Overweight and Obesity in Early Childhood A Systematic ... · (e.g., genetic risk factors). The purpose of this review is to examine risk factors for childhood obesity with the goal

Overweight and Obesity in Early Childhood A Systematic Review ofIndividual, Family, and Peer Risk Factors

Nadja Frate*, Brigitte Jenull, Heather M. ForanDepartment of Health Psychology, Alpen-Adria-Universität, Klagenfurt

Abstract

Purpose: The prevalence of children who are overweight or obese has been increasing worldwide. While avariety of biological as well as socio demographic correlates have been identified and reviewed, asystematic review of psychosocial factors, particularly among preschool aged children, is lacking. Thissystematic review synthesizes the research on individual, family and peer risk factors for overweight andobesity in preschool aged children.

Method: A systematic review of the recent literature on psychosocial factors and overweight or ratherobesity in the early stages of childhood was conducted.

Results: A total of 27 studies from 2011-2016 were identified that examined individual, family and socialrisk factors for obesity in children. Results indicate the importance of eating regulation as well as familyfactors in understanding early childhood risk for obesity. There was mixed support for associationsbetween behavioral and emotional symptoms and obesity among this age group. For other risk factorsexamined, too few studies exist to be able to make strong conclusions about their relevance forunderstanding preschool obesity risk.

Conclusions: Psychosocial factors are associated with overweight and obesity among preschool agedchildren. However, this review also highlights the dearth of research on several potentially important riskfactors for childhood obesity in this age group (such as family violence, parental and peer relationships).Longitudinal studies, which examine multiple risk factors simultaneously over this importantdevelopmental period, are sorely needed.

Citation: Frate N*, Jenull B, Foran HM, (2018) Overweight and Obesity in Early Childhood A Systematic Review of Individual, Family, and Peer Risk Factors. Adv Pediatr Res 5:5. doi:10.12715/apr.2018.5.5

Received: May 16, 2018; Accepted: May 23, 2018; Published: May 31, 2018

Copyright: © 2018 Bosteels et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Competing interests: The authors have declared that no competing interests exist.

Sources of funding: There is no source of funding for this article.*E-mail: [email protected]

Keywords:Overweight and obesity, Preschool age, Childhood,Psychological aspects, Risk factors

IntroductionThe prevalence of overweight and obesity hasdramatically increased during the last three decades– not only for adults, but for children as well [1].The greatest weight gain has been among childrenaged 6-12. Accordingly to the WHO, 10.1% ofboys and 10.5% of girls aged 4-6 are overweight,10.2% of boys as well as 7.9% of girls of the same

age group are obese. This means 20.3% of boys and18.4% of girls have a BMI >25 kg/m². For boysaged 6-9 this value increases to 29.3% and for girlsto 25.7%. Among children ages 9 - 12 years old,this total increases to 37.2% for boys and to 30.5%for girls [2].

Children who become overweight or obese duringpreschool age are likely to retain this weight gainthroughout childhood [3]. This is especiallyconcerning because overweight and obesity areaccompanied by a range of physical and mentalhealth risks. Overweight children experienceimpaired psychosocial functioning and quality of

Advances in Pediatric Research Frate et al. 2018 | 5 : 51

life compared to normal weight children of thesame age [4,5]. Accordingly, both earlyintervention programs, which address key riskfactors for obesity among preschool aged children,as well as indicated or selective preventionprograms, which address the negative psychosocialconsequences among overweight and obesechildren, are needed.

Prevention measures at preschool age areconvenient since parents and children are easilyavailable. Children gain their first step toindependence. Unhealthy habits can be guided intothe right direction before individual patternsstrengthen. In early years, nutrition, motion, etc. isinfluenced by the caregiver – therefore surveyprocedures and prevention measures had to bringparents and kindergartens into focus.

Although many biological risk factors have beenidentified, much of these are not easily modifiable(e.g., genetic risk factors). The purpose of thisreview is to examine risk factors for childhoodobesity with the goal of informing preventionefforts with young children. In summary individualdifferences in obesity risk must be emphasised andassumed [6,7,8].

From a psychological perspective, risk factors mayoperate at the individual, family and social levels(Figure 1).

Figure 1: Psychological risk factors

This framework is useful in illustrating how factorsinfluencing health behaviors in relation tochildhood overweight and obesity affect at

individual, family and social levels in a complexand interdependent way.

Individual factors

Body dissatisfaction already occurs before schoolage. The discontent with one’s own body leads todiets, restrictive eating behavior, as well as misuseof laxatives and diuretics. This is very often avicious circle accompanied by a higher risk ofoverweight and/or obesity [9,10].

In this context, a healthy eating regulation plays amajor role concerning overweight and obesity. Ahealthy eating behavior includes a well-balancedenergy intake, a healthy choice of foods and thedecision about starting and stopping eating [11].Several studies examine eating regulation (foodresponsiveness, food enjoyment, satietyresponsiveness, eating in the absence of hunger,reinforcing value of food and the capacity tovoluntarily inhibit eating [12,13,14].

Self-control and impulsivity is linked with eatingregulation and weight gain [15].

Behavioral or emotional problems, e.g. poorcapacity of self-regulation and the awareness of ahealthy sense of hunger are associated with a higherBMI and leads to a regulation of emotions throughfood [16,17,18].

Family factors

Insecure attachments become significantimportance. Insecure attachment is associated withpoorer body satisfaction [19,20,21]. Insecureattachement also shows significant effects onchildren’s regulation of emotions and of their eatingbehavior [22,23]. Two reviews show, that insecureattachment is associated with eating pathology inchildhood, adolescence and adulthood refer to therisk, that the association may be confounded bydepression and low self-esteem [24, 25,26].Empirical observations corroborate the belief thatchildren with secure attachment are better indealing with negative emotions and their eatingregulation [23,25,26]. As far as a secure attachmentis, concerned parents and their style of educationplay a major role in socializing children’s self-regulation and energy intake [27].

A meta-analysis of parent-child relationship andgeneral parenting showed - by means of 156 studies– an association between a higher level of parental


responsiveness and a positive parent-childrelationship with lower weight [28].

Family problems and negative life events increasethe risk of obesity and overweight (Lumeng,Gannon, Cabral, Frank, & Zuckerman, 2003).Negative life events are associated with poor sleep,child behavioral problems, stress, and changes inmetabolism, increased food consumption andreduced self-regulatory capacity [29].

Social factors

Obese children suffer from a higher risk ofstigmatization and peer victimisation – that also hasan impact on social, psychological and behavioralfunctioning [30].

The association of overweight and obesity withpeer relationship problems is not surprising-giventhe fact of stigmatization and victimization [31,32].

MethodA systematic literature search in the most importantdatabases PsycINFO, Wiley Online Library, Web ofScience and Pubmed has been implemented withthe help of individual search terms such as„preschool“ AND „overweight“ OR„obesity“ according to PRISMA guidelines [33].Additionally, a manual search of relevant referenceswas implemented. Studies, which were peer-reviewed, published in the last five years (2011–2016), and written in German or English, wereincluded. In addition, the following inclusionarycriteria were required:

The sample consists of preschool children, which isdefined as children who were not yet in school andtypically between the ages 3 to 6.

Studies focused on psychosocial factors at theindividual, family and social factors.

A direct relationship between weight status andpsychosocial factors was examined.

Exclusionary criteria were:

Clinical studies (such as those focused on clinicaleating behavioral problems, like picky eating, bingeeating, food neophobia …), clinical subgroupswithin interventional studies, or studies focused onparental mental health.

Biological risk factors, such as cognitive andtemperament factors, because these are inherent and

within behavioral prevention and interventionmeasures less modifiable.

ResultsFigure 2 describes the process of article selection.The literature research yields 8472 hits (onlyjournal articles, no books) from four databases:Wiley Online Library (n=1462), PsycINFO(n=138), Pubmed (n=5273), Web of Science(n=1599). The search for relevant information andfurther sources reveal 25 additional hits.

