Maternal feeding practices predict fruit and vegetable consumption in young children. Results of a 12-month longitudinal study

Appetite 57 (2011) 167–172

Contents lists available at ScienceDirect

Appetite journal homepage: www.elsevier.com/locate/appet

Research report

Maternal feeding practices predict fruit and vegetable consumption in young children. Results of a 12-month longitudinal study Jane E. Gregory, Susan J. Paxton *, Anna M. Brozovic School of Psychological Science, La Trobe University, Melbourne, Victoria 3086, Australia

A R T I C L E I N F O

A B S T R A C T

Article history: Received 22 April 2010 Received in revised form 3 April 2011 Accepted 20 April 2011 Available online 29 April 2011

This study aimed to explore the prospective relationship between maternal feeding practices and young children’s frequency of consumption of fruits, vegetables and sweets, and also child weight-for-height zscores. Participants were 60 mothers who completed questionnaires when their children were 1 year old and again when their children were 2 years old. Regression analyses were performed. After controlling for availability and prior child consumption of the target food, maternal use of pressure to eat at 1 year predicted lower child frequency of fruit consumption at 2 years and approached significance for lower vegetable consumption. Maternal modelling of healthy eating at 1 year predicted higher child frequency of vegetable consumption at 2 years. Restriction did not significantly predict child frequency of consumption of fruits, vegetables or sweets over time. Child weight-for-height scores at 2 years were predicted by weight-for-height at 1 year but not by feeding practices. The findings suggest that maternal feeding practices can influence child eating at a very young age. Interventions should focus on encouraging parents to model healthy eating to promote healthy eating in children. ß 2011 Elsevier Ltd. All rights reserved.

Keywords: Children Feeding practices Child eating behaviour Pressure to eat Modelling

Introduction Food preferences and eating patterns developed in early childhood appear to continue into adolescence and adulthood (Mikkila¨, Ra¨sa¨nen, Raitakari, Pietinen, & Viikari, 2005; Nicklaus, Boggio, Chabanet, & Issanchou, 2004; Northstone & Emmett, 2008; Rozin, 1990; Unusan, 2006). In the early years in particular, parents play an important role in shaping children’s food preferences and intake. As gatekeepers of the household food supply, parents are, to a certain extent, able to control which foods are accessible to their children. Additionally, parents use feeding strategies which may influence children’s food preferences and intake (Jansen, Mulkens, & Jansen, 2007; Vereecken, Legiest, De Bourdeaudhuij, & Maes, 2009). This study focuses on three such feeding practices: pressuring children to eat more healthy foods; restriction of unhealthy foods; and modelling of healthy eating. Research suggests that despite positive intentions some parental feeding practices may not achieve the desired effect of promoting healthful eating. For example, in cross-sectional studies, pressuring children to eat more healthy foods has been associated with lower child consumption of fruit (Fisher, Mitchell, Smiciklas-Wright, & Birch, 2002; Vereecken et al., 2009) and vegetables (Fisher et al., 2002), and with higher child consumption

* Corresponding author. E-mail address: [email protected] (S.J. Paxton). 0195-6663/$ – see front matter ß 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.appet.2011.04.012

of unhealthy foods (Campbell, Crawford, & Ball, 2006; Vereecken et al., 2009). Similarly, restriction of unhealthy foods is used by parents in an attempt to improve children’s diets, but experimental studies have found that restriction of a particular food may result in increased preference for (Jansen et al., 2007) and intake of the restricted food (Fisher & Birch, 1999). There is also a limited amount of research available regarding relationships between these feeding strategies and child weight in early childhood. In one prospective study, it was found that maternal use of pressure to eat and restriction with 1-year-old children significantly predicted lower child weight at 2 years, after controlling for the child’s initial weight (Farrow & Blissett, 2008). In older samples, pressure to eat has been consistently associated with lower child weight status (Birch et al., 2001; Carnell & Wardle, 2007; Powers, Chamberlin, van Schaick, Sherman, & Whitaker, 2006; Robinson, Kiernan, Matheson, & Haydel, 2001; Spruijt-Metz, Lindquist, Birch, Fisher, & Goran, 2002), while restriction has been related to higher child weight in some studies (Birch et al., 2001; Francis, Hofer, & Birch, 2001; Moens, Braet, & Soetens, 2007) and not associated with weight in others (Carnell & Wardle, 2007; Kro¨ller & Warschburger, 2008). One 3-year prospective study found that restriction at baseline was associated with lower BMI z-score in 5- to 6-year-old children but not in 10- to 12-year-old children (Campbell et al., 2010). Inconsistent findings and a limited number of longitudinal studies with regards to feeding practices and weight have highlighted the need for additional research in this area.

