HEALTH EDUCATION RESEARCH
Theory & Practice
Vol.19 no.3 2004
Pages 261±271
Children's eating attitudes and behaviour: a study of
the modelling and control theories of parental
in¯uence1
1
Rachael Brown and Jane Ogden
Abstract
1Department
of General Practice, Guys Kings and
St Thomas' School of Medicine, Kings College London,
5 Lambeth Walk, London SE11 6SP, UK
2Correspondence to: J. Ogden;
E-mail: Jane.Ogden@kcl.ac.uk
on the child's diet or on other aspects of their
behaviour. To conclude, a positive parental
role model may be a better method for improving a child's diet than attempts at dietary
control.
Introduction
Understanding children's eating attitudes and
behaviour is important in terms of children's
health. Evidence also indicates that dietary habits
acquired in childhood persist through to adulthood
(Kelder et al., 1994; Nicklas, 1995; Steptoe et al.,
1995). In addition, research also indicates a role for
childhood nutrition on adult health (Hales et al.,
1991; Moller et al., 1994; Berenson et al., 1998).
Much research also shows that many children's
diets in the Western world are unsatisfactory. For
example, the Bogalsua Heart Study in the US
showed that the majority of 10 year olds exceeded
the American Heart Association dietary recommendations for total fat, saturated fat and dietary
cholesterol (Nicklas, 1995). A survey in the UK
showed a similar picture, with 75% of children
aged 10±11 exceeding the recommended target
level for percentage of energy derived from fat
(Butriss, 1995). Comparable results have also been
reported by Wardle (Wardle, 1995) and Currie et al.
(Currie et al., 1997).
There has been a range of explanations offered
to understand why children eat what they eat. Lack
of knowledge has been implicated as causing poor
diets, but is not explanation enough as health
education campaigns have had limited success in
changing eating habits (Gatherer et al., 1979).
Other research has focused on social cognition
Health Education Research Vol.19 no.3, ã Oxford University Press 2004; All rights reserved
DOI: 10.1093/her/cyg040
Downloaded from http://her.oxfordjournals.org/ by guest on June 22, 2016
The present study compared the modelling and
control theories of parental in¯uence on children's eating attitudes and behaviour with a
focus on snack foods. Matched questionnaires
describing reported snack intake, eating motivations and body dissatisfaction were completed
by 112 parent/child pairs. Parents completed
additional items relating to control in terms of
attempts to control their child's food intake
and using food as a tool for controlling behaviour. The results showed signi®cant correlations between parent and child for reported
snack intake, eating motivations and body dissatisfaction, indicating an important role for
modelling. Parents were then divided according to their control scores. Children whose parents indicated greater attempts to control their
child's diets reported higher intakes of both
healthy and unhealthy snack foods. In addition,
those children whose parents indicated a
greater use of food as a means to control their
child's behaviour reported higher levels of
body dissatisfaction. The results provide some
support for both the modelling and control theories of parental in¯uence. However, whereas
modelling appears to have a consistent impact,
parental control has a differential impact
depending upon whether this control is focused
1,2
R. Brown and J. Ogden
Parental attitudes must certainly affect their
children indirectly through the foods purchased
for and served in the household¼in¯uencing the
children's exposure and¼their habits and preferences.
Some evidence supports an important role for
parents. For example, Klesges et al. (Klesges et al.,
1991) showed that children selected different foods
when they were being watched by their parents
compared to when they were not. Olivera et al.
(Olivera et al., 1992) reported a correlation
between mothers' and children's food intakes for
most nutrients in pre-school children, and suggested targeting parents to try to improve children's diets. Likewise, Contento et al. (Contento
et al., 1993) found a relationship between mothers'
health motivation and the quality of children's
diets. Food preferences therefore change through
262
watching others eat. Research also indicates that
children may not only model their parents' food
intake, but also their attitudes to food and their
body dissatisfaction. For example, Hall and Brown
(Hall and Brown, 1982) reported that mothers of
girls with anorexia show greater body dissatisfaction than mothers of non-disordered girls.
Likewise, Steiger et al. (Steiger et al., 1994)
found a direct correspondence between mothers'
and daughters' levels of weight concern, and Hill
et al. (Hill et al., 1990) reported a link between
mothers' and daughters' degree of dietary restraint.
Research therefore emphasizes the role of observational learning with a particular role for parental
attitudes and behaviour.
Other studies have highlighted a role for parental
control. Some research has explored the impact of
controlling food intake by rewarding the consumption of `healthy food' as in `if you eat your
vegetables I will be pleased with you'. For
example, Birch et al. (Birch et al., 1980) gave
children food in association with positive adult
attention compared with more neutral situations.
This was shown to increase food preference.
Similarly an intervention study using videos to
change eating behaviour reported that rewarding
vegetable consumption increased that behaviour
(Lowe et al., 1998). The relationship between food
and rewards, however, appears to be more complicated than this. In one study, children were
offered their preferred fruit juice as a means to be
allowed to play in an attractive play area (Birch
et al., 1982). The results showed that using the
juice as a means to get the reward reduced the
preference for the juice and have been supported
by similar studies (Lepper et al., 1982; Birch et al.,
1984; Newman and Taylor, 1992). These examples
are analogous to saying `if you eat your vegetables,
you can eat your pudding'. Although parents use
this approach to encourage their children to eat
vegetables the evidence indicates that this may be
increasing their children's preference for pudding
even further as pairing two foods results in the
`reward' food being seen as more positive than the
`access' food. As concluded by Birch:
Downloaded from http://her.oxfordjournals.org/ by guest on June 22, 2016
models, but most studies using this approach have
focused on adults rather than children and those
which have explored children's diets have left
much of the variance in eating behaviour unexplained (Resnicow et al., 1997).
An alternative approach to children's diets
has focused on developmental theories, and
emphasizes the in¯uence of signi®cant others on
a child's development of food preferences and
eating habits. In line with Social Learning Theory
[e.g. (Bandura, 1977)], some research has highlighted the role of observational learning and
modelling. In one study, peer modelling was used
to change children's preference for vegetables
(Birch, 1980). The target children were placed at
lunch for 4 consecutive days next to other children
who preferred a different vegetable to themselves
(peas versus carrots). By the end of the study the
children showed a shift in their vegetable preference which persisted at a follow-up assessment
several weeks later. The impact of observational
learning has also been shown in an intervention
study designed to change children's eating
behaviour using video based peer modelling
(Lowe et al., 1998).
Research has also focused on the role of parents
and Wardle (Wardle, 1995) contended that:
Children's diets
¼although these practices can induce children
to eat more vegetables in the short run, evidence
from our research suggests that in the long run
parental control attempts may have negative
effects on the quality of children's diets by
reducing their preferences for those foods.
[(Birch, 1999), p. 10]
Birch also reviewed the evidence for the impact of
imposing any form of parental control over food
intake and concluded that:
For example, when food is made freely available
children will choose more of the restricted than the
unrestricted foods particularly when the mother is
not present (Fisher and Birch, 1999).
In summary, children's diets are often poor and
research has addressed reasons for this. Some
studies have focused on modelling, and indicate
that children may model both their parent's eating
behaviour and also their eating related attitudes
and body dissatisfaction. Other studies have highlighted a role for control, and have indicated that
whilst many parents impose control over their
child's intake and use food to control their child's
behaviour, this may not always have the desired
positive effect. However, these two theories have
been mainly addressed independent of each other.
