Rev Paul Pediatr.

2015;33(1):42–49

REVISTA PAULISTA
DE PEDIATRIA
www.rpped.com.br

ORIGINAL ARTICLE

Children’s eating behavior: comparison between normal

and overweight children from a school in Pelotas,
Rio Grande do Sul, Brazil

Darlise Rodrigues dos Passosa,*, Denise Petrucci Gigantea,

Francine Villela Maciela, Alicia Matijasevichb

a
Universidade Federal de Pelotas (UFPel), Pelotas, RS, Brazil
b
Universidade de São Paulo (USP), São Paulo, SP, Brazil

Received 14 March 2014; accepted 10 July 2014

KEYWORDS Abstract
Obesity; Objective: To evaluate differences in children’s eating behavior in relation to their nutri-
Nutritional assessment; tional status, gender and age.
Feeding behavior; Methods: Male and female children aged six to ten years were included. They were re-
Child cruited from a private school in the city of Pelotas, Rio Grande do Sul, southern Brazil,
in 2012. Children´s Eating Behaviour Questionnaire (CEBQ) subscales were used to assess
eating behaviors: Food Responsiveness (FR), Enjoyment of Food (EF), Desire to Drink (DD),
Emotional Overeating (EOE), Emotional Undereating (EUE), Satiety Responsiveness (SR),
Food Fussiness (FF) and Slowness in Eating (SE). Age-adjusted body mass index (BMI) z-sco-
res were calculated according to the WHO recommendations to assess nutritional status.
Results: The study sample comprised 335 children aged 87.9±10.4 months and 49.3% had
normal weight (n=163), 26% were overweight (n=86), 15% were obese (n=50) and 9.7%
were severely obese (n=32). Children with excess weight showed higher scores at the
CEBQ subscales associated with “food approach” (FR, EF, DD, EOE, p<0.001) and lower
scores on two “food avoidance” subscales (SR and SE, p<0.001 and p=0.003, respectively)
compared to normal weight children. Differences in the eating behavior related to gen-
der and age were not found.
Conclusions: “Food approach” subscales were positively associated to excess weight in
children, but no associations with gender and age were found.
© 2014 Sociedade de Pediatria de São Paulo. Published by Elsevier Editora Ltda. All rights
reserved.

*Corresponding author.
E-mail: [email protected] (D.R. Passos).
DOI of refers to article: http://dx.doi.org/10.1016/j.rpped.2014.11.007
1984-1462/© 2014 Sociedade de Pediatria de São Paulo. Published by Elsevier Editora Ltda. All rights reserved.
Children’s eating behavior: comparison between normal and overweight children from a school in Pelotas 43

PALAVRAS-CHAVE Comportamento alimentar infantil: comparação entre crianças sem e com excesso de
Obesidade; peso em uma escola do município de Pelotas, RS
Avaliação nutricional;
Comportamento Resumo
alimentar; Objetivo: Avaliar diferenças no comportamento alimentar infantil em função do estado
Criança nutricional, sexo e idade.
Métodos: O estudo incluiu crianças na faixa etária de seis a dez anos, de ambos os sexos,
de uma escola privada em Pelotas (RS), no ano de 2012. Para avaliar o comportamen-
to alimentar utilizaram-se as subescalas do questionário “Children’s Eating Behaviour
Questionnaire” (CEBQ): Resposta à Comida (FR), Prazer em Comer (EF), Desejo de Beber
(DD), Sobreingestão Emocional (EOE), Subingestão Emocional (EUE), Resposta à Sacieda-
de (SR), Seletividade (FF) e Ingestão Lenta (SE). Avaliou-se o estado nutricional por meio
do escore-z do IMC/idade.
Resultados: Participaram 335 crianças com idade de 87,9±10,4 meses. Apresentaram eu-
trofia 49,3% (n=163), sobrepeso 26% (n=86), obesidade 15% (n=50) e obesidade grave
9,7% (n=32). Crianças com excesso de peso tiveram maior pontuação nas subescalas de
“interesse pela comida” (FR, EF, DD, EOE, p<0,001) e menor pontuação nas subescalas de
“desinteresse pela comida” (SR e SE p<0,001 e p=0,003, respectivamente), se compara-
das às crianças com peso adequado. Não foram observadas diferenças no comportamento
alimentar segundo sexo e idade.
Conclusões: Observou-se que comportamentos alimentares que refletem “interesse pela
comida” estão associados positivamente ao excesso de peso, mas não foi encontrada
associação com o sexo e a idade da criança.
© 2014 Sociedade de Pediatria de São Paulo. Publicado por Elsevier Editora Ltda. Todos os
direitos reservados.