168

J.E. Gregory et al. / Appetite 57 (2011) 167–172

Another parental feeding strategy, modelling of healthy eating, is thought to promote healthful child eating behaviour. Research has consistently found that parents’ food consumption is positively related to their children’s food consumption (De Bourdeaudhuij et al., 2008; Vereecken, Keukelier, & Maes, 2004; Wardle, Carnell, & Cooke, 2005). Rozin and Millman (1987) proposed that environment was more likely than heredity to explain familial similarities in food preferences. One environmental influence on child food intake may be the availability of those foods in the home (Cullen et al., 2003; Jago, Baranowski, & Baranowski, 2007; Koui & Jago, 2008). One study found that children who eat meals with their family more regularly tend to consume more vegetables, after taking into account the availability of fresh food in the area and the extent to which parents do not buy fruit and vegetables due to a perception of high cost and low preference in the family of these foods (Campbell et al., 2006). It is possible that parental modelling of healthy eating may have a positive influence on children’s eating, distinct from the potential impact of availability of healthy foods in these households. The evidence described here indicates that there is a relationship between child food consumption patterns and parental feeding practices. However, the literature lacks longitudinal evidence that could provide an indication of causality in these relationships. It is possible that pressuring children to eat, for example, could lower children’s consumption of healthy foods; it is also possible that this strategy is used in response to the child’s reluctance to eat these foods. Another gap in the literature is an examination of these relationships in children aged 1–2 years. This age may be a particularly important period in the development of child eating behaviour, with evidence suggesting that early food experiences shape children’s later food preferences and intake (Nicklaus et al., 2004; Skinner, Carruth, Bounds, Ziegler, & Reidy, 2002). The primary aim of the present study was to examine prospective relationships between maternal use of pressure to eat, restriction, modelling and food availability with 1-year-old children, and children’s frequency of consumption of fruits, vegetables and sweets 1 year later. A further aim of this study was to test whether maternal feeding practices at 1 year were associated with child weight-forheight z-scores (WFHz) at 2 years. Methods Participants Participants were mothers of 1-year-old children from The Child and Family Health Study conducted through La Trobe University, Australia. Recruitment was conducted through community notices in local newspapers and via playgroup coordinators who invited mothers in their groups to participate. Participants were excluded if they reported that their child had a physical illness that severely impacted on their eating behaviour. Seventy-eight mothers returned questionnaires during the initial phase, and 63 of these mothers also returned the follow-up questionnaire 12 months later. For the final analyses, data from 3 participants were removed due to missing subscales, leaving a final sample of 60 (77% of those who participated initially). This study was approved by the Human Research Ethics Committee at La Trobe University. Participants received a $10 supermarket gift voucher for their participation at each time point. Measures A self-report questionnaire was used for mothers to record demographic and anthropometric information, as well as measures of maternal feeding practices, child food consumption, and food availability.

Maternal feeding practices All of the parental feeding items were measured on a five point Likert scale, and scores for each feeding practice were calculated using the mean of all items in the subscale. Feeding variable scores ranged from 1 to 5, with a higher score indicating more frequent use of that practice. Two parental feeding subscales were taken from the Child Feeding Questionnaire (CFQ) (Birch et al., 2001), a widely used instrument in the child feeding literature (Ventura & Birch, 2008). The subscale ‘‘pressure to eat’’ contains four items, asking parents the degree to which they pressure their child to eat more food (e.g. ‘‘If my child says ‘I’m not hungry,’ I try to get him/her to eat anyway’’, a = .61). The ‘‘restriction’’ subscale measures how much parents attempt to control their child’s eating by restricting access to food (e.g. ‘‘I intentionally keep some foods out of my child’s reach’’ and ‘‘If I did not guide or regulate my child’s eating s/he would eat too many junk foods’’). The original scale is comprised of eight items; however, recent research has questioned the inclusion of two of these on both theoretical and statistical grounds (Anderson, Hughes, Fisher, & Nicklas, 2005; Corsini, Danthiir, Kettler, & Wilsona, 2008). These two items, regarding the use of food as a reward for good behaviour, were excluded from our measure, resulting in a six-item scale (a = .79). Modelling of healthy eating was measured using three items (a = .75) written for the purposes of this study: ‘‘I try to eat only healthy foods in front of my child’’; ‘‘My child sees me eating fast food’’ (reversed item), and; ‘‘My child sees me eating healthy snacks (e.g. fruit, yoghurt, nuts, toast).’’ Child frequency of food consumption and food availability The child’s consumption of fruits, vegetables and sweet foods was measured with the Child Food Frequency Questionnaire. This scale was developed by Campbell and colleagues (Campbell et al., 2006), and was based on data from the 1995 Australian National Nutrition Survey. On a 7-point scale from ‘‘not eaten’’ to ‘‘4 times a day’’, mothers were asked to record how often their child ate each of the listed fruits (14 items), vegetables (13 items) and sweets (13 items, e.g. ice-cream, chocolate biscuits) over the past week. The authors of the scale listed more common food items separately (e.g. ‘‘banana’’ or ‘‘chocolate’’), while grouping other foods together (e.g. ‘‘honeydew/watermelon/rockmelon’’). Individual items were recoded to represent an average daily frequency of consumption for that item (e.g. 4 times a day = 4, once a day = 1, once a week = 1/ 7 = .14), then the items in each food group were added together for each case, with the final frequency scores indicating the average number of times per day that the participant’s child ate fruits, vegetables and sweets. Additionally, parents were asked for each food listed: ‘‘If this food was NOT eaten, was this food available in your house last week (yes/no)?’’ Each food listing was then given a ‘‘food availability’’ score. If the food was eaten by the child at all during the week (and therefore was available to him or her), or not eaten but available in the house, then that food was given an availability score of 1. If it was neither eaten nor available, it was given a score of zero. The overall availability for fruits, vegetables and sweets was then calculated by summing together the availability scores within each food group. The scores therefore reflected the variety of fruits, vegetables and sweets that were available to the child over the week, rather than the total volume of food. It should be noted that the measure did not account for food which was made available to the child (but was not eaten) outside the home. Statistical analyses Body mass index (BMI) was calculated for the mothers. Ageand gender-adjusted weight-for-height z-scores (WFHz) were