8472 abstracts were screened according to theinclusion criteria (preschool children,socioecological factors, and direct relationship tothe weight status). 8405 studies were excluded and92 articles passed the selection process. Accordingto the duplicate removal and full-text screening atotal of 27 articles were taken into account.

Figure 2: Search history

The included studies were divided into threedomains: individual factors (n=22; body perception,n=4; eating regulation, n=11, behavioral/emotionalproblems, n=7), family factors (n=4; attachment,n=2; family problems and adversity, n=2), andsocial factors (n=5; stigmatization, n=3; peerrelationships, n=2). Four studies examining twodomains were included.

The majority of studies had a cross-sectional (n=22)or longitudinal design (n=3). Two studies combinedcross-sectional and longitudinal study designs.Most studies took place in the US American area(n=14), followed by Europe (n=9; Asia, n=2 andAustralia, n=2). The study’s sample sizes rangefrom 17 (Broome & Brugess, 2012) to 11 202probands (Griffiths, Dezateux, & Hill, 2011).


Gender relations were mostly balanced, withdifferences in the number of girls and boys usuallyless than 10% (n=27). Only three surveys displayedlarger variations ranging between 20% and 30%.

Body perception

Body perception was examined in four studies andthe results mostly support children’s early self-

awareness about their body size and their desiredsocial standards. Results indicate that bodydissatisfaction can already be detected amongpreschool aged children and body perceptions areinfluenced by gender, weight status, and parent’sexpectations regarding thinness.

Table 1: Body perception

Author(s) Data, sample size,sex

Location, datastructure

Measureanthropometric data,BMI by definition

Measure bodyperception

Results

Broome &Burgess, 2012

Children aged 4–5years, n=17, 65%female, 35% male

Indiana, pilot study Weight and heightassessed within thestudy, Centers forDisease Control andPrevention (CDC)

Body satisfactionwas assessed usinga Figure RatingScale

Children correctlyidentified the bodyshapes representative ofsomeone who isoverweight (100%) andunderweight (53%).

Jenull & Salem,2015


Austria, cross-sectional Study

Weight and heightassessed by doctors ina medical check-up,age- and sex-specificBMI percentiles.

Body satisfactionwas assessed usinga Figure RatingScale

22% of preschoolchildren were satisfiedwith their bodies. 43%selected a thinner and36% a fatter idealfigure. Boys (χ²=13.418,df=6, p=.037) andyounger children(χ²=31.917, df=12, p=.001) often chose a morecorpulent figure asideal.

Tremblay, Lovsin,Zecevic, & Larivie´re, 2011


United States,cross-sectionalStudy

Weight and heightassessed within thestudy, CDC

Concepts of bodyimage wasassessed using aFigure RatingScale

Overweight children(χ²=[1, N=144]=9.0, p< .01) and their parents(χ²=[1, N=144]=34.9, p=< .0001)underestimated thechilds´s body size.Normal weight girlswere less satisfied thanoverweight girls (χ²=[1,N=144] =5.05, p=.03)

Wong, Chang, &Lin, 2013


Taiwan, cross-sectional

Study

Weight and heightassessed in school,Taiwan Department ofHealth

Body satisfactionwas assessed usingthe Figure RatingScale by children,parents andteachers

If caregiver hoped thatthe child would bethinner, 58% of thechildren wanted to bethinner and 31% wantedto keep their weight.Contentedness with thechild weight, 40%wanted to be thinnerand 38% wanted to keeptheir weight (χ²; p < .0001)


All four surveys used Figure Rating Scales, whichshowed images of females/males ranging fromunderweight to obese. The results show, that theideal of a thin body is highly valued among theyoungest generation of our society. With theexception of a pilot study (Broome & Burges,2012) 40% to 58% of preschool aged childrendesired a thinner body [34,36,37]. If parents aredissatisfied with their children’s weight status, 2/3of these children wanted to be thinner as well [37].

Regarding gender differences, Tremblay et al.(2011) found out those girls are more likelydissatisfied with their bodies [38]. One possibleexplanation for those gender differences could bethe fact, that boys associate a larger body size withbeing muscular. Jenull’s study (2015) showed thedesire of boys and younger children to have theirdream body with a thicker silhouette becausethicker in this case is associated with strongerand/or older.

Jenull et al. (2015) as well as Wong et al. (2013)claim a limitation of their studies in a rural or urbanarea – therefore the studies outcome is notrepresentative for the remaining population [36,37].

Broome and Burges (2012) pilot study was limitedby a smaller sample. Racial and gender differencescould therefore not be examined, furthermore theoutcome’s generalization were thereby limited aswell [34]. The authors discussed the effect of thechildren’s self-evaluation report and reliability. Yetthe questionnaire tool used in the study was nottested for their validity and reliability.

Eating regulation

Eating regulations that are related to children’sweight status are subject to intensive research whileresearching 11 surveys could be found.

Table 2: Eating regulation

Author(s) Data, samplesize, sex



Measure eatingregulation Results

Bergmeier,Skouteris,Horwood, Hooley,& Richardson,2014

Children aged2-5 years,n=201, 58%female, 42%male

Australian, cross-sectional andprospective study

Weight and heightreported by mothers,CDC

Child EatingBehaviorQuestionnaire(CEBQ)

Enjoyment of food wasassociated with child´sBMI (β=.29, p < .01) att1 but not for t2

Braungart-Rieker,Moore, Planalp, &Lefever, 2014


United States, pilotstudy

Weight and heightassessed by a femaleexperimenter, CDC

CEBQ

Children who scoredhigher in foodapproach had higherBMIs (r=.49, p < .001)

Cross, Hallett,Ledoux, OĆonnor,& Hughes, 2014

AfricanAmericanchildren aged4-5 years,n=140, 53%female, 47%male Hispanicchildren aged4-5 years,n=159, 49%female, 51%male

Caucasian, cross-sectional study

Weight and heightassessed by trained staffmembers, CDC

Observation parent–child interactions inhome visits CEBQChild FeedingQuestionnaire (CFQ)

In African Americanchildren satietyresponsivenessmediated theassociation betweenpressure to eat andchildren’s weight (B(SE)=−0.073 (0.036),P < .05)

Domhoff, Miller,Kaciroti, &Lumeng, 2015

Children aged3-4 years,n=1002, 51%

United States, cross-sectional study

Weight and heightassessed by researchassistants, CDC

CEBQ

Food responsiveness(r=.10, p < .01) ,emotional overeating(r=.08, p < .05) and


female, 49%male

enjoyment of food (r=.18, p < .01) werepositively correlatedwith child´s BMI z-scores. Satietyresponsiveness (r=-.18,p < .01), slowness ineating (r=-.16, p < .01),emotional undereating(r=-.08, p < .05), andfood fussiness (r=-.07,p < .05) were negativecorrelated

Frankel et al., 2014



Weight and heightassessed within thestudy, CDC

CEBQ

Satiety (F[2,291]=7.19,p < .001), foodresponsiveness(F[2,289]=6.16, p < .01), and enjoyment offood (F[3,290]=8.43, p< .001) wereassociated with higherweight

Hughes, Power, OĆonnor, & Fisher,2015



Weight and heightassessed by trainedstaff, CDC

CEBQ Observationand parent report tomeasure child eatingbeyond satiationsnack-time to asseseating in absence ofhunger.

