for high energy intakes to promote weight gain. If the energy needs are lower and the micronutrient needs are the same, then the micronutrient density of the diet becomes critical. Thus, both normative standards will act in synergy to prevent excess weight gain and to promote good linear growth in accordance with the only proven gold standard “the breast-fed infant.” This prescription has stood the test of successful human evolution for the past several million years. We still face the challenge of defining the optimal energy intakes and growth rates in early life, conducive to long-term health across the life course.


Basic Considerations

To address food insecurity and protect vulnerable groups from the scourge of hunger and malnutrition, developing countries have implemented various forms of food supplementation and/or income transfer programs to prevent the consequences of malnutrition (18). It has been estimated that close to 1 of 5 Latin Americans lives in poverty conditions and can be considered undernourished/food insecure (19). However, these programs have the potential to induce energy excess unless they consider critical issues in the selection of beneficiaries and in the type of food supplement provided. In the case of income transfer, choices made by recipients of the programs may favor energy-dense, nutrient-poor foods that may also contribute to overweight and obesity in women of reproductive age and young children. A recent FAO survey with data from 19 Latin American countries found that >20% of the population—approximately 83 million people of an estimated 414 million in these countries—receive some level of food assistance benefits from nutrition-related programs (19,20). In contrast, the number of malnourished people in the studied countries was 10 million, 12% of the total beneficiaries of food programs. This indicates that nearly 9 of 10 beneficiaries may be of normal weight or even overweight. Nutrition programs have evolved beyond the immediate needs of the malnourished and have become part of the social and economic benefits demanded by populations living in poverty. The Brazil Zero Hunger Program, which is directed at those suffering from hunger, is predominantly an income-transfer program that provides food and other commodities and does not include biological or nutritional vulnerability in the criteria to select beneficiaries. In fact, the program has been criticized recently because the poor in Brazil are mostly stunted and overweight, rather than suffering from wasting (21).

The nature of the problem is exemplified in a simple manner in Fig. 2A, which presents the weight-for-age distribution of children in a typical developing country with high prevalence of underweight children. As depicted in Fig. 2B, if food is provided for all children, not just those who are underweight, then the whole curve will be displaced to the right, generating a significant proportion of overweight and obese children. The desirable outcome, if targeting is efficient, is the elimination of the tail below −2 SD without increasing the proportion of those >2 SDs; in this case, the distribution of weight-for-age for the population is normal, with few individuals underweight or obese. The evidence emerging from countries undergoing the epidemiological transition indicates that if food supplements are provided without careful targeting, the situation becomes similar to that depicted in Fig. 2B. Some planners may even consider this phenomenon to be an unavoidable consequence, whereas others suggest it as a desirable outcome because it serves as evidence of the impact of a successful intervention (22).

FIG. 2
FIG. 2:
Model examining the effect of universal food supplementation of children on the distribution of weight-for-age. A, Baseline with a significant group of underweight (<−2 SD). B, Shift in distribution after universal food supplementation. Undernutrition has been reduced considerably, whereas overweight and obesity have increased.

A critical issue in defining the nature of this problem is recognizing that underweight children are usually stunted, thus most malnourished children will be of low weight- and length-for-age, but will have a near-normal weight-for-length/height. In other words, they are underweight and stunted, but not wasted. Because recovery in length-for-age is incomplete if nutrition improvement occurs after 24 to 36 months of age, these children will gain significantly more weight-for-age than length-for-age when given additional food. In older children, little or no gain in length-for-age is observed (23). Later in the article, we examine these issues in various settings where additional food energy is provided to children who are underweight and evaluate the consequences in terms of obesity prevalence. We also analyze how supplementary feeding programs may contribute to rising obesity trends, what factors may explain this phenomenon and the potential strategies needed to avoid a rise in obesity as a byproduct of malnutrition prevention efforts.

Most programs do not have targeting criteria based on actual anthropometry but rather on socioeconomic condition. The benefits are given to all, independent of whether there is evidence of low body weight. Significant reductions in underweight and wasting have occurred in most countries; undoubtedly supplementary feeding programs have played an important role in this decline. Unfortunately, they may have also contributed to the rising trend in obesity (1). Stunting remains a problem in most developing countries, despite the virtual eradication of wasting in many (24). In this setting, providing food supplements may be beneficial for some, whereas it may be detrimental for others. Careful selection of beneficiaries of food assistance programs and determination of the right combination of nutrients/foods, education, and lifestyle interventions are required to optimize nutrition and health at each stage of the life cycle. This is a crucial problem that should be addressed before initiating specific programs (25).

