O

MBRS

AOrcaCDaSibuSaimwRMtaiodm

SaLNaoiCvglpC

MvB

A

2

RESEARCH

riginal Research

aternal Perceptions of Early Childhood Idealody Weight Differ among Mexican-Origin Mothersesiding in Mexico Compared to California

YLVIA GUENDELMAN, PhD, MSW; LIA C. H. FERNALD, PhD; LYNNETTE M. NEUFELD, PhD; ELENA FUENTES-AFFLICK, MD, MPH

tbnfCwltbsdtiJ

Dod(ft“aa12r(Sms

qcaossasrppsad

BSTRACTbjective To assess maternal perceptions of children’s cur-ent and ideal body sizes, and the meaning of and factorsontributing to overweight in infancy and early childhoodmong Mexican-origin mothers living in Mexico and inalifornia.esign A quali-quantitative study combining focus groupsnd a self-administered questionnaire.ubjects/setting A purposive sample of 84 low-income, Mex-can-origin mothers of 4- to 6-year-old children recruitedetween March 2006 and January 2008 from rural andrban communities in Mexico and California.tatistical analyses Bivariate, multivariate, and qualitativenalyses of maternal perceptions of children’s actual anddeal body size supplemented by qualitative analyses of

eaning of and factors contributing to childhood over-eight/obesity.esults Ideal child body size was considerably lower amongexican-origin mothers living in California (3.86�0.56)

han it was among mothers living in Mexico (4.32�0.83),nd this difference was significant (P�0.001) after adjust-ng for sociodemographic covariates. Among mothers ofverweight children, 82% of mothers in California wereissatisfied with their child’s weight compared with 29% ofothers in Mexico (P�0.003). Focus-group results suggest

. Guendelman is professor of Maternal and Child Healthnd of Community Health and Human Development, and. C. H. Fernald is associate professor of Public Healthutrition and Martin Sisters Chair in medical researchnd public health, Community Health and Human Devel-pment, University of California, Berkeley. L. M. Neufelds chief technical advisor, Micronutrient Initiative, Ottawa,anada; at the time of the study, she was director, Di-isión de Epidemiología de la Nutrición, Centro de Investi-aciones en Nutrición y Salud, Instituto Nacional de Sa-ud Pública, Morelos, México. E. Fuentes-Afflick isrofessor of Pediatrics and Epidemiology, University ofalifornia, San Francisco.Address correspondence to: Sylvia Guendelman, PhD,SW, Community Health and Human Development, Uni-

ersity of California, Berkeley, 207-J University Hall,erkeley, CA 94720-7360. E-mail: sylviag@berkeley.eduManuscript accepted: August 13, 2009.Copyright © 2010 by the American Dietetic

ssociation.0002-8223/10/11002-0004$36.00/0

doi: 10.1016/j.jada.2009.10.033

22 Journal of the AMERICAN DIETETIC ASSOCIATION

hat these differences in the perception of children’s idealody size can be attributable to differences in body sizeorms among mothers and awareness of the negative ef-

ects of obesity that occur after migration to California.onclusions Maternal perceptions of early childhood over-eight appear to differ among Mexican-origin women

iving in Mexico and California. Recognition of the nega-ive health consequences of obesity and identification ofarriers to achieving weight control are important firstteps toward childhood obesity prevention. Interventionsirected at Mexican-origin mothers should focus on cul-urally acceptable ways of transmitting weight-controlnformation.

Am Diet Assoc. 2010;110:222-229.

evelopment of overweight begins in early childhood(1,2). Preschool- and elementary-aged children whoare overweight are five times more likely to become

verweight adolescents (3). Among 2- to 5-year-old chil-ren in the United States, the prevalence of overweightbody mass index [calculated as kg/m2] �95th percentileor age) is 13.9% (4). Among Mexican-American children,his prevalence is 19.2% and an additional one-third areat risk for overweight” (body mass index for age �85thnd �95th percentile) (4). Prevalence of overweightmong young children in Mexico is also high (5-7). The999 Mexican National Nutrition Survey indicated that7.5% of children aged 2 to 5 years were classified as atisk of overweight or overweight based on US definitions8). By 2006, the Mexican National Health and Nutritionurvey reported that this prevalence had increased toore than one third among 5- to 11-year-olds, and was

lightly lower in 2- to 5-year-olds (9).Parents play an important role in the quantity and

uality of food consumed and activity patterns of theirhildren by controlling the amount and types of foodvailable, the context and schedule of eating, and thepportunities for physical activity (10-12). From thistandpoint, parental perceptions of children’s ideal bodyize can influence child-feeding practices and physicalctivity. Although the determinants of overweight/obe-ity in Mexican-origin children are not well-understood,esearchers have postulated that cultural factors are im-ortant (5). Anecdotal evidence suggests that feedingractices and weight perceptions might contribute to obe-ity in Mexican-origin children in the United States (13)nd parents who prefer a larger body size for their chil-ren might feed them more (14).