In line with this, the present study aimed to explore
the relationship between parents and children's
eating attitudes and behaviour, and to assess and
compare the modelling and control theories of
parental in¯uence. Much previous research has
explored children's diets in terms of major food
groups with an emphasis on staple foods such as
bread, pasta and vegetables. These foods make up
the content of a child's main meals throughout the
day and are part of a child's daily routine. In
contrast, snack foods such as sweets, chocolate,
grapes and toast are often eaten in between meals,
and can be sources of either con¯ict or pleasure.
Further, such snack foods often play an important
Method
Participants
Children aged between 9 and 13 were recruited
from two junior schools and one secondary school
in southern England. Children were approached by
either the researcher or a teacher at the school and
asked to give a consent form to their parents. It is
estimated that about 260 children were asked for
their consent and that 50% of children agreed to
take part. Questionnaires were then given to 137
parents and children and 112 pairs of completed
questionnaires were returned (response rate =
81.7%).
Design
The study used a dyadic design looking at pairs of
parents and children.
Procedure
A letter was sent to the head teachers of the
schools, explaining the nature of the project and
requesting participation from one of their year
groups. Information was then sent to parents
requesting consent for their children to participate
and asking if they would take part. On receiving
consent from parents, questionnaires were administered in the schools. Administration varied
according to the requests of the Head Teacher.
The questionnaires were administered in small
groups in one junior school, whole-class administration in the other junior school, with items being
read out verbally to both. In the secondary school
the questionnaires were sent home. The children's
weight and height data was collected from the
schools either by the class teachers or the
researcher (R. B.).
263
Downloaded from http://her.oxfordjournals.org/ by guest on June 22, 2016
¼child feeding strategies that restrict children's
access to snack foods actually make the
restricted foods more attractive. [(Birch, 1999),
p. 11]
role as the currency central to the interaction
between parent and child. Therefore, the present
study aimed to explore the relative role of modelling and control with a focus on the intake of
snack foods.
R. Brown and J. Ogden
Measures
Children and parents received matched questionnaires consisting of the following items. The
internal reliability of the items was explored
where appropriate using Cronbach's a.
Pro®le characteristics
Participants were asked to state their age and
gender.
Reported snack food intake
Motivations to eat
Participants rated eight items using a ®ve-point
Likert scale, following the statement `How
often do you feel like eating when...' [never
(1), not very often (2), sometimes (3), very often
264
Body dissatisfaction
This was measured using a shortened Body
Satisfaction Questionnaire (Cooper et al., 1987).
Ten items were rated using a ®ve-point Likert scale
(never, not very often, sometimes, very often,
always), e.g. `Have you ever felt unhappy about
your body?', `How often do you refuse food
because you are worried about your weight?', `Has
eating sweets or cakes ever made you feel fat',
`Have you been afraid of becoming fat (or fatter)?
A higher score re¯ected greater body dissatisfaction (range 1±5) (children's a = 0.81, parent's a =
0.87).
Body difference
Using body silhouettes of adult and children
(Stunkard et al., 1986) participants were asked to
circle one of the nine ®gures they felt closest to
their own size and then rate the ®gure they would
most like to look like. The discrepancy between the
two ®gures was scored. A more positive score
re¯ected a desire to be fatter and a more negative
score re¯ected a desire to be thinner.
Downloaded from http://her.oxfordjournals.org/ by guest on June 22, 2016
Much previous research has focused on a wide
range of foods consumed including staple foods
such as pasta and bread, and snack foods such as
sweets and fruit. The present study aimed to
focus solely on the intake of snack foods as
these play a central role in the interaction between
parent and child, and are often the place where
many of the problems with food intake occur. In
line with this, participants circled the number of
times that a snack food item was eaten both
`yesterday' and `in general' using ®ve-point Likert
scales. For `yesterday' the scale ranged from 0 (0)
to 4+ times (4) and for `in general' the scale ranged
from `never' (0) to `every day' (4). The foods were
presented in a random order. They were analysed
both in terms of the individual food items and in
terms of total scores which were summated to
re¯ect `unhealthy' and `healthy' foods (range 0±4).
These foods were selected to provide some
preliminary insights into the kinds of snacks that
children and their parents eat. The measure was not
designed to comprehensively assess all foods
eaten.
Unhealthy snack foods: chocolate, crisps, pastries, ice cream, sweets, cakes and biscuits
(children's a = 0.71, adult's a = 0.63).
Healthy snack foods: grapes, oranges, peaches,
yoghurt, toast and apples (children's a = 0.69,
adult's a = 0.48).
(4), always (5)]. These were derived from the
Dutch Eating Behaviour Questionnaire (Van Strien
et al., 1986) and summated to create total scores
(range 1±5).
Internal motivation. Four items measured
internal motivation: `you have nothing to do',
`are frightened', `are feeling upset', `are feeling
cross'. A higher score re¯ected greater levels of
internal motivation (children's a = 0.59; parent's
a = 0.69).
External motivation. Four items measured external motivation: `If you walk past a bakery sweet
shop or cafe how often do you really want to eat
something?', `If food looks, smells or tastes good,
how often do you eat more of it than usual?', `If
you have got something delicious how often do you
want to eat it straight away?', `If you see other
people eating, how often do you eat more of it than
usual'. A higher score re¯ected greater levels of
external motivation (children's a = 0.77, parent's
a = 0.69).
Children's diets
Additional parental items
Control
This assessed two aspects of control and food.
Items were rated on a ®ve-point Likert scale and
summated to create a total score (range 1±5).
Control over their child's diet.
Control over their child's behaviour using
food
Examples of items include: `How often do you
treat your child with food for good behaviour?', `If
your child is unhappy how often do you use food to
cheer them up?', `Is a snack between meals
considered a treat for good behaviour?', `Do you
use food as a way of distracting your child (e.g. if
they are preventing you from doing your chores)?'.
A higher score re¯ected a higher use of food to
control the child's behaviour (eight items, a =
0.81).
Pro®le characteristics:
Parents were also asked to record their: Subjective
class (working class, lower middle class, upper
middle class, upper class), ethnicity (black, white,
Asian, other), weight and height.
Data analysis
The results were analysed to (1) describe the
participants' pro®le characteristics and their reported snack food intake using descriptive statistics, and (2) to assess the modelling theory of
parental in¯uence, parents and children's reported
snack food intake, motivations for eating and their
body dissatisfaction were compared using
Spearman's correlation coef®cients. Finally, the
pairs were then divided into groups based upon
Results
Pro®le characteristics
Participant's pro®le characteristics are shown in
Table I. There were 112 pairs of parents and
children. Children's ages ranged from 9 to 13;
adults from 23 to 53. The majority of the parents
were female, but there was almost an equal split
of boys and girls. The majority of parents were
white and described themselves as lower middle
class. Children's and adults BMI were within the
healthy range.
