Birth weight and obesity in children and adolescents: a systematic review



Camila Elizandra Rossi; Francisco de Assis Guedes de Vasconcelos

Programa de Pós-Graduação em Nutrição do Centro de Ciências da Saúde do Departamento de Nutrição da Universidade Federal de Santa Catarina - UFSC





AIM: To verify scientific and epidemiologic evidences of the hypothesis of association between birth weight and overweight/obesity in childhood and adolescence based on a systematic review of the literature.
METHOD: A systematic review was performed in the MedLine/Pubmed, Scielo-Brasil and Lilacs electronic bases. Articles were classified into two categories of analysis based on the Human Development Index of the country where the study was performed: a) birth weight and overweight/obesity in countries with a high Human Development Index; b) birth weight and overweight/obesity in countries with a high, but still ascending Human Development Index, and medium Human Development Index. Downs & Black checklist was adapted and used to evaluate the methodological quality of the fourteen articles chosen.
RESULTS: In both categories the association between high birth weight and overweight/obesity was predominant. Additionally, one of the seven articles in the first category found low birth weight as a predictor of body and abdominal fat. In the second category, three articles identified the association between catch-up growth and overweight/obesity and another found low birth weight as a protector for overweight (including obesity). Among the four Brazilian studies found, the association between birth weight and overweight or obesity was not statistically significant in two articles.
CONCLUSIONS: High birth weight was associated with overweight/obesity in the majority of articles. The association between low birth weight and overweight/obesity needs ongoing investigation.

Key words: Overweight, Obesity, Birth Weight, Literature Review.




As of the 1990's, there have been reports in the literature on the association between low birth weight (LBW), characterized by birth weight below 2500 g, and the presence of overweight or obesity during childhood1, adolescence2 and adult life3, 4. Another study suggested that the amount of fat-free mass during adolescence is lower in newborns with LBW5.

On the other hand, high birth weight also (HBW), characterized by birth weight equal or over 4000 g, has been associated with the development of excess body weight, during childhood and adolescence6. However, results that challenge significant evidence of statistical significance with HBW have also been reported, especially in studies performed in countries in economic ascension7,8.

Knowing that LBW is more prevalent in impoverished populations and in developing countries9, while HBW has an increased prevalence in some developed countries like the United States (US) and Canada,10 and some European countries11, the socioeconomic level of samples assessed has been appointed as an important confounding variable in the association of birth weight (BW) and obesity. Therefore, this association seems to remain inconsistent in studies performed in children in developing countries12.

Based on what has been said, it is important to identify in the literature if LBW and HBW have been reported as factors associated with overweight and/or to obesity, so that preventive measures may be planned in mother and infant-juvenile health, focusing on the priorities of each population. In this fashion, the objective of the present study was to verify the level of scientific and epidemiological evidence of the assumption of association between birth weight (BW) and overweight/obesity in childhood and adolescence, based on a systematic review of the literature. The initial questions of the investigation were: Is there an association between BW and overweight/obesity in children and adolescents? Does the association between BW and overweight/obesity in children and adolescents differ according to the human development index of the country in which the study was performed?



A systematic literature review was performed, trying to find scientific articles describing the association between birth weight and overweight and/or obesity. Articles published as of 1998 were selected. The electronic databases searched were: Scielo-Brasil (Scientific Electronic Library Online), Lilacs and Medline/Pubmed by National Library of Medicine, in July 2008. The keywords used in the Scielo-Brasil and Lilacs databases were compliant with the definitions in DecS (Health Descriptors) and were: a) birth weight and obesity; and low birth weight and obesity. In the Medline/Pubmed database keywords defined according to their description on the MeSH (Medical Subject Headings) were used, searching: a) birth weight (or low birth weight) and obesity (or overweight or adiposity), and b) fetal programming and obesity, and establishing the following inclusion criteria: studies performed in humans of both genders; age group between 6-18 y; and articles published and indexed in the database in the past 10 years. In the latter, articles stored in the "related articles" links were also attained. It should be underscored that different keywords were used in each database because of the definitions that each database proposes for keywords. By adopting this procedure, possibly more articles related to the topic of interest were found in each database. Other inclusion criteria were the condition that the variable birth weight was the exposure, and overweight and obesity the outcomes; and those studies considered BMI measurements to quantify fat-free mass or fat mass and anthropometric indexes such as weight-for-height, weight-for-age and related indexes as outcome also be included. As exclusion criteria, the authors defined not to analyze: a) articles with results exclusive for outcomes in adult or pre-school ages, as these were not age groups of interest; b) review articles, due to the proposal of analyzing only original articles, that is, based on empirical data; and c) clinical trials, because only observational studies were included.