Two US studies have examined perceptions of weight

© 2010 by the American Dietetic Association

sfinsmMroMd(poPsMnaoit

woosncsaio

rwCcrC

MAcmalcmbccg(dfpitcra

o

lCaiegcwraUqsdvUN

tcjNSrttr(

mdrosttbi

SWwstMvdglWfdzcwpwtlmfid

tatus by Latino parents of their preschool children. Therst study reported that 35% of Latino parents of diverseational origins did not believe their overweight childhould be labeled “overweight” (15), while only 65% ofothers in the other study (which included mothers ofexican-origin and white and black non-Latinas) cor-

ectly identified that their children could be defined asverweight (16). In a small study conducted in Veracruz,exico, half of the parents of obese school-aged children

id not identify their children as having a weight problem17). Overweight in early childhood may not be viewed asroblematic if parents believe that their child looks goodr that children can outgrow their high body mass (18).arents might not accurately perceive their child’s weighttatus despite health providers’ educational efforts (19).ore research is needed to identify why some parents do

ot recognize excess weight in their preschoolers and tossess variations in weight perceptions by cultural orther social-environmental factors. Understanding thesessues is critical to designing successful obesity-preven-ion interventions for Mexican-origin children.

To assess maternal perceptions of their children’seight and investigate factors associated with childhoodverweight, we conducted focus groups with Mexican-rigin mothers who had children 4 to 6 years old. Weelected this age group because it is a particularly vul-erable period for development of fatty tissue (20); be-ause children have not yet entered the formal schoolingystem, where external factors influence their behavior;nd because there is a lower prevalence of distorted bodymages and eating disorders, which is important givenur interest in examining perceived body weight (21).Study participants included low-income mothers from

ural and urban areas in Mexico and Mexican immigrantomen who resided in rural and urban communities inalifornia. We postulated that maternal perceptions of

hildren’s body size would be comparable in urban andural sites within each country, but would vary betweenalifornia and Mexico.

ETHODSqualitative study was designed to assess maternal per-

eptions of their child’s actual and ideal body size; what iteans to be overweight during infancy and childhood;

nd factors contributing to childhood overweight amongow-income women. Eight focus-group discussions wereonducted with a purposive sample of 84 Mexican-originothers who were recruited from urban and rural sites

etween March 2006 and January 2008. In the San Fran-isco Bay area and Salinas Valley, mothers were re-ruited through the Special Supplemental Nutrition Pro-ram for Women, Infants, and Children and in MexicoJalisco and Guanajuato) through Oportunidades, a con-itional cash transfer program that provides money andood supplements for low-income young children andregnant and lactating women. Selected women residedn the study area, were Mexico-born, and had a child inhe specified age group at the time of recruitment. Uponompleting the focus group, participants in Californiaeceived $50 in cash and participants in Mexico receivedsmall gift, worth about US$5.Focus-group discussions were conducted in Spanish;

ccurred in clinics, community centers, or schools, and

asted 2 to 21⁄2 hours. Child care was provided on site. Inalifornia, discussions were facilitated by the first authornd in Mexico by an experienced social worker. The facil-tators, both trained in qualitative methods and experi-nced in focus-group methodology (22,23), followed auide that explored four content areas: what makes ahild grow healthily; mothers’ perceptions about over-eight in infants and young children; parental behaviors

egarding the type and amount of food that children eat;nd what changes occur when people immigrate to thenited States (see Results section for exact wording ofuestions relevant to this analysis). Focus-group discus-ions were audiorecorded, transcribed, and followed by aebriefing. The study was approved by Institutional Re-iew Boards at the University of California, Berkeley;niversity of California, San Francisco; and the Institutoacional de Salud Pública in Mexico.Quantitative data were collected from a self-adminis-

ered questionnaire that asked about the mother’s andhild’s ages and maternal educational attainment. Sub-ective social status was assessed using the MacArthuretwork on Socio-Economic Status and Health Scale ofubjective Social Status, which is a drawing of a ladderepresenting a hierarchy on which respondents are askedo place themselves socioeconomically or socially in rela-ion to members of their self-defined community or inelation to people living in the country where they live24).