Latin American Food and Nutrition Programs Targeting Children and Adolescents

Information on those programs from selected countries (Colombia, Costa Rica, Cuba, Chile, Guatemala, Mexico, Peru and Venezuela) can be found at A total of 40 nutrition or food programs target pregnant women and children, however, only 12 include anthropometric indices to select their beneficiaries. Most are targeted to the low socioeconomic status groups, based on biodemographic characteristics independent of whether there is objective evidence that people require extra energy or are underweight for stature. Moreover, FAO and other international bodies have defined hunger to meet the mandate of the World Food Summit and the Millenium Development Goals based on achievement of energy intake to meet theoretical energy needs. This has been labelled “undernourishment” and its the assessment is based on energy available from food balance sheets, energy needs of the population and distribution according to socioeconomic levels, with no actual measurement of any indicator of nutritional status. This in itself tends to overestimate the number of undernourished people, and because no provision to assess body weight or physical activity is made, is often in sharp contrast to the direct assessment of BMI of these populations (9). In areas of the world, such as Latin America and southeast Asia, where populations are progressively becoming urban and sedentary and energy availability is not a limiting factor, the FAO definition of undernourished misguides policy (it assumes that the problem is solved by providing food energy to poor populations). Under these conditions, poverty is related to adequate or even excess energy relative to physical activity levels and diet is of poor quality in terms of healthy foods, rather than the amount of energy (26). This requires that we incorporate the need to improve the quality of diets as a complement to food security. We analyzed these programs derived from data collected by Uauy and Rivera in the context of obesity prevention in developing countries and the considerations made in the introduction to this section.

Colombia has 6 widescale programs targeting children, adolescents and pregnant women; these programs are under the technical supervision of the Instituto Colombiano de Bienestar Familiar. Beneficiaries are from the lowest socioeconomic level from both urban and rural areas. The objectives of these programs include family support, improvement in nutritional status, promotion of breast-feeding and healthy lifestyle. All of these programs have national coverage and for targeting criteria use age and biological vulnerability. Four programs also include anthropometry to target beneficiaries who are mostly malnourished. The food that is distributed bienestarina provides between 30% and 80% of recommended energy units and proteins and can include ≥1 of the energy and protein requirements, respectively. The usual ingredients of bienestarina include a fortified vegetal mix made from powdered milk, soy flour, any cereal flour (rice, wheat or corn), legumes, sugar, molasses and oil.

Costa Rica has 4 programs with national coverage that target pregnant women, children and adolescents; 1 program gives away money to buy a specific menu and another 1 tries to reduce micronutrient deficiencies. None of the 4 programs include anthropometric indices in the targeting criteria. Cuba has 6 programs of national coverage, of which 5 have as main objective to reduce micronutrient deficiencies. Foods provided by these programs include either fortified pureed fruit or flour. Guatemala has 8 national programs and 3 regional programs that target pregnant women and children. They predominantly target malnourished populations; only 1 of them considers anthropometry (weight-for-height) for targeting purposes. These programs provide beneficiaries a food ration consisting of legumes, cooking oil and flour (a corn-soy blend). Other programs provide either micronutrient-fortified sugar or flour. Mexico has 6 large-scale food programs. One of the most important was called Progresa (the Spanish word for progress) but was renamed Oportunidades (the Spanish word for opportunities) (27). This is an incentive-based welfare program whose main objective is to invest in the nutrition, health and education of young children from poor families to improve their chances of obtaining better opportunities later in life. Apart from receiving a food supplement, the family receives money. Low-income families who are not beneficiaries of Oportunidades may participate in a program that provides a daily breakfast and a food package that provides 20% of the daily energy recommendations (includes cooking oil, legumes and corn flour to prepare tortillas). Oportunidades is an exception among the large-scale programs in Latin America because it includes anthropometric measurements as part of its targeting criteria.