J.E. Gregory et al. / Appetite 57 (2011) 167–172

calculated for children with the computer program NutStat, which calculates z-scores using United States based norms from the Centre for Disease Control and Prevention (CDC) 2000 growth charts (CDC, 2000). All variables were examined for normality. Skewness and kurtosis were corrected using a logarithmic transformation for child fruit and sweets consumption. A square root transformation was used for child vegetable consumption and vegetable availability. Bivariate correlation analyses were used to examine whether maternal feeding practices at 1 year were associated with child consumption of fruits, vegetables and sweets or child WFHz at 1 and 2 years. Hierarchical multiple regression analyses were used to find whether maternal feeding practices predicted child food consumption or WFHz over time. The WFHz analysis was conducted using the smaller subsample of mothers (83%, n = 50) who reported complete child and weight data for both time points. Results Participant characteristics At the point of initial recruitment, mothers had a mean age of 33.87 (SD = 4.39), and 75% were tertiary educated, with a further 13% having completed high school only. Self-reported height and weight data were provided by 87% of mothers, 46% of whom had a body mass index (BMI) greater than 25, which is the generally accepted cut-off point for overweight (World Health Organization, 1995). This proportion of overweight is comparable with other Australian data for women in this age group (Australian Bureau of Statistics, 2009). The children (32 male, 28 female) had a mean age of 1.46 years (SD = 0.26). Statistics for rates of overweight in Australian children are not available for 1-year-old children using weight-for-height z-scores, and so for comparison purposes, age

169

and gender adjusted BMI z-scores (BMIz) at age 2 years were calculated from CDC growth charts (CDC, 2000). Of these children, 10% could be considered overweight or obese at 2 years using international standard cut-off points (Cole, Bellizzi, Flegal, & Dietz, 2000), which is around half the rate found in other data for agematched peers (Commonwealth Scientific Industrial Research Organisation, 2008). The period between completing the two questionnaires ranged from 52 to 71 weeks, with an average follow up of 55 weeks. For all the feeding and eating variables, independent sample ttests found no significant differences according to child gender, to whether or not mothers provided complete child height and weight information, and to whether or not mothers returned the second questionnaire (data not shown). Table 1 presents the means, standard deviations and possible range for maternal feeding, child eating and food availability variables at 1 and 2 years. All feeding, consumption and food availability variables at 1 year were significantly correlated with their 2-year counterparts (Table 1). Simple associations between maternal feeding practices at 1 year and child food consumption and weight-for-height at 1 and 2 years are presented in Table 2. Longitudinal predictors of child food consumption and weight-for-height A series of hierarchical multiple regressions was used to explore the impact of maternal feeding practices on child frequency of food consumption over time, after controlling for earlier child consumption of the target food (fruits, vegetables and sweets). The child’s consumption of the target food at age 2 years was entered as the outcome variable. At step 1 in the regression we controlled for the child’s prior consumption of the target food at 1 year. To minimise the number of variables entered into the model,

Table 1 Mean and standard deviation (SD) scores for maternal feeding, child food consumption and food availability variables at 1 and 2 years of age. Factor

Possible range

Feeding practices Pressure to eat Restriction Model healthy Frequency of food consumptiona Child fruit consumption/day Child vegetable consumption/day Child sweets consumption/day Food availabilityb Fruit availability Vegetable availability Sweets availability a b y

ry

Mean (SD) 1 year old

2 years old

1–5 1–5 1–5 Observed range 0–15 0–16 0–5

2.5 (0.9) 3.6 (0.9) 3.9 (0.5)