A focus-group activity was conducted that used theodified Stunkard Body Rating Scale adapted for chil-

ren, a visual scale of seven male and female silhouettesepresenting body images that range from very thin tobese (25,26). Body rating figures have been shown to betrongly correlated with body mass index percentiles ineens (27). Each mother was asked to identify the figurehat best approximated her child’s appearance (actualody size) and then to identify the figure that showed thedeal body size for the same child.

tatistical Analysise used STATA version 9.2 (2006, Stata Statistical Soft-are, StataCorp LP, College Station, TX) to examine

ample characteristics and examine maternal percep-ions of their children’s actual and ideal body sizes.eans and standard deviations were computed for each

ariable by study site (Mexico or California). We con-ucted tests of difference using multivariate linear re-ression that controlled for child age and sex, rural/urbanocality, maternal education, and subjective social status.

e then calculated a score (actual weight subtractedrom ideal weight), and labeled the variable “maternalissatisfaction” if the difference score was different fromero (ie, if the mothers indicated that they wanted theirhildren to be either heavier or lighter than they actuallyere). Using tests of proportions, we compared the pro-ortion of mothers in both groups (California vs Mexico)ho were dissatisfied with their child’s weight. Half of

he sample was missing data on maternal age because ofack of self-report and so we excluded this variable from

ultivariate analyses. Additional analyses were strati-ed by urban/rural locality and did not yield substantialifferences; these findings are not reported.

We analyzed focus-group transcripts to identify the

February 2010 ● Journal of the AMERICAN DIETETIC ASSOCIATION 223

tsdn

RCa(lerP

QIdrPifiisCaocmvbevi

wawm

ustowcPwd2

AWoMlmb“wfifi

1

2

3

Ctfdatemcgwa

m

2

hemes that were discussed most frequently or exten-ively. Transcripts from each site were reviewed indepen-ently and compared across sites by all authors. Keyarratives were extracted verbatim.

ESULTShildren’s age, sex, and maternal age did not differmong participants in Mexico (n�32) and Californian�52) (Table). Compared with Mexican-origin mothersiving in California, those sampled in Mexico had lowerducational attainment (5.4 vs 7.5 years; P�0.004) andeported higher perceived social status (11.8 vs 8.9;�0.001).

uantitative Analyses of Child’s Body Sizen bivariate analyses, maternal perceptions of their chil-ren’s current actual body size did not differ by country ofesidence (Mexico: 4.24�1.00 vs California: 3.85�0.96;�0.08, Figure 1); these differences were also not signif-

cant with the inclusion of covariates (���.36, 95% con-dence interval: �0.89 to 0.17; P�0.2). Mothers in Mex-

co, however, preferred a significantly larger ideal bodyize for their children (mean�4.32�0.83) than mothers inalifornia (mean�3.86�0.56; P�0.004). In multivariatenalyses, these results remained consistent; Mexican-rigin mothers living in California preferred a signifi-antly lower ideal body size for their children than didothers living in Mexico (���.57, 95% confidence inter-

al: �0.92 to �0.23; P�0.001), while controlling for ur-an/rural residence, sex of child, age of child, maternalducation, and maternal perceived social status. No co-ariates were independently associated with perceiveddeal body size for the child.

Among Mexican-origin mothers in California, 44%ere dissatisfied with their child’s weight (ie, their child’sctual perceived weight was higher or lower than theeight they perceived to be ideal) compared with 30% of

Table. Selected sociodemographic characteristics of Mexican-ori-gin mothers participating in focus groups in Mexico and California

Mexico (n�32) California (n�52) P value

4™™™™ mean (standard deviation) ™™™3Mother’s age (y)a 30.7 (5.7) 32.1 (6.1) 0.49Mother’s education

(y) 5.4 (3.1) 7.5 (3.4) 0.004Mother’s perceived

social statusb 11.8 (3.4) 8.9 (3.1) �0.001Child’s age (y) 4.7 (1.3) 5.0 (1.3) 0.24

4™™™™™™™™™™™™ % ™™™™™™™™™™™™3Child’s sex female 61.1 56.3 0.66Living in rural area 42.6 51.5 0.42

aBecause of missing data, the sample size for mother’s age was 31 in California and12 in Mexico.bAssessed with the MacArthur Scale of Subjective Social Status, which asks eachrespondent to rank him/herself on two ladders representing social hierarchy withinhis/her community and within his/her society. Scores on these two ladders weresummed for a total score range from 0 to 20, with a higher score signifying a higherperceived social status and a lower score representing lower status.

others in Mexico; this difference was not significant l

24 February 2010 Volume 110 Number 2

sing a test of proportions (P�0.19). When restricting theample to just those children whose mothers perceivedhat their actual weight was overweight (ie, ranked as �5n the perceived weight scale), mothers in Californiaere significantly more likely to be dissatisfied with their

hild’s weight than mothers in Mexico (82.3% vs 28.6%;�0.002). Among children judged to have an actualeight that was ranked �4, there were equal numbers ofissatisfied mothers in each group (31.6% in Mexico and7.0% in California; P�0.72).

nalyses of Focus-Group Questionshat Do You Think and Feel When You See an Overweight Baby

r Child?exico. Among rural and urban participants, three re-

ated themes emerged (Figure 2, column 1). Among theothers in the focus groups, the concept of an overweight

aby was interpreted as synonymous with a large orchubby” baby and viewed as acceptable, while over-eight in children was perceived more negatively. Therst two themes related specifically to babies and thenal to children.