In the search performed on the Medline/Pubmed database, a total of 33 articles were found, using the first combination of keywords, 3 of which review articles; and 37 articles using the second combination, 8 of which review articles. In the Scielo-Br database, nine articles were found using the terms birth weight and obesity (1 review article) and 1 with the terms low birth weight and obesity. In the Lilacs database, 32 studies were found, totaling articles and summaries of post-graduate papers (1 review article). After reading abstracts, 14 articles were selected for analysis. It should be underscored that the reference lists of the articles selected were not consulted.

The methodological quality of each article was assessed based on criteria established by Downs & Black13, which allow guiding the reader/reviewer as to the limitations of the article assessed, enabling critical reading of the studies published. The original questionnaire had 27 questions, divided in four groups: a) presentation (assessment of items such as clarity to describe objectives, confounding variables, probability values); b) external validation (related to extrapolation of data to the population from which the sample was taken); c) internal validation (analysis of biases, reliability of exposure and outcomes, and use of confounding variables); and d) power of the study. For the present study, questions 8, 13 to 15, 23, 24 and 27 were removed, because they referred to clinical trials. The questions of the scale enabled, objectively, to assess if the article under analysis complied or not with what was being asked. For each question, score zero was given to an article that did not comply with what was being assessed and one (1) if it complied. Only question 5 received score 2 if the question was answered by the article. In this fashion, the maximum score that an article of the present review could reach was 21.

As a synthesis strategy, a script to describe the characteristics of each article was designed, highlighting: authorship, country, year of publication, study design, sample characteristics and size, diagnostic criteria, cut-off points to define overweight and/or obesity, outcome, exposure, main results found, main statistical analyses, and scores on the methodological quality of the studies. Data attained for each article are described in Charts 1 and 2.

As analysis strategy, the fact that the distribution of BW changes according to the socioeconomic level of each nation was taken into account, and the 14 studies were grouped into two categories, according to the classification proposed by the United Nations Development Program (UNDP), based on the Human Development Index (HDI)14 of the country in which the study was performed: a) BW and overweight/obesity in countries with an HDI above 0.900 (high human development - 7 articles)5, 6, 15-19, including United Kingdom, Australia, Denmark, Finland, US and Germany; and b) BW and overweight/obesity in countries with an HDI between 0.800 and 0.899 (high but still ascending human development index - six articles) 7, 8, 20-23, including Brazil, Mexico, and the Seychelles Isles and BW and overweight/obesity in countries with an HDI below 0.800 (middle human development index - one article)24, carried out in China. It should be pointed out that the HDI is calculated based on data of life expectancy at birth, schooling at the elementary, secondary and higher education levels, literacy among adults, and per capita Gross Domestic Product in American dollars14, which explains the methodological approach chosen, as the HDI can be considered as proxi of a nation's socioeconomic level. Moreover, the utilization of socioeconomic variables in the articles analyzed was also assessed, as it is an important confounding variable in the relationships with outcomes such as overweight and obesity.



According to what is presented in Charts 1 and 2, 14 articles that met eligibility criteria were identified in the time period studied. Of these, 6 were published as of 20056, 17, 20-22, 24, demonstrating that the theme is still current.

Among the 14 articles, the utilization of different diagnosis criteria to assess overweight and obesity (outcome of the present investigation) was perceived, as well as different classifications of birth weight (exposure variable). BMI/Age was the anthropometric index most frequently used (9 articles) to diagnose overweight/obesity, but different criteria and different cut-off points for each criterion were used. The criteria of Cole et al. (2000), CDC (2000) and Must et al. (1991) were the ones most frequently used. Regarding birth weight, the utilization of z-scores or tertiles and quartiles prevailed, while the categories recommended by the World Health Organization (WHO)25 were used in a minority of articles (2 in 14) (LBW < 2500 g; Sub-Optimal Birth Weight (SOBW) = 2500 - 2999 g; Optimal Birth Weight (OBW) = 3000 - 3999 g).

The age groups of the samples were very heterogeneous, in that 8 articles assessed children and adolescents, 2 only assessed adolescents (> 10 y) and 4 only assessed children (< 10 y).

