HEALTH EDUCATION RESEARCH Vol.26 no.

2 2011
Pages 296–307
Advance Access publication 24 February 2011

What helps children eat well? A qualitative exploration of

resilience among disadvantaged families

Lauren K. Williams*, Jenny Veitch and Kylie Ball

Centre for Physical Activity and Nutrition Research, School of Exercise and Nutrition Sciences, Deakin University, 221
Burwood Highway, Burwood, Victoria 3125, Australia.
*Correspondence to: L. K. Williams. E-mail: lauren.williams@deakin.edu.au
Received on July 21, 2010; accepted on December 19, 2010

Abstract potential avenues for nutrition promotion

among disadvantaged children.
It is well known that persons of low socioeco-
nomic position consume generally a less healthy
diet. Key determinants of unhealthy eating
among disadvantaged individuals include Introduction
aspects of the family and external environment.
Much less is known about family and environ- A large number of studies have demonstrated that
mental determinants of healthy eating among poor nutrition is disproportionally experienced by
social disadvantaged children. The aim of this those with socioeconomic disadvantage (e.g. low
study was to gain insight into the family and income, low educated). For instance, individuals
environmental factors underlying resilience to of lower socioeconomic position (SEP) reportedly
poor nutrition among children and their moth- consume diets higher in fat, lower in micronu-
ers living in disadvantaged neighbourhoods. trient density and have lower intakes of fruit and
Semi-structured interviews were conducted vegetables than those of higher SEP [1–4]. In
with 38 mother–child pairs (N 5 76) from dis- children, an inadequate diet is associated with
advantaged neighbourhoods. Children were se- higher body mass index (BMI) and disease, both
lected if they were a healthy weight, consumed in childhood and adulthood [5–7]. A key prereq-
uisite to promoting good nutrition among low
adequate intakes of fruit and vegetables and
SEP children is to better understand the mecha-
were physically active. Two main themes
nisms underlying healthy eating behaviours
emerged from the interviews: active strategies
among this group.
from parents to promote healthy eating and ex- Social ecological models suggest that aspects
ternal barriers and supports to healthy eating. of the home or neighbourhood environment, as well
Mothers believed that exercising control over as personal factors, are likely to predict health
access to unhealthy food, providing education behaviours such as healthy eating [8, 9]. It is well
and encouragement for consumption of healthy recognized that the family environment plays a key
food and enabling healthy food options aided role in the development of eating behaviours and
their child to eat well. Children did not perceive food consumption among children [10–13]. Parents
food advertisements to be major influences on are primary socialization agents who transmit
their eating preferences or behaviour. The messages to children about their eating [14]. Over
results of the current study offer insight into the past decade, the association between parental

Ó The Author 2011. Published by Oxford University Press. All rights reserved. doi:10.1093/her/cyr004
For permissions, please email: journals.permissions@oup.com
 What helps children eat well? A qualitative exploration

feeding style and practices and children’s eating been consistently reported as exerting a negative
behaviour has received increased empirical focus. influence on children’s eating and weight status.
For instance, adolescents who describe their parents For example, children exposed to food advertising
as authoritative (strict yet involved and supportive) of unhealthy foods have been found to show in-
have been found to have healthier dietary behav- creased preference for promoted foods and higher
iours than those who describe their parents as consumption and purchasing requests to parents for
authoritarian (also strict yet uninvolved) [15, 16]. advertised foods [24, 25]. However, although the
Parenting practices that encompass both direct majority of food advertisements promote unhealthy
(e.g. verbal encouragement to eat specific foods) foods [25], there is some evidence to suggest a re-
and indirect (e.g. modelling consumption of spe- lationship between advertisements promoting nutri-
cific foods) communications have been found to tious foods and positive attitudes and beliefs about
influence children’s eating [17]. For instance, pro- these foods [26, 27]. Friends and peers have also
hibiting palatable foods and exerting excessive con- been found to influence children’s food preferences
trol over access to unhealthy foods has been found and eating behaviour. For instance, a recent quali-
to increase taste preference and desirability for that tative study highlighted parental reports that peer
food, leading to weight gain and poor eating habits consumption of certain foods (i.e. from their school
[18]. In addition, implementing ‘food rules’, such lunchboxes) encouraged preferences for obtaining
as offering dessert as a reward for consumption of and consuming these foods [23].
vegetables, have been found to increase the child’s While there is good evidence for an association
preference for the reward food [19–21]. Further- between various family and environmental influen-
more, a recent review of qualitative data on parental ces on children’s eating behaviours, the majority of
perceptions regarding healthy behaviours for pre- studies have focussed on parent report measures
venting overweight in young children revealed that only and on child unhealthy BMI or risk of over-
several parenting strategies such as lack of time, weight/obesity as the main outcome. Rather than
lack of motivation and decreased perception of focussing on obesity-inducing behaviours, it may
responsibility for child weight management acted be fruitful to examine ‘resilient’ children; socioeco-
as barriers to behaviours for promoting healthy eat- nomically disadvantaged families with children
ing and preventing overweight [22]. In addition to who manage to eat well, despite increased risk of
parental feeding styles and practices, parents’ own poor nutrition. To our knowledge, no study has
eating styles have also been show to influence simultaneously assessed parent and child percep-
children’s eating behaviours. Parents’ food prefer- tions of barriers and supports to healthy eating ex-
ences, the foods they consume and make available clusively among a sample of low SEP children of
to their children and their restrained eating normal body weight and with a diet including
(e.g. dieting) influence children’s eating behav- aspects of good nutrition. Enhancing our under-
iours, potentially as a result of social modelling standing of supports for healthy eating and methods
and by the choice of foods made available to their employed to overcome unhealthy eating among dis-
children [14]. advantaged children who eat adequate intakes of
In addition to the family environment, many fac- fruit and vegetables may inform nutrition interven-
tors within the broader environment have also been tions and obesity prevention among this group. The
found to encourage or inhibit healthy eating among current study was designed to gain a better under-
children. For instance, healthy eating behaviours standing of the barriers and supports assisting chil-
among children have been associated with good dren to eat well. Specifically, we aimed to gain in-
accessibility to quality healthy food options in their depth insights into the family and environmental
local neighbourhood and involvement in the grow- factors underlying resilience to poor nutrition from
ing process of fruits and vegetables (e.g. from their children and their mothers living in disadvantaged
home vegetable garden) [23]. The media has also neighbourhoods.