Introduction investigating eating behaviors at early ages and suggest

that actions aimed at promoting healthy eating habits
The study of eating behavior plays a central role in the should focus with greater emphasis on children. Given the
prevention and treatment of chronic diseases associated above, the aim of this study was to evaluate differences in
with poor nutrition.1 Among them is obesity, a main nutri- eating behavior according to nutritional status, gender and
tional problem that constitutes a challenge in both devel- age of children aged 6 to 10 at a private school in the city
oped and developing countries.2 In 2008, approximately of Pelotas.
33% of Brazilian children between five and nine years old
were overweight, of which about 14% were already obese.3
Studies have shown differences in several dimensions of Method
eating behavior among children with and without excess
weight.4-17 It is believed that overweight children are more This is a cross-sectional study in a private school in the city
responsive to external stimuli in the environment (e.g., of Pelotas, Rio Grande do Sul, carried out from May to June
flavor and color of food), demonstrate greater pleasure in 2012. All children included in the study were 6 to 10 years
eating and have lower responsiveness to satiety when com- old, enrolled in the 1st, 2nd or 3rd year of elementary school,
pared to children with healthy weight, which causes them whose parents or guardians gave their consent to partici-
to eat larger amounts, and in the absence of hunger, thus pate in the study (total of 359 children). Children without
demonstrating a greater interest in food.4,5,11,13,16 Moreover, anthropometric data were excluded from the sample.
they have the habit of eating in order to deal with different The assessed outcome was eating behavior, assessed
emotional states (happiness, anxiety and stress),4,5,13 they through the subjective perception of parents about their
often drink sugary beverages during the day and eat more children’s behavior by answering the CEBQ questionnaire
quickly.14,15 On the other hand, underweight children seem (Children’s Eating Behaviour Questionnaire),4 translated
to be more selective in relation to food, consuming small and validated for a sample of Portuguese children.5 This
meals, with a limited number of foods and more slowly, questionnaire contains 35 questions divided into eight sub-
thus reflecting a lack of interest in food.4,13 scales, so that four subscales investigate behaviors that
It is known that eating behaviors are formed in the first reflect “interest in food” – Food Response (FR), Enjoyment
years of life,18 and eating habits in adulthood are related to of Food (EF), Desire to Drink (DD) and Emotional Overeating
those learned in childhood.19 Additionally, changes in (EOE) – and the other four subscales reflect behaviors relat-
behavior with advancing age tend to be more difficult to be ed to “lack of interest in food” – Emotional Undereating
achieved.20 These situations demonstrate the importance of (EUE), Satiety Responsiveness (SR), Slowness in Eating (SE)​​
44 Rodrigues dos Passos D et al