Regarding collection and use of skin-fold data, since 2006, the WHO26 has recommended that the same individual be assessed twice and by different evaluators, so as to, by means of inter and intra-evaluator variation, calculate the technical error of measurement (TEM), an important strategy for internal validation of the study. None of the articles published after this date, however, have quoted or published the TEM value.

Below, the results of the studies analyzed are briefly described.

Birth weight and overweight/obesity in countries with a high HDI

In the US, the influence of HBW, of gestational diabetes mellitus (GDM) and of mother's BMI on overweight of children and adolescents were observed. Gillmann et al.15 observed that birth weight was associated with overweight in children with ages between 9-14 y, and, those born from mothers with GDM, an association that was also significant. Frisancho16 did not observe the effect of birth weight on the BMI of children and adolescents, given newborns with high birth weight only became obese adolescents when one of the parents was obese. In Berlin17, the negative influence of GDM was also observed, given mothers with this condition had children with higher BMI/gestational age at birth, which in turn, was a predictor of overweight at 6-8 y.

In these studies it is possible to verify that the association between BW and BMI of children and adolescents should be controlled by maternal characteristics, especially those related to pregnancy, to avoid for possible confounding effects.

Two studies assessed subjects of the United Kingdom, presenting divergent results. Reilly et al.6 observed that each 100 g increase in birth weight increased the risk of obesity in children at 7 years of age. In addition to birth weight, the catch-up weight growth (compensatory growth and above normal growth standards for weight and/or height) between birth and 2 years and high growth rates in the first year of life were identified as potential risk factors for obesity. In the study of Singhal et al.5, results differed, in that for each standard deviation increase in birth weight there was a 3% increase in the amount of adolescents' fat-free body mass. For fat mass and for BMI, no significant associations with birth weight were observed. Authors adjusted analysis models for socioeconomic variables, height, stages of sexual maturation, and exercise.

In Australian children18, those born with lower weights and who had higher weights at 7-8 y were observed to have significantly more abdominal fat and a higher percentage of total fat than those born with higher weights. Moreover, it was observed that at each reduction of 1 kg in birth weight, the amount of abdominal fat increased at 7-8 y. Some flaws were identified in the article, such as important confounding variables that were included in other studies but were not included in the article, it is not clear if subjects eligible for the study represent the population of interest (only 24% of the population) and the sample was comprised by subjects who did not belong to the population of interest (29 were volunteer candidates).

In the capital of Finland19, the association between birth weight/gestational age and overweight of adolescents was observed not to remain after controlling for the other study variables. The article had flaws in the description of characteristics of subjects lost to follow-up and in the validation of the sample studied in regard to the source population (representativeness). Moreover, the fact that authors used the BMI recommended for adults as the cut-off point in the assessment of adolescents (BMI>25 kg/m2) draws attention. Such procedure seems wrong and leads to overweight prevalence quite below those expected (in the study, around 4%).

We perceived that among the 7 articles assessed, in 4 (57.2%), there was a significant association between BW and overweight or obesity, in that the score attributed to most of these articles (3) was very near the maximum score they could attain. Among the 4 articles, 3 had associations of HBW with outcomes, and one had results related to association of LBW with high abdominal and body fat deposits. One of the articles also presented association between catch-up weight growth and high growth rate in the first year of life and overweight/obesity. The remaining 3 studies (42.8%), did not identify a significant association of BW with overweight/obesity, in that one of these showed a positive relation between higher birth weights and fat-free mass.

Birth weight and overweight/obesity in countries with high, ascending HDI, and birth weight and overweight/obesity in countries with middle HDI

Four Brazilian studies were assessed, and results differed. Goldani et al.20 observed a positive association between BW and obesity in individuals at 18 y, even after adjusting for socioeconomic and biological variables. However, the article had four flaws, three of which compromised the external validation of the study. In the article of Tomé et al.21, birth weight in the different categories below 3500 g protected for the occurrence of overweight including obesity in school children of Ribeirão Preto (SP) (reference category > 4000g), but also resulted in higher occurrence of low weight (BMI < percentile 5).

In the study of Dutra et al.22 the prevalence ratio for overweight was proportional to the increase of BW, but the association was not statistically significant. The study of Monteiro et al.7, identified an association of catch-up weight growth with overweight and obesity, and the association of catch-up height growth with overweight during adolescence. Birth weight and birth weight / gestational age, however, were not associated with outcomes in multivariate analyses.

In Mexico23, birth weight above 2890 g was observed to be associated with overweight schoolchildren. In the association with obesity, only BW >3110 g had a significant association.

