7/28/2019 635228_obi_china

1/7

Original Research Article

Determinants of High Weight Gain and High BMI Status in the First ThreeMonths in Urban Chinese Infants

JIANDUAN ZHANG,1* JINGXIONG JIANG,2 JOHN H. HIMES,3 JING ZHANG,1 GUOYAN LIU,4XIAONA HUANG,2YUAN GUO,1

JUNXIN SHI,1 AND SHUHUA SHI11Department of Woman and Childs Care and Adolescence Health, School of Public Health, Tongji Medical College,

Huazhong University of Science and Technology. Wuhan 430030, Hubei, China2Department of Child Health, National Center of Women and Childrens Health Care, Chinese Center for D isease Control a nd Prevention,

Beijing 100013, China3Department of Epidemiology and Community Health, School of Public Health, University of Minnesota, Minneap olis, Minnesota 554544Department of Preschool Education, Normal College, Shenzhen University, Shenzhen 518060, Guangdong, China

Objectives: Investigate the potential factors associated with high weight gain and high BMI status in the first

three months of life.

Methods: Totally, 930 healthy neonates (489 boys and 441 girls) were involved in this community-based, prospec-

tive study in China. Data on body weight and length were collected at birth, and the 1st and 3rd months. A self-adminis-

tered questionnaire was used to collect data regarding social demography, gestational status, delivery, and the feeding

patterns of children.

Results: Prevalences of high BMI status (BMI = 85th p, re WHO BMI standards) increased over time in both sexes,

reaching 24.5% and 12.0% for boys and girls, respectively. General linear mixed models indicate high BMI status at 3

months is significantly and inversely associated with breastfeeding, as a proportion of feeding occasions [OR 0.74

(95%CI: 0.560.98)] and positively with lower birth weight [OR 2.07 (95%CI: 1.233.49)]. High weight gain (=85th p, re

WHO velocity standards) in the first 3 months is also significantly associated with breastfeeding [OR 0.76 (95%CI:

0.590.96)] and sex, with boys at a higher risk than girls [OR 1.44 (95%CI: 1.071.97)]. Living in an extended family is

associated with both high weight gain and high BMI status, but with marginal statistical significance.

Conclusion: Analyses indicate an increasing trend of high BMI status in early infancy. Breastfeeding provides a

protective effect for both high weight gain and high BMI status. The results concerning birth weight suggests a target

for intervention. Am. J. Hum. Biol. 24:633639, 2012. ' 2012 Wiley Periodicals, Inc.

Childhood obesity is one of the most important globalpublic health concerns and it has been spreading to many

parts of the world (Ogden, et al., 2006; Sanigorski, et al.,2007; Spurgeon, 2002; Stamatakis, et al., 2005). With rapidsocioeconomic development since the 1980s, China hasbeen experiencing a significant transition in life style andnutritional status. Accordingly, the prevalences of child-hood overweight and obesity have risen dramatically. Inurban areas the prevalence of obesity in preschool childrenincreased from 1.5% in 1989 to 12.6% in 1997, and that ofoverweight increased from 14.6% to 28.9% in the same pe-riod (Luo and Hu, 2002). A recent report indicates thatcombined overweight and obesity prevalence, and only obe-sity prevalence among school age children reached 34.2%and 30.3%, respectively, in 2005 (Ji and Cheng, 2009).

Concurrent and long-term adverse health outcomes of ex-cessive adiposity in childhood are well documented, includ-ing asthma, Type 2 diabetes and elevated cardiovasculardisease risks (Brophy, 2009; Dietz and Gortmaker, 2001;WHO, 2000). As these conditions track into adulthood, theyare associated with greater middle-age mortality and mor-bidity(Dietz, 1998; Gunnell et al., 1998). Obese children aremore likely to suffer from emotional and social conditions,e.g., depression, low self-esteem, eating disorders, negativebody image,and socialstigma (Stockton et al., 2009).

