UNCORRECTED PROOF Research report Parent feeding restriction and child weight. The mediating role of child disinhibited eating and the moderating role of the parenting context Jessica L. Joyce *, Melanie J. Zimmer-Gembeck School of Psychology, Griffith University, Gold Coast 4222, Australia Few researchers or practitioners would dispute that parents are a major determinant of young children’s physical and social environ- ments. Because of this, parents influence children’s behaviours, habits, and attitudes through multiple socialisation processes (e.g., Darling & Steinberg, 1993; Maccoby, 1992). Such parental influences have been found to extend to children’s development of weight problems (e.g., Krebs & Jacobson, 2003) and understanding these influences can identify important avenues for prevention and intervention of children’s eating problems and overweight. Although there are multiple studies that have found associa- tions between parent feeding behaviours, such as prompting, encouraging or pressuring children to eat, restricting a child’s access to food or groups of foods, using food to control, and modelling and children’s weight (e.g., Birch & Fisher, 2000; Golan & Crow, 2004), recent evidence suggests that the influence of parenting on children’s overweight and obesity is likely to be more complex. For example, theory and empirical evidence suggest that the association between parenting and children’s weight may be mediated by children’s eating behaviours (Ventura & Birch, 2008), and that the association between specific parent feeding beha- viours and children’s eating may depend on the quality of the socioemotional climate of parent–child interactions (i.e., the parenting context) (Rhee, 2008; Rosenkranz & Dzewaltowski, 2008; Ventura & Birch, 2008). In this study, we tested these two possibilities. Our first aim was to test a mediational model of the indirect association between parent feeding restriction and child weight via child disinhibited eating. Given evidence that parent restriction may be implicated in the aetiology of children’s disinhibited eating (e.g., Birch, Fisher, & Davison, 2003; Fisher & Birch, 1999a, 1999b, 2002) and its association with child weight (e.g., Faith et al., 2004a), we focused on this feeding behaviour. Our second aim in this study was to examine how parenting dimensions may provide a context that can change the impact of parent restriction on children’s eating behaviour. In this case, we expected a negative parenting context to increase the negative impact of restriction in children’s eating behaviour, whereas a positive parenting context might reduce this association. Hence, parenting dimensions were expected to serve as moderators of the association between parent restriction and children’s disinhibited eating. Appetite xxx (2009) xxx–xxx 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 ARTICLE INFO Article history: Received 12 November 2008 Received in revised form 27 March 2009 Accepted 28 March 2009 Keywords: Feeding Parent restriction Feeding styles Parent–child relations Disinhibited eating Eating style Child overweight ABSTRACT Objective: We had two aims in this study of parenting and young children’s eating and weight. The first aim was to test whether the association of parental restriction with young children’s higher BMI is dependent on the intervening (i.e., mediating) role of their disinhibited eating. The second aim was to test how the parenting context may change the influence of parent restriction on children’s eating. Parenting dimensions of supportiveness and structure were expected to attenuate the positive association between parent restriction and child disinhibited eating, whereas parenting dimensions of coerciveness and chaos were expected to strengthen this association. Methods: Caregivers of children aged 4–8 years (N = 230, 48% female) completed questionnaires and children’s height and weight were measured to calculate BMI z-scores. Structural equation modelling and hierarchical regression analyses were conducted. Results: Children’s disinhibited eating partially mediated the association between parent restriction and children’s BMI. However, restriction was found to be directly associated with children’s BMI, and this direct association was stronger than the indirect one. Associations between restriction and children’s disinhibited eating differed depending on the parenting context in the feeding domain, including supportiveness, coerciveness and chaotic parenting. ß 2009 Published by Elsevier Ltd. * Corresponding author. E-mail addresses: [email protected], j.joyce@griffith.edu.au (J.L. Joyce). G Model APPET 782 1–9 Please cite this article in press as: Joyce, J. L., & Zimmer-Gembeck, M. J. Parent feeding restriction and child weight. The mediating role of child disinhibited eating and the moderating role of the parenting context. Appetite (2009), doi:10.1016/j.appet.2009.03.015 Contents lists available at ScienceDirect Appetite journal homepage: www.elsevier.com/locate/appet 0195-6663/$ – see front matter ß 2009 Published by Elsevier Ltd. doi:10.1016/j.appet.2009.03.015
Transcript
Appetite xxx (2009) xxx–xxx
12
3
4
5
6
789
10111213141516171819202122232425
G Model
APPET 782 1–9
EC
TED
PR
OO
F
Research report
Parent feeding restriction and child weight. The mediating role of childdisinhibited eating and the moderating role of the parenting context
Jessica L. Joyce *, Melanie J. Zimmer-Gembeck
School of Psychology, Griffith University, Gold Coast 4222, Australia
A R T I C L E I N F O
Article history:
Received 12 November 2008
Received in revised form 27 March 2009
Accepted 28 March 2009
Keywords:
Feeding
Parent restriction
Feeding styles
Parent–child relations
Disinhibited eating
Eating style
Child overweight
A B S T R A C T
Objective: We had two aims in this study of parenting and young children’s eating and weight. The first
aim was to test whether the association of parental restriction with young children’s higher BMI is
dependent on the intervening (i.e., mediating) role of their disinhibited eating. The second aim was to
test how the parenting context may change the influence of parent restriction on children’s eating.
Parenting dimensions of supportiveness and structure were expected to attenuate the positive
association between parent restriction and child disinhibited eating, whereas parenting dimensions of
coerciveness and chaos were expected to strengthen this association.
Methods: Caregivers of children aged 4–8 years (N = 230, 48% female) completed questionnaires and
children’s height and weight were measured to calculate BMI z-scores. Structural equation modelling
and hierarchical regression analyses were conducted.
Results: Children’s disinhibited eating partially mediated the association between parent restriction and
children’s BMI. However, restriction was found to be directly associated with children’s BMI, and this
direct association was stronger than the indirect one. Associations between restriction and children’s
disinhibited eating differed depending on the parenting context in the feeding domain, including
supportiveness, coerciveness and chaotic parenting.
� 2009 Published by Elsevier Ltd.
Contents lists available at ScienceDirect
Appetite
journa l homepage: www.e lsev ier .com/ locate /appet
262728293031323334353637383940414243
UN
CO
RRFew researchers or practitioners would dispute that parents are a
major determinant of young children’s physical and social environ-ments. Because of this, parents influence children’s behaviours,habits, and attitudes through multiple socialisation processes (e.g.,Darling & Steinberg, 1993; Maccoby, 1992). Such parental influenceshave been found to extend to children’s development of weightproblems (e.g., Krebs & Jacobson, 2003) and understanding theseinfluences can identify important avenues for prevention andintervention of children’s eating problems and overweight.
Although there are multiple studies that have found associa-tions between parent feeding behaviours, such as prompting,encouraging or pressuring children to eat, restricting a child’saccess to food or groups of foods, using food to control, andmodelling and children’s weight (e.g., Birch & Fisher, 2000; Golan &Crow, 2004), recent evidence suggests that the influence ofparenting on children’s overweight and obesity is likely to be morecomplex. For example, theory and empirical evidence suggest thatthe association between parenting and children’s weight may be
Please cite this article in press as: Joyce, J. L., & Zimmer-Gembeck, M. J.child disinhibited eating and the moderating role of the parenting c
0195-6663/$ – see front matter � 2009 Published by Elsevier Ltd.
doi:10.1016/j.appet.2009.03.015
mediated by children’s eating behaviours (Ventura & Birch, 2008),and that the association between specific parent feeding beha-viours and children’s eating may depend on the quality of thesocioemotional climate of parent–child interactions (i.e., theparenting context) (Rhee, 2008; Rosenkranz & Dzewaltowski,2008; Ventura & Birch, 2008).