297
L. K. Williams et al.

Methods participation in moderate-to-vigorous physical

activity measured using objective accelerometry
Participants physical activity measures). The criteria for healthy
The participants were 38 mother–child pairs. A eating were based on the distribution of fruit and
summary of the sociodemographic characteristics vegetable consumption for this group. Increasing
for the sample is provided in Table I. Briefly, the the criteria further (i.e. to include more than 2 serves
per day), the size of the eligible sample would have
mean age of mothers and children was 38 and 9
been reduced too much to recruit sufficient partici-
years, respectively, and the majority of the sample
pants for this study. Fruit and vegetable consumption
(79%) resided in rural regions of Victoria, Aus-
was used as a marker for a generally healthier diet in
tralia. Participants were drawn from a larger study
selecting participants and was assessed from survey
of 4349 women who participated in the Resilience
data from an item that asked ‘How many serves of
for Eating and Physical Activity Despite Inequality
fruit (in a separate item—vegetables) do you usually
(READI) study, a longitudinal cohort study exam-
consume each day?’. The child age range was se-
ining resilience to obesity among socially and eco-
lected, as we were interested in prepubescent chil-
nomically disadvantaged women and children
dren who were considered old enough to participate
residing in rural and urban areas of Victoria, Aus-
in an interview session. An analysis of qualitative
tralia [28–30]. Disadvantage was defined using data relating to physical activity and sedentary
area-level disadvantage, which has been shown behaviours and their determinants in mother–child
to have associations with poorer health outcomes, pairs is to be the subject of a separate paper. From
independent of individual markers of disadvantage these 67 eligible children, 12 parents did not respond
[31]. Women in the READI study were randomly to the study invitation, 8 women refused participa-
selected from neighbourhoods ranked in the lowest tion and 6 women had relocated, and following re-
Victorian tertile of relative disadvantage, an index cruitment, there was a dropout rate of 3 (7%) due to
that considers area-level income, education and em- work commitments. The remaining 38 children (re-
ployment [32]. For the READI study, all women sponse rate 57%), along with their mother, partici-
completed a baseline survey that assessed individ- pated in the current study.
ual, social and environmental factors potentially
associated with physical activity, diet and weight. Procedure
From the READI baseline cohort, women with In July 2009, mothers were mailed a letter explain-
young children (n = 1680) were invited to complete ing the study and inviting them (and their child) to
a survey about the health and lifestyle of their child participate. After receiving the letter, each potential
(n = 685, response rate = 59%). participant was contacted by phone (no more than
For the current study, participants were selected twice) by the field manager in order to arrange
using purposive criterion sampling. From the co- a time and date for the interview. Interviews were
hort of women who had completed only the base- conducted from July to October 2009.
line survey and a survey for their child (n = 685), Two trained research staff conducted each inter-
67 children aged 8–12 years were identified as re- view at participant’s homes. At each visit, two sep-
silient. For the current study, resilience was defined arate interviews were conducted, one for the mother
as: healthy weight range (using mother reported and one for the child. The child interview was con-
height and weight to calculate BMI for age percen- ducted without the mother immediately present (i.e.
tiles where children rated between the 5th and the mother was in the family home but not sitting in
<85th percentiles were considered within the on the interview). The child was not present for the
healthy weight range [33]), regular consumers of mother interview. With the participant’s permission,
fruit (>1 serves per day) and vegetables (>2 serves an electronic dictaphone was used to record each
per day) and physically active (ranked top 50% for interview and hand written notes were also taken