Child eating self-regulation waspositively associatedwith eating in theabsence of hunger (r=.20, p < .01 and foodresponsiveness (r=.15,p < .05); a negativecorrelation was foundwith satietyresponsiveness (r=-.24,p < .01)

Jansen et al., 2012

Children aged4 years,n=4987, 50%female, 50%male

Netherlands, cross-sectional study

Weight and heightassessed by trainedstaff, Dutch referencecurves

CEBQ, CFQ

Enjoyment of food (r=.155, p < .001), foodresponsiveness (r=.219, p < .001) andrestriction (r=.087, p< .001) were positivelyassociated with meanBMI; Fussiness (r=-.079, p < .001), satietyresponsiveness (r=-.236, p < .001),emotional undereating(r=-.102, p < .001) andpressure to eat (r=.186,p < .001) werenegatively related

Leung et al., 2015


United States,longitudinal study

Weight and height (at 4,5 and 6 years) assessedby trained researchassistants, CDC

CEBQ

Food responsivenessand enjoyment of food,were associated withhigher BMIs, foodresponsiveness (β=.21,p < .001), and


enjoyment of food(β=-.27, p < .001)

Mackenbach et al.,2012


Netherlands,population-basedcohort

Height and weightrepeatedly assessedduring regular visits tothe Child HealthCentres, Dutchreference curves

CEBQ

Positive correlationsbetween BMI and foodresponsiveness (r=.22,p <.01) as well asenjoyment of food (r=.16, p <.01). Negativecorrelations werefound with higherlevels of satietyresponsiveness (r=-.24,p <.01), fussiness (r=-.07, p <.01), andemotional undereating(r=-.10, p <.01)

Remy, Issanchou,Chabanet, Boggio,& Nicklaus, 2015

children aged3–6 years,n=236, 46%female, 54%male

France,experimental study

Weight, height andwaist circumferenceassessed by medicaldoctors, Frenchreference data

Observation of eatingin absence of hungerduring threesequential conditionsettings during threesessions (lunch,preload and lunch,lunch and post-mealsnack)

Eating in absence ofhunger was not relatedto z-BMI or waistcircumference

Spence, Carson,Casey, & Boule,2011


Canada, crosssectional study

Weight and heightassessed by trainedhealth assistant, CDC

CEBQ

Positive associationsbetween BMI and foodresponsiveness (F23.26, p < 0.01),enjoyment of food (F17.51, p < 0.01) andemotional overeating(F 6.19, p < 0.01);negative associationsfor satietyresponsiveness (F26.32, p < 0.01),slowness in eating (F17.57, p < 0.01), andfood fussiness (F 5.27,p=0.01)

The most commonly used survey instrument wasthe Child Eating Behavior Questionnaire (CEBQ)(n=10). The CEBQ covers eight scales (foodresponsiveness, enjoyment of food, emotionalovereating, desire to drink, satiety responsiveness,slowness in eating, emotional underrating, andfussiness), which however were not usedconsistently. The CEBQ was combined twice withan observation and another two times with the CFQ(Child Feeding Questionnaire) [39,40,41]. Only oneexperimental study focused on eating in absence ofhunger using observations within three sessions[42].

Positive correlations were found in sixinterdisciplinary surveys concerning the weightstatus from children and enjoyment of food[41,43-46]. Bergmeier et al. (2014) also confirmeda significant interdisciplinary context between childenjoyment of food and child BMI, but thisassociation changed over time (from t1 to t2 [47].

Furthermore, food responsiveness was associatedpositively in six studies with a higher BMI ofpreschool aged children [40,41,43-46]. A pilotstudy of 40 children showed a relation betweenhigher food approach scores and the weight status[48]. A longitudinal study confirmed a positive


context between food responsiveness, enjoyment offood and a higher BMI range (Leung et al., 2015).Additionally, emotional over-eating showed apositive overall correlation with a higher BMI[43,46].

Hughes et al. (2015) found out the positivecorrelation between BMI and child eating self-regulation. It showed, that eating in absence ofhunger was associated with self-regulation, e.g.,delay of gratification. Remy et al. (2015) on theother hand could not find a relation between eatingin absence of hunger and children’s BMI or thewaist circumference - however comparing girls andboys eating in absence of hunger was moredeveloped for the boys [42].

Satiety responsiveness and children’s weight statusshowed a negative correlation in five studies[40,41,43,45,46]. By contrast, Frankel et al. (2014)found positive associations between satiety andhigher weight. Cross et al. (2014) could not verify arelationship between appetitive characteristics ofthe child and child weight – however showed that agreater maternal restriction predicted a higherresponsiveness to satiety and satiety responsivenessmediated the association between children’s weightand pressure to eat, whereas pressure to eat wasassociated with children’s weight status [44,49].

The studies limitations concerning the eatingregulations are relevant for the studies sample sizeand design. Due to Remy’s large sample, a surveyconcerning the interaction between parents and

children’s feeding by video recording could not beimplemented [42]. Frankel et al. summarizednormal and underweight children – according tothat fact, no statement about underweight childrencan be made [44]. The presented studies focus onhomogenous samples, e.g. low income (Frankel etal., 2014; Leung et al., 2015), mothers with upperincomes and education, lack of multiple ethnicities(Hughes et al., 2015) – therefore the results arenon-transferable [40,44,47,49].

Mackenbach et al. implemented a nonresponseanalysis and showed some selective attrition under-representation of children from low socioeconomicbackground [44]. Also Bergmeister et al. andJansen et al. described a selective response [42,47].Remy et al. (2015) noted a limitation about asubject design and three sessions within his analysisthat lead to a boredom effect [42].

Cross et al. noted that the CFQ is an insufficientculturally sensitive questionnaire [39]. The CEBQrequirements when read aloud in the study ofDomhoff et al. (2015) could have influenced theresults while participants were more or less likely toendorse certain behaviors due to social desirabilitybias [43].

Behavioral/emotional problems

Seven studies analysed behavioral/emotionalproblems and the association of overweight orobesity in preschool children.

Table 3: Behavioral/emotional problems



Measureanthropometricdata, BMI bydefinition

Measurebehavioral/emotional problems

Results

Griffiths et al.,2011

Children aged 3 and5 years, n=11202,50% female, 50%male

United Kingdom,cross-sectional andlongitudinalstudy

Weight and heightassessed by trainedinterviewers at 3 and5 years, IOTF(InternationalObesity Taskforce)

Behavioral/emotional problemswere assessed usingthe SDQ at age 3and 5 years

Overweight boysshowed higherscores for conductproblems,hyperactivity,inattention, totaldifficulties andemotional problems.Overweight girlsscored higher fortotal difficulties andpeer relationshipproblems


Hughes et al., 2015 Children aged 4-6years, n=187, 48%female, 52% male


Weight and heightassessed by trainedstaff, CDC

Self-regulation(executivefunctioning) wasobserved using theFlexible ItemSelection Task andthe Tapping Task.

Emotional regulationwas assessed byobserving the giftdelay tasks and thedelay ofgratification.

The results show nocorrelations withChild BMI z-scoresand self-regulationvariables

Mackenbach et al.,2012

Children aged 3-4years, n=3137, 50%female, 50% male

Netherlands,population-basedcohort

Height and weightrepeatedly assessedduring regular visitsto the Child HealthCentres, Dutchreferencecurves

Behavioral problemswere assessed usingthe CBCL bymothers and fathersat age 3 years

The results show nosignificantassociations withbehavioral problemsand a higher childBMI, butinternalizingproblem scores wasassociated withlower mean BMI

Nagata, Hagan,Heyman, &Wojcicki, 2015

Children aged 3years, n=174, 51%female, 49% male

San Francisco, cross-sectional study

Height and weightwere assessed at agethree years withinthe study, CDC

Behavioral problemswas assessed usingthe preschool CBCL(1½–5) by mothers

The results show noassociations betweenobesity andpervasivedevelopmental,affective, anxiety,and attention deficithyperactivityproblems

Pieper & Laugero,2013


United States, pilotstudy

Weight, height andwaist circumferenceassessed byresearchers, noinformation – BMIfor age percentile

Emotional arousalwas measuredthrough affective Q-sensors via skinconductance.Teacherscored e.g.,impulsive control

CBQ reportet byparents

The results shownegative correlationsbetween BMI forage percentile andimpulse control aswell as betweenwaist circumferenceand inibitory control

Rollins, Loken,Savage, & Birch,2014


Pennsylvania,experimental study

Weight and heightassessed by trainedstaff members, CDC

Parents report usingthe CBQ

There were no sign.correlations betweeninhibitory controland BMI

Suglia, Duarte,Chambers, &Boynton-Jarrett,2013

Children aged 3 and5 years, n=1589,49% female, 51%male

United states, birthcohort study

Height and weightassessed by trainedinterviewers at agefive years, CDC

Behavioral problemswas assessed usingthe preschool CBCLat age five years

Externalizingbehavioral problemswere associated withobesity among boysand girls

Pieber and Laugero (2013) assumed that a reducedexecutive function (e.g., a lower self-regulation anda reduced capacity of emotion regulation) is

connected with an unhealthy eating behavior, whichleads to emotional-based overeating [50]. Incontrast, Hughes et al. outcomes showed


correlations amongst eating in absence of hunger,emotional, as well as self-regulation but nocorrelation between BMI and self-regulationvariables [39].