In Peru, there are 4 programs targeting low-income pregnant women and/or children. Apart from socioeconomic considerations, 3 of them include anthropometry to target beneficiaries. In general the supplement provides 30% of the energy recommendation and includes fortified specific foods such as baby food, cookies or a food package. Venezuela has 5 programs with national coverage, all of which target according to socioeconomic level. One of these programs applies subsidies to what are considered staple foods (corn flour, margarine, cooking oil, sugar, whole milk, canned fish and legumes). The other programs provide daily meals for vulnerable groups.

From this summary we conclude that only 1 of 4 food programs in Latin America includes anthropometry to target beneficiaries. The most commonly used anthropometric index for targeting purposes is weight-for-age. Because most underweight children in Latin America are stunted and not wasted, the target group may in fact be subjected to energy excess. Most energy-dense food supplements provided to children after 3 y of age, when there is a limited chance to gain length, may in fact lead to overweight and obesity (28). About 75% of the programs provide micronutrient-fortified supplements that are important in promoting linear growth, but unless bioavailability of micronutrients is taken into account, this may also be of limited value. Our analysis suggests that Latin American countries should reexamine the type of interventions in place and consider provision of micronutrient-ich supplements with limited energy except for populations or individuals who are wasted.

Chile, a Case Study for the Effect of the Nutrition Transition in Young Children

Chile is often presented as a paradigm of the success of supplementary feeding programs. Indeed, there is a close association between the presence of these massive and expensive interventions and the decline in malnutrition in all age groups. Close examination of these data demonstrate that for infants, preschool and schoolchildren, undernutrition based on weight-for-age was virtually eradicated by the late 1980s, and stunting rates remained low but significant until the mid-1990s. It is exactly at this stage of the transition that obesity needs to be considered in the implementation of programs and necessary changes need to be incorporated. In fact, over the past 2 decades the impact of the programs in reducing malnutrition was progressively lost, whereas the association with rising obesity prevalence has become notable (29).

The National Nursery Schools Council Program (JUNJI) created in 1971 under the Ministry of Education is of special relevance to our discussion of nutrition of preschool children. JUNJI provides child care as well as supplementary food for low-income infants and preschoolers. Coverage is close to 70% of those in need. In 2003, approximately 120,000 children under 5 y of age attended JUNJI. Of those, 95% were preschoolers from 2 to 5 y of age; the rest were infants under 2 y. The food distributed is programmed to cover 58% to 75% of children's daily energy needs, depending on whether they attend a half-day or a full day. The energy supplied by age grouping is 800 kcal/d from 2 to 3 y and 900 kcal/d from 3 to 5 y. If a nutritional deficit (underweight) is detected, then an additional 100 kcal/d are provided. This program most likely has contributed to the notable decline in underweight and stunting observed in the preschool population during the past decade, but it may also beassociated with the increase in obesity rates (29). Figure 3A illustrates the dramatic decline in undernutrition evaluated by height-for-age, weight and weight-for-height (note that the cutoff in this case is <1 SD, so the expected prevalence would be about 16%). The rise in overweight and obesity shown in Fig. 3B depicts the parallel increase in the height of children. We have recently addressed this issue in school-age children in Chile, noting a significant association between the rise in obesity and the improved length of 6-y-old children; nearly half of the increase in obesity could be accounted for by the increase in length (30).

FIG. 3
FIG. 3:
A, Dramatic decline in undernutrition (evaluated by height-for-age, weight and weight-for-height). B, Rise in overweight and obesity and the parallel increase in the height of children. Prevalence (%) anthropometric indices below -1 SD (A) and above 1 SD (B).

In addition, as shown in Fig. 4, we evaluated the change in the prevalence of obesity (BMI for age ≥95th percentile; CDC/NCHS) of children between March (school entry) and November (end of school year) for 1992 and 1996. The figure illustrates a marked rise in obesity in all age groups during this period. The comparison between March and November for the same year showed that in 1992, the largest increment in obesity was observed in 3-year-olds, whereas in more recent years, this was observed in 2-year-old children. The prevalence of obesity has remained stable from 1996 onward until recently, when a possible drop was observed.

FIG. 4
FIG. 4:
Cross-sectional comparison of obesity (BMI >95th percentile) prevalence at start of school year (March) and end of school year (November) according to age in 1992 and 1996.

The rise in obesity during the past decade has led to significant changes in the energy prescribed and the foods provided in an effort to curb the rise in obesity prevalence; these changes are summarized next.