2.4 (0.9) 3.5 (0.9) 3.9 (0.6)

.65 .44 .61

4.3 (2.5) 4.2 (3.1) 1.2 (1.0)

4.2 (2.6) 3.2 (1.9) 1.4 (1.1)

.47 .54 .56

3–13 3–13 0–12

8.2 (2.4) 9.7 (2.0) 5.7 (2.7)

8.0 (2.7) 9.3 (2.3) 5.5 (2.2)

.47 .46 .49

Frequency with which the food was consumed by the child per day. Number of different foods within each food group (fruit, vegetable sweets) available in the child’s home during the week measured. Pearson’s r correlation from 1 year to 2 years. For all variables, p  .001.

Table 2 Pearson’s correlations between maternal feeding practices and food availability at 1 year, and child consumption of fruit, vegetables and sweets at 1 and 2 years. Feeding practice

Pressure to eat Restriction Modelling Food availabilityb a

1 year

2 years

Fruit

Veg

Sweets

WFHz

Fruit

Veg

Sweets

WFHz

.33** .03 .39** .42**

.23a .16 .45** .51**

.23a .24a .29* .59**

.14 .47** .15

.10 .14 .27* .30*

.27* .04 .51** .45**

.26* .12 .20 .34**

.10 .31* .08

Approaches significance, p < .10. Availability of target food, i.e. fruit, vegetables or sweets. p < .05. ** p < .01 b *

J.E. Gregory et al. / Appetite 57 (2011) 167–172

170

Table 3 Results of hierarchical multiple regressions measuring longitudinal predictors of child frequency of consumption of fruit, vegetable and sweets. Predictors at 1 year

Fruit consumption

Step 1 Consumptiony Step 2 Consumptiony Pressure to eat Restriction Model healthy Food availabilitya

R2 = .22, p < .001 .47 <.001 R2Change ¼ :09, p = .088 .49 .001 .28 .023 .06 .11

b

a

p

.611 .406

Vegetable consumption Semipartial

b

p

Sweets consumption Semipartial

.40 .26

R2 = .29, p < .001 .54 <.001 R2Change ¼ :15, p = .005 .26 .045 .21 .058

.21 .20

.06 .10

.34 .16

.29 .13

.47

.005 .195

.54

b

p

R2 = .31, p < .001 .56 <.001 R2Change ¼ :02, p = .756 .50 .001 .15 .193 .02 .864 .06 .622 .02 .890

Semipartial .56 .38 .15 .02 .06 .02

Availability and consumption of the target food, i.e. fruits, vegetables or sweets; b standardized Beta.

at step 2 we included a feeding practice (pressure to eat, restriction, modelling and availability of the target food) only if the correlation between that feeding variable and the target food consumption variable had been significant or approached significance (p < .10) in the bivariate analysis. Standardized beta scores and significance levels for the regressions are presented in Table 3. The child frequency of consumption of fruits, vegetables and sweets at 2 years old was independently predicted by the frequency of consumption of the respective food at 1 year old. Additionally, frequency of fruit consumption at 2 years was predicted by lower maternal use of pressure to eat at 1 year. Child frequency of vegetable consumption at 2 years was significantly predicted by more frequent maternal use of healthy modelling, and approached significance for less frequent use of pressure to eat. Frequency of child sweets consumption at 2 years was not predicted by any of the feeding practices after controlling for initial sweets consumption at 1 year. Finally, we used a hierarchical multiple regression to test whether maternal use of restriction when children were aged 1 year predicted child weight status over time, after controlling for initial weight status. At step 1, child WFHz at 1 year significantly predicted child WFHz at 2 years (R2 = .30, p < .001) and restriction did not significantly add to the variance for child WFHz at step 2 (R2Change ¼ :00, p = .926). We did not include pressure to eat or modelling of healthy eating in the regression model as they were not found to be correlated with child WFHz at 1 or 2 years. Discussion This study found that 2-year-old children’s frequency of consumption of fruits, vegetables and sweets was significantly predicted by their consumption of each of those foods 1 year earlier. These findings remained significant after accounting for the availability of these foods in the home, suggesting that habit may play an important role in eating behaviour from a very early age. This may be particularly relevant to the consumption of sweet foods, which was not influenced by any of the feeding strategies measured in this study. Fruit and vegetable consumption, on the other hand, appeared to have been impacted by both prior consumption and maternal feeding practices. We found that maternal use of pressure to eat with 1 year old children was associated with decreased frequency of consumption of fruit and vegetables 1 year later. Cross-sectional studies have found negative relationships between pressure to eat and child consumption of fruit (Fisher et al., 2002; Vereecken et al., 2009) and vegetables (Fisher et al., 2002), and the present study adds to this literature by showing that there may be a longitudinal impact of the use of pressure on children’s consumption of fruits and vegetables. This is consistent with the findings of Galloway, Fiorito, Francis, and Birch (2006), who conducted an experiment to test the impact of pressure to eat on 3- to 5-year-old children’s (N = 27) soup consumption. Over time, children consumed less of the soup