. “A bigger baby is a better baby.” Mothers in Mexicofavored chubbier babies who were perceived as“healthy, cute, and growing appropriately.” Severalmothers believed that chubby babies were geneticallypredisposed to be plump, rather than attributable toparenting practices. Only a few mothers in Mexicostated that they disliked overweight babies and thatthis condition needed to be prevented.

. “Babies outgrow their fat.” Mothers in Mexico repeat-edly mentioned that “baby fat does not pose a healthproblem” or “fat babies tend to grow out of it.” Severalparticipants cautioned that “diets are not good forchildren’s growth and development,” and disapprovedof a physician’s recommendation of food restriction.

. “Overweight children look bad, but not that bad.” Ex-cess weight in children was viewed more negativelythan in babies: “overweight children look bad or funnyand cause distress.” Nonetheless, several mothersfailed to view overweight in children as a problem,mentioning that “some fat kids really do not look thatfat,” or “considering their age, they are fine.” Somemothers seemed forgiving of overweight children, par-ticularly if their own children were overweight.

alifornia. Among California participants, the samehemes emerged among mothers in rural sites and theyrequently expressed that “some babies and young chil-ren are chubby and healthy, while others are chubbynd unhealthy” (Figure 2, column 2). Despite commentshat suggested knowledge of the negative health andmotional effects of infant and childhood overweight,any mothers thought that excess weight was of little

onsequence because “baby fat is baby fat . . . they out-row it.” Furthermore, several mothers discussed excesseight in the context of thinness, claiming that thinnesslso poses health problems.In urban sites in California, the opinions expressed byothers were quite different. None of the mothers be-

ieved that babies would outgrow their fat as they got

owmeasasulcl

dc

W

Mtamhd

FC hn Wi

lder and very few agreed with the perception that over-eight youngsters “look good” or “not too bad.” Severalothers qualified that “if a six-to eight-month old baby is

ating too much, we feel this is nice (“que bonito”); butfter this age this is not right but [somehow] we continueaying “que bonito!” implying that this is the culturallycceptable response. Mothers admitted that people as-ume thin babies are not fed enough or they have a bad ornloving mother. The discussion in urban areas focused

ess on the way the child looked and more on fears andonsequences of childhood overweight because obesity

igure 1. Maternal perception (%) of child’s actual and ideal body sizealifornia (n�52). Figure drawings reprinted with permission from Jo

eads to “sickness, child suffering, heart disease, and c

epression” and “to breaking a mother’s back when youarry them.”

hat Makes Children Become Overweight?

exico. Most mothers in rural and urban Mexico thoughthat children gain weight because they eat excessivelynd/or the wrong foods. Urban participants, particularlyothers who rated their child as being thin or having a

ealthy weight, blamed the parents of overweight chil-ren for feeding them the wrong foods or excessively. In

rding to figure drawings developed by Collins (25). Mexico (n�31) andley & Sons, Inc (25).

acco

ontrast, mothers in rural communities commented that

February 2010 ● Journal of the AMERICAN DIETETIC ASSOCIATION 225

c‘oaOcCwtRwcom

CT

iim

ihfrmW

Fc

2

hildren grow fat because of stress: “It is a problem ofnerves’ and of tensions that arise from being mistreatedr being made fun of.” Others acknowledged that geneticsnd parental body mass influence children’s body mass.nly one mother acknowledged the importance of exer-

ise: “my son has a bicycle and that helps to keep him fit.”alifornia. Mothers in rural and urban areas of Californiaere more likely to mention a lack of exercise as a con-

ributing factor, in addition to a poor diet and excess food.ural mothers believed that children in their communityatch too much TV, play too many video games, and are

rowded into small apartments: “in Mexico people don’twn cars, many don’t have phones, so you have to

Mexico (Column 1)

Theme 1. A bigger baby is a better baby● Chubby babies look cute. (R,U)● They look healthy and good. (R,U)● My child was born preterm. He was so small and thin that we

thought he would die. Now he is heavy, but he is also big. (R)● My baby was big at birth and the doctor wanted to do somethin

about it until she saw that we [the parents] are tall. (U)● Chubbiness does not pose a problem to the organism. (U)● It’s because of colostrom and the rich breast milk. (U)

Theme 2. Babies outgrow their fat● My children as infants were chubby; now they are thin. (R)● Babies grow and become thinner. (U)● Children should not be placed on a diet before age 6 because

they get blemishes and diets are not good for children’s growthand development. (U)