Changes in Criteria to Prescribe Energy Provided by the Program

Initially, the program provided a fixed amount of calories which was not differentiated by age. Between 1970 and 1977, the energy content of the meals amounted to 1500 kcal/d, that is, every preschool child independent of age received the full daily energy recommendation based on FAO/WHO 1973 or 1985. In addition, children who presented a nutritional deficit received an extra 150 kcal/d. From 1987 onward, the energy content of the meals has been adjusted according to age and the child's weight status. If the child presented a deficit (low weight-for-height), an extra supplement was provided. This remained the case until 1995, when the energy content of this additional supplement was reduced to 100 kcal/day and was used for the few children who required this extra food based on being underweight for their length.

These adjustments occurred between 1977 and 1987; first, the energy content of the meals was reduced to 1200 kcal/d (between 75% and 80% of daily energy needs) because it was determined that the children only stayed awake about two thirds of school time. In addition, the energy content of the meals was differentiated according to 3 age groups. This last classification only lasted 1 y because of implementation problems, and since 1988, only 2 age groups are in place: 2 to 3 and 3 to 5 y old. From 1988 until 1995, the energy content was further reduced to 1000 kcal/d and 1200 kcal/d, respectively.

Adjustment of Energy Content

In 1996, evaluation of the nutritional status of preschool children under JUNJI's supervision showed high rates of overweight and obesity. In addition, there was a high prevalence of inactivity. These 2 considerations led to a further reduction of the energy content to 900 kcal/d for children ages 2 to 3 y and 1000 kcal/d for those from 3 to 5 y old. These modifications remained until 2001, when further reductions were made to 800 kcal/d and 900 kcal/d, respectively. This corresponds to an average of 60% of energy recommendations based on FAO/WHO/UNU 1985 (16) and ≈72% of the recommendations published in 2004. (17). These changes were implemented initially in 2001 in one third of the country, and the process was completed by 2004. As stated above, if a child presents a nutritional deficit, then an additional 100 kcal/d is provided.

The potential impact of the nutrition programs in the rise of obesity was examined comparing the nutritional status of children at 2 y old when they entered JUNJI with that observed 3 y later when they were 5 y old. The analysis for 2 longitudinal follow-up time periods (1992–1994 and 1996–1998) are summarized in Table 3. The first study considered 8086 children. Of the 6238 children who entered the program with normal nutritional status, 1503 (24.1%) and 506 (8.1%) became overweight and obese, respectively, over the 3 y. Of the 1440 who were initially overweight, 730 (50.7%) became obese. In total, there was a 3-fold increase in the number of obese children, from 408 to 1236. The second study period shown in Table 3 included preschool children from Santiago studied from 1996 to 1998. The proportion of normal and overweight children becoming obese was lower than was previously observed. However, the baseline data indicated that >10% were already obese on entry, and this figure increased by about half. Therefore, there was less of a rise during the program, but obesity was already high on entry.

Comparison of nutritional status at entry and exit of preschool children participating in JUNJI during 3-y longitudinal follow-up

The dramatic increase in obesity of young children during the past 15 years across the world, including Chile, is likely due to multiple factors acting in synergy. To document the effect of trends of economic and social changes affecting diet and physical activity on obesity prevalence of 6-y-old children in Chile, we conducted a correlation analysis of these data. The study revealed that social indicators such as gross national product per person, percentage living in poverty and social expenditure per person plus energy and macronutrient availability, car ownership, low birth weight prevalence and deaths caused by diarrheal disease were highly associated with obesity in childhood (31). The multivariate regression of these data indicated that changes in income and car ownership explained virtually all of the changes observed in the univariate model. The only 2 additional factors that entered into the multivariate analysis beyond income were percentage of fat energy units in the diet and the number of deaths from diarrhea as an index of enteric infections with a negative coefficient. In other words, less infection was linked to higher obesity prevalence.

The effect of food and nutrition programs could not be isolated from other factors, but because they were maintained stable or increased while the number of people living in poverty decreased, the net result was a major increase in the amount of food provided per beneficiary. This effect was further amplified because recent economic growth led to self-exclusion of high- and middle-income beneficiaries and thus there was a significant association between the increased availability of food provided by the programs with the increase in obesity. This association by no means proves causality. The necessary changes in the programs to decrease energy content while maintaining or increasing micronutrient supply were slow to come, but went in the right direction overall. Recent evaluations of obesity prevalence and incidence in preschool children suggest that the problem is being controlled although it has not been reversed.