they were pressured to eat, and made more negative comments about it than when they were not pressured (Galloway et al., 2006). The current study also adds to the literature by showing negative associations between the use of pressure and children’s eating behaviour from a very early age. This is interesting considering that the way a parent pressures 3–5 year old child to eat more is likely to be very different from how a parent pressures a 1-year-old, who may still require assistance with feeding. Despite these likely differences, a negative impact of pressure is observed in both age groups. Batsell and Brown (1998) and Batsell, Brown, Ansfield, and Paschall (2002) suggested that children who perceive they are being forced to consume certain foods develop a ‘‘cognitive aversion’’ for those foods because they associate the food with the negative feeding experience. This was noted in a retrospective account from 140 university students who were surveyed about their experience of being pressured to eat. Many students recalled feelings of hopelessness in the pressure situation, with consequent aversions often lasting into adulthood (Batsell et al., 2002). It was interesting to find that the use of pressure at 1 year was positively correlated with child frequency of fruit consumption in the cross-sectional analysis, and then negatively associated with fruit consumption 1 year later. One explanation for the distinction is that this strategy may be effective at increasing fruit consumption in the first instance, but children learn to resist the pressure, and over time it has a negative impact on consumption. This premise is also supported by Galloway et al.’s (2006) study, in which they found that children whose parents usually used minimal levels of pressure at home consumed more soup in the pressure condition and less in the no pressure condition, when compared with children whose parents used high levels of pressure at home. Similarly children may initially eat more fruit in response to their parents pressuring them to do so, thus unhelpfully reinforcing the parents’ use of this strategy. However, this was finding was not consistent with child frequency of vegetable consumption, which was negatively associated with pressure to eat in both the cross-sectional and the longitudinal analysis. This may reflect an increased use of pressure by parents in response to concern about their children’s low vegetable intake. The results of the present study suggest that a more effective means of increasing children’s consumption of vegetables may be modelling of healthy eating. The finding that modelling was associated with increased child frequency of vegetable consumption over the 12-month period is consistent with experimental studies finding that children are more likely to taste a new food if other people are eating the same food than if the food is simply presented to them (Addessi, Galloway, Visalberghi, & Birch, 2005; Harper & Sanders, 1975; Salvy, Vartanian, Coelho, Jarrin, & Pliner, 2008). This process of social facilitation may have developed as a useful way of increasing children’s taste exposure to novel foods. Observing a trusted model eating the food item provides children with evidence that this is a ‘‘safe’’ food for them to eat, and thus it is more likely that they will try the food themselves. After repeated tasting of the food,

J.E. Gregory et al. / Appetite 57 (2011) 167–172

children are more inclined to accept the food into their diets (Birch, McPhee, Shoba, Pirok, & Steinberg, 1987; Wardle, Herrera, Cooke, & Gibson, 2003). Our results suggest that modelling had an effect on general vegetable consumption levels over time. We are unable to determine whether mothers modelled the consumption of each individual vegetable, or if the effects of modelling transferred to other vegetables, thereby increasing general consumption. Unlike vegetables, children’s frequency of consumption of fruits and sweets were not significantly predicted by modelling of healthy eating. One reason for this could be the difference in palatability of these foods compared with vegetables. It is thought that infants are predisposed to accept foods with a sweeter taste, such as fruits, which contain valuable calories and are less likely to be toxic than bitter foods (Birch, 1999). If it is inherent knowledge that sweet foods are ‘‘safe’’, then children may be less influenced by modelling when it comes to eating these foods. Further research examining the differential impact of modelling on children’s consumption patterns would be useful. Considering experimental research with 3–6-year-old children finding that restriction of specific foods led to subsequent increased selection and intake of the restricted food (Fisher & Birch, 1999; Jansen et al., 2007), we had anticipated that restriction would impact on children’s sweets consumption over time. However, the only significant predictor of sweets consumption at age 2 was the child’s prior consumption of sweets, and this relationship could not be explained by the availability of sweets in the home. This finding suggested that habit may play a very important role in young children’s sweet-eating behaviour. While the experimental studies described above found that restricting foods increased children’s consumption of those foods immediately after, our findings suggest that this effect may not be maintained over time. Another explanation is that there was an element of ongoing restriction which prevented the children from increasing their sweets intake. At the age of 2 years, it is more than likely that children’s diets are still bound by the food their parents make available to them. The impact of restriction therefore may not be manifested in eating behaviour until later in life when the children have more freedom to choose the foods they eat. Finally, research suggests that ‘‘covert’’ restriction strategies such as not keeping sweets in the house may be more effective than ‘‘overt’’ strategies where the child knows that those foods are being withheld (Ogden, Reynolds, & Smith, 2006). It is possible that at the age of 1 or 2 years, children might not actually be aware of the restrictions being imposed, and thus the use of restriction may not affect either their intake of or their preference for sweet foods. Simple associations showed a positive relationship between restriction at age 1 year and child WFHz at both 1 and 2 years. However, the use of restriction did not predict child weight over time after controlling for the child’s weight at baseline. This suggested that it is likely that parents used restriction in response to the child’s higher weight status, but that the use of restriction did not influence the child’s weight 1 year later. This finding was inconsistent with prior research finding negative associations between restriction and weight over time in 1-year-old (Farrow & Blissett, 2008) and 5- to 6-year-old children (Campbell et al., 2010). The latter of these studies found no prospective relationship between restriction and child BMI z-scores in 10- to 12-year-old children (Campbell et al., 2010), and another study found no longitudinal association between restriction and fat mass in 11year-old children (Spruijt-Metz, Li, Cohen, Birch, & Goran, 2006). An emerging idea from these studies is that restrictive feeding practices may actually have a protective effect over time in younger but not older children (Campbell et al., 2010), although this is not supported by the present study. The inconsistency found across longitudinal studies restriction and child weight appears to be similar to that found in cross-sectional research (Birch et al.,