Infancy is understood as a critical period of developingadiposity and the relative weight gain from birth to 3months has positive relationships with childhood fat mass(Chomtho et al., 2008). Rapid infancy weight gain is con-sistently associated with increased subsequent obesity risk

as well (Ong and Loos, 2006; Vogels et al., 2006; Whitaker1997). This makes infancy and early childhood an excellent

opportunity for obesity prevention and, consequently, itshould be a primary focus for intervention against futureobesity problems (Birch and Ventura, 2009).

Childhood obesity is a multifactorial disorder influencedby genetic, environmental and cultural factors. Neverthe-less, the rapid temporal changes in prevalences of obesitywithin populations are unlikely to have resulted primarilyfrom genetic factors. Because of the wide variation in envi-ronmental and cultural factors that may foster obesity, itis important to examine the particular subset of factorswithin a country associated with the development of childobesity so that determinants can be understood andappropriate interventions can be tailored.

China has several distinctive characteristics thatmay increase the risk of early development of obesity in

Contract grant sponsor: National Natural Science Fund (NNSF) ofChina; Contract grant number: 30800917; Contract grant sponsor: TheSpecial Fund for Basic Scientific Research of Central Colleges; Contractgrant number: 2011TS017.

*Correspondence to: Jianduan Zhang, Department of Woman andChilds Care and Adolescence Health, School of Public Health, Tongji Medi-cal College, Huazhong University of Science and Technology, 13 HangkongRd., Wuhan 430030, Hubei, China. E-mail: jd_zh@hotmail.com or

jd_zh@mail.hust.edu.cn

Received 21 July 2011; Revision received 18 February 2012; Accepted 8April 2012

DOI 10.1002/ajhb.22284

Published online 24 May 2012 in Wiley Online Library (wileyonlinelibrary.com).

AMERICAN JOURNAL OF HUMAN BIOLOGY 24:633639 (2012)

VVC 2012 Wiley Periodicals, Inc.

7/28/2019 635228_obi_china

2/7

children, e.g., the highest Cesarean-section rate in theworld (Li et al., 2006), extended family structure, and alow proportion of exclusive breastfeeding within 46months (Huang et al., 2009). Yet the understanding of thehigh weight gain and high BMI status of early childhoodis limited in China to findings from mostly small studies

and cross-sectional surveys, and extant studies usuallyfocused on older children (Chen et al., 2006; Yi et al.,2002). Moreover, we are aware of no longitudinal studiesavailable from birth.

This study reports results from a longitudinal study ofweight gain and BMI status from birth to three months ofage in Chinese infants from three cities, where demo-graphic, cultural, and feeding variables were collected fora large sample of children.

SUBJECTS AND METHODS

A total of 930 infants (489 boys and 441 girls) wererecruited through a multistage sampling scheme withinthe first month following birth and were followed subse-quently at least until three months of age. For the sam-

pling, three cities were selected to represent three distinctadministrative and geographic regions of China: Shenyang,capital city of Liaoning province (Northern China); Wuhan,capital city of Hubei Province (Central China) and Guangz-hou, capital city of Guangdong Province (Southern China).These are relatively affluent cities and the prevalence ofobesity was thought to be relatively high.

Within each of the three urban areas CommunityHealth Service Centers/Stations were randomly selected.Finally, the participant infants were then randomlyselected from among newborns to local residents thatwere registered and who would receive health surveil-lance and routine follow-up. All protocols were standar-dized across the three field sites and data collectors weretrained and monitored centrally. Informed consent was

obtained from the parents.Birth weight and crown-heel length at birth were

obtained from the Perinatal Health Booklet (PHB), whichwas kept by individual parents with health informationrecorded related to pregnancy from the first prenatalcheckup until 42 days following delivery. These data werecollected by consistently trained health staff. Weight andlength were measured within 10 days of the exact monthlytarget ages. Only in Wuhan were children required to bemeasured at 2 months of age. Body weight was measuredto the nearest 50 g with electronic scales and with subjectswearing light indoor clothing. Recumbent length was meas-ured to the nearest 0.1 cm using an infantometer. Bothweight and length were taken in duplicate and means ofthe two measurements were used in the analyses.