298
 What helps children eat well? A qualitative exploration

Table I. Sociodemographic characteristics of interview participants

Mean (SD) or N (%) Range

Mother
Mean age (years) 38.56 (4.64) 28.79–46.12
Education
Low: did not complete high school 8 (21.1%)
Medium: completed high school/trade/certificate/diploma 17 (44.7%)
High: completed tertiary education 13 (34.2%)
Employment
Full time 12 (31.6%)
Part time 8 (21.0%)
Not currently employed 18 (47.4%)
Marital status
Single 0
Married/DeFacto 36 (94.7%)
Separated/divorced/widowed 2 (5.3%)
Country of birth
Australia 35 (92.1%)
Other 3 (7.9%)
Mean BMI 26.18 (6.29) 17.31–43.97a
Region
Rural 30 (79%)
Provincialb 5 (13%)
Metro 3 (8%)
Child Mean (SD) or N (%) Range
Mean age (years) 9.37 (1.56) 7.08–12.96
Gender
Male 20 (52.6%)
Female 18 (47.4%)
Mean BMI 17.03 (1.83) 14.30–21.73
a
Six (15.8%) parents were classified as obese (BMI > 30). bProvincial refers to a town with a large city within a rural area.

by a staff member. Interviews followed a specific included questions that assessed mother and child
interview schedule and lasted approximately 20–45 perceptions on supports and barriers to fruit and
min. Participants were advised that they had been vegetable consumption, patterns of healthy eating
selected because the child consumed relatively more more generally and physical activity. The existing
fruit and vegetables compared with other children in evidence, our own previous research and a series of
the READI cohort and that we were interested in pilot interviews were used to develop the interview
finding out how and why this was the case. At the schedule items. Only the healthy eating questions
end of the interview, mothers’ were presented with were examined for this paper. Different interview
a $25 gift voucher and children a $10 voucher in schedules were used for mothers and children. Sam-
recognition of and gratitude for their time. Ethics ple questions from the interview schedule include
approval for this study was granted by the Deakin (mother) ‘What do you think helps your child eat
University Human Research Ethics Committee. a healthy diet?’, ‘Is there anything that you think
makes it difficult for your child to eat a healthy
Materials diet?’, ‘Do you actively do anything that you think
The social ecological model guided the develop- helps your child to eat a healthy diet?’, (child)
ment of a semi-structured interview schedule that ‘What do your parents do to help you eat healthy

299
L. K. Williams et al.

food?’ and ‘Does what your friends eat make a dif- inter-coder agreement. No major discrepancies in
ference to what you eat?’, if yes ‘How?’. Probing coding or interpretation were observed.
questions were also employed when responses were
dichotomous (i.e. yes/no) and more in-depth infor- Results
mation was required.
The children’s interview also involved the re- The results are presented around two main themes
searcher presenting the child a slide show on the that emerged from the data: (i) active strategies
computer that displayed various pictures, such as from parents to promote healthy eating behaviours
family meals, breakfast, fruit choices and after- and (ii) external barriers and supports to healthy
school snacks. Pilot interviews with children eating. Active strategies from parents to promote
indicated that the inclusion of visual aids, such as healthy eating encompassed four sub-themes: pa-
a computer slide show, helped to make the rental control and moderation, support for healthy
interviews feel less threatening and also provided eating, eating rules and parental role modelling.
children with a prompt that helped them to think External barriers and supports to healthy eating in-
and comment on their eating habits and food cluded three sub-themes: access, advertising and
preferences. The mothers’ interview included similar friends/family. Each of the above main themes is
types of questions regarding what they believed illustrated with excerpts from participants. Excerpts
was helping (and making it difficult) for their child from children include their code (calculated as their
to eat a healthy diet and what they did as parents to mothers code plus 0.1), gender and age in paren-
influence their child’s eating behaviours. theses. Excerpts from mothers include their code
Coding and analysis and the gender and age of their child in parentheses.
The code is used in replacement of pseudonyms
Interviews were transcribed verbatim. The first au- (pseudonyms are used in excerpts where names
thor then read all the transcripts to develop a de- have been provided by participants) and to identify
tailed hierarchical numerical coding scheme that mother–child pairs (e.g. participant 10 is the mother
was used to code all transcripts. Open, axial and of participant 10.1).
selective coding, utilizing NUD*IST (QSR Inter-
national, 2002) version six qualitative software
Theme 1: active strategies from parents to
program, was used to code, subcategorize and
promote healthy eating behaviours
unify coding of transcript text [34]. Transcripts
were coded to identify mother–child pairs [e.g. 1 Parental control and moderation: limiting and
(mother) and 1.1 (child of mother 1)]. Template controlling access to unhealthy food
analysis [35] and inductive thematic analysis [36] The most salient theme that emerged from the inter-
were used to develop and interpret the themes [37]. views from both mothers and children was the per-
For instance, some themes emerged from a list of ception that children consumed a healthy diet
codes (template analysis) identified in the textual because parents had control over the amount and
data of the interview schedule, while our type of food the children consumed. In some instan-
knowledge of the literature shaped a further set ces, this was directly related to the child’s age, with
of themes that emerged from transcript data, not mothers articulating an anticipated decrease in con-
directly related to the interview schedule items trol as their children became older. Parental control
(thematic analysis). Researcher triangulation [38] over food was enforced by limiting access to un-
was employed to increase the validity of the data healthy food, limiting food choice, emphasizing
and its interpretation. This was achieved by having moderation and restricting unhealthy food options
the second author read the transcripts and then to special occasions (e.g. weekends, celebrations).
select a random sample of 10 mother–child For many mothers, the responsibility of consuming
transcripts that were cross-coded to check for a healthy diet was reportedly in their control and not