Theme 3. Overweight children look bad, but not that bad● Chubby infants look cute, but once they grow they are no longer

cute. (R)● Some overweight children are not too fat. (R)● I cannot say they look good or bad because they are my

children. (R)● One day I felt really bad because someone told me that my

daughter was thin because I did not feed her enough. (R)● Does the overweight have to do with “nerves” or with her not

feeling right? (R)● Being fat does not mean being sick; it has to do with the

parents’ complexion. (R)● Some eat too much junk food. (R)● Overweight children eat too much food. (R,U)● Overweight children look bad, they cause distress. (R,U)● Overweight children look bad, they look funny. (R,U)● Considering their age, they are fine, they are healthy. (U)● An overweight child looks full because (s)he eats right and is

healthy. (U)● They are healthy, they eat right. (U)● Overweight children are not fed right. (U)● My sister thinks that it is alright for her daughter to be fat

because she looks like her. (U)

igure 2. Selected quotes related to maternal perceptions of infant andhildren of Mexican mothers living in Mexico or California.

ove. . . .”; in contrast, “everything is done by car [in b

26 February 2010 Volume 110 Number 2

alifornia]; even children are pushed in their strollers!hat is why you see so much obesity.”Participants from rural California perceived that fam-

ly separations create maternal stress, which can triggernadequate eating: “when you are sad you are not in the

ood to cook, you just give your kids any old thing. . . .”Mothers in California and Mexico agreed that families

n the United States are smaller, purchasing power isigher, and food scarcity is less common, which results inamilies eating bigger portions and going to fast-foodestaurants more frequently: “in California there is souch to eat, that is why we are the way we are . . . big.”here Is It Easier to Control Your Child’s Weight? Mothers in

alifornia (Column 2)

Some chunky babies are healthy, others not. (R)One child I know is heavy and pretty. (R)Some fat babies are full of life. (R)How cute! (R,U)If they are thin, they think the child is sick or mom does notfeed him. (R,U)A baby that eats too much is cute but not later. (U)It is not right for the baby to be plump. (U)

Baby fat is just baby fat, they outgrow it. (R)My babies were chubby but since they were a year old, they arethin. (R)My daughter is too thin and that worries me. When I see anobese infant, I think he is doing fine. (R)It is not ok to have a fat baby because it hurts mom’s back. (U)Babies that are chubby stay that way. (U)

Overweight children do not look good. (R)There are chubby children that are healthy. (R)Our grandparents taught us that chubby children look better. (R)Chunky children do not grow well, they feel bad, don’t sleep welland develop diabetes and other health problems. (R)Fat children tire more easily. (R)Obesity comes from the family; it is inherited. (R)Fat children develop flab and they don’t look good. (R)My girl since birth has been overweight and she worries me. (R)Even if they are fat, they should continue eating. (R)Obesity brings illness including heart attacks, high cholesterol,poor self esteem. (R)Poor child, they should put him on a diet. (U)Poor child, he will suffer. (U)As a parent, you suffer when you see overweight children. (U)Overweight is not healthy; children can’t move, they get diabetes.(U)Some parents overfeed their kids or don’t make them doexercise, that is why they are fat. (U)It could also be nerves. (U)Poor child, he is not at fault; it is his metabolism, just like me,he gains weight easily. (U)

overweight among rural (R) and urban (U) mothers of 4- to 6-year-old

C

g

●●●●●

●●

●●

●

●●

●●●●

●●●●●●

●●●●

●

●●

child

oth California and Mexico perceived weight manage-

mhhUtsfwi“i

irecltnthr

DOCociciscfmwt

stsrmaidmlhppymwtof

dnaMi

wih(lsfdstcbrhh

(wtpbcrmMp

CbawiawasdbprMclfffodfc

kttstsl

spp

ent as easier in Mexico because they cook fresh meals atome and walk more: “Mexican people cook, they don’tave money for restaurants, but fast food is cheap in theS.” Farm-worker mothers in rural California explained

hat children eat microwaved food during the busy workeason. As one mother explained, “we don’t come backrom the fields before 7 pm and the kids eat chips orhatever is easy.” Mothers in urban California further

dentified the contribution of their partners, who oftengive them anything they ask for . . . hamburgers, chips,ce cream.”

Although mothers in Mexico and California agreed thatt is easier to control children’s diets in Mexico, severalecognized unhealthful aspects of the Mexican diet: “Weat far too many tortillas. . . .” “and if you leave yourhildren with their grandparents, they are even moreikely to be fed lard and fatty foods.” Several mothers felthat there is a lot of information about nutrition that theyeed to unlearn: “We tell our kids early on that they haveo finish what is on their plates, even if they are notungry” . . . “we often push them [to eat] too much”; “weeally never were taught how to eat.”