As mentioned in the study of Hughes et al. (2015),Pieper and Laugero (2013) the CBQ as a parent-report was used to capture the inhibitory control ona cross functional basis of a further survey[40,50,51]. The author’s evaluated children’sbehavioral response to parent’s use of restrictivefeeding practices and the risk for weight gain.Children with a lower inhibitory control had ahigher food intake in response to restriction. Norelations between inhibitory control and thechildren’s BMI could be found.

The CBCL was applied as a parent report withinthree studies concerning behavioral problemsrelating to children’s weight status. Mackenbach etal. found a correlation between, using a cross-sectional survey with 3137 preschool children,internalizing problems and higher levels ofemotional problems with a lower BMI [44]. Nagataet al. (2015) found no correlations between obesityand psychological problems in Latino preschoolchildren, using a smaller sample (n=174). Asopposed to Suglia et al. (2013), who reported acorrelation between externalizing behavioralproblems and obesity.

Griffiths et al. found higher scores using 3-year-oldobese boys in a cross functional survey for totaldifficulties, as well as peer relationships,hyperactivity and inattention and conduct problems[35].

At the age of 5 the outcomes with the same –furthermore obese boys showed a higher value foremotional problems. Obese girls at the age of 5showed problems solely in peer relationships and

total difficulties. Longitudinal it only showed, thatobesity at age 3 was predictive for peer relationshipproblems at age 5.

Pieper and Laugero and Rollins et al. both mentiona small sample size as a limitation [50, 51]. Thiscould lead to reduced power to detect individualdifferences in effects. Rollins et al. note, that therecould be a discrepancy between the responsebehavior on restriction in preschool and parentalhome [51]. Preschool children could reactdifferently on these restrictions, because of theirknowledge about it, over a long period at home.

Suglia et al. (2013) claim the CBCL as a validinstrument to assess child behavioral problems, butit was not designed to measure specific behaviorsassociated with obesity, for example rewardsensitivity or impulsivity [53]. Other factors thatrelate to social pressure should be taken intoconsideration. Emotional eating and emotionalregulation is undocumented. The survey of physicalactivity and nutrition was also limited. Mackenbachet al. discussed the limitation of children fromfamilies with lower education and young mothersand their underrepresentation [45].

Furthermore, there was a lack of multipleethnicities in Hughes et al. study. Rollins usedsamples from white families with high educationand high income [40,51].

Another limitation was the survey of a givencultural sample and the fact, that the results arenon-transferable on population elsewhere [52].

Attachment

Two studies focussed on children’s’ weight statusand secure attachment.

Table 4: Attachment

Author(s) Data, sample size, sex Location, datastructure


Measureattachment

Results

Anderson &Whitaker, 2011

Children aged 24 monthsand their risk for obesityat 4 1⁄2 years of age, n=6650, 49% female, 51%male

United States,longitudinal birthcohort study

Weight and heightassessed duringobservation in thechild’s home,age- andsex-specific BMIpercentiles.

ToddlerAttachment Sort

The odds of obesitywere higher forchildren withinsecure attachmentOdds of obesity forchildren with


insecure attachmentwere 1.30 [95% CI,1.05-1.62] timeshigher

Keitel-Korndörfer etal., 2015

Children aged 2-5 years(at commencement of thestudy), n=62, 58%female, 42% male

Germany, cross-sectional study

Weight and heightassessed in laboratory,child’s birth weighttaken from the birthreports, age- and sex-specific BMI percentiles

Attachment Q-Set(AQS)

The attachmentquality predictschild`s BMIpercentiles AQStotal score on theBMI percentile ofthe child (B=−26.44, SEB=13.59, β=-.24,p=.03, one-tailed,R2=.06). That is,the lower thequality of themother–childattachment thehigher the BMIpercentile of thechild.

Throughout both studies, correlations betweeninsecure attachment and a higher BMI were found.Half of the respondents in Keitel-Korndörfers studywere obese mothers and their children showed asignificant lower bonding security [7]. The authorsnote that decreased fitness affects the interactionwith the child or they are ashamed of their ownweight, which leads to mothers who cannot offertheir children a secure base for explorativebehavior. In a cross-sectional study in Germany,Keitel-Korndörfer and colleagues (2015) assessed62 preschool children from 31 normal weight and31 obese mothers and showed that attachmentquality predicts child’s BMI percentile [7].

Keitel-Korndörfer et al. justified the small samplesize with expenses and time exposure [7].

Another possible source of errors could bemissing’s, so Anderson and Whitaker argued [54].The authors noted a huge variety of potentiallyconfounding variables has been controlled; a biasdue to uncontrolled confounding or measurementcannot be excluded. In addition, too manyvariables, which could be part of the causalpathway between obesity and attachment security,lead to over controlled potential. Finally, theypresume that the observation of attachment security

is not representative in only one day when it comesto a child’s typical behavior.

Suglia and collegues study’s subject concerned therisk for obesity using maternal reports of intimatepartner violence, maternal substance use anddepressive symptoms, father's incarceration, foodand housing insecurity [53]. The results show acorrelation between more than one social risk factorand higher BMI scores for girls, but not for boys.

Although the death of a parent is the most stressfulevent in a child´s life, the results don´t show anassociation with an increased risk of overweightand the loss of a parent in preschool age [54].

Suglias survey considered limited health behaviormeasures [53]. Eating and exercise habits couldonly be included in the five-year follow up study.Other factors, such as emotional or eatingregulation were not captured. As a limitation of Li’sstudy (2012) it is to say, that children were notasked about the characteristics of their bereavementexperience, nature of bereavement, quality andbond to the deceased parent [54].


Family problems and adversity

Two studies analysed the correlation betweenweight status and family problems.

Table 5: Family problems and adversity




Measure familyproblems

Results

Li et al., 2012 Children withbereavement bydeath of a parentduring the first 6years of life;exposed cohort,n=492, 50% female,50% male;Unexposed cohort,n=45 909, 49%female, 51% male.

Denmark,population-basedcohort study

Weight and heightassessed by schooldoctors and nurses,IOTF.

Death of a parentwas asked about theDanish CivilRegistrationSystem.

Bereavement duringthe first 6 years oflife was notassociated with anincreased risk ofoverweight (–0.03[95% confidenceinterval [CI] –0.20to 0.14]) or averageBMI levels at 7-13years (–0.01 [95%confidence interval[CI] –0.40 to 0.38])

Suglia et al., 2013 Children obtainingat age 3 and age 5,n=1589, 49% girls,51% boys

United States, birthcohort study

Weight and heightassessed by trainedinterviewers when thechild was 5 years old,CDC

On the basis ofmaternal reportswhen the child was3 years old (father´sincarceration,partner violence,maternal substance,etc.) a social riskscore was assessed

More than onesocial risk factorincreased the riskfor obesity for girlsbut not for boys(40.4% vs 31.8%,girls, p < 0.05)

Stigmatization

Three cross-sectional studies explored the role ofstigmatization and overweight at preschool age.