300
 What helps children eat well? A qualitative exploration

the responsibility of the child. This involved much Well I would like ice-cream [for breakfast] but
more than simply providing and encouraging con- I know I wouldn’t be able to have that. (28.1:
sumption of healthy food, it was also about refusing female aged 10)
requests for less health food choices, emphasizing
moderation and maintaining an ‘eat this or nothing’ Support for healthy eating: encouragement to
approach to food and healthy eating. consume fruit and vegetables, healthy eating
education and provision of fruit and vegetables
I do give them treats occasionally, . . . I try and
get them to eat really healthy. I do buy the odd Some mothers reported that they verbally encour-
treat for them. Like, there’s a box of Nutrigrain aged their children to eat fruit and vegetables,
up on the shelf . but the kids know that during a theme that was reiterated by a small number of
the week they’re to eat Weetbix or something children.
healthier and on the weekend that’s a treat
for them. So, it’s still . they still get their Maybe if we go down the street and she wants
treats. But, in moderation. (11: mother of son afternoon tea I try and coach her into choosing
aged 7) something that’s healthier than something else.
(2: mother of female aged 8)
The impact of such parental control of food was If I’m asking for something to eat they’ll say eat
clearly articulated by many children. For the major- a banana or something. (19.1: daughter aged 10)
ity of children, food preferences (whether unhealthy
or healthy) were directly dictated by taste. Although Mothers reported that encouragement to eat well
a small number of children reported eating certain was also communicated by educational messages
foods to reduce the risk of weight gain or to sustain about the benefits of healthy eating. However, al-
energy, most children reported taste as the deciding though many children reported that they received
factor for preferred foods rather than health bene- encouragement to eat well, none of the children
fits. However, while taste strongly dictated prefer- specifically reported receiving messages of an edu-
ences, preferences did not always dictate cational nature. Although children did not report
consumption. Food consumption was reportedly receiving messages of an educational nature, many
largely governed by parental control. Hence, while children appeared to have a good knowledge of
a large proportion of children articulated a strong foods, in addition to fruit and vegetables that were
preference for unhealthy foods, they reported that and were not healthy.
access to these foods was restricted or limited by
their parents.
And yeah we had that discussion, the content of
sugar in cereals and I showed them on the side
Chocolate balls (cereal), they’re not allowed to
of the packet the rice bubbles and they were
have that as breakfast because I consider them
amazed at that at the time. (24: mother of son
lollies because there’s so much sugar in them . . .
aged 8)
we buy a jar of Nutella and it’s for school holi-
days . . . (24: mother of son aged 8)
Support for healthy eating from mothers was not
My mum sometimes buys chocolate balls which always communicated through verbal encourage-
are these round things that are chocolate. It’s ment and healthy eating education. Both mothers
actually a type of cereal that we’re not allowed and children reported the positive impact parental
to have as cereal, we’re just allowed to have them provision of adequate fruit, vegetables and healthy
as snacks, treats. A treat, it’s very rare. (24.1: food options had on healthy eating. Mothers
male aged 8) reported that their child maintained a healthy diet

301
L. K. Williams et al.

because they provided healthy food: in lunchboxes, The rule is you’ve got to eat three different col-
for dinner and by having a well-stocked supply of oured vegetables . . . You can choose which three
fruit and vegetables for snacking. When children but you’ve got to have three. (24: mother of son
were asked what their parents did to help them eat aged 8)
healthy food, almost all respondents reported that
Mum says you have to eat one piece of broccoli
their parents provided them with fruit and added
to have dessert. (10.1: female aged 8)
vegetables to their meals.

Look there’s a big fruit bowl I keep on the bench Parental role modelling
and the kids know they can eat that any time. Compared with direct influences on healthy eating,
There are certain foods they can graze on when- indirect influences, namely role modelling (both
ever but then there are other foods that it’s, you positive and negative) reported by mothers emerged
know, off limits. With fruit I mean yeah always as a less salient theme. Only one child reported an
two or three pieces in their lunch box. (37: awareness of her parents’ eating habits.
mother of daughter aged 10)
They usually make healthy food like vegetable [I] lead by example. I think we eat a fairly healthy
soup and when they make food they put in peas diet. (23: mother of son aged 9)
and carrots and lettuce. (7.1: female aged 8) Mum is very healthy. Dad’s not completely
healthy . . . Dad’s nowhere near as healthy as
Eating rules Mum. Mum’s always having the healthy food
Many mothers reported that they mandated eating . And Dad’s just happy with a sandwich or if
certain meals, namely breakfast. Some mothers also he’s down the street he might get a pie. (20.1:
mandated three meals a day. While almost all the female aged 10)
children reported consuming breakfast, lunch and
dinner, they did not articulate an awareness of this
Theme 2: external barriers and supports to
eating rigidity or report any concerns with eating
healthy eating
the standard three meals a day.
Access
Sometimes Jessica will be ‘‘I’m running late for The majority of mothers reported good access to
school, I haven’t had breakfast yet’’, or ‘‘I’ve fruit and vegetables within their local community.
still got to have breakfast’’. ‘Yes you do have Some mothers reported the poor quality and range
to eat breakfast before you leave, you don’t skip of fruit and vegetables available at supermarkets,
breakfast’’. You can skip other things if you’re yet overcame this barrier by growing their own
running late but yeah you don’t skip your break- fruit and vegetables or travelling further to access
fast or your tea. (19: mother of daughter aged 10) better quality produce. Some of the mothers re-
sided in rural and provincial areas where fruit
In addition, many mothers reported implementing and vegetables are the towns primary industry
specific rules to promote healthy eating. These in- and hence access to fresh seasonal produce (e.g.
cluded prohibiting dessert until vegetables were con- from orchardists and ‘pick your own’ produce out-
sumed, consuming leftover fruit from lunchboxes lets) acted as an environmental support to healthy
before further food is made available, prohibiting eating. Residing in these rural areas, for some, also
‘junk food’ before lunch and mandating a minimum resulted in less access to mainstream take-away
fruit and vegetable daily consumption quota (e.g. options and fast food, which mothers reported re-
must eat two pieces of fruit a day). Many children duced their child’s consumption of these un-
also articulated an awareness of these eating rules. healthy options.