Socioeconomic and health-related variables wereobtained from the PHB and from standardized adminis-tered interviews. Maternal weight gain, diseases duringpregnancy, and type of delivery were provided by thePHB. Child sex, date of birth, parental age, parental occu-pations, and education level, as well as family structureand income came from the interviews. The reported occa-sions of bottle and breast feeding were collected using aspecially developed feeding questionnaire at monthlyinterviews with caregivers.

We examined the prevalence of high and very high BMIstatus, defined as equal to or greater than the 85th and95th percentiles of the age- and sex-specific standards

respectively, using the World Health Organization (WHO)Child Growth Standards (2006). High weight gain wasdefined as weight gain in the first 3 months = 85th percen-tile of the WHO velocity standards (2008) specific to sex.Choosing which reference data or standards to use forcomparisons, and how to appropriately characterize the

children was not straight forward. The WHO standardshave been recommended for use in all children (WHO,2006). Nevertheless, the referent children are all consid-ered healthy even though the standards have been used tocharacterize overweight and obesity in very young chil-dren (Di Onis et al., 2010). The terms overweight and obe-sity are not customarily applied to children less than 2years of age, even though early rapid weight gain is astrong predictor of adiposity later in childhood. Z-scorecut-offs relative to the WHO standards have been used forBMI and weight-for-length to characterize overweightand obesity in very young children, although conventionalcriteria have used sex- and age-specific percentiles (WHO,1995). Our decisions to use BMI percentiles relative to theWHO standards should correctly identify those childrenwith the highest adiposity, and comport with traditional

cut-offs and interpretation (WHO, 1995).For the analysis, 3-month weight gains were calculated

by linear adjustment of the data at the measured intervalsto exactly 3.0 months in duration (91 days). So, for exam-ple, for a child who weighed 3.0 kg at birth and 6.0 kg atthe 3-month visit at 100 days of age, the adjusted 3-monthweight gain 5 (6.03.0 kg) 3 (91/100) 5 2.73 kg. Pretermbirth was defined as born prior to 37 weeks of gestation.

For analysis, maternal and paternal education levelswere recorded into two categories (= middle school;>middle school) because this dichotomy was more discrimi-nating than more complicated schemes or stratifying onuniversity education. The occupations of parents were cate-gorized into those typically reported in Chinese surveys.The monthly household income was recoded into ordinal

categories from 1 to 6 (15 < 1,000 RMB, 2 5 1,000-

7/28/2019 635228_obi_china

3/7

small amounts of missing data so the actual number of

cases may vary slightly in some descriptive summaries.Chi-square tests were used to evaluate the differencesbetween prevalences, and the differences of mean weightgains between groups were compared by t-tests. Age prev-alences in the figure were smoothed across ages usingcubic splines in Excel software. To accommodate inter-citydifferences and intra-class correlations among cities, gen-eral linear mixed models were used via the SAS procedurePROC MIXED to examine factors associated with highweight gain and BMI status at 3 months, with city as arandom effect. The resulting odds ratios (OR) and 95%confidence intervals are presented. All data were analyzedusing SAS v9.2.

RESULTS

Tables 1 and 2 summarize socio-demographic, perinatalstatus and feeding patterns of the infants and their fami-lies. Some of these highlight distinctive aspects of Chinesepopulations, e.g., high proportion of delivery by C-section(67%) and single-child families (87.3%); a very low preva-

lence of low birth weight (1.5%), and about 40% of childrenliving in an extended family. The parents tend to be ratheryoung and well educated with over half of mothers andfathers having university or advanced degrees. The ratesof exclusive breastfeeding at the 1st and 3rd month are39.2% and 40.4%, respectively.

Table 3 presents the means and standard deviations ofbody weight and length by monthly clinic visit and sex.The age-specific means exceed the medians of the WHOgrowth standards and approximate the 60th to 70th per-centiles for each measurement.

The increasing prevalences of high and very high BMIstatus at each age are presented in Figure 1. Prevalencesof high BMI status rises from 9.6% to 24.5% for boys andincreases from 7.7% to 20.9% for girls (X2cmn1 5 86.7, P 1 SD and 2SDs from the median among children aged 0to 5 with were calculated (de Onis, 2010). Given the normaldistribution of weight status, the values of 1 SD approxi-

mately equal to 85th percentile. Therefore, in our finalregression models, 85th percentile was adopted as the crite-ria to distinguish children with high weight gain and withhigh BMI status from the overall participants.