Table 6: Stigmatization

Author(s) Data, samplesize, sex


Measureanthropometric data,

BMI by definition

Measurestigmatization

Results

Holub, Tan, &Patel, 2011


United States,cross-sectional

study

Weight and heightassessed within thestudy, age- and sex-specific BMI percentiles

Weight stereotypeswere measured usinga thin, average andobese figure from aFigure Rating Scaleand ratings acrossfive (negative andpositive) adjectivepairs

Children showed fewerpositive adjectives forthe obese figure, thanfor the average or thinfigure. Older childrenrated the average figuremore positively thanyounger children did.Children´s BMI showedno association withtheir ratings


Kornilaki, 2014 Children aged 4-5years, n=85(average n=48,obese n=37) , 51%female, 49% male

Greece, cross-sectional study

Weight and heightassessed by a trainedassistant, age- and sex-specific BMI percentiles

Stigmatization wasassessed by choosingpreferred playmateusing a drawn figure(thin, average, obese).Additionally theypicked a figure(drawn byprofessional; thin,average, obese) whichpresented positive ornegativecharacteristics in 13short stories

The obese figure wasless chosen as preferredplaymate. Normalweight as well as obesechildren addressed thepositive attributesmostly to the thin andaverage figure, by themajority of children thenegative characteristicswere addressed to theobese figure

Spiel, Paxton, &Yager, 2012

Children aged 3-5years, n=118, 60% female, 40%male

Australia, cross-sectional study

Weight and height wasreported, age- and sex-specific BMI percentiles

Stigmatization wasassessed using aFigure Rating Scaleand questions aboutwhat child would beinvited/not invited toa party and whichchild has the most/theleast friends.Additionally theypicked a figure (formthin to very large)which presentedpositive or negativecharacteristics ineight stories

Negative characteristicswere addressed tolarger figures. Age ofthe children has a lowinfluence onattributions but not onBMI, the child´sperceived body size orgender.

All of the three studies used drawn figures orfigueres of a rating scale for the childrens interview.

This review’s results show- despite differentapproaches –preschool children’s preference of thinand normal weight peers as well as negativecharacteristics and attributions towards overweightand obese figures came to almost identicalconclusions [55- 57]. Age played a role 5 year oldsassigned negative characteristics significantly andmore frequently to the larger figure compared to 3year olds. With regard to children’s own perceivedbody size the authors found out, that it is predictivefor positive, but not for negative attractions.However, for the children the maternal preference(body image attitudes in this case) played a role fornegative, as well as positive attribution. Holub et al.(2011) came to a similar conclusion, whereas thematernal fear of fat affected the children’s negativeattribution [55].

Holub et al. noted a limitation concerning themeasurement of anti-fat attitudes influenced bysocial desirability [55]. Furthermore, mothers werenot explicitly asked about beliefs toward

overweight adults but about people, therefore thereis no difference between overweight adults andoverweight children. Implicit attitudes were notmeasured. There is evidence, that parents’ implicitattitudes are more predictive of children’s earlydeveloping racial prejudice than explicit attitudes.Ethnic differences could not be evaluated accordingto the small sample size.

Kornilaki (2014) adhere to the statement thatchildren were grouped according to their BMI, butwere not asked about their perceived body size[56]. Children could misjudge their body weightand that would lead to a distortion of the resultssince they didn’t identify with their respectiveweight group. Spiel’s survey (2012) showed anunderrepresentation of children with a higher BMIand the participants mainly came from highersocio-economic areas which reduce thegeneralization of results [57]. Measurement of childattitudes, making a forced choice about a figuremay not necessarily mean that a child wouldattribute this fact towards a real person.


Peer relationships

Solely two studies dealt with peer problems withinthe elected age group relating to preschool childrenand their weight status.

Table 7: Peer relationships




Measure bodyesteem andsatisfaction

Results

Griffiths et al.,2011

Children aged 3 and5 years, n=11 202,50% female, 50%male

United Kingdom,cross-sectional andlongitudinal study

Weight and heightassessed by trainedinterviewers at 3 and 5years, IOTF

Peer relationshipproblems wasassessed using theSDQ by parents at 3and 5 years

Obese boys at age 3and 5 years, as wellas obese 5 years oldgirls had more peerrelationshipproblems thannormal weightchildren.

Pérez-Bonaventura,Granero, &Ezpeleta, 2014

Children at age 3years ( n=611), 4years (n= 596) and5 years (n= 564),50% female, 50%male

Spain, cross-sectional andlongitudinal study

Height and weightassessed by nurses,WHO reference curves

SDQ, conductproblems scale byparents report

Cross-sectional theresults show asignificantassociation betweenhigher mean scoresfor peer relationshipproblems andoverweight at age 5years. Higher BMIz-cores wereassociated withconduct problemsand low prosocialbehavior scores atage 4 years. HigherBMI z-scores at age3 years predictedmore peerrelationshipproblems at age 4and 5 years

Griffiths et al. (2011) cross-sectional analysesshowed that obese boys compared to normal weightboys at the age of three demonstrate more peer andconduct problems whereas girls reached more meanscores for prosaically behaviors [35]. At the age offive peer problems could be found for both genders.The longitudinal study demonstrates that obesity atage 3 for boys is a good predictor for peer problemsat age 5. Beyond this, emotional and behavioralproblems are seen as particular risk factors for boysat that age.

In addition, Perez-Bonaventura and colleaguesfound a correlation between overweight and peerrelationship problems. Longitudinal aspects argued

that a higher BMI at age 3 predicted more peerrelationship problems at age 4 and 5 years [50].

As a limitation it should be noted, that the SDQ is areliable and valid instrument though based onparents statements. This may lead to an over- orunderestimation compared to children’s self-assessments. A possible bias could be parents’childhood memories relating to their own weighthistory transferring these memories to theirchildren. Gender stereotypes could distort theresults as well. Stigmatization, victimization andthe predominant ideal of beauty should beassociated with peer relationships as a moderatingvariable.


DiscussionThis review integrates the various risk factors forchildhood obesity into three levels with relevancefor prevention planning.

Individual factors

Body perception studies showed that very youngchildren already judge their bodies in an accurateway. Empirical results are interesting because of thefact that overweight children as well as their parentsunderestimate the children’s body size. Thismisjudgement could also be found in a large-scalestudy of parents with 2-5 year old children from1988 to 1994 and 2007 to 2012. Familialinfluences, environmental factors, mediation ofbody ideals play a major role for young people andare in need of serious attention.

Intervention and prevention programs should makebody satisfaction and body esteem a subject ofdiscussion in those early ages. Furthermore, itwould be important for parents and educators tothink critically on their own body weight concept;because primary caregivers could pass their ownconcepts to their children and a healthy body sizeconcept affect the eating behavior, weight andhealth.

The findings to eating regulation refer to a contextamongst food responsiveness, enjoyment of foodand a higher BMI. Emotional over-eating andeating self-regulation were less often associatedwith a higher weight status. Negative relationsconcerning the BMI level were especiallyassociated with satiety responsiveness. A veryinteresting connection shows parental restrictionthat affects satiety responsiveness. Hughes et al.(2015) found out, that higher levels of satietyresponsiveness are linked with lower weight andhigh levels of emotional regulation [40].

To avoid overweight and the prevention of gainingweight affects the self-regulation capability of thechildish food intake. Parents are key figures duringinfantile development. Besides model effectconcerning healthy food, especially all areas of theself-regulation capability. When it comes tooverweight there are hints that show parent’sreaction to their children food intake. Wheneverfood is used as a punishment or a reward, it affectsthe self-regulation capability of a child’s energyintake.

In the future, it would be important to examinedifferent eating behavior dimensions and theiroverlaps, as well as interdependencies andinteractions more precisely. Higher valuesconcerning food responsiveness and enjoyment offood could be cancelled due to a higher satietyresponsiveness. Studies should take a closer look atethnic/racial differences concerning feeding habits.Eating practices over time should also be examinedand evaluated.