302
 What helps children eat well? A qualitative exploration

We basically have to buy our fruit and vegetables Advertising

from a supermarket whereas if we went to much The majority of mothers and children did not report
of the bigger towns that are nearby you’ve got that advertising negatively influenced their food
access to the mini-markets and more of a range. purchasing or consumption behaviours. In fact,
I think that we’re very limited. So we’re going to many mothers and children reported the positive
extend the veggie patch this year. influence that some advertisements had on healthy
Yeah, we’ve got lettuce, at the moment we’ve got eating. Some children also reported that while
silver beet in there, garlic, we’ve got the herbs out some of the unhealthy advertised foods appeared
appealing, they did not pursue attaining advertised
the front . and peas, yeah. We grow it to eat it.
products due to knowledge of them being unavail-
(20: mother of daughter aged 10, rural resident)
able to them (due to geographical or parental
The school have the Stephanie Alexander pro- restrictions).
gram so they cook and they’ve got the garden
and they grow everything at school as well. Actually we both enjoy watching . there’s a Safe-
And living out here you’ve got the market gar- way ad that tells you about what fruit and vegies to
dens, and the fruit stalls and everything out there. eat now and we both watch it. If one of us misses
So the special treat when they go shopping is us the other will come and say, ‘‘Look this is the
they used to get given an apple. (24: mother of one that’s in season. Let’s get this and try this’’.
son aged 9, rural resident) Or if it’s a new fruit we haven’t seen we always try
to get it. (34: mother of daughter aged 12)
Yeah seasonal fruit is fantastic and so this area for
seasonal fruit is really good. It’s free and it’s fresh Cause they’re just advertisements trying to sell
and it’s off the tree so you can’t get better than you stuff and at school we’ve been shown what
that . A good thing about this region is there’s advertisements do so it kind of puts you off ads
and stuff. [34.1: female aged 12]
beautiful peaches, apricots, like all our friends are
mostly orchardists so we just duck around and I’ve seen a couple of the burger ones [advertise-
pick some peaches or pick some apricots. Yeah ments] they look nice. Like the McMuffins.
so seasonal is a big deal. We try and avoid having They’re always looking good. But never really
apples or things that are out of season. (29: had one. We don’t really eat a lot of McDonalds
mother of daughter aged 13, provincial resident) or anything anyway . , so. [20.1: female aged
10]
For those children who had access to a school
canteen, it was reported that canteens acted as a sup- Friends/Peers
port for healthy eating with almost all school can- Children and mothers reported both the negative
teens banning or limiting unhealthy food items. and positive influence of their siblings, friends
and peers on healthy and unhealthy eating predom-
There’s no lollies or chocolates or chips or inantly through verbal comments, peer modelling
anything like that [at the canteen] anymore, it’s and from the presence of desirable healthy and un-
all just . healthy options now. [1: mother of son healthy food items in friends’ lunchboxes.
aged 11]
If you’re over at a friend’s house they’ll go and
[The] school canteen, which is just starting this,
get chips or something . . . Then I’ll normally
just the end of last semester. It’s, it’s pretty
have chips with them. (25.1: male aged 12)
healthy . . . I’m sort of part in there, helping
out. And we’re all doing healthy food. [My child] is in prep and he’ll see something in
[16: mother of daughter aged 8] someone’s lunch box and he’ll come home and