Table I. Pro®le characteristics
Variable
Parent (n = 112)
Child (n = 112)
Age [years (range)]
Sex
male
female
Ethnicity
black
white
Asian
other
Subjective social class
working
lower middle
upper middle
upper = 0 (0%)
Height (cm)
Weight (kg)
BMI
40.8 (23±53)
11.23 (9±13)
15 (12.4%)
106 (87.6%)
49 (44%)
63 (56%)
1
92
6
1
(1%)
(82%)
(6%)
(1%)
31 (32.3%)
44 (45.8%)
21 (21.9%)
164.7 6 9.8
67.7 6 12.9
24.8 6 4.3
150.7 6 9.9
44 6 12
19.2 6 3.53
265
Downloaded from http://her.oxfordjournals.org/ by guest on June 22, 2016
Examples of items include: `How often are you
®rm about what your child should eat?', `How
often do you allow your child a free choice of what
to eat?', `How often are you ®rm about when your
child should eat?', `How often do you allow your
child to eat between meals?'. A higher score
re¯ected a greater degree of control placed on the
child's food intake by the parent (eight items, a =
0.67).
median splits on the parent's ratings of control. The
children's reported snack food intake, eating
motivations and body dissatisfaction were then
assessed as to whether their parents showed high or
low control over their child's diet and high or low
use of diet to control their child's behaviour as a
means to test the control theory of parental
in¯uence. Analysis used independent t-tests and
Levene's test for homogeneity of variance.
R. Brown and J. Ogden
Table II. Parents' (n = 112) and childrens' (n = 112) food intake (yesterday and in general)
Parent's
general
Child's
yesterday
Child's
general
M
SD
M
SD
M
SD
M
SD
Yesterday
General
Yesterday
General
0.2
0.4
0.1
0.4
0.7
0.6
1
0.6
0.3
0.6
0.7
0.6
1.6
1.9
1.2
2.1
2.5
2.3
0.8
1.1
0.9
1.2
0.9
1.0
0.4
0.6
0.1
0.6
0.8
0.7
1
0.9
0.4
0.7
0.8
0.8
1.8
1.7
0.9
2.3
2.6
2.5
1.1
1.0
1.0
1.2
1.1
1.1
10
6
13
5
1
4
9
6
13
5
1
3
10
8
13
7
4
6
8
10
13
7
2
4
0.7
0.3
0.1
0.2
0.3
0.3
0.7
0.8
0.5
0.3
0.4
0.5
0.5
0.9
2.1
1.8
1.2
1.3
1.7
1.6
2.4
0.9
0.9
0.7
0.7
0.9
0.7
1.1
1.0
0.7
0.2
0.3
1.1
0.6
0.8
0.1
0.7
0.5
0.6
1.1
0.9
1.8
2.6
2.7
1.3
1.7
2.3
1.8
2.4
0.9
1.0
0.9
0.9
1.0
1.1
1.1
2
9
12
11
7
8
3
4
7
12
1
8
10
2
2
5
12
11
1
8
3
3
1
12
11
6
9
5
Parents' and children's reported snack
food intake
Table II describes the reported snack food intake of
parents and children. Children's most common
snack foods for `yesterday' were sweets, chocolate,
biscuits, toast and crisps, and for `in general' were
crisps, toast, chocolate, apples and biscuits. Many
of the foods being eaten by the children could be
considered `unhealthy'. Adults ate more healthily,
with three out of their ®ve most common foods
being categorized as `healthy'. The foods most
often eaten for both `yesterday' and `in general'
were toast, chocolate, biscuits, apples and yoghurt,
although ranked in a different order for these
different time points.
Comparison of parents and children's
snack food intake, eating motivations and
their body dissatisfaction
Parents and children were analysed to explore the
correlation between their snack food intake, eating
motivations and body dissatisfaction. The results
are shown in Table III. For snack food intake the
266
Parent's rank
Child's rank
results showed a signi®cant correlation between
parent's and child's snack food intake in general,
and between parents and child's unhealthy snack
food eaten yesterday, indicating that a more
healthy or unhealthy diet shown by the parents
was associated with a similar diet by their child.
There was no association for healthy snack food
eaten yesterday. In terms of motivations, the results
showed a signi®cant correlation between parent
and child for internal motivations, but not for
external motivations, indicating that a child was
more likely to state that they ate for reasons such as
feeling upset or cross if their parent also stated
likewise. The results also showed a signi®cant
correlation between parent and child for both
measures of body dissatisfaction, indicating that a
higher dissatisfaction by the parent was re¯ected in
a higher level in the child.
Role of parent's level of control over their
child's diet
Parents who exercised high control over their
children's diets were compared with parents who
exercised lower control to see if control levels
Downloaded from http://her.oxfordjournals.org/ by guest on June 22, 2016
Healthy food
grapes
oranges
peaches
yoghurt
toast
apples
Unhealthy food
chocolate
crisps
pastries
ice cream
sweets
cakes
biscuits
Parent's
yesterday
Children's diets
Table III. Comparisons between parents' and their childrens' food intake
Variable
Parent (n = 112)
M
aSigni®cant
SD
M
Spearman's correlation
SD
r
P
1.93
1.96
0.351
1.96
1.44
0.59
0.27
0.59
0.54
0.40
0.651
1.99
0.39
0.28
0.41
0.61
0.15
0.392
0.234
0.317
0.119
0.01
0.01
0.001
1.10
1.93
0.67
0.85
1.10
2.16
0.77
0.85
0.352
0.17
0.01
0.069
1.57
±1.6
0.74
1.14
1.1
±0.43
0.71
0.89
0.22
0.195
0.01
0.03
correlation between parent and child.
Table IV. Impact of parent's control over diet on children's food intake, eating motivations and body dissatisfaction
Variable
Food intake
healthy fooda yesterday
unhealthy fooda yesterday
healthy food general
unhealthy food general
Eating motivations
internal motivation
external motivation
Body dissatisfaction
body dissatisfaction
body difference
aSigni®cant
Low control (n = 64)
High control (n = 56)
M
M
SD
t
P
CI
SD
0.4645
0.5500
2.0054
2.0468
0.3417
0.3377
0.5921
0.4995
0.6161
0.8010
1.9851
2.2041
0.483
0.4747
0.6767
0.5842
±2.067
±3.299
0.173
±1.569
0.04
0.001
0.863
0.119
±0.2629
±0.4018
±0.2111
±0.3558
to
to
to
to
± 0.006
±0.1003
0.2516
0.041
1.0742
1.9883
0.7672
0.7604
1.2054
2.2593
0.8217
0.9391
±0.904
±1.732
0.368
0.086
±0.4185 to 0.1562
±0.5809 to 0.039
1.1063
±0.4762
0.6841
0.8203
1.1164
±0.3571
0.7702
0.9987
±0.75
±0.713
0.941
0.477
±0.2752 to 0.2552
±0.4495 to 0.2114
main effect of group.
in¯uenced their child's snack intake, motivations
for eating and body dissatisfaction. The results are
shown in Table IV. The results showed no effect of
parental control over diet on snack food intake in
general, eating motivations or body dissatisfaction.
However, those children whose parents reported
higher levels of control over their children's diet
reported eating more of both the unhealthy and
healthy snack foods yesterday, indicating that
attempts to restrict a child's food intake may be
paradoxically associated with its increase.
Role of parent's level of control over their
child's behaviour using food
Parents who exercised high control over their
child's behaviour using food were compared with
parents who exercised lower control to see if
control levels in¯uenced their children's snack
food intake, motivations for eating and body
dissatisfaction. The results are shown in Table V.