There is an urgent need to improve the evaluation of nutritional status and the categorization of undernutrition. The systematic measurement of stature needs to be incorporated into the assessment of nutritional status in all countries. This provides a way to separate individuals who are underweight from those who are stunted, of normal weight or overweight for their stature. The assessment of height provides a clear indication that the main nutritional problem of children in most developing countries is stunting and not underweight. This realization should lead to a rethinking of the interventions needed, placing a greater emphasis on the improvement of linear growth.

The recognition that stunting is the major nutritional problem should lead to strengthening the prescription of exclusive breast-feeding as a norm for young infants up to 6 months old. There is also a need to improve the micronutrient content of complementary foods. Iron, zinc, vitamin A and other micronutrients should be incorporated in the foods provided by government programs for infants and young children. Programs targeted at malnourished infants and children (defined based on insufficient weight for age or in poor weight gain) need to include assessment of weight-for-length as criteria for discharge after a trial of food supplementation. If a child continues to be underweight after a trial of food supplementation but weight for stature is normal or in excess of normal, then the need to keep the child in the program should be reconsidered, especially for children older than 18 months. Further energy intake may lead to weight gain in excess of the norm.

The acknowledgement of the rising rates of obesity in young children should lead to a careful reassessment of the energy needs of children participating in the feeding programs and to the improvement of the quality of the foods provided. Progressive increases in fruits and vegetables and lower fat content milk should be added to the specifications of preschool and school feeding programs. Health promotion, nutrition education and physical activity programs need to be an integral part of preschool and school programs. A clear strategy for the promotion of healthy foods, physical activity, reduction of psychosocial stress, tobacco control, and healthy environment is needed (32).


  1. Promote and support exclusive breast-feeding for 6 months, with the introduction of complementary foods and continued breastfeeding thereafter—up to 2 y of age or longer, as mutually desired by the mother and infant.
  2. Promote the appropriate introduction of safe, nutritionally adequate and developmentally appropriate complementary foods. This is critical after 6 months and must continue until 3 y of age to support optimal linear growth. Stunting is difficult to modify after 3 y.
  3. Ensure that the needs of nutritionally at risk infants and children are met, giving special attention to linear growth of preterm and/or low birth weight infants. Prevent excess weight gain to decrease risk of obesity later in life.
  4. Monitor growth and avoid rapid weight gain at all stages of life, especially in infancy. Children should maintain a healthy weight. Create an environment that supports healthy food choices and promotes physical activity patterns concordant with a healthy lifestyle.
  5. Identify early on children at schools and health centers with individual and/or social risks for overweight and obesity, establishing dietary and physical activity measures to prevent unhealthy weight gain.
  6. Implement programs for treatment of childhood obesity early to prevent present and future adverse consequences.
  7. Monitor growth using appropriate standards and take necessary actions to prevent stunting in all children and avoid unhealthy weight gain in all children.