In this study, we tested these two possibilities. Our first aim wasto test a mediational model of the indirect association betweenparent feeding restriction and child weight via child disinhibitedeating. Given evidence that parent restriction may be implicated inthe aetiology of children’s disinhibited eating (e.g., Birch, Fisher, &Davison, 2003; Fisher & Birch, 1999a, 1999b, 2002) and itsassociation with child weight (e.g., Faith et al., 2004a), we focusedon this feeding behaviour. Our second aim in this study was toexamine how parenting dimensions may provide a context thatcan change the impact of parent restriction on children’s eatingbehaviour. In this case, we expected a negative parenting contextto increase the negative impact of restriction in children’s eatingbehaviour, whereas a positive parenting context might reduce thisassociation. Hence, parenting dimensions were expected to serveas moderators of the association between parent restriction andchildren’s disinhibited eating.
Parent feeding restriction and child weight. The mediating role ofontext. Appetite (2009), doi:10.1016/j.appet.2009.03.015
Research suggests that emotional and external food cues mayoperate together to trigger overeating (e.g., van Strien, Schippers, &Cox, 1995), a concept often referred to as disinhibited overeating.The psychosomatic and externality theories of obesity linkovereating in response to negative emotions (emotional over-eating) and food cues in the external environment (external eating)to children’s increased energy intake and weight gain over time(e.g., Kaplan & Kaplan, 1957; Rodin & Slochower, 1976).
Laboratory studies have demonstrated positive associationsbetween children’s eating in the absence of hunger, a behaviouralmeasure of disinhibited overeating, and weight gain, overweightand obesity (e.g., Fisher & Birch, 2002; Francis, Ventura, Marini, &Birch, 2007; Moens & Braet, 2007; Shunk & Birch, 2004). Similarly,positive associations have been found between questionnairemeasures of emotional and external eating and children’s weight(e.g., Carnell & Wardle, 2006; Jahnke & Warschburger, 2008;Tanofsky-Kraff et al., 2008).
Parenting behaviours and styles related to children’s eating andweight
One aim of this study was to examine how parentingbehaviours and parenting dimensions are associated with chil-dren’s eating and weight. We focused on the parent feedingbehaviour of restriction and parenting dimensions specific to thecontext of parent interactions with children in the eating/feedingdomain (parenting dimensions). First, parent feeding behaviourshave been shown to influence children’s eating and parentrestriction often is the strongest correlate of children’s disinhibitedeating behaviour (Faith, Scanlon, Birch, Francis, & Sherry, 2004b;Ventura & Birch, 2008). Nevertheless, only in a small number ofrecent studies have parent restriction, children’s eating, andchildren’s weight been examined simultaneously (e.g., Birch et al.,2003; Fisher & Birch, 1999a; Francis & Birch, 2005; Powers,Chamberlin, van Schaick, Sherman, & Whitaker, 2006).
Positive associations between restriction, children’s eating in theabsence of hunger and children’s weight have been reported in threestudies (Birch et al., 2003; Fisher & Birch, 1999a; Francis & Birch,2005). For example in one longitudinal study (Francis &Birch, 2005),overweight mothers who reported more parent restriction haddaughters who showed greater increases in eating in the absence ofhunger between the ages of 5 and 9 years. Additionally, thesedaughters had greater increases in BMI from age 5 to 9 years. Thissuggests that it is problem eating behaviour that mediates theassociation between parent restriction and young children’s BMI,but no previous study has examined this possibility.
A second aim of this study was to examine the role of parentingdimensions when accounting for the association between restric-tion and children’s disinhibited eating. Parenting dimensionsinclude general, contextual features of parenting within whichmore proximal processes of parent–child interactions take place.These can include the emotional climate, predictability in theenvironment, and understood family patterns and practices. Thecontext is an important consideration, because it is the context thatmay identify family differences in the meaning of specific parentalfeeding practices. Hughes, Power, Fisher, Mueller, and Nicklas(2005) have developed the Caregiver’s Feeding Styles Question-naire (CFSQ), a measure of feeding-specific parenting thatdistinguished patterns of parent feeding style along the twoparenting style dimensions demandingness and responsiveness.
Among parenting researchers, the assessment of parentingdimensions beyond warmth and control or demandingness andinvolvement has been advocated (Darling & Steinberg, 1993).Dimensions that have been proposed are autonomy support or
Please cite this article in press as: Joyce, J. L., & Zimmer-Gembeck, M. J.child disinhibited eating and the moderating role of the parenting c
CTE
D P
RO
OF
autonomy granting, chaos in the environment, and hostility orrejection separate from warmth and involvement. Integratingacross many studies of parenting, Skinner, Johnson, and Snyder(2005) argued that there are six family context dimensions, threeof these are positive and three are more negative in tone. Thepositive dimensions are warmth/involvement, structure, andautonomy support. The negative dimensions are rejection,coercion, and chaos. A measure has been developed to tap thesedimensions with specific reference to the feeding and eatingcontext (Joyce & Zimmer-Gembeck, in press). In this study, we usedthis measure to assess these six parenting feeding dimensions.
Most studies examining the influence of more global parentingdimensions using general measures of parenting style (i.e., notspecific to the feeding and eating context) have provided tests ofdirect associations of style with young children’s eating and weight(e.g., Blissett & Haycraft, 2008; Patrick, Nicklas, Hughes, & Morales,2005; Rhee, Lumeng, Appugliese, Kaciroti, & Bradley, 2006). Thesehave conceptualised and measured parenting style in a variety ofways, which may in part explain the inconsistent findings acrossthe studies. For example, Rhee et al. used coded observations ofplay sessions to obtain a measure of general maternal sensitivityand expectations for self-control (to classify authoritative,authoritarian, permissive and neglectful parenting styles). Theyfound children (4–5 years) of authoritarian, permissive orneglectful mothers were more likely to be overweight 2 yearslater, compared to children with authoritative mothers. In contrast,other studies have failed to find associations between generalparenting style and children’s BMI and/or eating (Agras, Hammer,McNicholas, & Kraemer, 2004; Brann & Skinner, 2005; Chen &Kennedy, 2005; Wake, Nicholson, Hardy, & Smith, 2007b). Forexample, Blissett and Haycraft using the Parenting Styles andDimensions Questionnaire (PSDQ; to classify authoritative,authoritarian and permissive parenting styles) found no associa-tion between parenting style and children’s BMI (aged 2–5 years).
When studies of other child behaviours are examined, it seemsthat parenting dimensions may moderate the efficacy of specificparenting behaviours—that is, the parenting context influences theeffectiveness or meaning of specific parenting practices orbehaviours (e.g., Darling & Steinberg, 1993). For example,restriction might not have negative effects on children’s eatingwhen the parenting context is generally warm and supportive, butmay have such effects when parental supportiveness is low.Nevertheless, only one previous study has examined the parentingcontext as a moderator of the association between restrictivefeeding practices and child eating (Musher-Eizenman & Holub,2006). Using a measure of general parenting style, they found thatauthoritarian and authoritative parents did not differ in theamount of restriction reported. However, authoritarian motherswho used more restriction had children who had more externaleating behaviour. Yet, authoritative fathers who used morerestriction had children with less external eating behaviour. Inthis study, our second aim was to examine the parenting context asa moderator of the association between the parent restriction andchildren’s disinhibited eating, with a broader range of parentingdimensions than has been previously assessed.
In sum, this study tested a mediational model in order toidentify whether the association between parent restriction andchild weight is an indirect one, which is mediated by childdisinhibited eating behaviour. Also, feeding-specific parentingdimensions were examined as moderators of the associationbetween parent restriction and child disinhibited eating. It waspredicted that the association between restriction and children’seating would be weaker when parents reported a warmer, moresupportive and more structured parenting context, but theassociation would be stronger when parents used more coercionand were more chaotic in their parenting.