No differences were found between the two groups
for snack food intake, eating motivations or body
267
Downloaded from http://her.oxfordjournals.org/ by guest on June 22, 2016
Food intake
t healthy yesterday
t healthy generala
t unhealthy yesterdaya
t unhealthy generala
Eating motivations
t internal motivationa
t external motivation
Body dissatisfaction
t body dissatisfactiona
t body differencea
Child (n = 112)
R. Brown and J. Ogden
Table V. The impact of parent's use of diet to control the child's behaviour on children's food intake, eating motivations and
body dissatisfaction
Variable
aSigni®cant
High control (n = 51)
M
M
SD
P
CI
SD
0.531
0.6585
1.9949
2.0593
0.4092
0.3632
0.6384
0.5433
0.6088
0.6822
1.9932
2.1603
0.3961
0.4985
0.6217
0.5238
±1.123
±0.293
0.015
±0.988
0.264
0.770
0.988
0.320
±2.371 to 0.065
±0.1844 to 0.1369
±0.2320 to 0.2355
±0.3016 to 0.0995
1.0227
2.0731
0.767
0.8049
1.2500
2.1750
0.7794
0.8440
±1.567
±0.659
0.12
0.511
±0.5146 to 0.060
±0.4083 to 0.2045
1.0094
±0.3788
0.7449
0.9075
1.3275
±0.3788
0.6774
0.9075
±2.367
0.768
0.02
0.444
±0.5843 to ±0.0519
±0.2069 to 0.4689
main effect of group.
difference. However, those children whose parents
reported a greater use of food to control their
child's behaviour showed higher levels of body
dissatisfaction.
Discussion
The present study aimed to explore the ways in
which parents effect their child's eating attitudes
and eating behaviour, and to test both the modelling and control and control theories of parental
in¯uence. There are some problems with the study,
however, which need to be addressed. First, many
children did not consent to take part in the study.
Because consent was needed to opt into the study
rather than to opt out, it is possible that some
parents did not receive the consent form and were
not told about the study. Non-consent may therefore re¯ect the child's forgetfulness rather than an
objection to a study on eating behaviour. However,
it is also possible that those who did not consent
were different to those who did in terms of eating
attitudes and food intake. It is not possible to
estimate the impact of this on the results as eating
control and consent could be related in either
direction. Therefore, care must be taken in
generalizing from the results of this study to all
children in general. Second, the measure of food
268
t
intake focused only on snack foods rather than
providing a comprehensive description of all
aspects of the children's diets. In addition, the
measure relied upon self-report which was not
supported by any objective assessment. However,
all measures of food intake are problematic.
Researcher observation can change food intake,
laboratory assessments offer an unnatural environment, and diary measures cause self-monitoring
and either increase or decrease eating. Further,
fully comprehensive food checklists can yield data
which is unmanageable and unsynthesized. The
measure used in the present study was designed to
be simple and short so that it could be understood
by both parents and children. It was also designed
to be focused on a child's consumption of snack
foods which are often at the centre of eating-related
negotiations between the parent and child. The
limitations of the measure are acknowledged, but it
is believed that it provides some insights into the
impact of modelling and control.
The results indicated that both parents' and
children's diets consisted of many unhealthy snack
foods such as crisps, chocolate and biscuits,
con®rming previous ®ndings from larger-scale
surveys [e.g. (Butriss, 1995; Nicklas, 1995;
Wardle, 1995)]. The results also indicated a strong
association between a parent's and their child's
Downloaded from http://her.oxfordjournals.org/ by guest on June 22, 2016
Food intake
healthy food yesterday
unhealthy food yesterday
healthy food general
unhealthy food general
Eating motivations
internal motivation
external motivation
Body dissatisfaction
body dissatisfactiona
body difference
Low control (n = 66)
Children's diets
reported higher levels of body dissatisfaction.
Food is embedded with a complex set of meanings
removed from hunger and satiety (Ogden, 2003). It
is possible that using food to change behaviour
detaches food further from its role in satiating
hunger and promotes a more problematic relationship with eating. The body dissatisfaction reported
in the present study may be a re¯ection of such a
relationship.
In summary, previous research has described
theories concerned with both modelling and control. The results from the present study offer
support for both these theories in pairs of parents
and children. In particular, modelling was found to
have a clear in¯uence on how children both think
and behave around food, with consistent associations found between parent's and children's
eating behaviours and attitudes. Parental control
was also found to exert an impact. Whilst control
over diet, however, in¯uenced the child's food
intake and not their attitudes, using food to control
behaviour was found to have the reverse effect.
Accordingly, whilst modelling may have a consistent impact upon a child's diet and attitudes, the
impact of parental control depends upon whether
this control is focused directly at the child's diet or
indirectly uses diet as a means to modify other
aspects of the child's behaviour. To date, the
theories of modelling and control have been
developed as independent perspectives using different populations and different methodologies.
The results from the present study provide some
insights into the relative impact of these different
forms of parental in¯uence and indicate that a
positive parental role model may be a more
effective means to facilitate change than parental
attempts to impose control over their child's food
intake. Such results have direct implications for the
development of health education interventions.
Recent recommendations for the development of
intervention programs have called for the provision
of information to parents concerning meal content,
size and timing, and details about the potentially
damaging in¯uence of coercive feeding practices
(Birch and Davison, 2001). The results from the
present study indicate that in addition to simply
269
Downloaded from http://her.oxfordjournals.org/ by guest on June 22, 2016
snack food intake for all snacks in general and
unhealthy snacks eaten yesterday. This provides
support for the modelling theory of parental
in¯uence and indicates that children's diets are
effected by the types of food eaten by their parents
[e.g. (Klesges et al., 1991; Olivera et al., 1992;
Wardle, 1995)]. Modelling also appeared to have a
role in the transmission of eating-related attitudes,
with the results showing associations between
parents' and their childrens' internal motivations
and body dissatisfaction. This re¯ects previous
research which has reported a correspondence
between mothers' and daughters' degree of weight
concern (Hall and Brown, 1982; Hill et al., 1990;
Steiger et al., 1994). The results from the present
study therefore support a modelling theory of
parental in¯uence, and indicate that parents' eating
behaviours and attitudes closely correspond to
those of their children.
The present study also explored the role of
control. This was ®rst assessed in terms of parental
attempts to control their child's diet, and the results
showed that children whose parents reported
greater attempts to restrict their child's food intake
indicated eating more of both the healthy and
unhealthy foods. These parents may have been
imposing such control because their child had a
pre-existing tendency to overeat. However, in line
with previous studies of parental control (Birch,
1999; Fisher and Birch, 1999), it is also possible
that parentally enforced food restriction has the
paradoxical effect of triggering overeating in
children. Such an impact is in accordance with
experimental studies of children's diets (Fisher and
Birch, 1999), and also ®nds re¯ection in studies of
dietary restraint and the impact of self-imposed
food restriction on eating behaviour [e.g. (Herman
and Mack, 1975; Ogden, 2003)].
The role of control was also explored in terms of
the use of food to modify behaviour. The results
showed that such parental control had no impact
upon the child's diet, which is in contrast to
previous research [e.g. (Birch et al., 1980, 1982;
Newman and Taylor, 1992)]. However, those
children whose parents reported a greater use of
food as a tool for behavioural modi®cation
R. Brown and J. Ogden
educating parents what to feed their children, the
parent's diet itself should also be the focus on
change. If parents can be encouraged to recognize
that their own eating behaviour is the most
important source of information for their children
then maybe parents can be encouraged to adopt a
`do as I do not what I say' approach to their
children's food intake.
References
270
Downloaded from http://her.oxfordjournals.org/ by guest on June 22, 2016
Bandura, A. (1977) Social Learning Theory. Prentice-Hall,
Englewood Cliffs, NJ.
Berenson, G.S., Srinivasan, S.R., Bao, W., Newman, W.P., III,
Tracy, R.E. and Wattigney, W.A. (1998) Association
between multiple cardiovascular risk factors and
atherosclerosis in children and young adults. New England
Journal of Medicine, 338, 1650±1656.