171

2001; Carnell & Wardle, 2007; Francis et al., 2001; Kro¨ller & Warschburger, 2008; Moens et al., 2007), indicating the importance of further research in this area using large, diverse samples. It would also be interesting to explore whether parental use of restriction influences eating behaviour in situations where parents have less control over their child’s food intake, such as outside the home or in the transition to adulthood. It is important to note some limitations to our study. First, the findings should be interpreted carefully due to the relatively small sample size (Tabachnick & Fidell, 2000). Also, our sample consisted of predominantly tertiary educated women, therefore the results cannot be generalised to populations with lower levels of education. We did not control for education or other potential co-variates in our regression analyses, which may have affected results. Second, measures were based on mothers’ reports rather than direct observation or objective physical measurements. Some mothers’ responses may have been influenced by their perceptions of the ‘‘right’’ behaviour with regards to feeding their children, and social desirability with regards to maternal and child weight status. Furthermore, there is some evidence to suggest that mothers tend to under-report the weight of overweight children (Scholtens et al., 2007), which could partly explain the lack of significant results for analyses related to weight-for-height and the lower rate of overweight found in our sample compared with other Australian data. Third, we need to point out some potential issues with the measures used. Both the CFQ and the FFQ were designed for use in slightly older children than the present study (Birch et al., 2001; Campbell et al., 2006) and may not have accurately captured the feeding and eating behaviour in our young sample. Mothers of 1- or 2-year-old children may differ in their interpretation of pressure and restriction when compared with mothers of older children. The Food Frequency Questionnaire was limited to measuring the reported instances of children eating a particular type of food, but did not measure portion sizes or actual energy intake. With a young sample who may still require assistance with feeding, there is a possibility that this measure was reflective of the frequency with which these foods were offered to the child rather than an indication of the child’s preference for or selection of that food. A food was considered ‘‘available’’ if it been eaten by the child, or if it was not eaten but was available in the house at any point during the week. We therefore had no way of establishing the total amount of the target food in the house, only the variety. The measure of availability also does not incorporate food that might have been made available to the child outside the home but was not eaten. Finally, our measure of modelling of healthy eating was one written for the purposes of this study. Items were written from a theoretical basis but the measure has not been validated. Children are likely to have more novel foods introduced to them in the early years than any other time in their life. The results of this study stress the importance of parents’ behaviour on facilitating the acceptance of healthy foods. Children often require several tastings before they accept an initially unpalatable food into their diets (Birch, 1982; Sullivan & Birch, 1990; Wardle, Cooke, et al., 2003), and our results suggest that pressuring young children to eat healthy foods is not likely to be an effective strategy over time. Parents should focus instead on modelling positive eating behaviour. By enjoying a healthy and varied diet themselves, parents could help their children to learn healthy eating habits from a very early age. References Addessi, E., Galloway, A. T., Visalberghi, E., & Birch, L. L. (2005). Specific social influences on the acceptance of novel foods in 2–5-year-old children. Appetite, 45(3), 264–271. Anderson, C. B., Hughes, S. O., Fisher, J. O., & Nicklas, T. A. (2005). Cross-cultural equivalence of feeding beliefs and practices. The psychometric properties of the Child Feeding Questionnaire among Blacks and Hispanics. Preventive Medicine, 41(2), 521–531.