1. Uauy R, Kain J. The epidemiological transition: need to incorporate obesity prevention into nutrition programmes. Public Health Nutr 2002; 5:223–229.
2. Garza C, de Onis M. Rationale for developing a new international growth reference. Food Nutr Bull 2004; 25:S5–S14.
3. World Health Organization Working Group on Infant Growth. An evaluation of infant growth: the use and interpretation of anthropometry in infants. Bull WHO 1995; 73:165–74.
4. Growth patterns of breastfed infants in seven countries. WHO Working Group on the Growth Reference Protocol and WHO Task Force on Methods for the Natural Regulation of Fertility. Acta Paediatr 2000; 89:215–22.
5. De Onis M, Garza C, Victora C, et al. The WHO Multicentre Growth Reference Study: planning, study design, and methodology. Food Nutr Bull 2004; 25:S1–S86.
6. Popkin B, Gorden-Larsen P. The nutrition transition: worldwide obesity dynamics and their determinants. Int J Obes (Lond) 2004; (Supp 3):S2–9.
7. Kuczmarski RJ, Ogden Cl, Guo SS, et al. CDC Growth Charts for the United States: methods and development. Vital Health Stat 2002; 246:1–190.
8. Cole TJ, Bellizzi MC, Flegal KM, et al. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000; 320:1240–1243.
9. Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, et al. CDC Growth Charts: US Advance Data from Vital and Health Statistics, No. 314. Atlanta: National Center for Health Statistics; 2000.
10. Must A, Dallal GE, Dietz WH. Reference data for obesity: 85th and 95th percentiles of BMI (wt/ht2) and triceps skinfold thickness. Am J Clin Nutr 1991; 53:839–846.
11. Kain J, Uauy R, Vio F, et al. Trends in overweight and obesity prevalence in Chilean children: comparison of three definitions. Eur J Clin Nutr 2002; 56:200–204.
12. Hardy LR, Harrell JS, Bell RA. Overweight in children: definition, measurements, confounding factors, and health consequences. J Pediatr Nurs 2004; 6:376–384.
13. Flegal KM, Ogden Cl, Wei R, et al. 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. Am J Clin Nutr 2001; 73:1086–1093.
14. Lobstein T, Baur L, Uauy R, et al. Obesity in children and young people: a crisis in public health. Obes Rev 2004; (Suppl 1):4–104.
15. Onyango AW, Pinol AJ, de Onis M, et al. Managing data for a multicountry longitudinal study: experience from the WHO Multicentre Growth Reference Study. Food Nutr Bull 2004; 25:S46–S52.
16. FAO/WHO/UNU. Energy and Protein Requirements. WHO TRS 724. Geneva: World Health Organization, 1985.
17. Food and Agriculture Organization. Human Energy Requirements. Report of a Joint Expert Consultation, TRS 1. Rome: FAO, 2004.
18. Food and Agriculture Organization. Targeting for Nutrition Resources for Advancing Nutritional Well-Being. Rome: FAO, 2001.
19. Uauy R, Oyarzún MT. Subnutrición y estado nutricional en Chile. Análisis técnico de informe FAO: estado sobre la inseguridad alimentaria en el mundo. Rev Chil Nutr 2005; 32:262–270.
20. Peña M, Bacallao J. Obesity among the poor: an emerging problem in Latin America and the Caribbean. In: Peña M, Bacallao J, eds. Obesity and Poverty. A New Public Health Change. Sci Publ No. 576. Washington, DC: Pan American Health Organization, 2000.
21. Brazil's hunger program, FOME ZERO. Available at:
22. Uauy R, Albala C, Kain J. Obesity trends in Latin America: transiting from under to overweight. J Nutr 2001; 131:893S–899S.
23. Martorell R. Results and implications of the INCAP Follow-up Study. J Nutr 1995; 125:1127S–1138S.
24. Allen L, Uauy R. Guidelines for the study of mechanisms involved in the prevention or reversal of linear growth retardation in developing countries. Eur J Clin Nutr 1994; 48:S212–S216.
25. Food and Agriculture Organization. Targeting for Nutrition Resources for Advancing Nutritional Well-Being. Rome: FAO, 2001.
26. Monteiro CA, Conde WI, Popkin BM. Is obesity replacing or adding to undernutrition? Evidence from different social classes in Brazil. Public Health Nutr 2002; 5:105–112.
27. Rivera JA, Sotres-Alvarez D, Habicht JP, et al. Impact of the Mexican program for education, health, and nutrition (Progresa) on rates of growth and anemia in infants and young children: a randomized effectiveness study. JAMA 2004; 291:2563–2570.
28. Albala C, Vio F, Kain J, et al. Nutrition transition in Chile: determinants and consequences. Public Health Nutr 2002; 5:123–128.
29. Albala C, Vio F, Kain J. Nutrition transition in Latin America: the case of Chile. Nutr Rev 2001; 59:170–176.
30. Kain J, Uauy R, Lera L, et al. Trends in height and BMI of 6 year-old children during the nutrition transition in Chile. Obes Res 2005; 13:2178–2186.
31. Uauy R, Atalah E, Kain J. The nutrition transition: new nutritional influences on child growth. In: Martorell R, Haschke F, eds. Nutrition and Growth. Nestlé Nutrition Workshop Series Vol. 47. Philadelphia: Lippincott, Williams and Wilkins; 2001.
32. Salinas J, Vio F. Health promotion in Chile. Rev Chil Nutr 2002; 29:166–173.
© 2006 Lippincott Williams & Wilkins, Inc.