Birch, L.L. (1980) Effects of peer models' food choices and
eating behaviors on preschoolers' food preferences. Child
Development, 51, 489±496.
Birch, L.L. (1999) Development of food preferences. Annual
Review of Nutrition, 19, 41±62.
Birch, L.L. and Davison, K.K. (2001) Family environmental
factors in¯uencing the developing behavioural controls of
food intake and childhood overweight. Pediatric Clinics of
North America, 48, 893±907.
Birch, L.L., Birch, D., Marlin, D. and Kramer, L. (1982)
Effects of instrumental eating on children's food preferences.
Appetite, 3, 125±134.
Birch, L.L., Marlin, D. and Rotter, J. (1984) Eating as the
`means' activity in a contingency: effects on young
children's food preference. Child Development, 55, 431±439.
Butriss, J. (1995) Nutrition in General Practice, 1st edn, vol. 2.
Promoting Health and Preventing Disease. Royal College of
General Practitioners, London.
Contento, I.R., Basch, C., Shea, S., Gutin, B., Zybert, P.,
Michela, J.L. and Rips, J. (1993) Relationship of mothers'
food choice criteria to food intake of pre-school children:
identi®cation of family subgroups. Health Education
Quarterly, 20, 243±259.
Cooper, P.J., Taylor, M.J., Cooper, Z. and Fairburn, C.G.
(1987) The development and validation of the Body Shape
Questionnaire. International Journal of Eating Disorders, 6,
485±494.
Currie, C., Todd, J. and Thompson, C. (1997) Health
Behaviours of Scottish School Children: Report 5:
Comparisons of National Patterns in 1990 and 1994.
Research Unit in Health and Behavioural Change and
Health Education Board for Scotland, Edinburgh.
Fisher, J.O. and Birch, L.L. (1999) Restricting access to a
palatable food affects children's behavioral response, food
selection and intake. American Journal of Clinical Nutrition,
69, 1264±1272.
Gatherer, A., Par®t, J., Porter, E and Vessey, M. (1979) Is
Health Education Effective? An Overview of Evaluated
Studies. Health Education Council, London.
Hales, C.N., Barker, D.J.P., Clark, P.M.S., Cox, L.J., Fall, C.,
Osmond, C. and Winter, P.D. (1991) Fetal and infant growth
and impaired glucose tolerance at age 64. British Medical
Journal, 303, 1019±1022.
Hall, A. and Brown, L.B. (1982) A comparison of the attitudes
of young anorexia nervosa patients and non patients with
those of their mothers. British Journal of Psychology, 56,
39±48.
Herman, P. and Mack, D. (1975) Restrained and unrestrained
eating. Journal of Personality, 43, 646±660.
Hill, A.J., Weaver, C. and Blundell, J.E. (1990) Dieting
concerns of 10 year olds girls and their mothers. British
Journal of Clinical Psychology, 29, 346±348.
Kelder, S.H., Perry, C.L., Klepp, K.-I. and Lytle, L.L. (1994)
Longitudinal tracking of adolescent smoking, physical
activity and food choice behaviours. American Journal of
Public Health, 84, 1121±1126.
Klesges, R.C., Stein, R.J., Eck, L.H., Isbell, T.R. and Klesges,
L.M. (1991) Parental in¯uences on food selection in young
children and its relationships to childhood obesity. American
Journal of Clinical Nutrition, 53, 859±864.
Lepper, M., Sagotsky, G., Dafoe, J.L. and Greene, D. (1982)
Consequences of super¯uous social constraints: effects on
young children's social inferences and subsequent intrinsic
interest. Journal of Personality and Social Psychology, 42,
51±65.
Lowe, C.F., Dowey, A. and Horne, P. (1998) Changing what
children eat. In Murcott, A. (ed.), The National's Diet: The
Social Science of Food Choice. Addison Wesley Longman,
London, pp. 57±80.
Moller, J.H., Taubert, K.A., Allen, H.D., Clark, E.B. and Lauer,
R.M. (1994) Cardiovascular health and disease in children:
current status. Circulation, 89, 923±930.
Newman, J. and Taylor, A. (1992) Effect of a means-end
contingency on young children's food preferences. Journal
of Experimental Psychology, 64, 200±16.
Nicklas, T.A. (1995) Dietary studies of children and
young adults (1973±1988): the Bogalusa heart study.
American Journal of Medical Science, 310(Suppl. 1),
S101±S108.
Ogden, J. (2003) The Psychology of Eating: From Healthy to
Unhealthy Behaviour. Blackwell, Oxford.
Olivera, S.A., Ellison, R.C., Moore, L.L., Gillman, M.W.,
Garrahie, E.J. and Singer, M.R. (1992) Parent±child
relationships in nutrient intake: the Framingham
Children's Study. American Journal of Clinical Nutrition,
56, 593±598.
Resnicow, K., Davis-Hearn, M., Smith, M., Baranowski, T.,
Lin, L.S., Baranowski, J., Doyle, C. and Wang, D.T. (1997)
Social cognitive predictors of fruit and vegetable intake in
children. Health Psychology, 16, 272±276.
Steiger, H., Stotland, S., Ghadirian, A.M. and Whitehead, V.
(1994) Controlled study of eating concerns and
psychopathological traits in relative of eating disorders
probands: do familial traits exist? International Journal of
Eating Disorders, 18, 107±118.
Steptoe, A., Pollard, T.M. and Wardle, J. (1995) Development of a measure of the motives underlying the selection
Children's diets
of food: the food choice questionnaire. Appetite, 25,
267±284.
Stunkard, A.J., Stinnett, J.L. and Smoller, J.W. (1986)
Psychological and social aspects of the surgical treatment
of obesity. American Journal of Psychiatry, 143, 417±429.
Van Strien, T., Frijters, J.E.R., Bergers, G.P.A. and Defares,
P.B. (1986) The Dutch eating behaviour questionnaire for
assessment of restrained, emotional and external eating
behaviour. International Journal of Eating Disorders, 5,
747±755.
Wardle, J. (1995) Parental in¯uences on children's diets.
Proceedings of the Nutrition Society, 54, 747±758.
Received on December 20, 2002; accepted on May 7, 2003
Downloaded from http://her.oxfordjournals.org/ by guest on June 22, 2016
271
Targeting Parents Exclusively in the Treatment
of Childhood Obesity: Long-Term Results
Moria Golan* and Scott Crow†
Abstract
GOLAN, MORIA, AND SCOTT CROW. Targeting
parents exclusively in the treatment of childhood obesity:
long-term results. Obes Res. 2004;12:357–361.
Objective: To report the long-term change in children’s
overweight following a family-based health-centered approach where only parents were targeted compared with a
control intervention where only children were targeted.
Research Methods and Procedures: Fifty of the 60 children
who participated in the original study were located 7 years
later, and their weight and height were measured. At the
point of the 7-year follow-up, the children were 14 to 19
years of age. Repeated measure ANOVA was used to test
differences between the groups in percent overweight at
different time-points.
Results: Mean reduction in percent overweight was greater
at all follow-up points in children of the parent-only group
compared with those in the children-only group (p ⬍ 0.05).
Seven years after the program terminated, mean reduction in
children’s overweight was 29% in the parent-only group vs.
20.2% in the children-only group (p ⬍ 0.05).
Discussion: Over the long term, treatment of childhood
obesity with the parents as the exclusive agents of change
was superior to the conventional approach.