172

J.E. Gregory et al. / Appetite 57 (2011) 167–172

Australian Bureau of Statistics. (2009). National Health Survey. Summary of results, 200708 (Reissue). Cat. No. 4364.0. Canberra: ABS. Batsell, W. R., & Brown, A. S. (1998). Human flavor-aversion learning. A comparison of traditional aversions and cognitive aversions. Learning and Motivation, 29(4), 383–396. Batsell, W. R., Brown, A. S., Ansfield, M. E., & Paschall, G. Y. (2002). You will eat all of that!’’ A retrospective analysis of forced consumption episodes Appetite, 38(3), 211–219. Birch, L. L. (1982). I don’t like it; I never tried it. Effects of exposure on two-year-old children’s food preferences. Appetite, 3(4), 353–360. Birch, L. L. (1999). Development of food preferences. Annual Review of Nutrition, 19, 41–62. 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 measure of parental attitudes, beliefs and practices about child feeding and obesity proneness. Appetite, 36(3), 201–210. Birch, L. L., McPhee, L., Shoba, B. C., Pirok, E., & Steinberg, L. (1987). What kind of exposure reduces children’s food neophobia? Looking vs. tasting. Appetite, 9(3), 171–178. Campbell, K. J., Andrianopoulos, N., Hesketh, K., Ball, K., Crawford, D., Brennan, L., et al. (2010). Parental use of restrictive feeding practices and child BMI z-score. A 3-year prospective cohort study. Appetite, 55(1), 84–88. Campbell, K. J., Crawford, D. A., & Ball, K. (2006). Family food environment and dietary behaviors likely to promote fatness in 5–6 year-old children. International Journal of Obesity, 30(8), 1272–1280. Carnell, S., & Wardle, J. (2007). Associations between multiple measures of parental feeding and children’s adiposity in United Kingdom preschoolers. Obesity, 15(1), 137–144. CDC (2000). EpiInfo’s Nutrition Program, NutStat Retrieved January, 2009, from http:// www.cdc.gov/epiinfoS. Cole, T. J., Bellizzi, M. C., Flegal, K. M., & Dietz, W. H. (2000). Establishing a standard definition for child overweight and obesity worldwide. International survey. British Medical Journal, 320, 1–6. Commonwealth Scientific Industrial Research Organisation. (2008). 2007 Australian National Children’s Nutrition and Physical Activity Survey. Canberra: Australian Government. Corsini, N., Danthiir, V., Kettler, L., & Wilsona, C. (2008). Factor structure and psychometric properties of the Child Feeding Questionnaire in Australian preschool children. Appetite, 51(3), 474–481. Cullen, K. W., Baranowski, T., Owens, E., Marsh, T., Rittenberry, L., & de Moor, C. (2003). Availability, accessibility, and preferences for fruit, 100% fruit juice, and vegetables influence children’s dietary behavior. Health Education and Behavior, 30(5), 615– 626. De Bourdeaudhuij, I., te Velde, S., Brug, J., Due, P., Wind, M., Sandvik, C., et al. (2008). Personal, social and environmental predictors of daily fruit and vegetable intake in 11-year-old children in nine European countries. European Journal of Clinical Nutrition, 62(7), 834–841. Farrow, C., & Blissett, J. (2008). Controlling feeding practices. Cause or consequence of early child weight? Pediatrics, 121(1), e164–e169. Fisher, J. O., & Birch, L. L. (1999). Restricting access to palatable foods affects children’s behavioral response, food selection, and intake. American Journal of Clinical Nutrition, 69(6), 1264–1272. Fisher, J. O., Mitchell, D. C., Smiciklas-Wright, H., & Birch, L. L. (2002). Parental influences on young girls’ fruit and vegetable, micronutrient, and fat intakes. Journal of the American Dietetic Association, 102(1), 58–64. Francis, L. A., Hofer, S. M., & Birch, L. L. (2001). Predictors of maternal child-feeding style. Maternal and child characteristics. Appetite, 37(3), 231–243. Galloway, A. T., Fiorito, L. M., Francis, L. A., & Birch, L. L. (2006). ’Finish your soup’. Counterproductive effects of pressuring children to eat on intake and affect. Appetite, 46(3), 318–323. Harper, L. V., & Sanders, K. M. (1975). The effect of adults’ eating on young children’s acceptance of unfamiliar foods. Journal of Experimental Child Psychology, 20(2), 206–214. Jago, R., Baranowski, T., & Baranowski, J. C. (2007). Fruit and vegetable availability. A micro environmental mediating variable? Public Health Nutrition, 10(07), 681–689. Jansen, E., Mulkens, S., & Jansen, A. (2007). Do not eat the red food! Prohibition of snacks leads to their relatively higher consumption in children Appetite, 49(3), 572–577. Koui, E., & Jago, R. (2008). Associations between self-reported fruit and vegetable consumption and home availability of fruit and vegetables among Greek primaryschool children. Public Health Nutrition, 11(11), 1142–1148.