303
L. K. Williams et al.

say ‘‘I want that tomorrow’’ .he definitely does Perhaps, this was because mothers also offered ed-
that. [1: mother of son aged 11] ucation and explanations about unhealthy food
items and promoted the importance of being
I made some dairy gluten free vegetable slice
healthy. It is also possible that the children in the
and it was lovely cold. We sat there and had it
current study, particularly those who were younger,
together. She came home from school, ‘‘I’m
were also accustomed to this parenting style (or
never having that again’’. ‘‘Why not?’’ ‘‘The
unaware of anything different) and shared similar
kids said it was disgusting and they was teasing
attitudes about food and eating.
me’’. I said, ‘‘That’s okay, you just tell them that
These parenting attributes (i.e. strict, restrictive,
they’re missing out on good stuff’’. [2: mother of
yet encouraging and supportive) are consistent with
daughter aged 8]
an authoritative parenting style, which has been
demonstrated as a positive predictor of healthy eat-
Discussion ing and weight status [16]. The emotional climate
created by authoritative parents is one of the high
The purpose of the current study was to explore the strictness and involvement, warmth, emotional sup-
family and environmental factors underlying resil- port, appropriate granting of autonomy and clear,
ience to unhealthy eating. Individual interview dis- bidirectional communication [39]. As parenting is
cussions with mothers from disadvantaged a learned process, teaching parents to implement
neighbourhoods and their children revealed the practices inherent in an authoritative parenting style
presence of parental strategies and external barriers (e.g. by accepting responsibility for their child’s
and supports to promoting healthy eating behav- diet, prohibiting unhealthy foods and encouraging
iours. This is one of the first studies to include both and modelling healthy food choices) is a viable av-
mother and child reports and focus exclusively on enue for nutrition promotion interventions among
low SEP families of children who eat well. Our low SEP families.
study underscores the importance of focussing spe- In addition, children in our study had a good
cifically on ‘resilient’ children (i.e. those eating rel- knowledge of healthy and prohibited foods, despite
atively well) to further elucidate potentially not articulating an active awareness of specific
effective parent–child attitudes and behaviours in food-related education. The finding that transfer-
preventing unhealthy eating. ence of healthy eating education from mother to
Compared with previous studies that have pre- child was likely internalized by children highlights
dominantly focussed on the barriers to healthy eat- potential limitations of relying on children’s self-
ing and a healthy weight status, our results report in evaluation of health promotion interven-
highlighted the active role mothers from disadvan- tions.
taged neighbourhoods played in promoting healthy The results from our study also highlighted some
eating. For instance, almost all the mothers in the environmental influences as both barriers and sup-
current study believed that parents were the main ports to healthy eating among families residing in
vehicle for influencing healthy eating and as a disadvantaged neighbourhoods. Previous research
result, they exercised significant control over their has indicated the negative impact on eating of ad-
child’s food by implementing ‘food rules’, providing vertising and poorer access and availability of
access to fruit and vegetables and restricting healthy food options [40, 41]. Some mothers from
unhealthy food items. our study reported instances of poor availability and
Although there is some evidence to suggest that quality of healthy produce yet many had developed
excessive control over access to certain foods and strategies to overcome these barriers, namely,
implementing food rules has a negative effect on through responding to and creating more sustain-
eating and weight [18, 19, 21], the children did not able access to fruit and vegetables and other healthy
report their mothers to be too strict or controlling. food options. For instance, many families had their

304
 What helps children eat well? A qualitative exploration

own fruit and vegetable garden, a practice consis- that schools are limiting their options to healthy
tently associated with increased fruit and vegetable choices is positive.
consumption [23, 42]. Although a number of fam- Strengths of this qualitative study are the large
ilies in the current study benefited from residing in sample, inclusion of both mother and child reports
rural or provincial areas where fruit and vegetables and a focus on healthy eating among resilient chil-
were the town’s primary industry and accessibility dren from low SEP families. Some limitations of
to larger garden space was more available, it is the current study warrant consideration. Reports of
possible that provision of skills and resources for external supports and barriers particularly, rely on
home-grown produce is a potential avenue for in- participants’ perceptions of their own local environ-
creasing fruit and vegetable consumption among ment, which may differ from objective supports/
low SEP families. barriers. They may also differ across samples, pop-
Previous research has consistently highlighted ulation groups or geographic areas. This limitation
the negative impact advertising has on children’s was most notable given the large proportion of par-
eating, yet our results indicated that although some ticipants residing in rural and provincial areas. It is
children reported an awareness of the negative in- interesting to observe that the majority of children
fluence of food advertising, most children did not identified as resilient to overweight, obesity and
feel negatively influenced by televised food adver- unhealthy eating from the larger study resided in
tisements. There are three plausible explanations rural areas. Although there is no clear evidence that
for this finding. Firstly, many children reported a difference in BMI exists between Australian rural
low access to outlets selling unhealthy food, a likely and urban children [30, 45, 46], the results from the
outcome given the large proportion of rural residen- current study suggest certain factors that may be
cies (e.g. limited fast food outlets, small supermar- protective of unhealthy eating and potentially over-
kets). Therefore, children may not be affected by weight and obesity among rural children from dis-
advertising of foods unavailable to them. Secondly advantaged neighbourhoods.
and unsurprisingly, given the high control over un- Given that the study focused exclusively on
healthy food practices exhibited by mothers when women and mothers, it does not offer any insight
children reported desirability for unhealthy adver- into the role that fathers and other extended family
tised foods, they perceived these foods as pro- members have on their child’s eating. For instance,
hibited by their parents. Thirdly, parents who almost 95% of the sample was married, and al-
restrict access to unhealthy food may also be more though only one child mentioned her father, it is
likely to restrict television viewing time, making possible that fathers are potential sources of influ-
exposure to these advertisements less prominent. ence on children’s healthy eating. A further limita-
It is possible that these factors offer protection tion present in most research that involves
against the negative influence of unhealthy food interviews is that participants may answer in a so-
advertising. cially desirable manner. This limitation is perhaps
Another external support for healthy eating artic- more salient in the current study where participation
ulated in the interviews was school canteens. At in the larger survey had occurred and exposed adult
present Victorian Government only stipulates can- participants to the broad aims and objectives of the
teen ‘guidelines’, which are not consistently imple- research conducted. In the current study, effects of
mented at each government school [43]. Our data social desirability were minimized by conducting
suggest little evidence of canteens serving un- interviews independently for parents and children.
healthy food or acting as a negative influence on In addition, children were not privy to the main
children’s healthy eating. Although it has been pre- aims and objectives of the project. Finally, we only
viously reported that consumption of foods from assessed barriers and supports to healthy eating
Australian canteens contributes a small amount of among children who consumed a relatively healthy
daily food intake for children [44], the suggestion diet rather than to draw comparisons of those who