Little evidence was found for a relationshipbetween behavioral/emotional problems andoverweight concerning preschool aged children.Externalizing behavioral problems – as opposed tointernalizing problems – are related to weight gain[35,45,52,53].

Another three studies analysed self- and eatingregulation according to the risk of a higher BMI. Itshowed a correlation between a lower impulsecontrol and a higher weight status. The resultsdocument, that the child eating self-regulation isassociated with the emotional regulation (whichexpresses itself through an unhealthy eatingbehavior) [40,50,51].

Early deficiencies in self-regulation andexternalizing problems show a higher risk indeveloping weight problems in preschool age andshould be examined more precisely. A successfulself-regulation contributes to a positive self-effectiveness and plays a major role in the eatingregulation. To prevent unhealthy eating behavior itis important to detect behavioral and emotionalproblems in early years.

Family factors

Attachment security presents a mental andemotional resource. Throughout both studiespresented here, insecure bonds claim a higher riskfor overweight and obesity. On a critical note, itmust be pointed out that the interview and theobservation of mothers could be biased and may notreflect the typical daily bonding situation.

Overall, attachment security is paid little attentionup to now.

Previous studies refer to a direct bonding effectconcerning the dissatisfaction of one’s own bodyand an indirect eating behavior effect [20,21]. Along-term study analysed the influence of an earlymother-child-bond on the risk of obesity in young


age [58]. The results show a correlation between aninsecure attachment and a higher risk ofoverweight. Additionally, the authors described acumulative effect between poor maternal-childrelationship and an insecure bond that leads to ahigher risk of obesity. Further research projectsshould also focus on fathers, grandfathers and otherattachment figures because it is assumed thatrelationships with adults are either protective orrepresent a risk.

For parenting styles for this review no study wasenclosed. Sleddens, Gerards, Thijs, de Vries, andKremers (2011) review showed for infants,toddlers, pre-schoolers, school-aged child’s, oradolescent with an age below 18 years, thatchildren from authoritative homes had a lowerBMI, were eating healthier and were more activecompared to children from authoritarian,permissive/indulgent, uninvolved/neglectfulparenting style homes [59].

Hancock, Lawrence, and Zubrick found in a long-term study, that motherly protectiveness could leadto a risk of overweight at the age of 10 to 11, butnot for younger children [60]. “Overprotective”defines a parenting behavior, where parentalmonitoring prevails unlike an independent childlikebehavior and the separation of a child causesproblems. The relations between a higher weightand highly protective parenting could result fromlimited motion behavior because ofoverprotectiveness but also from mothers who reactdifferently when it comes to food preferences.

Highly protective parenting, such as indulgent anduninvolved parenting and feeding styles were stateda risk factor. Cross-sectionally it was shown thatfeeding practices at children aged 4–12 years oldsuch as pressure to eat and restriction were linkedto BMI most strongly, whereas longitudinalparenting style was the strongest and mostconsistent association with child’s BMI [61].

The studies to family problems show, that violence,neglect, abuse etc. increase the risk of overweightand obesity. Future studies should analyse themechanism of association on growth trajectories ofBMI in order to embed childhood educationprograms as well as prevention and interventionmeasures and to understand potential mediators anddevelopmental mechanism. The possible higher riskof negative life events for children should also betaken into account.

Mental and physical consequences of life events onfamily members and important caregivers should beanalysed more precisely. Finally yet importantlypositive coping strategies should receive moreattention in future studies.

Social factors

The stigmatization of overweight people is widelyspread and well proven in numerous surveys.Preschool children hold children’s negativeperceptions towards overweight peers [62]. Positivecharacteristics are attributed to thinner or normalweight children -whereas negatives are attributed tooverweight children [63]. This could holdunfavorable effects on the social development.Affected children fall victim to forms of peeraggression and victims of overt [30].

Sikorski, Luppa, Luck, and Riedel-Heller considerweight stigmatization and discrimination ofoverweight people as chronic stressor [64]. Theauthors analysed the influence on psychologicalwell-being of children, adolescents and adultsconcerned and found a positive mediation throughpsychological risk factors on mental health outcomethroughout all studies. The highest negativeassociation was found relating to self-esteem inadults and children. At the same time, it leads frompoorer psychological functioning to unhealthyweight-control strategies [30].

It would be important, that parents and earlychildhood educators provide protective factors forchildren in order to protect and prevent them fromthe negative influence of weight-basedstigmatization and victimization. Negative bodyshape attitudes and stereotypies should be correctedin order to avoid stigmatization [62]. Therefore,prevention measures should not concentrate only onweight loss but address body satisfaction and theacceptance of diverse body shapes [65]. Futurestudies should include further influence factors, forexample father’s attitudes and negative stereotypingand media experiences.

The results of two studies, which analysedoverweight/obesity in correlation with peerproblems, concluded that, a higher weight status isrelated to peer relationship problems. As early asthe age of 3, boys and obesity are linked with peerand conduct problems. At the age of 5 both gendersare affected [35]. Longitudinal both studies foundout, that a higher weight at the age of 3 predicted


more peer relationship problems at the age of 5[35,66].

Prevention should focus on self-confidence and areduction of stigmatization in order to reduce themisery of obese children. Further studies shouldfocus on the relationship between affectivedisorders, emotional problems and behavioralproblems of overweight and obese children.Relationships and friendships interact with thefurther development and could -in the event ofoverweight and obesity - have an impact or result inan affective disorder.

LimitationsThe reported results give a first overview on therelevance of psychosocial factors influencing therisk of overweight and obesity in preschool age.The majority of selected studies analysedcorrelations within a cross-sectional design (n=22;there of 3 pilot, 2 experimental and 2 birth cohortstudies), that do not allow causal relationships,mediation effects or bidirectional effects,directional associations or interpretations, andconclusions about changes and differences in thedevelopmental course – therefore aetiologicalmechanisms could not be explained. Three of thestudies used a longitudinal design and another twocombined cross- and longitudinal designs (wherebythe age category remained preschool agethroughout all surveys concerning the interpretationof results).

Dropout rates and the sample size’s representativestatus should be considered. Rural and urbanstudies do not show a generalization when it comesto population in other geographic areas. The sameprocedure applies with homogenous samplesaccording to socio-economic characteristics.

There was a balanced gender ratio throughout themajority of surveys, therefore differentially affectsbetween female and male pre-schoolers could bereasonably investigated, however the BMI data wasdefined by means of different guidelines. Fewstudies analyse musculature and body build. Thesole use of the BMI leads to misclassificationsbecause no information about the body-fat-distribution exists. Another opportunity could beother measures of body composition such ashydrostatic weighing or skin fold ratio. On a criticalnote, it must be pointed out that the physical

activity and eating behavior of children was oftenunconsidered.

The collection of psychosocial factors occurred insix studies through an investigation of children, 13studies depended on parents’ information and eightstudies gathered their relevant aspects through amix (children, parents and/or teachers,observation…).

Information based on parents’ assumptions couldreduce validation, which represents a knownlimitation of this type of research. Reporter biascontain a risk of over- or underestimatingpsychological problems. Mothers who are moreaware in their children problems may also reportmore difficulties in eating behavior (Mackenbach etal., 2012). Social desirability could distort theresults. Triangulation of different perspectives mayprovide a more complete picture of thepsychological factors associated with child obesity(Griffiths et al., 2011).

However, refraction on external data seemsnecessary especially for the preschool age. On theone hand, there is a lack of high-quality surveyprocedures for children of that age and on the otherhand, young children do not often have theopportunity to describe and rate their ownintentions.

A meta-analysis could not be executed for variousreasons. First, there was a big heterogeneityconcerning the studies, whereby a statisticalsummary was not possible or reasonable.