Kro¨ller, K., & Warschburger, P. (2008). Associations between maternal feeding style and food intake of children with a higher risk for overweight. Appetite, 51(1), 166–172. Mikkila¨, V., Ra¨sa¨nen, L., Raitakari, O., Pietinen, P., & Viikari, J. (2005). Consistent dietary patterns identified from childhood to adulthood. The Cardiovascular Risk in Young Finns Study. British Journal of Nutrition, 93(6), 923–931. Moens, E., Braet, C., & Soetens, B. (2007). Observation of family functioning at mealtime. A comparison between families of children with and without overweight. Journal of Pediatric Psychology, 32(1), 52–63. Nicklaus, S., Boggio, V., Chabanet, C., & Issanchou, S. (2004). A prospective study of food preferences in childhood. Food Quality and Preference, 15(7–8), 805–818. Northstone, K., & Emmett, P. (2008). Are dietary patterns stable throughout early and mid-childhood? A birth cohort study. British Journal of Nutrition, 100(5), 1069–1076. Ogden, J., Reynolds, R., & Smith, A. (2006). Expanding the concept of parental control. A role for overt and covert control in children’s snacking behaviour? Appetite, 47(1), 100–106. 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 lowincome African-American preschoolers. Obesity, 14(11), 2026–2033. Robinson, T. N., Kiernan, M., Matheson, D. M., & Haydel, K. F. (2001). Is parental control over children’s eating associated with childhood obesity? Results from a population-based sample of third graders. Obesity Research, 9(5), 306–312. Rozin, P. (1990). Development in the food domain. Developmental Psychology, 26(4), 555–562. Rozin, P., & Millman, L. (1987). Family environment, not heredity, accounts for family resemblances in food preferences and attitudes. A twin study. Appetite, 8(2), 125– 134. Salvy, S. J., Vartanian, L. R., Coelho, J. S., Jarrin, D., & Pliner, P. (2008). The role of familiarity on modeling of eating and food consumption in children. Appetite, 50(2– 3), 524–528. Scholtens, S., Brunekreef, B., Visscher, T., Smit, H., Kerkhof, M., De Jongste, J., et al. (2007). Reported versus measured body weight and height of 4-year-old children and the prevalence of overweight. European Journal of Public Health, 17(4), 369– 374. Skinner, J. D., Carruth, B. R., Bounds, W., Ziegler, P., & Reidy, K. (2002). Do food-related experiences in the first 2 years of life predict dietary variety in school-aged children? Journal of Nutrition Education and Behavior, 34(6), 310–315. Spruijt-Metz, D., Li, C., Cohen, E., Birch, L., & Goran, M. (2006). Longitudinal influence of mother’s child-feeding practices on adiposity in children. Journal of Pediatrics, 148(3), 314–320. Spruijt-Metz, D., Lindquist, C., Birch, L., Fisher, J., & Goran, M. (2002). Relation between mothers’ child-feeding practices and children’s adiposity. American Journal of Clinical Nutrition, 75(3), 581–586. Sullivan, S. A., & Birch, L. L. (1990). Pass the sugar, pass the salt. Experience dictates preference. Developmental Psychology, 26(4), 546–551. Tabachnick, B. G., & Fidell, L. S. (2000). Using multivariate statistics (4th ed.). Boston: Allyn and Bacon. Unusan, N. (2006). University students’ food preference and practice now and during childhood. Food Quality and Preference, 17(5), 362–368. Ventura, A. K., & Birch, L. L. (2008). Does parenting affect children’s eating and weight status? International Journal of Behavioral Nutrition and Physical Activity 5(15). Vereecken, C. A., Keukelier, E., & Maes, L. (2004). Influence of mother’s educational level on food parenting practices and food habits of young children. Appetite, 43(1), 93–103. Vereecken, C. A., Legiest, E., De Bourdeaudhuij, I., & Maes, L. (2009). Associations between general parenting styles and specific food-related parenting practices and children’s food consumption. American Journal of Health Promotion, 23(4), 233–240. Wardle, J., Carnell, S., & Cooke, L. (2005). Parental control over feeding and children’s fruit and vegetable intake. How are they related? Journal of the American Dietetic Association, 105(2), 227–232. Wardle, J., Cooke, L., Gibson, E. L., Sapochnik, M., Sheiham, A., & Lawson, M. (2003). Increasing children’s acceptance of vegetables; a randomized trial of parent-led exposure. Appetite, 40(2), 155–162. Wardle, J., Herrera, M. L., Cooke, L., & Gibson, E. L. (2003). Modifying children’s food preferences. The effects of exposure and reward on acceptance of an unfamiliar vegetable. European Journal of Clinical Nutrition, 57(2), 341. World Health Organization. (1995). Physical status: the use and interpretation of anthropometry. Geneva: WHO.