Parent feeding restriction and child weight. The mediating role ofontext. Appetite (2009), doi:10.1016/j.appet.2009.03.015
Participants were 247 caregivers (94% female) and theirchildren aged 4–8 years (M = 5.74 years, SD = 0.89; 48% female).Children were attending preschool or grades 1 or 2 in two primaryschools in Australia. Parents who participated had primaryresponsibility for the planning and preparation of family meals.Overall, 49% of parents returned questionnaires. Thirteen partici-pants were excluded from data analyses due to excessive missingdata. In addition, four parents completed a questionnaire for two oftheir children (four sets of twins). One of each was randomlyremoved, leaving a final sample size of 230.
Of these 230 parents, child BMI and parent BMI were availablefor 211 and 227 participants, respectively. Nineteen children werenot at school on days of measurement or parents did not consentand three parents failed to self-report both their weight and height.The average age of caregivers was 37.2 years (SD = 4.2, Median = 37years). The majority of the sample (94%) reported their race/ethnicity as White/Caucasian/European. Approximately a third ofcaregivers (30%) reported completing a university degree, a ratecomparable with the Australian population.
Measures
Child weight
Children’s height (m) and weight (kg) measurements wereobtained using a calibrated digital scale and a fixed measuringtape. Weight was recorded to the nearest 0.1 kg and height wasrecorded to the nearest 0.5 cm. BMI z-scores were calculated usingCDC criteria (Centers for Disease Control and Prevention, 2000).Based on CDC definitions, 8.1% of children were classified as at riskfor overweight (�85th to <95th percentiles) and 4.7% of childrenwere classified as overweight (�95th percentile). Based on theInternational Obesity Task Force (IOTF) standard definitions (Cole,Bellizzi, Flegal, & Dietz, 2000), 10% of children were classified asoverweight or obese. These figures are lower than Australianoverweight and obesity prevalence statistics for these age groups(e.g., Vaska & Volkmer, 2004; Wake, Hardy, Canterford, Sawyer, &Carlin, 2007a), but more similar to those reported for the this agegroup and region of Australia (Abbott et al., 2007).
Child disinhibited eating
In accordance with Ouwens, van Strien, and van der Staak(2003) and Moens and Braet (2007), a composite score ofemotional and external eating was used to measure disinhibitedeating. Children’s external and emotional eating were measuredwith the food responsiveness (5 items; e.g., ‘‘Even if my child is fullup, s/he finds room to eat his/her favourite food’’) and emotional
overeating (4 items; e.g., ‘‘My child eats more when worried’’)subscales of the Children’s Eating Behaviour Questionnaire (CEBQ:Wardle, Guthrie, Sanderson, & Rapoport, 2001). All items had a 5-response option ranging from 1 (never or almost never) to 5 (always
or almost always). See Carnell and Wardle (2007b) for behaviouralvalidation data of the CEBQ.
In the current study, Cronbach’s a’s were .70 for foodresponsiveness and .60 for emotional overeating. Based on thecriteria that a subscale item should be eliminated when thesubscale reliability was <.65 and removal of a subscale itemimproved reliability by>.05, one item on the emotional overeatingsubscale (‘‘My child eats more when s/he has noting else to do’’)was removed. Removal of this item increased the Cronbach’s a to.83. The remaining eight items were averaged with higher scoresindicating more disinhibited eating. The Cronbach’s a for the eightitems was .72.
Please cite this article in press as: Joyce, J. L., & Zimmer-Gembeck, M. J.child disinhibited eating and the moderating role of the parenting c
TED
PR
OO
F
Parent feeding restriction
The restriction subscale of the Child Feeding Questionnaire(CFQ: Birch et al., 2001), a parent report questionnaire designed foruse with parents of children ranging in age from 2 to 11 years, wasused to measure the parent feeding behaviour of restriction. Itemsassessed the extent to which a parent restricted her/his child’saccess to foods (8 items; e.g., ‘‘If I did not guide or regulate mychild’s eating, s/he would eat too much of her favourite foods’’). Theoriginal 5-point response scale was modified to a 7-point scale thatranged from 1 (extremely untrue) to 7 (extremely true) to allow forgreater variability in responses. Items were averaged and higherscores indicated more feeding restriction. In the current study theCronbach’s a was .82.
Parent feeding dimensions
The Parent Feeding Dimensions Questionnaire (PFDQ; Joyce &Zimmer-Gembeck, in press) was used to assess multiple parentalfeeding-specific dimensions. The PFDQ is a 32-item measuredesigned for completion by parents of children aged 4–7 years. ThePFDQ includes six dimensions of the family context that collapseinto four subscale scores, including supportiveness (2 dimensions ofwarmth and autonomy support), structure, coerciveness (2 dimen-sions of rejection and coercion), and chaos.
The supportiveness scale assesses a parent’s expression ofaffection, kindness, enjoyment, regard, and support within thefood domain and the extent to which a parent supports her/hischild to make good decisions about eating by providing appealingoptions, and socialises healthy eating without being overlycoercive (10 items; e.g., ‘‘I often take my child grocery shoppingand ask him/her to help choose what healthy foods to buy’’). Thestructure scale assesses the extent to which a parent providesinformation to his/her child about expectations for behaviour,maintains consistent guidelines, and sets appropriate limits withregard to eating (6 items; e.g., ‘‘I maintain consistent expectationsfor my child regarding eating’’).
The coerciveness scale assesses a parent’s overreactivity,irritability, and communication of negative feelings such asdisapproval of her/his child’s eating behaviour and the extent towhich a parent is extremely restrictive and controlling in thefeeding domain (10 items; e.g., ‘‘Sometimes to get my child to eatthe healthy foods I expect him/her to eat, I have to raise my voice’’).The chaos scale measures inconsistent, unpredictable, arbitrary,and/or undependable parenting in the feeding and eating context(6 items; e.g., ‘‘When my child does not follow my advice or rulesabout healthy eating, my reaction is not very predictable’’).
All items have response options ranging from 1 (extremely
untrue) to 7 (extremely true). Subscale items were averaged to formcomposite measures. Higher scores indicated a higher degree ofthe parenting dimension in question. In this study, the suppor-tiveness, structure, coerciveness, and chaos subscales of the PFDQhad interitem correlations of a = .81, .72, .92, and .80, respectively.Face, content and construct validity of the PFDQ have beenestablished (Joyce & Zimmer-Gembeck, in press).
Parent body mass index (BMI)
Parent weight and height were obtained via self-report. In total,22.9% of parents were classified as overweight (�25 and <30 kg/m2) and 7.9% were classified as obese (�30 kg/m2) (World HealthOrganisation, 1998).
Procedure
A consent and questionnaire package including a reply paidenvelope for the return of the questionnaire was mailed to eachcaregiver to complete at home. The Human Subjects ResearchCommittee of the university in which this research was conducted,
Parent feeding restriction and child weight. The mediating role ofontext. Appetite (2009), doi:10.1016/j.appet.2009.03.015
approved the data collection procedure. Following parentalconsent, children’s height and weight measurements werecollected during school hours. Data were collected from individualchildren in a quiet and private space. Measurements were alwayscompleted with two researchers present, both who held childsafety cards for working with children.
Overview of the analytic strategy
Mediation model. Structural equation modelling (SEM: Kaplan,2000) was used to test the mediational model. SEM was completedusing maximum likelihood estimation available within AMOSsoftware (SmallWaters Corporation, 1999). Model fit was assessedwith commonly available fit indices, including the x2 test statisticand the Comparative Fit Index (CFI) (Bentler & Bonett, 1980) andthe root mean square error of approximation (RMSEA) (Browne &Cudek, 1999).