In the Seychelles Isles 8 (Indian Ocean), the positive association between birth weight and overweight and obesity in children was not confirmed in a multivariate analysis. In this case, the faster weight gain in the first year of life (catch-up weight growth), regardless of birth weight, was shown to be significantly associated with the outcome variable.

In China24, the catch-up weight growth was observed to be more frequent in LBW newborns and with gestational age in the lowest tertile. High birth weight and catch-up weight growth between zero and 3 months of age was associated to overweight (including obesity) of children at seven years of age. The accelerated weight gain up to 3 months of age had a higher effect on BMI at a later age of boys who were born with lower weights in comparison with girls in the same condition.

Among the 7 articles assessed, in 4 (57.2%) there was a significant association between BW and overweight or obesity, in that the score attributed to these articles was very near to the maximum score they could reach. Among the 4 articles, 3 had association of higher weights with outcomes, and one had results related to the association of LBW with low weight. Moreover, in 3 articles there was association of catch-up growth and high growth rate in the first year of life and overweight/obesity. The remaining 3 studies (42.8%) did not identify significant association of BW with overweight/obesity.



Based on the analysis of studies of the present revision, it is possible to observe that birth weight has currently become an important theme to Public Health, due to evidence of a possible association to the phenomenon of obesity in childhood and in adolescence.

In brief, of the 14 articles analyzed, in 5 (35.7%) the authors perceived results pointing out that higher birth weights were associated with higher BMI, percentage of fat or overweight/obesity. Of these studies, 3 were performed in developed countries of North America and Europe, in which an increase in HBW rates is being observed, probably due to the occurrence of mother obesity and gestational diabetes, which were also factors associated with overweight and obesity in children and adolescents. Among the 14 studies, 3 identified association between birth weight and overweight/obesity only in analyses not adjusted for confounding factors, indicating multivariate analyses as essential in the associations for the outcomes overweight and obesity. The association between catch-up growth and overweight/obesity was observed in 4 studies, in that 3 were carried out in countries of ascending and middle HDI. Catch-up growth is characterized as fast growth in weight and/or height, compensatory and above normal growth standards for a specific age, which occurs during rehabilitation, resulting from a nutritional disease or deficiency25, 27. Given situations of nutritional deficiency are more common in countries of lower human and economic development, there is a probable association between this fact and obesity in childhood in these countries, while in those more developed, the prevailing association is the one between high birth weight and obesity.

Only one article reported positive results for the association between lower birth weights and higher BMI in childhood. According to a review study performed by Barker28, three main physiological mechanisms are mentioned in the literature as being mediators of the effects of LBW in the later development of obesity and even in other conditions. The first mechanism would be the change in the phenotype expression generated by the insufficient replication of cells, which seems to lead the body to store energy, as an adaptive response. A second mechanism would be the change generated in metabolism through hormone expression, highlighting an association between higher resistance to insulin and LBW. Another hypothesis still, would be that LBW predisposes individuals to be more vulnerable to environmental influences present in posterior phases of the life cycle. Sawaya29 observed some evidence that recovery of LBW through catch-up growth results in higher reserves of fat mass and lower protein reserves in muscles. Singhal et al.5 also identified that children born with low weight had less fat-free mass during childhood and adolescence. Power et al.30, in turn, observed in children from 7-11 y born with Sub-Optimal Birth Weight, tended toward early sexual maturation when compared to children born with Optimal Birth Weight. Theses pieces of evidence seem to indicate that low birth weight does not influence the occurrence of overweight/obesity directly, but results in adaptation mechanisms of the body, such as catch-up growth and hormone variations, which could predispose individuals to develop overweight/obesity. That is why the assumption of the association between low birth weight and obesity needs to be better studied, especially assessing the composition of the body in terms of fat-free mass and fat mass.

Still, on low birth weight, the assessment of birth weight by gestational age was identified as an important variable of interest in studies, as it can determine if newborns underwent of intra-uterine growth restriction (IUGR) or if the pre-term newborn had adequate uterine development in relation to gestational age.

In relation to the methodological quality of articles, those with a cross-sectional design had higher scores attributed to them. This may be explained because this kind of study is of easier operation and also by the scale utilized, which was developed specifically for cohort and intervention studies. As a reason for lower scores in cohort studies, susceptibility of sample loss is mentioned and, for this reason, there is a methodological need to compare characteristics of subjects not assessed to those who remain in the study, and stricter in the item external validation.