305
L. K. Williams et al.

consume less adequate diets. There has been much Conflict of interest statement
research that has focussed on predictors of un-
healthy behaviours and by using a qualitative study None declared.
design and restricting our sample in this way, it
enabled us to generate hypotheses from those who
manage to remain resilient about potential avenues References
for nutrition promotion among disadvantaged 1. De Irala-Estevez J, Groth MV. A systematic review of
children. socioeconomic differences in food habits in Europe: con-
The results of the current study offer insight into sumption of fruit and vegetables. Eur J Clin Nutr 2000;
potential avenues for nutrition promotion among 54: 706–14.
2. Barker M, Lawrence W, Woadden J et al. Women of lower
disadvantaged children residing in socioeconomi- educational attainment have lower food involvement and eat
cally disadvantaged areas. Further research should less fruit and vegetables. Appetite 2008; 50: 464–8.
focus on comparisons of the barriers and supports to 3. Giskes K, Turrell G, Patterson C et al. Socioeconomic
differences among Australian adults in consumption of
healthy eating identified from the current sample fruit and vegetables and intakes of vitamins A, C and
among disadvantaged families with children who folate.J Hum Nutr Diet 2002; 15: 375–85. discussion 87–90.
donot manage to eat well. Although our findings 4. Ball K, Mishra GD, Thane CW et al. How well do
need to be replicated in a larger sample using sur- Australian women comply with dietary guidelines? Public
Health Nutr 2004; 7: 443–52.
vey methods, the results tentatively suggest that 5. Stang J, Taft Bayerl C, Flatt MM. Position of the American
active parental strategies such as exercising control Dietetic Association: child and adolescent food and nutrition
over access to unhealthy food, provision of educa- programs. J Am Diet Assoc 2006; 106: 1467–75.
6. Nobili V, Alisi A, Raponi M. Pediatric non-alcoholic
tion and encouragement for consumption of healthy fatty liver disease: preventive and therapeutic value of
food, provision of healthy food options and positive lifestyle intervention. World J Gastroenterol 2009; 15:
role modelling may aid the prevention of unhealthy 6017–22.
7. Corvalan C, Uauy R, Kain J et al. Obesity indicators and
eating among disadvantaged children. Lessons
cardiometabolic status in 4-y-old children. Am J Clin Nutr
learnt from families equipped with strategies to 2010; 91: 166–74.
overcome environmental barriers to healthy eating, 8. Bronfenbrenner U. The Ecology of Human Development.
such as skills in sourcing local and home-grown Canbridge, MA: Harvard University Press, 1979.
9. Davison KK, Birch LL. Childhood overweight: a contextual
produce, may also facilitate healthy eating among model and recommendations for future research. Obes Rev
disadvantaged children. 2001; 2: 159–71.
10. Sothern MS. Obesity prevention in children: physical activ-
ity and nutrition. Nutrition 2004; 20: 704–8.
Funding 11. Strauss RS, Knight J. Influence of the home environment on
the development of obesity in children. Pediatrics 1999;
103: e85.
National Health and Medical Research Council 12. Dietz WH, Gortmaker SL. Preventing obesity in children
(374241, 425845 to L.W., 479513 to K.B.); and adolescents. Annu Rev Public Health 2001; 22: 337–53.
13. Fulkerson JA, Story M, Neumark-Sztainer D et al. Family
National Heart Foundation of Australia (08M3912 meals: perceptions of benefits and challenges among parents
to J.V.). of 8- to 10-year-old children. J Am Diet Assoc 2008; 108:
706–9.
14. Birch LL, Davison KK. Family environmental factors
Acknowledgements influencing the developing behavioral controls of food
intake and childhood overweight. Pediatr Clin North Am
2001; 48: 893–907.
Authors acknowledge the support and guidance 15. Kremers SP, Brug J, de Vries H et al. Parenting style and
from Professor David Crawford, Dr Clare Hume adolescent fruit consumption. Appetite 2003; 41: 43–50.
and Dr Michelle Jackson who contributed to the 16. Pearson N, Atkin AJ, Biddle SJ et al. Parenting styles,
projects inception, coordination and execution and family structure and adolescent dietary behaviour. Public
Health Nutr 2010; 13: 1245–53.
to the Research Fellows who conducted the 17. Pearson N, Biddle SJ, Gorely T. Family correlates of
interviews. fruit and vegetable consumption in children and