By using the above criteria, the prevalence of high BMIstatus in our study at 3 months reaches 24.5% for boys and20.9% for girls, and 12.1% boys and 7.9% girls reached orsurpassed 95th percentile of BMI-for-age, that is, with veryhigh BMI status. These levels clearly exceed those expectedusing the WHO standards, however, are lower than the15.8% reported for a recent combined-sex sample of infantsless than 4 months of age from Yunnan, China (Li et al.,2007), where obesity" was defined as at least 20% heavierthan the older WHO reference of weight for height.

We found a pattern of significant and progressive

increase in prevalence of high BMI in both sexes. Thereare few data available for direct comparisons at theseearly ages. A similar pattern of increasing overweight andobesity was observed in Iranian infants from birth to 6months of age (Heidari et al., 2008). Certainly, in olderchildren prevalences often increase with age (Hillieret al., 2007; Rudolf et al., 2006; Sanigorski et al., 2007).Thus, the development of high BMI status in our sampleappears to be primarily a postnatal phenomenon.

Infants with birth weight less than 3 kg had approxi-mately twice the risk of having high BMI status at 3months compared with those born at larger birth weights.This pattern of lower birth weight and subsequent higher

weight in infancy has been observed elsewhere (McMillenand Robinson, 2005; Oken and Gillman, 2003; Stettleret al., 2002). Our results indicate that those lower BWchildren must have experienced a period of relative catch-ing-up, although not all of their weight gains exceeded the85 percentile of the WHO velocity standard, because theassociation with lower birth weights was not significant inthe analysis of high weight gains.

It is interesting that our sample of Chinese infantsexceeded the expected frequencies above the cut-offs forBMI relative to the WHO standards, given that breastfeed-ing was less than ubiquitous at these early ages. The pro-portion of feeding occasions that were breastfeeding at 3months showed significant protective effects in our studyfor both excessive 3-month weight gains and high BMI sta-tus at 3 months. In the preliminary analysis we found nei-ther consistent nor significant associations between exclu-sive breastfeeding or ever breastfeeding and high weightgains or BMI status at 3 months, presumably because ofthe relatively high breastfeeding rate. Accordingly, we con-structed the continuously distributed variable to capturethe proportion of feeding occasions that were breast fed.

Our findings are consistent with many studies of breast

feeding and infant growth (WHO, 2002), and the associa-tions of breastfeeding with increased obesity, fatness, andmetabolic syndrome later in childhood and in adults(Bergmann et al., 2003; Bogen et al., 2004; Grummer-Strawn and Mei, 2004).

We did not find significant associations between ourbreastfeeding measure at 1 month and either high BMIstatus at 3 months or excessive 3-month weight gains. Wehave speculated that the potential impact of breastfeedingat 1 month may be diminished in our population becauseof the very high proportion of C-section deliveries (67%).In practice, the immediate initiation of breastfeeding inthe hospital is made more difficult and is delayed because

TABLE 4. Crude and adjusted odds ratios (OR) and 95% confidence intervals (CI) for high BMI status at 3 months

Effects Crude Odds Ratio (95% CI) Adjusted Odds Ratio (95% CI)

Sex 0,Girls (ref.); 1,Boys 1.23 (0.91, 1.68) 1.16 (0.83, 1.63)Birth weight (kg) 0,!3.0 (ref.); 1,Middle school (ref.)1 Middle school

0.96 (0.66, 1.40) 0.90 (0.59, 1.39)

Monthly household income (RMB) 1.03 (0.91, 1.17) 1.04 (0.93, 1.89)Breastfeeding proportion at 1 mo (Z) 0.78 (0.50, 1.21) 1.13 (0.85, 1.49)Breastfeeding proportion at 3 mo (Z) 0.63 (0.42, 0.95) 0.74 (0.56, 0.98)

Notes: Adjusted for gestational age, maternal weight gain during pregnancy, paternal education level.