Key words: childhood obesity, family-based treatment,
targeting parents
Introduction
The prevalence of obesity in children and adolescents is
increasing rapidly (1). The treatment of obesity is one of the
Received for review July 3, 2003.
Accepted in final form December 18, 2003.
The costs of publication of this article were defrayed, in part, by the payment of page
charges. This article must, therefore, be hereby marked “advertisement” in accordance with
18 U.S.C. Section 1734 solely to indicate this fact.
*School of Nutritional Sciences, The Hebrew University of Jerusalem, Rehovot, Israel and
†Department of Psychiatry, University of Minnesota, Minneapolis, Minnesota.
Address correspondence to Moria Golan, Shahaf, Kibbutz Naan, Mobile Post, Soreq 76829,
Israel.
E-mail: moriag@netvision.net.il
Copyright © 2004 NAASO
most difficult challenges that multidisciplinary pediatric
health care providers face. Coping with obesity requires
lifelong attention to healthy eating and an active lifestyle. It
is necessary to alter the home environment rather than be
preoccupied with the child’s weight, because such a preoccupation may damage the child’s self-esteem as well as the
relationship between the parent and the child (2).
The home environment has the potential to affect children’s energy balance and diet composition in numerous
ways. Parents may influence the family environment by
exposing family members to certain foods, by actively encouraging the family members to eat certain foods, or by
passively allowing certain foods in the regular diets. Foodrelated parenting practices and the physical and emotional
environment in which eating and activity behaviors are
developed are also important variables (3–5). Parents may
also serve as role models for their children’s eating and
activity behavior, although various studies have shown low
family resemblance in food preference (6).
In nearly all prevention and targeted intervention work
with young children (regardless of the specific health outcome being addressed), two-generation programs are considered essential for improving the child’s outcome (7). In
the treatment of childhood obesity, clear benefits have been
demonstrated from parental involvement (8 –11). State-ofthe-art childhood obesity programs are based on family.
Family-based intervention is implemented on the premise
that parental support, family functioning, and home environment are important determinants of treatment outcomes.
These interventions emphasize the necessity of making
changes in the home and family environment (12,13).
Epstein et al. (13) provided evidence that treatments
designed to target and reinforce a change in habits and
weight loss in obese parents and children together are superior over 10 years to treatments that focus on the child’s
habits and weight change independent of parental success,
as well as to a control treatment that targets and reinforces
the family members for attendance only. Epstein et al.
concluded from the 10-year outcome data that weight regulation in children can be achieved and maintained over
OBESITY RESEARCH Vol. 12 No. 2 February 2004
357
Targeting Parents in Childhood Obesity Treatment, Golan and Crow
extended periods from childhood through adolescence to
adulthood. They suggested that parent modeling of children’s eating and exercise behaviors was not a mechanism
that could account for the outcome because children maintained a decreased percent overweight, whereas their obese
parents were regaining weight.
It is likely that parental behavior is important during the
acquisition of the skills, whereas changes in the environment (stimulus variables) and reinforcement for new eating
and exercise behaviors may be operative over extended
follow-up periods (13).
While it is apparent that parents should be involved in
treatment of obese children, there is no consensus on the
“how.” It has been suggested that strategies that emphasize
parental control over the quality and pattern of the food
environment should be encouraged, whereas strict parental
control over a child’s food intake should be discouraged
(14 –16). Authoritative parenting (in which parents are both
firm and supportive and assume a leadership role in the
environmental change with appropriate granting of child’s
autonomy) rather than authoritarian style (which controls
child-feeding practices) was found to be the effective parental child-feeding modality (4,16,17).
The Expert Committee for Obesity Evaluation and Treatment recommended that treatment of childhood obesity
begin early and involve the family. Parenting skills were
mentioned as the foundation for successful intervention that
includes gradual targeted increases in activity and a targeted
reduction in high-fat, high-calorie foods (2).
In most family-based behavioral weight loss programs for
children, the obese child is the main agent of change, with
varying degrees of parental involvement. We suggested
previously that, if parenting style is the focus of treatment,
the parents are the main agent of change, and they, rather
than the child, should be targeted by the intervention. Exclusively targeting the parents in the interventions that had
a family health-centered approach led to weight loss in
obese children (12). Parents served as both a source of
authority and a role model for their children. Parents provided an environment with fewer “obesogenic” factors and
more self-regulation and healthy behavior practice. A program that omitted the obese child from direct intervention
and targeted parents only was associated with greater weight
loss and higher consumption of healthy foods compared with a
program that treated children with the child-only condition,
where they were the main agents of change (18).
The purpose of this paper is to report the long-term
outcome in overweight children following a family-based
health-centered approach where only parents participated in
the group sessions (parent-only group) compared with a
person-based intervention where only the children participated in the group sessions and the family environment was
not directly targeted (children-only group).
358
OBESITY RESEARCH Vol. 12 No. 2 February 2004
Research Methods and Procedures
Subjects
Fifty of the 60 children who were initially randomly
recruited for this study were located after 7 years. There
were five participants missing in each group. At the point of
the 7-year follow-up, the children were 14 to 19 years of age
(mean age, 16 ⫾ 0.5 years).
Initial Study Protocol
The Ethics Committee for Human Experimentation of Tel
Aviv University approved the research protocol, and all
parents provided written informed consent.
The children were randomly assigned to either the parentonly group (only parents were targeted) or the child-only
group (children were targeted). The two groups were
matched for sex and age. No significant differences in
socioeconomic status were found between the groups.
Parent-only Group. Only parents participated in the
group sessions, and all suggested changes were intended for
the entire family. Parents attended 14 1-hour support and
educational group sessions. The first four sessions were held
weekly, the next four were held biweekly, and the last six
were held once every six weeks. Clinical dietitians delivered
the sessions. Two similar groups were held, with 15 families
in each group. The topics discussed included limits of
responsibilities, nutrition education, eating and activity behavior modification, decreasing stimulus exposure, parental
modeling, problem solving, cognitive restructuring, and
coping with resistance. Parents were encouraged to practice
authoritative parenting style (11). A detailed description is
given elsewhere (12,18).
Child-only Group. Each child was prescribed a diet providing 1500 kcal/d. Thirty 1-hour group sessions were led
by a clinical dietitian. Two similar groups were held, with
15 children allocated to each group. The first seven sessions
were conducted weekly, and the remainder were held biweekly for a total of 1 year. The topics discussed included
physical activity, eating behavior modification, stimulus
control, self-monitoring, nutrition education, problem solving, and cognitive restructuring. Individual counseling sessions were held whenever a child missed the group session,
encountered difficulties in adhering to change, or asked for
a change in his or her diet prescription.
Data Collection at Follow-up
Weight and height were measured 1, 2, and 7 years after
the program’s termination. Measurements were to the nearest 0.1 kg and 1 cm, respectively, using a standard medical
balance-beam scale with a rigid vertical height rod (Shekel
Scales, Tel Aviv, Israel). Subjects were weighed while
wearing light clothing and no shoes. Percent overweight
was calculated by the following formula: 100 ⫻ (actual
weight ⫺ desirable weight)/desirable weight. Desirable
weight of the children was based on the 50th percentile
Targeting Parents in Childhood Obesity Treatment, Golan and Crow
weight for a particular age, sex, and height according to the
National Center for Health Statistics growth charts (U.S.)