ISCUSSIONur analyses of 84 low-income Mexican-origin mothers inalifornia and Mexico suggest that maternal perceptionsf children’s current actual body size do not differ byountry of residence, although there are large differencesn perceptions of what is ideal for children’s body size. Inomparison with Mexican women living in Mexico, Mex-can-origin mothers living in California expressed atronger preference for a smaller ideal body size for theirhildren, and these differences persisted after controllingor sociodemographic covariates. Furthermore, amongothers of overweight children, mothers in Californiaere almost four times as likely to be dissatisfied with

heir child’s weight compared with mothers in Mexico.Our qualitative analysis of eight focus-group discus-

ions rendered two testable hypotheses about the factorshat influence children’s ideal body size. One hypothesisuggests that differences in ideal body size may be aesult of a shift in body-size norms that occurs afterigration from Mexico to California. Evidence from

dults suggests that ethnicity protects against internal-zation of a “thin ideal” (28), but that this associationepends on acculturation level (29,30). In our study,others living in Mexico were more likely than those

iving in California to express traditional beliefs thateavier infants are healthier and reflect better parentingractices, and they were more accepting of overweight,erhaps because of their belief that infants and veryoung children outgrow excess weight. In contrast, urbanothers in California were likely to believe that lesseight is reflective of better lifestyle and parenting; from

heir perspective, infants and children do not necessarilyutgrow their excess weight. Mothers in rural Californiaell in between these two sets of beliefs.

Another hypothesis that emerged from the focus-groupiscussions was that Mexican-origin mothers in Califor-ia may be more aware of childhood obesity and its neg-tive health consequences than their counterparts inexico. Moreover, fewer mothers in California, especially

n urban sites, perceived that their children’s body size p

as an inherited (unmodifiable) trait. Obesity campaignsn California have underscored the high rates of child-ood overweight and negative health consequences14,31). In rural Mexico, this sense of urgency may beacking among the existing public health messages, de-pite the fact that large urban centers are beginning toocus on obesity-related factors associated with chronicisease (14). Furthermore, the coexistence of obesity,tunting, and micronutrient deficiencies in Mexico, some-imes within the same family, can affect body-size per-eptions among Mexican mothers (6). Some researcherselieve that the persistence of undernutrition in Mexicoeinforces traditional beliefs that heavier children areealthier and can lead to less recognition of obesity as aealth issue (17).Previous studies among low socioeconomic families

32,33) and Latina mothers in the United States, some ofhom are of Mexican-origin (15,18), have reported a ma-

ernal preference for heavier children and a failure toerceive overweight. The Mexican women in our studyelieved that children with higher body mass are per-eived to be healthy, which is consistent with anecdotaleports (14), yet our study delved further by comparingaternal perceptions of body size between women inexico and California and by documenting differences in

erceptions of ideal body size.Our findings also indicate that the mothers sampled in

alifornia reported less satisfaction with their children’sody sizes than mothers sampled in Mexico, but onlymong mothers of children who were considered over-eight. This finding requires additional research compar-

ng maternal perceptions of child’s body size with child’sctual body size. In a previous study of low-incomeomen in New York City, body size dissatisfaction wasssociated with more healthful diets (34). Whether dis-atisfaction with children’s body sizes leads to stricterietary practices and more focus on weight loss has noteen established; research is required to test such hy-otheses in Mexican-origin mothers. The focus group nar-atives identified multiple challenges that confrontedexican mothers sampled in California. Increased pur-

hasing power in the United States was reported to trans-ate into an abundance of food and greater consumption ofast foods and food prepared outside the home. Smalleramilies in the United States were reported to allow moreood consumption per person. Moreover, the combinationf a smaller social support network at home, increasedepression, and incorporation of mothers into the work-orce was identified as relating to less home cooking andontrol over their children’s diet in the United States.Mexican-origin mothers sampled in California ac-

nowledged a tendency to follow a combination of oldraditions and new dietary practices, and they recognizedhat some traditional practices might not be healthful,uch as cooking with too much fat or eating too manyortillas. Several participants indicated that they under-tand the need to undo deeply ingrained health beliefs,ike making children eat even when they are not hungry.

Our findings should be interpreted cautiously. Ourample was small and although we recruited from sup-lemental food programs to allow for cross-country com-arisons, these programs target low socioeconomic status

opulations and provide nutrition education. Hence the

February 2010 ● Journal of the AMERICAN DIETETIC ASSOCIATION 227

rIrftwoqgiwa

mlrtactmanAamsd(qos

otaimirifistcfTtahftntfii

SNt

fi

SBafn

mC

R

1

1

1

1

1

1

1

1

1

1

2

2

2

esults are not generalizable to all low-income mothers.n particular, both groups of mothers were recruited di-ectly from programs (the Supplemental Food Programor Women, Infants and Children and Oportunidades)hat included elements of nutrition education. Thus, theomen included in the analysis may be more aware thanther low-income mothers about the risks for and conse-uences of obesity. Furthermore, the differences betweenroups that we are reporting relating to maternal dissat-sfaction may reflect the quality and quantity of what theomen have learned in their nutrition-education sessionsnd might not be attributable to country of residence.Another limitation is that we did not obtain anthropo-etric measurements to assess actual body size and re-