TABLE 5. Crude and adjusted odds ratios (OR) and 95% confidence intervals (CI) for high weight gains in the first three month

Effects Crude Odds Ratio (95% CI) Adjusted Odds Ratio (95% CI)

Sex 0,Girls (ref.); 1,Boys 1.43 (1.09, 1.86) 1.44 (1.07, 1.97)Birth weight (kg) 0, !3.0(ref.); 1,Middle school(ref.)

1, Middle school

1.49 (1.07, 2.09) 1.23 (0.83, 1.23)

Monthly household income (RMB) 1.00 (0.90, 1.12) 1.08 (0.95, 1.23)Breastfeeding proportion at 1 mo (Z) 0.99 (0.87, 1.13) 1.18 (0.93, 1.50)Breastfeeding proportion at 3 mo (Z) 0.71 (0.50, 1.01) 0.76 (0.59, 0.96)

Adjusted for gestational age, maternal weight gain during pregnancy, paternal education level. Because of the measurements at 2 months were only obtained in onecenter (Wuhan), they were not included in this figure.

637INFANT BMI STATUS AND WEIGHT GAIN IN CHINA

American Journal of Human Biology

7/28/2019 635228_obi_china

6/7

of the surgical procedure. Moreover, the postoperativepain also increases the difficulty in breastfeeding prac-tices (de Armas et al., 2009). We find it encouraging thateven if immediate breast feeding cannot be initiated, per-sistence with breastfeeding until 3 months can still have aprotective effect on high weight gains and BMI status at 3

months of age.In the unadjusted analyses, family structure wasfound to be a significant determinant of both high weightgains and BMI status. Even though these associationsdid not remain statistically significant in the fullyadjusted models, we believe it is informative to commentabout this aspect of the findings because they departfrom the family patterns often found in many Westerncountries. Three-generation families or extended familiesare a common family structure in China, especially whenthe children are still very young. For example, it is esti-mated that 5070% of urban Chinese children live withtheir parents and grandparents (Li, 2005; Zheng et al.,2008). This is considered a traditional pattern and it isbecoming less common, especially in urban areas duringthe recent period of social transition. In our research,

36% of the children lived in an extended family. Giventhat 87% of the children had no siblings because of thepersistent family planning policy instituted in the1980s, one child is usually surrounded" by at least fouradults, who tend to be over-caring, over-concerned, andover-feeding the only child. According to Jiangs qualita-tive research, grandparents tended to urge children toeat more meals and larger portion at served meals forpreschool children (Jiang, 2007).

Our analyses did not indicate significant effects of meas-ures related to socioeconomic status like maternal educa-tion or income. Nevertheless, we chose to leave these varia-bles in the final models to ensure that these theoreticallyimportant covariates were accounted for. Lower socioeco-nomic status has been shown elsewhere to be a strong inde-

pendent risk factor of overweight and obesity in children(Lu, 2004). The lack of associations of socioeconomic factorswith overweight at 3 months or excessive 3-month weightgains may be due to the young ages of infants involved andthe relative homogeneity of diets at that age. Because ofthe one-child policy most Chinese parents value their chil-dren highly and spare no expense in picking the bestbrands of infant formula and foods so income may not bevery discriminating especially during infancy. Finally, thevalidity of the household income data is not clear and thecloseness of the OR to 1.00 in the logistic models may sim-ply reflect less than optimal quality of the data.

We are aware that high BMI status during infancymight not be strikingly predictive of overweight statuslater in childhood and adulthood because of the rapidly

changing growth and shifting of percentiles, especiallyduring the first year of life. Nevertheless, given the estab-lished longitudinal trajectories for weight growth andimpacts on later obesity of excessive weight gains ininfancy (Danielzik et al., 2004; Dennison, et al., 2006; Ongand Loos, 2006), and the increasing prevalences of over-weight and obesity in the preschool years (Demerathet al., 2009; Luo and Hu, 2002), we believe it is importantto understand the natural history of overweight from theearliest ages. This large and well-controlled community-based study in this cohort of Chinese infants provides newinsights regarding probable determinants in this impor-tant and rapidly changing population.