To assess the significance of mediational paths, we usedbootstrapping (see Shrout & Bolger, 2002). Bootstrapping is aresampling method that assists with making statistical inferencesabout a population from which a sample is drawn. It involvestaking repeated, random selections of cases from a study samplewith replacement after each sampling (Good, 1999). Bootstrappingtechniques can be used in SEM and are especially helpful in cross-sectional mediational path models when the sample size is small tomoderate (N < 400) and variables are not normally distributed.Bootstrapping has been found to increase power and accuracy bynot depending on normal theory assumptions but, instead,drawing estimates from the data (Shrout & Bolger, 2002). Inaddition to using bootstrapping, which increases power with thecurrent sample size, a number of guidelines suggest that the SEMmodel in this study had an adequate sample size for the number ofparameters estimated (Bentler & Chou, 1987; Loehlin, 1992;Stevens, 1996). Bootstrapping was used to estimate the standarderrors (SEs) and 95% confidence intervals (CIs) for all modelestimates (Shrout & Bolger, 2002). The bootstrapped SEs and 95%CIs were used to make conclusions about the full, partial, or nomediational role of child disinhibited eating.
Moderation models
Following procedures outlined by Holmbeck (1997) andJaccard, Turrisi, and Wan (1990), hierarchical regression analyseswere conducted to examine parenting dimensions as moderatorsof the association between parent restriction and child disinhibitedeating. In these models, child eating was the dependent variable.As suggested by Jaccard et al., centring (i.e., the mean of thevariable subtracted from raw scores for each participant) of theindependent variable (restriction) and moderators (parentingdimensions) was conducted to reduce the potential problem ofmulticollinearity. In Step 1, relevant covariates and the centred
UN
CTable 1Means, standard deviations, and zero-order correlations between study variables.
1 2 3
1. Child BMI z-score –
2. Child disinhibited eating .30** –
3. Parent feeding restriction .23** .35** –
Parenting dimensions
4. Supportiveness .03 .00 �.09
5. Structure �.11 .07 �.07
6. Coerciveness .05 .16* .43**
7. Chaos .06 .11 .36**
8. Parent BMI .08 �.02 .13
Mean (SD) 0.09 (0.88) 1.59 (0.48) 4.43 (1.13)
* p < .05.** p < .01.
Please cite this article in press as: Joyce, J. L., & Zimmer-Gembeck, M. J.child disinhibited eating and the moderating role of the parenting c
CTE
D P
RO
OF
parent restriction and parenting dimension scores were entered asindependent variables. In Step 2, an interaction term betweenparent restriction and the parenting dimension was added. Theinteraction term was computed by multiplying the centredrestriction and parenting dimension scores. Power estimated inthis study was very high (.99) for each regression model whenbased upon the number of independent variables and the observedtotal R2 after the final step (Hair, Anderson, Tatham, & Black, 1998;Soper, 2009).
Results
Descriptive statistics and zero-order correlations
Table 1 summarizes means, standard deviations, and zero-ordercorrelations between measurement variables. As expected, childBMI was significantly positively associated with disinhibitedeating, r = .30, p < .01. Hence, children with a higher BMI hadparents who reported that their children showed more disin-hibited eating. With regards to intercorrelations betweenmeasures of parenting, the parenting dimensions of coercion,chaos, supportiveness and structure were significantly intercorre-lated, in the expected directions (see Table 1), and parents’ reportof their restriction was positively associated with coerciveness andchaos, r = .43 and .36, respectively, both p < 01, but not sig-nificantly associated with the positive parenting dimensions (i.e.,supportiveness and structure). Of the five measures of parenting,only restriction was associated with child BMI, r = .23, p < .01. Yet,two parenting measures were associated with children’s disin-hibited eating; parents who reported that their children showedmore disinhibited eating behaviour reported that they were bothmore restrictive and coercive with their children, r = .35, p < .01and r = .16, p < .05, respectively.
Associations between study variables and demographic vari-ables (i.e., number of children under 18 years in the household,child age and gender, and parental BMI, education level, and racial/ethnic background) also were examined (data not shown). Apositive association between parent BMI and the parentingdimension chaos, r = .22, p < .01, was the only significant correla-tion and this association remained after controlling for parenteducation level and racial/ethnic background (partial r = .22 and.23, both p < .01, respectively). Therefore parent BMI was includedas a covariate in remaining analyses.
Child eating as a mediator
Two models were estimated to determine if child disinhibitedeating behaviour mediated the association between parentrestriction and child BMI. In Model 1, only the direct pathsbetween parent restriction and child BMI (X! Y) and between
Table 2Bootstrapped path estimates, SEs and 95% CIs for models predicting child BMI z-score with and without child disinhibited eating as a mediator (N = 208).
Model paths Unstandardised, B Standardised, b
Estimate (SE) Lower 95% CI Upper 95% CI Estimate (SE) Lower 95% CI Upper 95% CI
Echild disinhibited eating and child BMI (M! Y) were tested. Model2 included the hypothesized direct and indirect paths betweenparent restriction, child disinhibited eating and child BMI. ParentBMI was freed to have a direct effect on child BMI in both models.
Table 2 summarizes maximum likelihood estimates andbootstrapped SEs and CIs for both models. In Model 1, the directpath from parent restriction to child BMI was positive andmarginally significant, standardised path coefficient = .14, p = .05.However, the lower CI of the direct effect of restriction on child BMIwas .03, indicating that the bootstrapped direct effect ofrestriction! child BMI was significantly different from zero.The path from child disinhibited eating to child BMI was positiveand significant, standardised path coefficient = .26, p < .05. Themodel had a poor fit to the data, x2 (2, N = 208) = 31.1, p < .001,CFI = .46, RMSEA = .265 (90% CI .188–.351), however, and the pathfrom parent BMI to child BMI was not significant.
As can be seen in Table 2 and Fig. 1, Model 2 involved freeingone additional path from parent restriction to child disinhibitedeating. This path was freed in order to test the strength of theindirect path from parent restriction to child BMI via childdisinhibited eating. This model had a good fit to the data, x2 (1,N = 208) = 1.5, p = .223, CFI = .99, RMSEA = .048 (90% CI .000–.199).Supporting the hypothesized mediation model, the x2 differencetest showed that including the indirect path from parentrestriction to child BMI via child disinhibited eating in the modelsignificantly improved the fit of data to the model, x2 difference(1) = 29.6, p < .05.
UN
Fig. 1. Unstandardised and standardised SEM maximum likelihood estimates from testi
statistics: x2 (1, N = 208) = 1.5, p = .223, CFI = .99, RMSEA—.048 (90% CI .000–.199). *p <
Please cite this article in press as: Joyce, J. L., & Zimmer-Gembeck, M. J.child disinhibited eating and the moderating role of the parenting c
TED
As expected, the direct paths from (1) parent restriction to child
disinhibited eating and (2) child disinhibited eating to child BMIwere significant, standardised coefficient = .37 and .26, respectively,both p < .001. The direct path from parent restriction to child BMIremained unchanged from Model 1, standardised coefficient = .14,p = .05. The 95% CI of the indirect effect of restriction on child BMI viachild eating did not include 0, indicating that this path wassignificant and criteria for mediation was met. However the lower CIfor this indirect path was close to 0 (.01) indicating that this effectwas small and only marginally significant. In summary, consideringthe confidence intervals and indirect and direct estimates, children’sdisinhibited eating partially mediated the link between parentrestriction and children’s BMI. Parent restriction had a directassociation with children’s BMI, but also had a small, indirectassociation via disinhibited eating.