It is worth pointing out that different reference populations were used to identify the prevalence of overweight and obesity among children and adolescents studied, and there was a significant variability in the classification of birth weight. This difference among methods made it difficult to compare studies, and possibly changed the strength of the association between variables of interest in each study, because different BMI classifications interfere in the prevalence of outcome. Even in a same population, if different criteria for diagnosing overweight and obesity are used, distinct data can be generated31, 32 and, that is why it should be pointed out that a more extensive review, utilizing articles with the same diagnostic criteria, would lead to more consistent results on the association of interest.


Final Considerations

High birth weight appeared associated with overweight/obesity in most articles that reported association of birth weight with overweight/obesity. The need to develop more studies that assess the association between low birth weight and overweight/obesity should be underscored, given the number of articles that verified association with nutritional status were scarce. In relation to studies by level of development of countries, the authors observed that in those with high HDI, the relationship between high birth weight and overweight/obesity was more evident than in the remaining countries; and in developing countries (high, but ascending HDI and middle HDI) catch-up growth was associated with overweight/obesity.

Therefore, it is suggested that as a preventive measure, due care be taken during the gestational period to avoid the occurrence of obesity and diabetes, because both factors can cause newborns to have high birth weight, which in turn, stands out currently as a variable associated with overweight and obesity in childhood and in adolescence. Another measure would be the prevention of nutritional deficiencies that lead to slow intra-uterine growth and weight loss, which could lead to catch-up growth.



1. Ong KKL, Ahmed AL, Emmett PM, Preece MA, Dunger DB. Association between postnatal catch-up growth and obesity in childhood: prospective cohort study. BMJ 2000; 320: 967-71.         

2. Walker SP, Gaskin PS, Powell CA, Bennett F. The effects of birth weight and postnatal linear growth retardation on body mass index, fatness and fat distribution in mid and late childhood. Public Health Nutr 2002; 5(3): 391-6.         

3. Te Velde SJ, Twisk JWR, van Mechelen W, Kemper HCG. Birth weight, adult body composition and subcutaneous fat distribution. Obes Res 2003; 11(2): 202-7.         

4. Stettler N, Kumanyika SK, Katz SH, Zemel BS, Stallings VA. Rapid weight gain during infancy and obesity in young adulthood in a cohort of African Americans. Amer J Clin Nutr 2003; 77: 1374-8.         

5. Singhal A, Wells J, Cole TJ, Fewtrell M, Lucas A. Programming of lean body mass: a link between birth weight, obesity, and cardiovascular disease? Amer J Clin Nutr 2003; 77: 726-30.         

6. Reilly JJ, Armstrong J, Dorosty AR, Emmett PM, Ness A, Rogers I, Steer C, Sherriff A. Early life risk factors for obesity in childhood: cohort study. BJM 2005; 330: 1357-64.         

7. Monteiro PO, Victora CG, Barros FC, Monteiro LM. Birth size, early childhood growth, and adolescent obesity in a Brazilian birth cohort. Intern J Obes and Rel Metab Dis 2003; 27: 1274-82.         

8. Stettler N, Bovet P, Shamlaye H, Zemel BS, Stallings VA, Paccaud F. Prevalence and risk factors for overweight and obesity in children from Seychelles, a country in rapid transition: the importance of early growth. Intern J Obes 2002; 26: 214-9.         

9. UNICEF/WHO. United Nations Children's Fund and World Health Organization. Low birthweight: country, regional and global estimates. UNICEF: New York; 2004.         

10. Ananth CV, Wen SW. Trends in fetal growth among singleton gestations in the United States and Canada, 1985 throught 1998. Seminars in Perinatology 2002; 26(4): 260-7.         

11. Surkan PJ, Hsieh CC, Johansson ALV, Dickman PW, Cnattingius S. Reasons for increasing trends in large for gestational age births. Obstetrics & Gynecology 2004; 10(4): 720-6.         

12. Martins EB, Carvalho MS. Associação entre peso ao nascer e o excesso de peso na infância: revisão sistemática. Cad Saúde Pública 2006; 22(11): 2281-300.         

13. Downs SH & Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health 1998; 52: 377-84.         

14. Programa das Nações Unidas para o Desenvolvimento (PNUD). Relatório de Desenvolvimento Humano 2007/2008. Nova York, EUA. 2007. 402 p. Disponível em http://www.pnud.org.br/arquivos/rdh/rdh20072008/hdr_20072008_pt_complete.pdf. Acessado em 25 de agosto de 2008.         