LIMITATIONS

In this study, we used BMI as a measure of weight sta-tus in infancy which is not universally well tested. How-ever, BMI-for-age comparing a childs weight, adjusted forlength, among children of a similar age, is an acceptablesummary measure and is becoming more commonly used

by researchers to study early weight status (Himes, 2009)and was used in a recently published article (de Onis,2010). The WHO Growth Standard does not contain anyEast Asian populations which might affect the results tothe degree that genetic or population-specific factors mayplay roles in early growth patterns. Nevertheless, it isrecommended by the WHO for universal use althoughit includes data from six countries only.

The birth weight and crown-heel length were obtainedfrom the Perinatal Health Booklet which might not be asprecise as the data obtained under more carefully con-trolled research settings. There is no reason to suspect,however, that such data are biased even though they maybe less reliable. To the degree that measurement reliabil-ity contributed additional random error, our findings

should be relatively conservative.

ACKNOWLEDGMENTS

The authors extend their sincere thanks to the healthstaffs from Maternal and Child Health Centre/Hospitaland the Community Health Service Centers in Shenyang,Qiaokou and Jianghan districts in Wuhan, and Panyu andHaizhu districts in Guangzhou who provided support inthe project through data collection and services on thefront lines. Last but not least, they greatly appreciate theparents who have been willing to be involved in thisresearch project.

LITERATURE CITED

Bergmann KE, Bergmann RL, von Kries R, Bohm O, Richter R, Duden-hausen JW, Wahn U. 2003. Early determinants of childhood overweightand adiposity in a birth cohort study: Role of breastfeeding. Int J ObesRelat Metab Disord 27:162172.

Birch LL, Ventura AK. 2009. Preventing childhood obesity: what works?Int J of Obes 33:S74S81.

Bogen DL, Hanusa BH, Whitaker RC. 2004. The effect of breast-feedingwith and without formula use on the risk of obesity at 4 years of age.Obes Res 12:15271535.

Brophy S, Cooksey R, Gravenor MB, et al. 2009. Risk factors for childhoodobesity at age 5: Analysis of the Millennium Cohort Study. BMC PublicHealth 9:467473.

Chen C M. 2008. Overview of obesity in Mainland China. Obes Rev 9:1421.

Chen X, Qiu X, Yang S, et al. 2006. Risk factors of Obesity among Chil-dren Under 7 Years Old. Chin J Child Health Care 14:9394.

Chomtho S, Wells JCK, Williams JE, et al. 2008. Infant growth and laterbody composition: evidence from the 4-component model. Am J Clin

Nutr 87:17761784.Coordinating Study Group of Nine Cities on the Physical Growth and De-velopment of Children Capital Institute of Pediatrics. 2007. A nationalsurvey on growth of children under 7 years of age in nine cities of China,2005. Chin J Pediatr 45:609614 (in Chinese).

Danielzik S, Czerwinski-Mast M, Langnase K, Di lba B, Muller MJ. 2 004.Parental overweight, socioeconomic status and high birth weight are themajor determinants of overweight and obesity in 57 y-old children:baseline data of the Kiel Obesity Prevention Study (KOPS). Int J ObesRelat Metab Disord 28:14941502.

de Armas MG, Megas SM, Modino SC, Bolanos PI, Guardiola PD, AlvarezTM. 2009. Importance of breastfeeding in the prevalence of metabolicsyndrome and degree of childhood obesity. Endocrinol Nutr 56:400403.

Demerath EW, Reed D, Choh AC, et al. 2009. Rapid postnatal weight gainand visceral adiposity in adulthood: the FELS longitudinal study. Obe-sity 17:20602066.

638 J. ZHANG ET AL.

American Journal of Human Biology

7/28/2019 635228_obi_china

7/7

Dennison BA, Edmunds LS, Stratton HH, Pruzek RM. 2006. Rapid infantweight gain predicts childhood overweight. Obesity 14:491499.

Dietitians of Canada, Canadian Pediatric Society, The College of FamilyPhysicians of Canada, Community Health Nurses Association of Can-ada. 2004. A health professionals guide to using growth charts. PaediatrChild Health 9:174176.