306
 What helps children eat well? A qualitative exploration

adolescents: a systematic review. Public Health Nutr 2009; 31. King T, Kavanagh AM, Jolley D et al. Weight and place:
12: 267–83. a multilevel cross-sectional survey of area-level social
18. Birch LL, Fisher JO. Development of eating behaviors disadvantage and overweight/obesity in Australia. Int J Obes
among children and adolescents. Pediatrics 1998; 101: (Lond) 2006; 30: 281–7.
539–49. 32. Australian Bureau of Statistics ABS. National Nutrition
19. Birch LL. Development of food preferences. Annu Rev Nutr Survey User’s Guide. Canberra, Australia: Australian Gov-
1999; 19: 41–62. ernment Publishing Service, 1995.
20. Moore SN, Tapper K, Murphy S. Feeding strategies used by 33. CDC. Centre Disease Prevention and Health Promotion
primary school meal staff and their impact on children’s growth charts. 2000. Available at: www.cdc.gov Accessed:
eating. J Hum Nutr Diet 2010; 23: 78–84. 21 June 2010.
21. Birch LL. Children’s preferences for high-fat foods. Nutr 34. Rice PE, Ezzy D. Qualitative Research Methods: A Health
Rev 1992; 50: 249–55. Focus. South Melbourne, Australia: Oxford university
22. Pocock M, Trivedi D, Wills W et al. Parental perceptions Press, 1999.
regarding healthy behaviours for preventing overweight and 35. King N. Chapter 7. In: Symon G, Cassell C (eds). Template
obesity in young children: a systematic review of qualitative Analysis. Thousand Oakes: CA: Sage, 1998.
studies. Obes Rev 2009; 11: 338–353. 36. Boyatzis RE. Transforming Qualitative Information.
23. Campbell KJ, Crawford DA, Hesketh KD. Australian Thematic Analysis and Code Development. Thousand Oaks,
parents’ views on their 5-6-year-old children’s food choices. CA: Sage, 1998.
Health Promot Int 2007; 22: 1–1–8. 37. Green J, Thorogood N. Qualitative Methods for Health
24. Halford JC, Gillespie J, Brown V et al. Effect of television Research. London, UK: Sage, 2004.
advertisements for foods on food consumption in children. 38. Patton M. Qualitative Evaluation and Research Methods,
Appetite 2004; 42: 221–5. 2nd edn. Thousand Oaks, CA: Sage, 1990.
25. Aktas Arnas Y. The effects of television food advertisement 39. Darling N, Steinberg L. Parenting styles as a context: an
integrative model. Psychol Bull 1993; 113: 487–96.
on children’s food purchasing requests. Pediatr Int 2006;
40. Timperio A, Ball K, Roberts R et al. Children’s fruit and
48: 138–45.
vegetable intake: associations with the neighbourhood food
26. Dixon HG, Scully ML, Wakefield MA et al. The effects of
environment. Prev Med 2008; 46: 331–5.
television advertisements for junk food versus nutritious
41. Story M, French S. Food Advertising and Marketing
food on children’s food attitudes and preferences. Soc Sci Directed at Children and Adolescents in the US. Int J Behav
Med 2007; 65: 1311–23. Nutr Phys Act 2004; 1: 3.
27. Horne PJ, Tapper K, Lowe CF et al. Increasing children’s 42. Kamphuis CB, Giskes K, de Bruijn GJ et al. Environmental
fruit and vegetable consumption: a peer-modelling and determinants of fruit and vegetable consumption among
rewards-based intervention. Eur J Clin Nutr 2004; 58: adults: a systematic review. Br J Nutr 2006; 96: 620–35.
1649–60. 43. Bell AC, Swinburn BA. School canteens: using ripples
28. Macfarlane A, Abbott G, Crawford D et al. Personal, social to create a wave of healthy eating. Med J Aust 2005;
and environmental correlates of healthy weight status 183: 5–6.
amongst mothers from socioeconomically disadvantaged 44. Bell AC, Swinburn BA. What are the key food groups to
neighborhoods: findings from the READI study. Int J Behav target for preventing obesity and improving nutrition in
Nutr Phys Act 2010; 7: 23. schools. Eur J Clin Nutr 2004; 58: 258–63.
29. Cleland V, Ball K, Hume C et al. Individual, social and 45. Booth ML, Wake M, Armstrong T et al. The epidemiology
environmental correlates of physical activity among women of overweight and obesity among Australian children and
living in socioeconomically disadvantaged neighbourhoods. adolescents, 1995–97. Aust N Z J Public Health 2001; 25:
Soc Sci Med 2010; 70: 2011–8. 162–9.
30. Cleland V, Hume C, Crawford D et al. Urban-rural compar- 46. Booth ML, Macaskill P, Lazarus R et al. Sociodemographic
ison of weight status among women and children living in distribution of measures of body fatness among children and
socioeconomically disadvantaged neighbourhoods. Med J adolescents in New South Wales, Australia. Int J Obes Relat
Aust 2010; 192: 137–40. Metab Disord 1999; 23: 456–62.

307