Finally, it has to be noted, that this paper onlyincludes publications from 2011 until 2016. Maybeunpublished work or articles from another databasecould lead to a more complete understanding of thecurrent state of knowledge about the associationbetween psychological factors influencing the riskfor preschool overweight or obesity.

ConclusionThere are multiple risk factors and causes forchildhood overweight/obesity. Prevention programsneed to integrate these risk factors at the individual,family and social levels. To move forward inunderstanding child obesity, more longitudinalstudies are needed which take into accountdevelopmental processes, gender differences, aswell as differences in subgroups (e.g., based onethnicity, geographic region).


Considering the young audience parental inclusionis mandatory. Not only feeding and exercise habitsare influenced by parents, but also body image,body satisfaction, self-consciousness, medialiteracy and so on. Parents are role models and sothey play an essential role with their parentingstyles and communication patterns.

Prevention and intervention programs shouldinclude elements of all dimensions as defined bythe bio psychosocial models. Psychological factorsfor example emotional competencies, a positiveself-perception and self-esteem, should be drawnmore attention. As mentioned before, preventionstrategies should also include working with parentsto improve parenting skills. Bonding securityaffects the ability to understand the child’s needsand the reflective function. Therefore, we needinterventions for both children and parents (mothersand fathers). This would have an impact on thechild’s potential to overcome pressure and regulateemotions in an adequate way.

References1. Anderson SE, Gooze RA, Lemeshow S, Whitaker RC.

Quality of early maternal–child relationship and risk ofadolescent obesity. Pediatrics. 2011. doi:10.1542/peds.2011-0972.

2. Anderson SE, Whitaker RC. Attachment security andobesity in US preschool-aged children. Archives ofPediatrics & Adolescent Medicine. 2011;165(3):235-242.

3. Bergmeier H, Skouteris H, Horwood S, Hooley M,Richardson B. Child temperament and maternal predictorsof preschool children’s eating and body mass index. Aprospective study. Appetite. 2014;74:125-132.

4. Blundell JE, Stubbs RJ, Golding C, Croden F, Alam R,Whybrow S, et al. Resistance and susceptibility to weightgain: individual variability in response to a high-fat diet.Physiology & Behavior. 2005;86(5):614-622.

5. Braungart-Rieker JM, Moore ES, Planalp EM, LefeverJB. Psychosocial pathways to childhood obesity: A pilotstudy involving a high risk preschool sample. EatingBehaviors. 2014;15:528–531.

6. Broome ME, Burgess JN. Perceptions of weight and bodyimage among preschool children: A pilot study. PediatricNursing. 2015;38(3):147.

7. Calkins SD, Leerkes EM. Early attachment processes andthe development of emotional self-regulation. In K. D.Vohs, & R. F. Baumeister (Eds.), Handbook ofselfregulation. Researh, theory and applications. (2. ed.,pp. 355-373). New York: The Guilford Pres. 2011.

8. Carnell S, Wardle J. Appetite and adiposity in children:evidence for a behavioral susceptibility theory of obesity.Am J Clin Nutr. 2008;88(1):22-29.

9. Cross MB, Hallett AM, Ledoux TA, OĆonnor DP,Hughes SO. Effects of children’s self-regulation of eating

on parental feeding practices and child weight. Appetite.2014;81:76-83.

10. Domoff, S. E., Miller, A. L., Kaciroti, N., & Lumeng, J.C. Validation of the children's eating behaviorquestionnaire in a low-income preschool-aged sample inthe United States. Appetite. 2015;95:415-420.

11. Epstein LH, Leddy JJ, Temple JL, Faith MS. Foodreinforcement and eating: a multilevel analysis. PsycholBull. 2007;133(5):884.

12. Epstein LH, Temple JL, Neaderhiser BJ, Salis RJ, ErbeRW, Leddy JJ. Food reinforcement, the dopamine Dreceptor genotype, and energy intake in obese andnonobese humans. Behav Neurosci. 2007;121(5): 877.

13. Francis LA, Sussman EJ. Self-regulation and rapid weightgain in children from age 3 to 12 years. Arch PediatrAdolesc Med. 2009;163(4):297–302.

14. Frankel LA, Hughes SO, O'Connor TM, Power TG,Fisher JO, Hazen, N. L. Parental influences on children'sself-regulation of energy intake: Insights fromdevelopmental literature on emotion regulation. J Obes.2012. doi:10.1155/2012/327259

15. Frankel LA, O’Connor TM, Chen T A, Nicklas T, PowerTG, Hughes SO. Parents’ perceptions of preschoolchildren’s ability to regulate eating. Feeding styledifferences. Appetite. 2014;76:166-174.

16. Gray WN, Kahhan NA, Janicke DM. Peer victimizationand pediatric obesity: A review of the literature.Psychology in the Schools. 2009;46(8);720-727.

17. Griffiths LJ, Dezateux C, Hill A. Is obesity associatedwith emotional and behavioral problems in children?Findings from the Millennium Cohort Study. Int J PediatrObes. 2011. doi: 10.3109/17477166.2010.526221

18. Griffiths LJ, Page AS. The impact of weightrelatedvictimization on peer relationships: the female adolescentperspective. Obesity. 2008;16(2):39-45.

19. Griffiths LJ, Wolke D, Page AS, Horwood JP. Obesity andbullying: different effects for boys and girls. Arch DisChild. 2006;91(2):121-125.

20. Hancock KJ, Lawrence D, Zubrick SR. Higher maternalprotectiveness is associated with higher odds of childoverweight and obesity: a longitudinal Australian study.Plos One. 2014. doi:org/10.1371/journal.pone.0100686

21. Harriger JA, Thompson JK. Psychological consequencesof obesity: weight bias and body image in overweight andobese youth. Int Rev Psychiatry. 2012;24(3):247-253.

22. Holub SC, Tan CC, Patel SL. Factors associated withmothers' obesity stigma and young children's weightstereotypes. J Appl Dev Psychol. 2011;32(3):118-126.

23. Hughes SO, Power TG, OĆonnor TM, Fisher JO.Executive functioning, emotion regulation, eating self-regulation, and weight status in low-income preschoolchildren: How do they relate? Appetite. 2015;89:1-9.

24. Incledon M, Wake M, Hay M. Psychological predictors ofadiposity: systematic review of longitudinal studies. Int JPediatr Obes. 2011. doi:10.3109/17477166.2010.549491

25. Jansen PW, Roza SJ, Jaddoe VW, Mackenbach JD, RaatH, Hofman A, et al. Children's eating behavior, feedingpractices of parents and weight problems in earlychildhood: results from the population-based GenerationR Study. Int J Behav Nutr Phys Act. 2012;9(1):1.


26. Jenull B, Salem I. Body satisfaction among preschool-agechildren in Carinthia (Austria). Eur Sci J. 2005;11(10):152-162.

27. Jewell T, Collyer H, Gardner T, Tchanturia K, Simic M,Fonagy P, et al. Attachment and metallization and theirassociation with child and adolescent eating pathology: Asystematic review. Int J Eat Disord. 2016;49(4):354-373.

28. Karasu SR. Of mind and matter: psychologicaldimensions in obesity. Am J Psychother. 2012;66(2):111-128.

29. Keitel-Korndörfer A, Sierau S, Klein AM, Bergmann S,Grube M, von Klitzing K. Insatiable insecurity: maternalobesity as a risk factor for mother–child attachment andchild weight. Attach Hum Dev. 2015;17(4):399-413. doi:10.1080/14616734.2015.1067823

30. Kornilaki EN. Obesity bias in preschool children: do theobese adopt anti-fat views? Hellenic J Psychol.2014:11:26-46.

31. Koskina N, Giovazolias T. The effect of attachmentinsecurity in the development of eating disturbancesacross gender: The role of body dissatisfaction. J Psychol.2010;144(5): 449-471.

32. Kuhl ES, Rausch JR, Varni JW, Stark LJ. Impaired health-related quality of life in preschoolers with obesity. JPediatr Psychol. 2012;37(10):1148–1156.