Parenting dimensions as moderators
A second aim of the current study was to determine whetherthe association between parent feeding restriction and childdisinhibited eating depended upon the parenting dimensions ofsupportiveness, structure, coerciveness, and chaos. These associa-tions were tested by regressing child disinhibited eating oninteraction effects formed as products of restriction and parentingdimensions.
Both of the negative parenting dimensions played significantmoderating roles in the associations between restriction and
ng the mediational role of child disinhibited eating (also see Table 2, Model 2). Fit
.05. **p < .01.
Parent feeding restriction and child weight. The mediating role ofontext. Appetite (2009), doi:10.1016/j.appet.2009.03.015
children’s disinhibited eating (see Tables 3 and 4). As shown inTable 3, parent BMI, restriction, and coerciveness explained 13.0%of the variance in children’s eating in Step 1 of a hierarchicalregression, Fchg(3, 223) = 11.07, p < .001. A further significant 3.0%of the variance was explained when the interaction term of parentcoerciveness and restriction was entered in Step 2, Fchg (1,222) = 7.98, p < .01. This interaction was explored by using theregression equation to calculate child disinherited eating at high,medium, and low values of parent feeding restriction and parentcoerciveness. The medium value was 0 (i.e., the centred mean ofeach variable), one standard deviation above the mean was set asthe high value, and one standard deviation below the mean was setas the low value. As can be seen in Fig. 2, the association betweenparent restriction and children’s disinhibited eating was morepositive at higher levels of parent coerciveness.
Similar findings emerged for the parenting dimension of chaos.In Step 1 of a hierarchical regression model, parent BMI, restriction,and chaos explained 12.9% of the variance in children’s eating,Fchg(3, 223) = 11.03, p < .001 (see Table 4). A further significant3.3% of the variance was explained when the interaction term ofparent chaos and restriction were entered in step in Step 2, Fchg (1,222) = 8.68, p < .01. The pattern was similar to that found forparent coerciveness; the association between parent restrictionand children’s disinhibited eating was more positive at higherlevels of parent chaos (see Fig. 3). Neither supportiveness norstructure moderated the association between parent restriction
Please cite this article in press as: Joyce, J. L., & Zimmer-Gembeck, M. J.child disinhibited eating and the moderating role of the parenting c
and children’s disinhibited eating (data not shown). However, theinteraction between parent supportiveness and restrictionapproached significance, p = .08, with a weaker associationbetween restriction and children’s disinhibited eating withincreasing levels of parent supportiveness.
Discussion
The findings of this study identify some aspects of parentingassociated with children’s eating and weight. Although the resultsof the study are correlational and do not provide evidence thatparents play a socialising role in young children’s disinhibitedeating and BMI, they contribute to the evidence that some of theassociation of parenting with children’s BMI may be mediated by
Parent feeding restriction and child weight. The mediating role ofontext. Appetite (2009), doi:10.1016/j.appet.2009.03.015
children’s disinhibited eating behaviour, and some parentingbehaviours may have a differential association with children’seating depending on the parent feeding context.
More specifically, these results supported our expectation thatthe association between parent restriction and children’s weightwould be mediated through children’s eating. We also expectedand found that associations between restriction and children’seating would depend on more general parenting dimensions in thefeeding domain, including supportiveness, coerciveness andchaotic parenting.
The mediating role of children’s eating
First, regarding the mediating role of children’s disinhibitedeating behaviour in the association between restrictive parentingbehaviour and children’s weight, partial mediation was found.Hence, parents who are more restrictive have children who engagein more disinhibited eating. It is children’s eating behaviour that, inturn, is associated with higher BMI. However, we also found thatrestriction is directly associated with children’s weight, and thatthis direct association is stronger than the indirect one.
Although this study was correlational, our findings support theview that restrictive feeding practices may play a causal role in thedevelopment of children’s disinhibited eating. For example, usingan experimental design, Fisher and Birch (1999b) found thatchildren ate more and had an increased preference for a snack foodthat was experimentally restricted, and that children restricted athome were more responsive to the experimental restriction. It hasbeen hypothesised that restrictive practices teach the child toassociate eating with external (or other) cues rather than internalcues of hunger and satiety, which subsequently inhibits the child’sability to self-regulate his or her own energy intake. Longitudinaldesigns that establish earlier parent restriction to be associatedwith later change in child eating behaviour and BMI, over andabove initial child eating and BMI and relevant covariates, areneeded to make further conclusions about the nature of theserelationships.
It is important to highlight those associations betweenparenting and children’s behaviours outside the eating and foodcontext have been found to be bidirectional (e.g., Chamberlain &Patterson, 1995; Crouter & Booth, 2003). It is expected that suchbidirectional associations would also be found when studyingparents and children’s eating and weight, and that children’sbehaviours and weight status probably change parenting as muchas parenting behaviours socialise children. For example, collec-tively, data from a small number of studies suggest that parentrestriction of children’s eating is, at least, partially elicited bychildren’s weight status (Birch & Fisher, 2000; Birch et al., 2003;Faith, Berkowitz et al., 2004). The results of this study providesupport for the hypothesis that the influence on parent restrictingbehaviours on child weight may be mediated through child eating.However, the findings do not rule out that possibility thatchildren’s eating behaviours and weight change parents’ use offeeding restriction. Longitudinal research is needed to test suchbidirectional effects.
Associations between parent restriction and children’s weight,whether direct or indirect, were small to moderate. Hence,additional mechanisms or factors are involved in children’s weightthan were included here. Parents may play a role, but, as describedin recent reviews (Davison & Birch, 2001; Lobstein, Baur, & Uauy,2004), there are many other factors and mechanisms that may beinvolved. Additional factors that may be involved in the associationbetween parent restriction and children’s BMI include parentperception and concern about their child’s weight, both of whichhave been found to be associated with restrictive feeding practices(Fisher & Birch, 1999a; Francis, Hofer, & Birch, 2001; Ogden,
Please cite this article in press as: Joyce, J. L., & Zimmer-Gembeck, M. J.child disinhibited eating and the moderating role of the parenting c
TED
PR
OO
F
Reynolds, & Smith, 2006). For example, May and colleagues (2007)found maternal restriction to be associated with maternal concernabout children becoming overweight, but not children’s weightstatus.
In the context of a bidirectional parent–child feeding dynamic, apotential key variable involved in the association between parentrestriction and children’s weight may be a parents’ concern abouttheir children’s weight. How parents respond to children’s weightmay be influenced by their own subjective beliefs, perceptions,interpretations, and expectations for children (e.g., Bugental &Shennum, 1984; Dix, Ruble, Grusec, & Nixon, 1987; Miller, 1995),and a parent’s beliefs and attitudes about eating and acceptableweight may play a more important role in determining theirconcern about their children’s weight than the objective weightstatus of the child.
It is important to note that while a number of studies have failedto find positive relationships between parent restriction andchildren’s BMI (e.g., Carnell & Wardle, 2007a), or that thisassociation is moderated by factors such as maternal weightstatus (e.g., Faith, Berkowitz et al., 2004), the current study findingssupport those who have found a positive association in demo-graphically similar groups of parents and children (Fisher & Birch,1999a; Francis et al., 2001). Furthermore, although we controlledfor parent BMI in all models tested, it was not significantlyassociated with children’s BMI. Parent BMI, particularly maternalBMI, is a robust correlate of child BMI (e.g., Danielzik, Langnase,Mast, Spethmann, & Muller, 2002; Maffeis, Talamini, & Tato, 1998;Strauss & Knight, 1999). It is likely that the correlation betweenparent and child BMI in this study was not significant due to thesmall sample size, as the size of the correlation between parent andchild BMI in this study is comparable to that in other larger scalestudies (e.g., Williams, 2001). Moreover, our finding is consistentwith studies that suggest that parental BMI and child BMI may notshow high correlations until children are out of the preschool years(e.g., Safer, Agras, Bryson, & Hammer, 2001; Whitaker, Deeks,Baughcum, & Specker, 2000).