(1). Weight loss was determined by deducting the final
percent overweight from the percentage overweight at baseline. Eat-26 (Eating Attitudes Test) was used to screen for
individuals with significant symptoms of eating disorders at
the 7-year follow-up visit.
Data Analysis
Statistical analyses were performed using SPSS 10 for
Windows (SPSS Inc., Chicago, IL); p values of ⬍0.05 were
considered statistically significant. All results are given as
mean ⫾ SD.
To compare the children’s overweight status at the different time-points, a repeated measure ANOVA was used,
with group and gender as the between variables and time as
the within variable and linear contrasts used to follow-up on
significant main effects or interactions. This is an intentionto-treat analysis where the missing values at the last follow-up meeting were replaced with baseline percent overweight values. The missing in-between values were
replaced by values that were extrapolated from each child’s
overweight curve.
2 analyses were used to compare the groups in percent
of youth who achieved a nonobese status (⬍20% overweight).
Results
Weight Loss at Intervention Termination
The change in percent overweight during the intervention
was published earlier (12). The children in the parent-only
group achieved a significantly higher reduction in percent
overweight compared with the children in the child-only
group (14.6% vs. 8.43%; p ⬍ 0.03, analysis of covariance).
This difference in weight reduction was observed even
though there was no significant difference in increased
height between the two groups over the 1-year study.
At the end of the intervention, 35% of the children in the
parents-only group reached a nonobese status, whereas in
the child-only group, only 14% reached nonobese status. 2
analyses revealed a significant difference between the two
groups in the rate of the nonobese children at the end of the
intervention (p ⬍ 0.01).
The change in overweight status did not show a statistically significant interaction with gender (p ⫽ 0.23) (12).
Follow-up Visits
At the 1-year follow-up visit (1 year after program termination), the weight loss in the children of the parent-only
group was statistically significant compared with that of the
child-only group (⫺13.6 vs. 0, p ⬍ 0.05; Figure 1).
At the 2-year follow-up, there was a mean reduction in
overweight of 15% in children of the parent-only group and
an increase of 2.9% in children in the child-only group (p ⬍
0.01; Figure 1).
Figure 1: Change in children’s percentage overweight at 1, 2, and
7 years of follow-up. Mean ⫾ SD. Significant difference between
the parent only group (———) and the child only group (– – – –):
*p ⬍ 0.05; **p ⬍ 0.01.
At the 7-year follow-up (7 years after program termination), to our surprise, both treatment conditions demonstrated substantial weight loss. However, the mean reduction in overweight was 29% in children in the parent-only
group and 20.2% in those of the child-only group (p ⬍ 0.05;
Figure 1). At this point, 60% of the children in the parentonly group compared with only 31% of the children in the
child-only group were in a nonobese status. 2 analysis
revealed a significant difference between the two groups in
the rate of the children’s nonobese status 7 years after the
intervention terminated (p ⬍ 0.01). In the repeated measure
analysis, significant differences were found between the two
treatment groups (F ⫽ 64.5, p ⬍ 0.000). The effects of time
and gender were both significant in the model (time F ⫽
14.4, p ⬍ 0.000; gender F ⫽ 7.9, p ⬍ 0.007). In both
groups, boys’ overweight was greater than the girls’, and
this trend continued all along. A significant interaction was
found for time by group (F ⫽ 6.9, p ⬍ 0.000), suggesting
that the differences between groups in overweight reduction
at each time-point were significant (p ⬍ 0.000). Time by
gender and group by gender did not show a significant
interaction, suggesting that rate of overweight reduction and
treatment effect did not differ between boys and girls. No
interaction was found between percent overweight at baseline and weight loss.
Seven years after the program terminated, two (6.6%) of
the girls from the child-only group reported eating disorder
symptoms. (Both were binging and purging.) None of the
children from the parent-only group reported any eating
disorder symptoms.
Discussion
A family-based, health-centered approach that targeted
solely parents was found to induce greater weight loss in
OBESITY RESEARCH Vol. 12 No. 2 February 2004
359
Targeting Parents in Childhood Obesity Treatment, Golan and Crow
obese children at treatment termination and at 1-, 2-, and
7-year follow-up visits. The differences between the groups
were statistically and clinically significant at each timepoint. A similar pattern of weight status over years was
reported by Epstein et al. (13).
Family eating environments and child-feeding practices
may constitute an “obesigenic” environmental influence for
children. Improving food choice and reducing caloric intake
in obese children are required for a long-term change. The
higher percentage of weight reduction and better maintenance of reduced weight observed in children of the parentonly group compared with the children who actively participated in the sessions might be explained by the greater
change in the “obesogenic” factors in the child’s environment. At the 12-month follow-up, greater reduction in the
child’s problematic eating behaviors and in the presence of
stimuli at home (sweets and snacks) was observed in the
parent-only group (18).
A possible mechanism for the change in the obesogenic
environment, although not measured in this study, could be
the change in parenting practice. Practicing authoritative
rather than controlling parenthood might contribute to parents’ ability to maintain a healthier environment. One of the
main objectives of the parent-only group was to enhance
authoritative parenting style to improve parents’ ability to
create a healthy environment in the house and support their
child’s autonomy and self-esteem. Although the change in
parenting style was not tested in this study, other studies
have reported that fruit consumption and fruit-specific cognitions were most favorable among adolescents who were
being raised with an authoritative parenting style (4). Fisher
et al. reported that parental pressure to eat fruit and vegetables discouraged intake among young girls (19).
Parents who are concerned about their children’s food
intake behavior may adopt controlling child-feeding practices in an attempt to prevent overweight or negative health
consequences (15). Mothers reported using more restrictive
feeding practices when they were invested in weight and
eating issues, when they were concerned about daughters’
weight, and when daughters were heavier. Mothers reported
using more pressure in child feeding when daughters were
thinner and when they perceived daughters as underweight
(20).
Constructive functioning of parental authority instead of
parental controlling is crucial for preventing and treating
childhood obesity, because parental strategies for controlling their child’s food intake are counterproductive to
the development of the child’s self-regulation ability (14 –
17,21,22). Parental control efforts may increase children’s
preference for restricted foods as well as their intake of
such foods (15,20 –22) while diminishing self-control in
eating (14).
In our study, parents served both as a source of authority
and as role models for the obese child. They controlled the
360
OBESITY RESEARCH Vol. 12 No. 2 February 2004
quality and pattern of the food environment (offering a
variety of healthful foods, less food stimulus at home,
overseeing the planning and assembly of meals) as well as
limited sedentary behaviors. Nevertheless, parents were advised not to restrict the amount of food the child ate during
meals, thus promoting more self-regulation that fosters
healthy practices related to weight control issues (11).
Approaching parents exclusively shifts the focus of the
group from weight issues to parenting issues, which is
decisive in light of current evidence suggesting that the
family is becoming a more democratic unit and that children
are spending more on sweet snack items than ever before
(23).
Prospective studies of adolescent girls suggest that dietary restriction predates bulimic symptoms (24), with one
study reporting that adolescent girls who were dieting had
an 8-fold increased risk for being diagnosed with an eating
disorder compared with nondieters (25). In our study at the
7-year follow-up, 6.6% of the subjects in the children-only
group developed eating disorder symptoms (close to the 9%
rate of eating disorders in the general population seen in
studies that used self-reported diagnoses). Epstein et al.
found that at the 10-year follow-up, 4% of the subjects (all
girls) reported having been treated for eating disorders (26).
Targeting parents for parenting skills in the treatment of
childhood obesity is supported by wide research (27–29).