ied on Stunkard Body Rating Scales developed for whiteather than Mexican-origin populations. It is possiblehat mothers in California were more aware of sociallycceptable responses because of their exposure to obesityampaigns and this bias could have led them to reporthinner ideal body sizes for their children compared toothers in Mexico. Previous studies have documented an

ssociation between maternal and child obesity in whiteon-Latinos and Latinos from Mexico and Central/Southmerica due to shared genetic history, behavioral norms,nd parental attitudes; we did not assess maternal bodyass (2,35). Furthermore, because of the small sample

izes, we were unable to examine differences by child sex,espite evidence suggesting notable differences by sex36). Strengths of our study include the use of a quanti-ualitative methodology to assess maternal perceptionsf child body size among Mexican-origin women on bothides of the border.In conclusion, among sampled Mexican-origin mothers

f 4- to 6 year-old children in this study, we found thathose in California and Mexico perceived their child’sctual body size very close to what they perceived as thedeal, but these values overall were lower among Mexican

others in California when compared with those in Mex-co. We hypothesize that differences in ideal body size areelated to variation in cultural norms and /or access tonformation about the negative effects of obesity in Cali-ornia vs Mexico and that these influence perceptions ofdeal and actual child weight among mothers. Futuretudies are needed to explore these hypotheses in bina-ional populations. In California and Mexico, strong edu-ation is required to create awareness of the health ef-ects of excess weight in childhood and how to identify it.his effort should include health education messages,raining, and community support. Health care providersnd policy makers must find innovative ways to reinforceealthful beliefs, including messages geared toward otheramily members, as well as culturally acceptable ways toransmit weight-control messages. The recognition of theegative health consequences of obesity and identifica-ion of barriers to achieving weight control are importantrst steps toward prevention of childhood obesity in Mex-

can-origin children.

TATEMENT OF POTENTIAL CONFLICT OF INTEREST:o potential conflict of interest was reported by the au-

hors.FUNDING/SUPPORT: We gratefully acknowledge the

nancial support of the Robert Wood Johnson Health and

28 February 2010 Volume 110 Number 2

ociety Scholars Program at the University of Californiaerkeley and the University of California, San Francisco,s well as the Center for Health Research and the Centeror Latin American Studies at the University of Califor-ia, Berkeley.ACKNOWLEDGEMENTS: We thank Rosalinda Do-inguez for assistance with focus groups and Patriciarawford, PhD, for her initial suggestions.

eferences1. Kimbro RT, Brooks-Gunn J, McLanahan S. Racial and ethnic differ-

entials in overweight and obesity among 3-year-old children. Am JPublic Health. 2007;97:298-305.

2. Whitaker RC, Wright JA, Pepe MS, Seidel KD, Dietz WH. Predictingobesity in young adulthood from childhood and parental obesity.N Engl J Med. 1997;337:869-873.

3. Nader PR, O’Brien M, Houts R, Bradley R, Belsky J, Crosnoe R,Friedman S, Mei Z, Susman EJ. Identifying risk for obesity in earlychildhood. Pediatrics. 2006;118:e594-e601.

4. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, FlegalKM. Prevalence of overweight and obesity in the United States, 1999-2004. JAMA. 2006;295:1549-1555.

5. Kaiser L, Melgar-Quinonez H, Lamp C, Johns M, Sutherlin J, Har-wood J. Food security and nutritional outcomes of pre-school ageMexican-American children. J Am Diet Assoc. 2001;103:924-929.

6. Rivera JA, Barquera S, González-Cassío T, Olaiz G, Sepúlveda J.Nutrition transition in Mexico and other Latin American countries.Nutr Rev. 2004;62:S149-S157.

7. Sanchez-Castillo CP, Lara JJ, Villa AR, Aguirre J, Escobar M, Guti-errez H, Chavez A, James WP. Unusually high prevalence rates ofobesity in four Mexican rural communities. Eur J Clin Nutr. 2001;55:833-840.

8. Hernandez B, Cuevas-Nasu L, Shamah-Levy T, Monterrubio E,Ramirez-Silva CI, Garcia-Feregrino R, Rivera JA, Sepulveda-Amor J.Factors associated with overweight and obesity in Mexican school-agechildren: Results from the National Nutrition Survey 1999. SaludPublica Mex. 2003;45(suppl 4):S551-S557.

9. Olaiz-Fernández G, Rivera-Dommarco J, Shamah-Levy T, Rojas R,Villalpando-Hernández S, Hernández-Avila M, Sepulveda-Amor J.Encuesta Nacional de Salud y Nutrición 2006 [National survey ofhealth and nutrition 2006]. Cuernavaca: Instituto Nacional de SaludPública. http://www.insp.mx/ensanut/. Accessed December 12, 2008.