The moderating effect of parenting dimensions
Our second set of findings showed how parent feedingrestriction may have somewhat stronger or weaker associationswith children’s disinhibited eating depending on the quality of thesocioemotional climate of parent–child interactions (i.e., parentingdimensions). The findings were most clear for the moderating rolesof negative parenting dimensions of coerciveness and chaos.Stronger associations between restriction and children’s disin-hibited eating were found when parents reported more coercive-ness and chaos in the eating and feeding arenas. For example, morecoerciveness was indicated by parents’ overreactivity, irritability,and the extent to which they are controlling in feeding and eatingcontext, whereas more chaos was indicated by the extent to whicha parent is inconsistent and unpredictable in the feeding domain.
Multiple studies have reported direct associations betweenmore global parenting dimensions using general measures ofparenting style (i.e., not specific to the feeding and eating context)and young children’s eating and weight (e.g., Blissett & Haycraft,2008; Patrick et al., 2005; Rhee et al., 2006), but we are aware ofonly one previous study that has examined parenting style as acontextual variable in the association of parent restriction andchild eating behaviour (Musher-Eizenman & Holub, 2006). Thisstudy has expanded on this work with the assessment of additionalparenting dimensions founded in theory about the importantcontextual elements of parenting and the family (Skinner et al.,2005). Our finding of a stronger association between restrictionand children’s disinhibited eating when parents reported a morecoercive parenting style is similar to the finding that authoritarian
Parent feeding restriction and child weight. The mediating role ofontext. Appetite (2009), doi:10.1016/j.appet.2009.03.015
(high demandingness/low responsiveness) parenting may havenegative impacts on children’s ability to regulate their energyintake (Musher-Eizenman & Holub, 2006).
Particularly new with this study is the concept of parentingchaos in relation to children’s eating. Although the concept ofstructure regarding mealtimes and eating and general familyfunctioning such as daily routines has received some attention inthe literature (Baughcum et al., 2001; Cullen, Baranowski,Rittenberry, & Olvera, 2000; Sallis et al., 1995; Wilkins, Kendrick,Stitt, Stinett, & Hammarlund, 1998), chaos as a parentingdimension includes lack of structure but also directly incorporatesitems to assess whether parents are more inconsistent, erratic,unpredictable, and undependable rather than only having diffi-culties with structure (Skinner et al., 2005). Also, chaos is not just aproxy for adverse parental psychological and social circumstances(e.g., Dumas et al., 2005). Our finding of a stronger associationbetween restriction and children’s disinhibited eating whenparents reported more chaotic parenting suggests that homeenvironments where parents are inconsistent, erratic in theireating schedules, and harder to predict regarding healthy versusunhealthy eating patterns may have more adverse influence ontheir children’s disinhibited eating when they use more feedingrestriction.
This study found a marginally weaker association betweenrestriction and children’s disinhibited eating with increasing levelsof parent supportiveness in the feeding domain, but no moderationeffect for structure. Despite evidence suggesting the negativeeffects of parent restriction, it has been argued that given ourcurrent food environment of abundant access to highly palatable,energy dense foods, some restriction of children’s food intake maybe necessary (Wadden, Brownell, & Foster, 2002). Our findingssuggest that the negative effects of restriction on children’sdisinhibited eating may be buffered when parents socialise healthyeating in a supportive, warm, and autonomy granting way.Correspondingly, Satter (1995) and others (e.g., Birch, Fisher, &Grimm-Thomas, 1996) have discussed how children’s self-regula-tion of eating may be facilitated when they are able to activelyparticipate in the feeding process. In summary, however, it was thenegative parenting dimensions that are the contextual variableshaving a stronger influence on the association between parentfeeding restriction and child disinhibited eating.
Study limitations
Study limitations include the use of a cross-sectional researchdesign to test pathways and mediational relationships, the use ofall self-report (except child BMI) and the possibility of sharedmethod variance, and measurement issues. Furthermore, theseresults only reflect associations among a sample of predominantlyWhite participants with a relatively low prevalence of overweight/obesity. From a measurement perspective, further work needs tobe done in relation to the measurement of disinhibited eatingamong young children. Parents in this study and other studiesreport a relatively low frequency disinhibited eating among theirchildren (Carnell & Wardle, 2007b; Jahnke & Warschburger, 2008;Powers et al., 2006). This could mean that there is a low prevalenceof disinhibited eating among young children. Alternatively thiscould reflect difficulties in parents’ ability to recognise thisbehaviour among their children.
Similarly, evidence has been provided that work needs to bedone in the conceptualisation of the restriction scale of the ChildFeeding Questionnaire (Anderson, Hughes, Fisher, & Nicklas,2005; Corsini, Danthiir, Kettler, & Wilson, 2008). Furthermore,some have argued that this work also needs to include aconsideration of parents’ motivations underlying restrictionand how parents enact this parent behaviour. For example,
Please cite this article in press as: Joyce, J. L., & Zimmer-Gembeck, M. J.child disinhibited eating and the moderating role of the parenting c
CTE
D P
RO
OF
Musher-Eizenman and Holub (2006) have illustrated how theconceptualisation of restriction should take into account themotivation underlying this practice (e.g., health versus weightcontrol reasons) and Ogden and colleagues (2006) show that adistinction can be made between covert versus overt restrictivepractices. However, we anticipate that parenting dimensions,such as those measured here, are one way to gather informationthat accounts for these distinctions and differing motivations ofparents. While parenting behaviours can be described as what
parents do, parenting dimensions tap how parents do it (e.g., withwarmth or hostility). Parents who provide a warm and responsiveenvironment for their children may have different motivations forrestriction and use more covert restriction when compared toparents who are more coercive in their parenting. Future researchcould test these possibilities.
This study provides further support that child eating may be animportant factor in the association between parent restriction andchild weight. It is important that future studies include measuresof the family context. Doing so could add to the identification ofpathways to child eating and weight problems, and could provideinformation useful for prevention and intervention efforts.
References
Abbott, R. A., Macdonald, D., Mackinnon, L., Stubbs, C. O., Lee, A. J., Harper, C., et al.(2007). Healthy kids Queensland survey 2006—summary report. Brisbane: Queens-land Health.
Agras, W. S., Hammer, L. D., McNicholas, F., & Kraemer, H. C. (2004). Risk factors forchildhood overweight: a prospective study from birth to 9.5 years. Journal ofPediatrics, 145(1), 20–25.
Anderson, C. B., Hughes, S. O., Fisher, J. O., & Nicklas, T. A. (2005). Cross-culturalequivalence of feeding beliefs and practices: the psychometric properties of thechild feeding questionnaire among Blacks and Hispanics. Preventive Medicine, 41,521–531.
Baughcum, A. E., Powers, S. W., Johnson, S. B., Chamberlin, L. A., Deeks, C. M., Jain, A.,et al. (2001). Maternal feeding practices and beliefs and their relationships tooverweight in early childhood. Developmental and Behavioral Pediatrics, 22(6), 391–408.
Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness-of-fit in theanalysis of covariance structures. Psychological Bulletin, 88, 588–606.
Bentler, P. M., & Chou, C. P. (1987). Practical issues in structural modeling. SociologicalMethods and Research, 16(1), 78–117.
Birch, L. L., & Fisher, J. O. (2000). Mothers’ child-feeding practices influence daughters’eating and weight. American Journal of Clinical Nutrition, 71, 1054–1061.
Birch, L. L., Fisher, J. O., & Davison, K. K. (2003). Learning to overeat: maternal use ofrestrictive feeding practices promotes girls’ eating in the absence of hunger.American Journal of Clinical Nutrition, 78, 215–220.