Glenny et al. found that at 1-year follow-up, children whose
parents had participated in a short course in general behavior management had significantly better weight control than
children in an intervention program that focused only on
weight reduction (28).
Although not measured, the dietitian noted a lower level
of resistance to change among the children in the parentonly group compared with those in the child-only group
where the children had to be actively responsible for the
necessary change. Lerner (30) suggested that children often
resist change and express it by rebellion and acting oppositely when subjected to demands for change. Although the
children were encouraged to seek their parents’ help if
needed, success was still their responsibility. For a child
whose family did not accept the necessary environmental
changes, the child’s success was probably of short duration.
The long-term results were surprisingly positive, with
60% of the children in the treatment group vs. 31% in the
control group in nonobese status. At the 7-year follow-up,
some of the participants from both groups were soldiers in
the army (with increased physical activity level at this
period) and some were in their late adolescence, a period in
which many obese individuals take responsibility for controlling their overweight. This might explain part of the
reduction in overweight observed at this point in time.
This study does have some limitations. Sample size was
modest; thus, to preserve adequate statistical power, only
two conditions could be studied. We chose the child-only
Targeting Parents in Childhood Obesity Treatment, Golan and Crow
approach as the control group, because this is the prevalent
intervention treatment approach in Israel and in many other
countries. Future research should compare the parent-only
approach with the parent-child approach, both targeting a
health-centered approach rather than weight loss. Studies
should explore ways to enhance parent compliance and
partnership in implementing the changes needed in the
house, in accordance with the observation that some parents
(mainly morbidly obese) were less cooperative, claiming
that their personal freedom was threatened.
Parents should be the main change mediator in weightrelated interventions, because their involvement is crucial
for the induction of a healthy environment, modeling of
healthy eating and activity patterns, and improvement in the
child’s practices and weight status in the long term. A
health-centered rather than weight-centered approach may
be the most appropriate intervention for the treatment and
prevention of childhood obesity.
Acknowledgments
There was no outside funding/support.
References
1. National Centre for Health Statistics online information. http://www.cdc.gov/nchs/products/pubs/pubd/hestats/
obese/obese99.htm (Accessed on June 2, 2003).
2. Barlow SE, Dietz WH. Obesity evaluation and treatment:
Expert Committee Recommendations. Pediatrics. 1998;102:
1–11.
3. Hill JO, Peters JC. Environmental contributions to the obesity epidemic. Science. 1998;280:1371– 4.
4. Kremers SPJ, Brug J, Vries H, Engels RC. Parenting style
and adolescent fruit consumption. Appetite. 2003;41:43–50.
5. Satter EM. Should the obese child diet? In: Clark KL, Parr
RB, Castelli WP (eds). Evaluation and Management of Eating
Disorders. Champaign, IL: Life Enhancement Publications;
1988, pp. 61–75.
6. Rozin P. Family resemblance in food and other domains: the
family paradox and the role of parental congruence. Appetite.
1991;16:93–102.
7. Graber JA, Brooks-Gunn J. Prevention of eating problems
and disorders: including parents. Eat Disord. 1996;4:348 – 62.
8. Kirschenbaum DS, Harris ES, Tomarken AJ. Effects of
parental involvement in behavioral weight loss therapy for
preadolescents. Behav Ther. 1984;15:485–500.
9. Epstein LH, Wisniewski L, Wing R. Child and parent psychological problems influence child weight control. Obes Res.
1994;2:509 –15.
10. Coates TJ, Killen JD, Slinkard LA. Parent participation in a
treatment program for overweight adolescents. Int J Eat Disord. 1982;1:37– 48.
11. Golan M, Fainaru M, Weizman A. Familial approach to the
treatment of childhood obesity: a conceptual model. J Nutr
Edu. 2001;33:102– 6.
12. Golan M, Weizman A, Apter A, Fainaru M. Parents as the
exclusive agents of change in the treatment of childhood
obesity. Am J Clin Nutr. 1998;67:1130 – 8.
13. Epstein LH, Valoski A, Wing RR. Ten year follow-up of
behavioral, family-based treatment for obese children. JAMA.
1990;264:2519 –23.
14. Birch LL, Fisher JO. Development of eating behaviors
among children and adolescents. Pediatrics. 1998;101:539 – 49.
15. Birch LL, Davidson KK. Family environmental factors influencing the developing behavioral controls of food intake
and childhood overweight. Pediatr Clin North Am. 2001;48:
893–907.
16. Satter EM. Internal regulation and the evolution of normal
growth as the basis for prevention of obesity in children. J Am
Diet Assoc. 1996;96:860 – 86.
17. Schmitz KH, Lytle LA, Phullips GA, Murray DM, Birnbaum AS, Kubik MY. Psychosocial correlates of physical
activity and sedentary leisure habits in young adolescents: The
Teens Eating for Energy and Nutrition at School study. Prev
Med. 2002;34:266 –78.
18. Golan M, Fainaru M, Weizman A. Role of behavior modification in the treatment of childhood obesity with the parents
as the exclusive agents of change. Int J Obes Relat Metab
Disord. 1998;22:1– 8.
19. Fisher JO, Mitchell DC, Smiciklas-Wright H, Birch LL.
Parental influences on young girls’ fruit and vegetable, micronutrient, and fat intakes. J Am Diet Assoc. 2002;102:58 – 64.
20. Fisher JO, Birch LL. Restricting access to palatable foods
affects children’s behavioral response, food selection, and
intake. Am J Clin Nutr. 1999;69:1264 –72.
21. Birch LL, Marlin DW, Rotter J. Eating as the ‘means’
activity in a contingency: effect on young children’s food
preference. Child Dev. 1984;55:431–9.
22. Francis LA, Hofer SM, Birch LL. Predictors of maternal
child-feeding style: maternal and child characteristics. Appetite. 2001;37:231– 43.
23. Roberts BP, Blinkhorn AS, Duxbury JT. The power of
children over adults when obtaining sweet snacks. Int J Pediatr Dentis. 2003;13:76 – 84.
24. Thompson JK, Coovert MD, Richards KJ, Johnson S,
Cattarin J. Development of body image, eating disturbance,
and general psychological functioning in female adolescents:
covariance structural modeling and longitudinal investigations. Int J Eat Disord. 1995;18:221–36.
25. Patton GC, Johnson-Sabine E, Wood K, Mann AH,
Wakeling A. Abnormal eating attitudes in London school
girls-a prospective epidemiologic study: outcome at twelve
month follow-up. Psychol Med. 1990;20:383–94.
26. Epstein LH, Valoski AM, Wing RR, McCurley J. Ten year
outcomes of behavioral family-based treatment for childhood
obesity. Health Psychol. 1994;13:373– 83.
27. Israel AC, Stolmaker L, Andrian CA. The effects of training parents in general child management skills on a behavioral
weight loss program for children. Behav Ther. 1985;16:169 – 80.
28. Glenny AM, O’Meara S, Melville A, Sheldon TA, Wilson
C. The treatment and prevention of obesity: a systemic review
of the literature. Int J Obes Relat Metab Disord. 1997;21:715–37.
29. Epstein LH, Myers MD, Raynor HA, Saelens BE. Treatment of pediatric obesity. Pediatrics. 1998;101:554 –70.
30. Lerner RM. Child development: life-span perspectives. Hum
Dev. 1982;25:38 – 41.
OBESITY RESEARCH Vol. 12 No. 2 February 2004
361
Purchase answer to see full
attachment