0. Birch LL, Fisher JA. Appetite and eating behavior in children. PediatrClin North Am. 1995;42:931-953.

1. Dewey KG. Is breastfeeding protective against child obesity? J HumLact. 2003;19:9-18.

2. Melgar-Quiñonez HR, Kaiser LL. Relationship of child-feeding prac-tices to overweight in low-income Mexican-American preschool-agedchildren. J Am Diet Assoc. 2004;104:1110-1119.

3. Garcia RS. No come nada: A Mexican American pediatrician calls fornationwide backup in fighting childhood obesity among his patients.Health Aff. 2004;23:215-219.

4. Institute of Medicine. Joint US-Mexico Workshop on Preventing Obe-sity in Children and Youth of Mexican Origin: Summary. Washington,DC: The National Academies Press; 2007.

5. Myers S, Vargas Z. Parental perceptions of the preschool obese child.Pediatr Nurs. 2000;26:23-30.

6. Maynard LM, Galuska DA, Blanck HM, Serdula MK. Maternal per-ceptions of weight status of children. Pediatrics. 2003;111(Pt 2):1226-1231.

7. Brewis A. Biocultural aspects of obesity in young Mexican schoolchildren. Am J Hum Biol. 2003;15:446-460.

8. Crawford PB, Gosliner W, Anderson C, Strode P, Becerra-Jones Y,Samuels S, Carroll AM, Ritchie LD. Counseling Latina mothers ofpreschool children about weight issues: Suggestions for a new frame-work. J Am Diet Assoc. 2004;104:387-394.

9. Serdula MK, Collins ME, Williamson DF, Anda RF, Pamuk E, ByersTE. Weight control practices of US adolescents and adults. Ann InternMed. 1993;119(Pt 2):667-671.

0. Rolland-Cachera MF, Deheeger M, Bellisle F, Sempé M, Guilloud-Bataille M, Patois E. Adiposity rebound in children: A simple indica-tor for predicting obesity. AJCN. 1984;39:129-135.

1. Ricciardelli LA, McCabe MP. Children’s body image concerns and

eating disturbance: A review of the literature. Clin Psychol Rev.2001;21:325-344.

2

2

2

2

2

2

2

2

3

3

3

3

3

3

3

2. Morgan DL. Focus Groups as Qualitative Research. Thousand Oaks,CA: Sage Publications; 1997.

3. Wilkinson S. Focus groups in health research: Exploring the mean-ings of health and illness. Health Psychol. 1998;3:329-348.

4. Adler N, Epel E, Castellazzo G, Ickovics J. Relationship of subjectiveand objective social status with psychological and physical health inhealthy white women. Health Psychol. 2000;19:586-592.

5. Collins ME. Body figure perceptions and preferences among preado-lescent children. Int J Eat Disord. 1991;10:199-208.

6. Maximova K, McGrath JJ, Barnett T, O’Loughlin J, Paradis G, Lam-bert M. Do you see what I see? Weight status misperception andexposure to obesity among children and adolescents. Int J Obes.2008;32:1008-1015.

7. Peterson M, Ellenberg D, Crossan, S. Body-image perceptions: Reli-ability of a BMI-based silhouette matching test. Am J Health Behav.2003;27:355-363.

8. Warren CS, Gleaves DH, Cepeda-Benito A, Fernandez M, Rodriguez-Ruiz S. Ethnicity as a protective factor against internalization of a

thin ideal and body dissatisfaction. Int J Eat Disord. 2005;37:241-249.

9. Cachelin FM, Phinney JS, Schug RA, Striegel-Moore RH. Accultura-

tion and eating disorders in a Mexican-American community sample.Psychol Women Q. 2006;30:340-347.

0. Chamorro R, Flores-Ortiz Y. Acculturation and disordered eatingpatterns among Mexican American women. Int J Eat Disord. 2000;28:125-129.

1. Miller JL, Silverstein JH. Management approaches for pediatric obe-sity. Nat Clin Pract Endocrinol Metab. 2007;3:810-818.

2. Baughcum AE, Chamberlin LA, Deeks CM, Powers SW, WhitakerRC. Maternal perceptions of overweight preschool children. Pediat-rics. 2000;106:1380-1386.

3. Eckstein KC, Mikhail LM, Ariza AJ, Thomson JS, Millard SC, BinnsHJ. Parents’ perceptions of their child’s weight and health. Pediatrics.2006;117:681-690.

4. Fuentes-Afflick E, Hedesol NA. Overweight in young Latino children.Arch Med Res. 2008;39:511-518.

5. Contento IR, Basch C, Zybert P. Body image, weight, and food choicesof Latina women and their young children. J Nutr Educ Behav.2003;35:236-248.

6. Hall SK, Cousins JH, Power TG. Self-concept and perceptions of

attractiveness and body size among Mexican-American mothers anddaughters. Int J Obes (Lond). 1991;15:567-575.

February 2010 ● Journal of the AMERICAN DIETETIC ASSOCIATION 229