33. Leung CYY, Miller AL, Kaciroti NA, Chen YP,Rosenblum K, Lumeng JC. Lowincome preschoolers withhigher temperamental surgency enjoy and respond moreto food, mediating the path to higher body mass index.Pediatr Obes. 2015;11(3):181-186.

34. Li J, Olsen J, Vestergaard M, Obel C, Baker JL, SorensenTIA. Bereavement in early life and later childhoodoverweight. Obesity Facts. 2012;5:881-889.

35. Liem ET, Sauer PJ, Oldehinkel AJ, Stolk RP. Associationbetween depressive symptoms in childhood andadolescence and overweight in later life: review of therecent literature. Arch Pediatr Adolesc Med.2008;162(10):981-988.

36. Lumeng JC, Gannon K, Cabral HJ, Frank DA, ZuckermanB. Association between clinically meaningful behaviorproblems and overweight in children. Pediatrics.2003;112(5):1138-1145.

37. Mackenbach JD, Tiemeier H, van der Ende J, Nijs IMT,Jaddoe VWV, Hofman A, et al. Relation of emotional andbehavioral problems with body mass index in preschoolchildren: the generation r study. J Dev Behav Pediatr.2012;33(8):641-648.

38. Mata J, Munsch S. Adipositas von kindern undjugendlichen. risiken, ursachen und therapie auspsychologischer sicht [obesity of children andadolescents. risks, causes and treatment from apsychological perspective]. Bundesgesundheitsblatt.2011;54:548-554.

39. Mayer M, Gleiss A, Häusler G, Borkenstein M, KapelariK, Köstl G, et al. Weight and body mass index (BMI):current data for Austrian boys and girls aged 4 to under19 years. Annals of Human Biology. Ann Hum Biol.2015;42(1):45-55.

40. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferredreporting items for systematic reviews and meta-analyses:the PRISMA statement. Ann Intern Med. 2009;151(4):64-269.

41. Nader PR, O'Brien M, Houts R, Bradley R, Belsky J,CrosnoeR, et al. Identifying risk for obesity in earlychildhood. Pediatrics. 2006;118(3):594–601.

42. Nagata J M, Hagan MJ, Heyman MB, Wojcicki JM. NoAssociation between obesity and behavior in low-income, preschool latino children. J Health Care PoorUnderserved. 2015;26(2):410-420.

43. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalenceof obesity and trends in body mass index among USchildren and adolescents, 1999–2010. JAMA: J Am MedInform Assoc. 2012;307(5):83-490.

44. Pérez-Bonaventura I, Granero R, Ezpeleta L. Therelationship between weight status and emotional andbehavioral problems in spanish preschool children. JPediatr Psychol. 2014;40(4):455-463.

45. Pieper JR, Laugero KD. Preschool children with lowerexecutive function may be more vulnerable to emotional-based eating in the absence of hunger. Appetite.2013;62:103-109.

46. Pinquart M. Associations of general parenting and parent–child relationship with pediatric obesity: a meta-analysis.J Pediatr Psychol. 2014;39(4):381-393.

47. Remy E, Issanchou S, Chabanet C, Boggio V, Nicklaus S.Impact of adiposity, age, sex and maternal feedingpractices on eating in the absence of hunger and caloriccompensation in preschool children. Int J Obes.2015;39(6):925-930.

48. Rollins BY, Loken E, Savage JS, Birch LL. Effects ofrestriction on children’s intake differ by childtemperament, food reinforcement, and parent’s chronicuse of restriction. Appetite. 2014;73:31-39.

49. Roth B, Munsch S, Zumsteg U, Isler E. PsychologischeAspekte der kindlichen Adipositas und ihre Behandlung[Psychological aspects of childhood obesity and itstreatment]. Paediatrica, 2004;15:24-26.

50. Schwimmer JB, Burwinkle TM, Varni JW. Health-relatedquality of life of severely obese children and adolescents.JAMA: J Am Med Inform Assoc. 2003;289:1813-1819.

51. Shloim N, Edelson LR, Martin N, Hetherington MM.Parenting styles, feeding styles, feeding practices, andweight status in 4-12 year-old children: a systematicreview of the literature. Front Psychol. 2015;6:1-20.

52. Sikorski C, Luppa M, Luck T, RiedelHeller SG. Weightstigma “gets under the skin”-evidence for an adaptedpsychological mediation framework—a systematicreview. Obesity. 2015;23(2):266-276.

53. Sleddens EF, Gerards SM, Thijs C, Vries NK, KremersSP. General parenting, childhood overweight andobesityinducing behaviors: A review. Int J Pediatr Obes.2011;6:12-27. doi: 10.3109/17477166.2011.566339

54. Spence JC, Carson V, Casey L, Boule N. Examiningbehavioral susceptibility to obesity among Canadian pre-school children: The role of eating behaviors. Int J PediatrObes. 2011. doi.org/10.3109/17477166.2010.512087

55. Spiel EC, Paxton SJ, Yager Z. Weight attitudes in 3-to 5-year-old children: age differences and cross-sectionalpredictors. Body Image. 2012;9(4):524-527.

56. Stice E, Presnell K, Shaw H, Rohde P. Psychological andbehavioral risk factors for obesity onset in adolescentgirls: a prospective study. J Consult Clin Psychol.2005;73:195–202.


57. Su W, Aurelia DS. Preschool children’s perceptions ofoverweight peers. Journal of Early Childhood Research.2012;10(1):19-31.

58. Suglia SF, Duarte CS, Chambers EC, Boynton-Jarrett R.Social and behavioral risk factors for obesity in earlychildhood. J Dev Behav Pediatr. 2013;34(8):549–556.doi:10.1097/DBP.0b013e3182a509c0

59. Tasca GA, Balfour L. Attachment and eating disorders: Areview of current research. Int J Eat Disord. 2014;47(7):710-717.

60. Tremblay, L., Lovsin, T., Zecevic, C., & Larivie´re, M.(2011). Perceptions of self in 3–5-year-old children: Apreliminary investigation into the early emergence ofbody dissatisfaction. Body Image, 8, 287-292.

61. Vgontzas AN, Lin HM, Papaliaga M, Calhoun S, Vela-Bueno A, Chrousos GP, et al. Short sleep duration andobesity: the role of emotional stress and sleepdisturbances. Int J Obes. 2008;32:801-809.

62. Waters SF, Virmani EA, Thompson RA, Meyer S, RaikesHA, Jochem R. Emotion regulation and attachment:

Unpacking two constructs and their association. JPsychopathol Behav Assess. 2010;32(1):37-47.

63. WHO (World Health Organization). Childhoodoverweight and obesity. Retrieved from http://www.who.int/dietphysicalactivity/childhood/en/. 2013.

64. Wick K, Leipold-Haas S, Dye L, Strauß B. Bindungsmuster, figurzufriedenheit und auffälligesessverhalten bei 13- bis 18-jährigen weiblichen undmännlichen Jugendlichen [Attachment, body satisfactionand conspicuous eating behavior in 13- to 18-year-oldfemale and male adolescents]. Kindheit und Entwicklung.2012;21(4):219-226.

65. Wong Y Chang Y-J, Lin C-J. The influence of primarycaregivers on body size and self-body image of preschoolchildren in Taiwan. Asia Pac J Clin Nutr. 2013;22(2):283-291.

66. Worobey J, Worobey HS. Body-size stigmatization bypreschool girls: In a doll's world, it is good to be“Barbie”. Body Image. 2014;11(2):171-174.


http://www.who.int/dietphysicalactivity/childhood/en/

http://www.who.int/dietphysicalactivity/childhood/en/

http://www.psycontent.com/content/yh654303g52165t0/



Documents

Overweight and Obesity in Early Childhood A Systematic ... · (e.g., genetic risk factors). The purpose of this review is to examine risk factors for childhood obesity with the goal