Birch, L. L., Fisher, J. O., & Grimm-Thomas, K. (1996). The development of children’seating habits. In H. L. Meiselman & H. J. MacFie (Eds.), Food, choice, acceptance andconsumption (pp. 161–206). London: Blackie Academic and Professional.
Birch, L. L., Fisher, J. O., Grimm-Thomas, K., Markey, C. N., Sawyer, R., & Johnson, S. L.(2001). Confirmatory factor analysis of the Child Feeding Questionnaire: a measureof parental attitudes, beliefs and practices about child feeding and obesity prone-ness. Appetite, 36, 201–210.
Blissett, J., & Haycraft, E. (2008). Are parenting style and controlling feeding practicesrelated? Appetite, 50, 477–485.
Brann, L. S., & Skinner, J. D. (2005). More controlling child-feeding practices are foundamong parents of boys with an average body mass index compared with parents ofboys with a high body mass index. Journal of the American Dietetic Association, 105,1411–1416.
Browne, M. W., & Cudek, R. (1999). Alternative ways of assessing model fit. In K. A.Bollen & J. S. Long (Eds.), Testing structural equation models (pp. 136–162). NewburyPark, CA: Sage.
Bugental, D. B., & Shennum, W.A. (1984). Difficult children as elicitors and targets ofadult communication: an attributional behavioral transactional analysis. Mono-graphs of the Society for Research in Child Development, 49 (1, Serial No. 205).
Carnell, S., & Wardle, J. (2006). Associations between child eating behaviour andadiposity in 3–5 years old. Appetite, 47, 257–279.
Carnell, S., & Wardle, J. (2007a). Associations between multiple measures of parentalfeeding and children’s adiposity in United Kingdom preschoolers. Obesity, 15, 37–144.
Carnell, S., & Wardle, J. (2007b). Measuring behavioural susceptibility to obesity:validation of the child eating behaviour questionnaire. Appetite, 48(1), 104–113.
Centers for Disease Control and Prevention. (2000). EpiInfo’s Nutrition program,Nutstat. Retrieved June 09, 2008, from http://www.cdc.gov/epiinfo/community-health.htm.
Chamberlain, P., & Patterson, G. R. (1995). Discipline and child compliance in parenting.In Bornstein, M. H. (Ed.). Handbook of parenting: applied and practical parenting. Vol.4 (pp.205–225). Mahwah, NJ: Erlbaum.
Parent feeding restriction and child weight. The mediating role ofontext. Appetite (2009), doi:10.1016/j.appet.2009.03.015
Chen, J. L., & Kennedy, C. (2005). Factors associated with obesity in Chinese-Americanchildren. Pediatric Nursing, 31(2), 110–115.
Cole, T. J., Bellizzi, M. C., Flegal, K. M., & Dietz, W. H. (2000). Establishing a standarddefinition for child overweight and obesity worldwide: international survey.British Medical Journal, 320, 1–6.
Corsini, N., Danthiir, V., Kettler, L., & Wilson, C. (2008). Factor structure and psycho-metric properties of the Child Feeding Questionnaire in Australian preschoolchildren. Appetite, 51, 474–481.
Crouter, A. C., & Booth, A. (2003). Children’s influence on family dynamics: the neglectedside of family relationships. Mahwah, NJ: Erlbaum.
Cullen, K. W., Baranowski, T., Rittenberry, L., & Olvera, N. (2000). Social-environmentalinfluences on children’s diets: results from focus groups with African-, Euro- andMexican-American children and their parents. Health Education Research, 15(5),581–590.
Danielzik, S., Langnase, K., Mast, M., Spethmann, C., & Muller, M. (2002). Impact ofparental BMI on the manifestation of overweight 5-7 year old children. EuropeanJournal of Nutrition, 41, 132–138.
Darling, N., & Steinberg, L. (1993). Parenting style as context: an integrative model.Psychological Bulletin, 113, 487–496.
Davison, K. K., & Birch, L. L. (2001). Childhood overweight: a contextual model andrecommendations for future research. Obesity Reviews, 2, 159–171.
Dix, T., Ruble, D. N., Grusec, J. E., & Nixon, S. (1987). Social cognition in parents:inferential and affective reactions to children a three age levels. Child Development,57(4), 879–894.
Dumas, J. E., Nissley, J., Nordstrom, A., Smith, E. P., Prinz, R. J., & Levine, D. W. (2005).Home chaos: sociodemographic, parenting, interactional, and child correlates.Journal of Clinical Child and Adolescent Psychology, 34(1), 93–104.
Faith, M. S., Berkowitz, M. D., Stallings, V. A., Kerns, J., Storey, M., & Stunkard, A. J.(2004a). Parental feeding attitudes and styles and child body mass index: pro-spective analysis of a gene-environment interaction. Pediatrics, 114, e429–e436.
Faith, M. S., Scanlon, K. S., Birch, L. L., Francis, L. A., & Sherry, B. (2004b). Parent-childfeeding strategies and their relationships to child eating and weight status. ObesityResearch, 12(11), 1711–1722.
Fisher, J. O., & Birch, L. L. (1999a). Restricting access to foods and children’s eating.Appetite, 32, 405–419.
Fisher, J. O., & Birch, L. L. (1999b). Restricting access to palatable foods affects children’sbehavioral response, food selection, and intake. American Journal of Clinical Nutri-tion, 69, 1264–1272.
Fisher, J. O., & Birch, L. L. (2002). Eating in the absence of hunger and overweight in girlsfrom 5 to 7 years of age. American Journal of Clinical Nutrition, 76, 223–231.
Francis, L. A., & Birch, L. L. (2005). Maternal weight status modulates the effects ofrestriction on daughters’ eating and weight. International Journal of Obesity, 29,942–949.
Francis, L. A., Hofer, S. M., & Birch, L. L. (2001). Predictors of maternal child-feedingstyle: maternal and child characteristics. Appetite, 37, 231–243.
Francis, L. A., Ventura, A. K., Marini, M., & Birch, L. L. (2007). Parent overweight predictsdaughters’ increase in BMI and disinhibited overeating from 5 to 13 years. Obesity,15(6), 1544–1553.
Golan, M., & Crow, S. (2004). Targeting parents exclusively in the treatment of child-hood obesity: long-term results. Obesity Research, 12(2), 357–361.
Good, P. I. (1999). Resampling methods: a practical guide to data analysis. Boston, MA:Birkhauser.
Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data analysis(5th ed.). New Jersey: Prentice-Hill.
Holmbeck, G. N. (1997). Toward terminological, conceptual, and statistical clarity inthe study of mediators and moderators: examples from the child-clinical andpaediatric psychology literatures. Journal of Consulting and Clinical Psychology, 65,599–610.
Hughes, S. O., Power, T. G., Fisher, J. O., Mueller, S., & Nicklas, T. A. (2005). Revisiting aneglected construct: parenting styles in a child-feeding context. Appetite, 44, 83–92.
Jaccard, J., Turrisi, R., & Wan, C. K. (1990). Interaction effects in multiple regression. CA,USA: Sage Publications [90_TDDIFF]Inc.
Jahnke, D. L., & Warschburger, P. A. (2008). Familial transmission of eating behaviors inpreschool-aged children. Obesity, 16, 1821–1825.
Joyce, J. L., & Zimmer-Gembeck, M. J. The parent feeding dimensions questionnaire: ameasure of parenting style dimensions in the child feeding and eating context, inpress.
Kaplan, D. (2000). Structural equation modelling: foundations and extensions. ThousandOaks, CA: Sage.
Kaplan, H. I., & Kaplan, H. S. (1957). The psychosomatic concept of obesity. Journal ofNervous and Mental Disorders, 125, 181–201.