15. Gillman MW, Rifas-Shiman S, Berkey CS, Field AE, Colditzand GA. Maternal gestational diabetes, birth weight, and adolescent obesity. Pediatrics 2003; 111: 221-6.         

16. Frisancho AR. Prenatal compared with parental origins of adolescent fatness. Am J Clin Nutr 2000; 72: 1186-90.         

17. Schaefer-Graf UM, Pawliczak J, Passow D, Hartmann R, Rossi R, Bührer C. et al. Birth weight and parental BMI predict overweight in children from mothers with gestational diabetes. Diabetes Care 2005; 28: 1745-50.         

18. Garnett SP, Cowell CT, Baur LA, Fay RA, Lee J, Coakley J. et al. Abdominal fat and birth size in healthy prepubertal children. Intern J Obes 2001; 25: 1667-73.         

19. Pietläinem KH, Kaprio J, Räsänen M, Winter Torsten, Rissanen, Rose AR. Tracking of body size from birth to late adolescence: contributions of birth length, birth weight, duration of gestation, parents' body size, and twinship. Am J Epidemiol 2001; 154: 21-9.         

20. Goldani MZ, Haeffner LSB, Agranonik M, Barbieri MA, Bettiol H, Silva AAM. Do early life factors influence body mass index in adolescents? Braz J Med Biol Res 2007; 40(9): 1231-6.         

21. Tomé FS, Cardoso VC, Barbieri MA, Silva AAM, Simões VMF, Garcia CA et al. Are birth weight and maternal smoking during pregnancy associated with malnutrition and excess weight among school age children? Braz J Med Biol Res 2007; 40(9): 1221-30.         

22. Dutra CL, Araújo CL, Bertoldi AD. Prevalência de sobrepeso em adolescentes: um estudo de base populacional em uma cidade no Sul do Brasil. Cad Saúde Pública 2006; 22(1): 151-62.         

23. Moraes AS, Rosas JB, Mondini L, Freitas ICM. Prevalência de sobrepeso e obesidade e fatores associados em escolares de área urbana de Chilpancingo, Guerrero, México, 2004. Cad Saúde Pública 2006; 22(6): 1289-1231. [CONFERIR NÚMEROS DE PÁGINAS]         

24. Hui LL, Schooling CM, Leung SSL, Mak KH, Ho TH, Leung GM. Birth weight, infant growth, and childhood body mass index: Hong Kong's Children of 1997 Birth Cohort. Arch Pediatr Adolesc Med 2008; 162(3): 212-8.         

25. WHO (World Health Organization). Physical status: the use and interpretation of anthropometry. Geneva: WHO (Technical Report Series, 854); 1995.         

26. WHO MGRS (Multicentre Growth Reference Study Group). Reliability of anthropometric measurements in the WHO Multicentre Growth Reference Study. Acta Paediatr Suppl 2006; 450: 38-46.         

27. Boersma B, Wit JM. Catch-up growth. Endocrine Rev 1997; 18(5): 646-61.         

28. Barker DJP. The developmental origins of adult disease. J Amer College of Nutr 2004; 23(6): 588S-95S.         

29. Sawaya AL, Martins AP, Grillo LP, Florêncio TT. Long-term effects of early malnutrition on Body Weight Regulation. Nutr Rev 2004; 62(7): 127S-33S.         

30. Power C, Li L, Manor O, Smith GD. Combination of low birth weight and high adult body mass index: at what age is it established and what are its determinants? J Epidem Comm Health 2003; 57: 969-73.         

31. Assis MAA, Rolland-Cachera MF, Grosseman S, Vasconcelos FAG, Luna MEP, Calvo MCM et al. Obesity, overweight and thinness in schoolchildren of the city of Florianopolis, Southern Brazil. Eur J Clin Nutr 2005; 59: 1015-21.         

32. Rolland-Cachera MF, Castelbon K, Arnault N, Bellisle F, Romano RC, Lehingue Y et al. Body mass index in 7-9-y-old French children: frequency of obesity, overweight and thinness. Inter J Obes 2002; 26: 1610-16.         



Camila Elizandra Rossi
Rua Sete de Setembro, 695
Bouganville Residence, apto. 101, Centro
Biguaçu, SC CEP 88160-000
E-mail: camilarnutri@yahoo.com.br

Associação Brasileira de Pós -Graduação em Saúde Coletiva São Paulo - SP - Brazil
E-mail: revbrepi@usp.br