De Onis M, Blossner M, Borghi E. 2010. Global prevalence and trends ofoverweight and obesity among preschool children. Am J Clin Nutr

92:12571264.Dietz WH. 1998. Childhood weight affects adult morbidity and mortality.

J Nutr 128:411414.Dietz WH, Gortmaker SL. 2001. Preventing obesity in children and adoles-

cents. Annu Rev Public Health 22:337353.Fu JF, Liang L, ZOU CC, Hong F, Wang CL, Wang XM, Zhao ZY. 2007.

Prevalence of the metabolic syndrome in Zhejiang Chinese obese chil-dren and adolescents and the effect of metformin combined with lifestyleintervention. Int J Obes 31:1522.

Gibson R. 2005. Principles of nutritional assessment. 2nd ed. New York,NY: Oxford University Press.

Grummer-Strawn LM, Mei Z. 2004. Does breastfeeding protect against pe-diatric overweight? Analysis of longitudinal data from the Centers forDisease Control and Prevention Pediatric Nutrition Surveillance Sys-tem. Pediatr 113:e81e86.

Gunnarsdottir I, Thorsdottir I. 2003. Relationship between growth andfeeding in infancy and body mass index at the age of 6 years. Int J Obes27:15231527.

Gunnell DJ, Frankel SJ, Nanchahal K, Peters TJ, Smith GD. 1998. Child-

hood obesity and adult cardiovascular mortality: A 57-y follow-up studybased on the Boyd Orr cohort. Am J Clin Nutr 67:11111118.Heidari ST, Vakili MA Emamghorashi F. 2008. Longitudinal pattern and

reference values of obesity indices of infants in Jahrom (SouthernRegion of ), Iran. Iran J Pediatr 18:3846.

Hillier TA, Teresa, Pedula KL, Schmidt MM, Mullen JA, Charles M-A,Pettitt, DJ. 2007. Childhood obesity and metabolic imprinting. DiabCare 30:22872292.

Himes J. 2009. Challenges of accurately measuring and using BMI andother indicators of obesity in children. Pediatrics 124:S3S22.

Huang Q, Chen L, Ruan H, Liang J, Gu Y. 2009. Feeding status of 0$ 2years infants in suburb of Guangzhou. Int Med Health Guidance News15:58.

Ji CY. 2008. The prevalence of childhood overweight/obesity and the epi-demic changes in 1985-2000 for Chinese school-age children and adoles-cents. Obes Rev 9:7881.

Ji Ch, Cheng TO. 2009. Epidemic increase in overweight and obesity inChinese children from 1985 to 2005. Int J Cardiol 132:110.

Jiang J, Urban R, Wang H, et al. 2007. Influence of grandparents on eat-ing behaviors of young children in Chinese three-generation families.

Appetite 48:377383.Kimberly KM, Marlene BH, Hatim AO. 2009. Overweight children and

adolescents: impact on psychological and social development. Int J ChildHealth Hum Dev 1:377384.

Krebs NF, Jacobson MS. 2003. Prevention of pediatrics overweight andobesity. Pediatrics 112:424430.

Laitinen J, Power C, Jarvelin MR. 2001. Family social class, maternalbody mass index, childhood body mass index, and age at menarche aspredictors of adult obesity. Am J Clin Nutr 74:287294.

Li H. 2005. How grandparents educated children in three-generation fam-ily. Preschool Res 6:2830 (in Chinese).

Li L, Zhao A, Xu R. 2007. The association between birth weight and obesityof infants and toddlers. Chin Maternal Child Health 22:12051206 (inChinese).

Li XY, Wu Z, Wang T, Xu L, Gao J. 2006. Caesarean section rate and itsinfluence factors in China. Chin J Public Health 22:12 (in Chinese).

Lu L. 2004. Advantage and disadvantage for grandparents caring. ChildFamily Education 10:68 (in Chinese).

Luo J, Hu FB. 2002. Time trends of obesity in pre-school children in Chinafrom 1989 to 1997. Int J Obes Relat Metab Disord 26:553558.

McMillen C, Robinson JS. 20 05. Developmental origins of the metabolic syn-drome: Prediction, plasticity, and programming. Phys Rev 85:571633.