Krebs, N. F., & Jacobson, M. S. (2003). Prevention of pediatric overweight and obesity.Pediatrics, 112, 424–430.
Lobstein, T., Baur, L. A., & Uauy, R. (2004). Obesity in children and young people: a crisisin public health. Obesity Reviews, 5(Suppl. 1), 4–85.
Loehlin, J. C. (1992). Latent variable models. Hillsdale, NJ: Lawrence Erlbaum Publishers.Maccoby, E. E. (1992). The role of parents in the socialization of children: a historical
review. Developmental Psychology, 28(6), 1006–1017.Maffeis, C., Talamini, G., & Tato, L. (1998). Influence of diet, physical activity and
parent’s obesity on children’s adiposity: a four-year longitudinal study. Interna-tional Journal of Obesity, 22, 758–764.
May, A. L., Donohue, M., Scanlon, K. S., Sherry, B., Dalenius, K., Faulkner, P., et al. (2007).Child-feeding strategies are associated with maternal concern about childrenbecoming overweight, but not children’s weight status. Journal of the AmericanDietetic Association, 107, 1167–1174.
Please cite this article in press as: Joyce, J. L., & Zimmer-Gembeck, M. J.child disinhibited eating and the moderating role of the parenting c
TED
PR
OO
F
Miller, S. A. (1995). Parents’ attributions for their children’s behaviour. Child Develop-ment, 66, 1557–1584.
Moens, E., & Braet, C. (2007). Predictors of disinhibited eating in children with andwithout overweight. Behaviour Research and Therapy, 45, 1357–1368.
Musher-Eizenman, D. R., & Holub, S. C. (2006). Children’s eating in the absence ofhunger: the role of restrictive feeding practices. In R. K. Flamenbaum (Ed.),Childhood obesity and health research (pp. 135–156). New York: Nova-SciencePublishers .
Ogden, J., Reynolds, R., & Smith, A. E. (2006). Expanding the concept of parental control:a role for overt and covert control in children’s snacking behaviour? Appetite, 47,100–106.
Ouwens, M. A., van Strien, T., & van der Staak, C. P. F. (2003). Tendency towardovereating and restraint as predictors of food consumption. Appetite, 40, 291–298.
Patrick, H., Nicklas, T. A., Hughes, S. O., & Morales, M. (2005). The benefits of author-itative feeding style: caregiver feeding styles and children’s food consumptionpatterns. Appetite, 44(2), 243–249.
Powers, S. W., Chamberlin, L. A., van Schaick, K. B., Sherman, S. N., & Whitaker, R. C.(2006). Maternal feeding strategies, child eating behaviors, and child BMI in low-income African-American preschoolers. Obesity, 14, 2026–2033.
Rhee, K. (2008). Childhood overweight and the relationship between parent behaviors,parenting style, and family functioning. The Annals of the American Academy ofPolitical and Social Science, 615, 12–37.
Rhee, K., Lumeng, J. C., Appugliese, D. P., Kaciroti, N., & Bradley, R. H. (2006). Parentingstyles and overweight status in first grade. Pediatrics, 117(6), 2047–2054.
Rodin, J., & Slochower, J. (1976). Externality in the nonobese: effects of environmentalresponsiveness on weight. Journal of Personality and Social Psychology, 33(3), 338–344.
Rosenkranz, R. R., & Dzewaltowski, D. A. (2008). Model of the home food environmentpertaining to childhood obesity. Nutrition Reviews, 66(3), 123–140.
Safer, D. L., Agras, W. S., Bryson, S., & Hammer, L. D. (2001). Early body mass index andother anthropometric relationships between parents and children. InternationalJournal of Obesity, 25, 1532–1536.
Sallis, J. F., Broyles, S. L., Frank-Spohrer, G., Berry, C. C., Davis, T. B., & Nader, P. R. (1995).Child’s home environment in relation to the mother’s adiposity. InternationalJournal of Obesity and Related Metabolic Disorders, 19, 190–197.
Satter, E. (1995). Feeding dynamics: helping children to eat well. Journal of PediatricHealth Care, 9, 178–184.
Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimentalstudies: new procedures and recommendations. Psychological Methods, 7, 422–445.
Shunk, J. A., & Birch, L. L. (2004). Girls at risk for overweight at age 5 are at risk fordietary restraint, disinhibited overeating, weight concerns, and greater weight gainfrom 5 to 9 years. Journal of the American Dietetic Association, 104, 1120–1126.
Skinner, E., Johnson, S., & Snyder, T. (2005). Six dimensions of parenting: a motivationmodel. Parenting: Science and Practice, 5(2), 175–235.
Soper, D. S. (2009). The free statistics calculators website, Online Software, http://www.danielsoper.com/statcalc/.
Stevens, J. (1996). Applied multivariate statistics for the social sciences. Mahwah, NJ:Lawrence Erlbaum Publishers.
Strauss, R. S., & Knight, J. (1999). Influence of the home environment on the develop-ment of obesity in children. Pediatrics, 103(6), 85–93.
Tanofsky-Kraff, M., Ranzenhofer, L. M., Yanovski, S. Z., Schvey, N. A., Faith, M. S.,Gustafson, J., et al. (2008). Psychometric properties of a new questionnaire toassess eating in the absence of hunger in children and adolescents. Appetite, 51,148–155.
van Strien, T., Schippers, G. M., & Cox, W. M. (1995). On the relationship betweenemotional and external eating behaviour. Addictive Behaviors, 20, 585–594.
Vaska, V. L., & Volkmer, R. (2004). Increasing prevalence of obesity in South Australian4-year-old: 1995–2002. Journal of Pediatrics and Child Health, 40(7), 353–355.
Ventura, A., & Birch, L. L. (2008). Does parenting affect children’s eating and weightstatus? International Journal of Behavioral Nutrition and Physical Activity, 5, 15.
Wadden, T. A., Brownell, K. D., & Foster, G. D. (2002). Obesity: responding to the globalepidemic. Journal of Consulting and Clinical Psychology, 70, 510–525.
Wake, M., Hardy, P., Canterford, L., Sawyer, M., & Carlin, J. B. (2007a). Overweight,obesity and girth of Australian preschoolers: prevalence and socio-economiccorrelates. International Journal of Obesity, 31, 1044–1051.
Wake, M., Nicholson, J. M., Hardy, P., & Smith, K. (2007b). Preschooler obesity andparenting styles of mothers and fathers: Australian National Population Study.Pediatrics, 120, e1520–1527.
Wardle, J., Guthrie, C. A., Sanderson, S., & Rapoport, L. (2001). Development of thechildren’s eating behaviour questionnaire. Journal of Child Psychology and Psychia-try, 42(7), 963–970.
Whitaker, R. C., Deeks, C. M., Baughcum, A. E., & Specker, B. L. (2000). The relationship ofchildhood adiposity to parent body mass index and eating behavior. ObesityResearch, 8(3), 234–240.
Wilkins, S. C., Kendrick, O. W., Stitt, K. R., Stinett, N., & Hammarlund, V. A. (1998). Familyfunctioning is related to overweight in children. Journal of the American DieteticAssociation, 98, 572–574.
Williams, S. (2001). Overweight at age 21: the association with body mass index inchildhood and adolescence and parents’ body mass index. A cohort study of NewZealanders born in 1972–1973. International Journal of Obesity, 25, 158–163.
World Health Organisation. (1998). Obesity: preventing and managing the globalepidemic. Report of a WHO consultation on obesity, Geneva, 3–5 June, 1997. Geneva:World Health Organisation.
Parent feeding restriction and child weight. The mediating role ofontext. Appetite (2009), doi:10.1016/j.appet.2009.03.015