Mei Z, Grummer-Strawn LM, Pietrobelli A, Goulding A, Goran M, DietzW. 2002. Validity of body mass index compared with other body composi-tion screening indexes for the assessment of body fatness in childrenand adolescents. Am J Clin Nutr 75:978985.

Nash A, Secker D, Corey M, Dunn M, OConnor DL. 2008. Field testing ofthe 2006 World Health Organization growth charts from birth to 2 years:

Assessment of hospital undernutrition a nd overnutrition rates and theusefulness of BMI. JPEN J Parenter Enteral Nutr 32:145153.

Neeltje Vogels, Danielle LA Posthumus, Edwin CM, Mariman, et al.2006. Determinants of overweight in a cohort of Dutch children. Am JClin Nutr 84:717724.

Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM.2006. Prevalence of overweight and obesity in the United States, 19992004. JAMA 295:15491555.

Oken E, Gillman MW. 2003. Fetal origins of obesity. Obes Res 11:496506.Ong KK, Loos RJF. 2006. Rapid infancy weight gain and subsequent obe-

sity: Systematic reviews and hopeful suggestions. Acta Paediatrica95:904908.

Rudolf MCJ, Levine R, Feltbower R, Connor A, Robinson M. 2006. TheTRENDS Project: Development of a methodology to reliably monitor theobesity epidemic in childhood. Arch Dis Child 91:309311.

Sanigorski AM, Bell AC, Kremer PJ, Swinburn BA. 2007. High childhoodobesity in an Australian population. Obesity 15:19081912.

Spurgeon D. 2002. Childhood obesity in Canada has tripled in past 20years. BMJ 324:1416.Stamatakis E, Primatesta P, Chinn S, Rona R, Falascheti E. 2005. Over-

weight and obesity trends from 1974 to 2003 in English children: Whatis the role of socioeconomic factors? Arch Dis Child 90:9991004.

Stettler N, Zemel BS, Kumanyika S, Stallings VA. 2002. Infant weightgain and childhood overweight status in a multi-center, cohort study.Pediatr 109:194199.

Stockton MB, Lanctot JQ, McClanahan BS, Klesges LM, Klesges RC,Kumanyika S, Sherrill-Mittleman D. 2009. Self-perception and bodyimage associations with body mass index among 8-10-year-old African

American girls. J Pediatr Psychol 34:11441154.Waterlow J, Buzina R, Keller W, Lane J, Nichaman M, Tanner J. 1977.

The presentation and use of height and weight data for comparing thenutritional status of groups of children under the age of 10 years. BullWorld Health Organ 55:489498.

Whitaker C, Wright JA, Pepe MS, Seidel KD, Dietz WH. 1997. Predictionobesity in young adulthood from childhood and parental obesity. N EnglJ Med 37:869873.

WHO Working Group on the Growth Reference Protocol and the WHOTask Force on Methods for the Natural Regulation of Fertility. 2002.Growth of healthy infants and the timing, type, and frequency of comple-mentary foods. Am J Clin Nutr 76:620627.

WHO. 1995. Physical status: The use and interpretation of anthropometry.Report of a WHO Expert Committee. World Health Organ Tech Rep Ser,854.

WHO. 2000. Obesity: Preventing and Managing the Global Epidemic.World Health Organization Technical Support Series No. 894. Geneva,Switzerland: World Health Organization.

WHO Multicentre Growth Reference Study Group, 2006. WHO childgrowth standards based on length/height, weight and age. Acta PaediatrSuppl 450:7685.

Yi M, Sun D, Zhou X. 2002. A Case-control Study on Association b etweenInfant Breastfeeding and Obesity Preschool Children. Acta AcademicaMed Qingdao Univ 38:206208.

Zheng K, Zhang X, Li L, Shu Y, Qiu L. 2008. Impact factors of breastfeed-ing during hospital stay after neonate birth in Hangzhou city. Chin J ofChild Health Care 16:716718 (in Chinese).

639INFANT BMI STATUS AND WEIGHT GAIN IN CHINA

American Journal of Human Biology