and Food Fussiness (FF). Examples of questions contained tests were based on the chi-square test. The significance
in the questionnaire are: “Given the opportunity, my child level was set at 5%.
would spend most of the time eating” (FR); “My child loves The project was approved by the Research Ethics
to eat” (EF); “Given the opportunity, my child would spend Committee of the Faculdade de Medicina of the Universidade
the day continuously drinking (soft drinks or sweetened Federal de Pelotas, protocol number 26/2012. The free and
juices)” (DD); “My child eats more when he/she is anxious” informed consent form was signed by the children’s parents
(EOE); “My child eats less when he/she is tired” (EUE); “My or guardians.
child feels full before finishing the meal” (SR); “My child
eats more and more slowly throughout the meal” (SE), and
“When given new foods my child initially refuses them” Results
(FF).The questionnaires were sent by the teachers for par-
ents or guardians to fill them out, and the answers were A total of 335 children participated in the study, represent-
given using a Likert scale of five points, according to the ing 93.3% of the students enrolled in the grades included in
frequency in which their children presented each behavior, the study (total of denials: 6.7%). It was observed that
with the score ranging from 1 to 5: never (1), rarely (2), 51.3% of the children were females, 94.3% were Caucasians,
sometimes (3), often (4) and always (5). The scores of and most parents had a college/university level of school-
questions that belonged to the same subscale were added ing (75.7% of mothers and 63.7% of fathers of the total
up, so that each subscale had a mean value and standard number of parents/guardians that completed the question-
deviation. In cases of unanswered questions, telephone naire). The mean age of the study population was 87.9
contact was made to obtain the information. months (±10.4 months), and the three age groups were
At school, anthropometric measurements of weight and approximately the same size: <7 years (n=134), 7 to 7.9
height were collected by previously trained and standard- years (n=119) and ≥8 years (n=82).
ized nutritionists or nutrition students. Children were Regarding the children’s nutritional status, the preva-
weighed with light clothes and barefoot, on a digital bio- lence of overweight children was 26%, followed by 15% of
impedance scale (Tanita Corporation of America, Inc., IL, obese and 10% of severely obese children. Thus, it is note-
USA) with a capacity of 150 kg and precision of 100 g. worthy that half of the study population (51%) had some
Height was measured with a portable vertical stadiometer degree of excess weight. Overweight and severe obesity
(Alturexata®, MG, Brazil) with 213 cm in length and preci- were more common in boys than in girls (28% of overweight
sion of 0.1 cm. The nutritional status of children was versus 24%, p=0.002, and 15% of severe obesity versus 5%,
assessed using the z-score of body mass index for age p=0.001, in boys and girls, respectively) (Fig. 1). It was
(BMI/A) and classified according to the cutoffs proposed by observed that 49% of the children had normal weight, and
the World Health Organization (WHO)21 into five categories: this condition was more common in girls than in boys (57%
underweight, normal weight, overweight, obesity and versus 41%, p=0.002). None of the children were classified
severe obesity. The WHO Anthro Plus software (WHO Anthro as thin according to the BMI/age assessment in our sample.
Plus®, World Health Organization, GE, Switzerland) was Table 1 shows the scores on the CEBQ subscales accord-
used to calculate the z-score.21 ing to categories of BMI/age z-score, gender and age group
The age variable was collected continuously in months of the children. It was observed that all subscales showed
and subsequently categorized into the following age groups: significant association with nutritional status, except for
<7 years, between 7 to 7.9 years and 8 years or older. To the subscales “Food Fussiness” and “Emotional Undereating”
evaluate parental educational level, we determined the (p=0.254 and p=0.637, respectively). It was observed that
percentage of parents who had finished college/university. all subscales of “interest in food” had higher scores in the
The collected data were entered in duplicate in EpiData obesity and severe obesity categories (Fig. 2). Conversely,
(EpiData ® Software and Templates, World Health among the subscales that reflected “lack of interest in
Organization, GE, Switzerland), and all analyses were per- food”, two of them – “Satiety Responsiveness” and
formed using Stata® software, version 12.0. The descriptive “Slowness in Eating” – had the highest scores in the catego-
analysis of the data was performed using mean and stan- ry of normal weight children, whereas for the other two –
dard deviation for continuous variables, and proportions for “Food Fussiness” and “Emotional Undereating” – no signifi-
categorical variables. Analysis of variance (ANOVA) was cant trend was observed in the variation of scores accord-
used to compare the mean score obtained in each of the ing to the BMI z-score/age categories (Fig. 2).
CEBQ subscales according to the categories of different In general, the eating behavior was very similar in boys
exposure variables (nutritional status, gender and age). and girls. The only subscale that was different between the
Linear trend test was performed to demonstrate the varia- genders was the “Desire to Drink”, with the mean score
tion in the scores of the CEBQ subscales in different cate- being significantly higher in boys than in girls (2.80±1.11
gories of BMI z-score. Linear regression was performed to versus 2.47±1.07, respectively, p=0.005) (Table 1). Eating
assess the association between BMI z-score/age and sub- behavior was very similar in all age groups, and only the
scale means, controlling for potential confounders: gender subscale “Slowness in Eating” showed a significant differ-
and age of children and parents’ level of schooling. The BMI ence between age groups, with a decrease in the score of
z-score (categorized) was correlated to the variable gender this subscale with increasing age (Table 1).
with contingency tables. The respective prevalence of chil- A multivariate analysis of the association between each
dren was calculated for each category of BMI z-score and of the subscales and the BMI z-score categories adjusted for
then compared between males and females. Statistical the children’s gender and age and parents’ level of school-
Children’s eating behavior: comparison between normal and overweight children from a school in Pelotas 45

p = 0.002
Male
Female

57%

Assessed children (%)

p = 0.002
41%

p = 0.574 p = 0.001
28%
24%

16% 15%
14%

5%

Normal weight Overweight Obesity Severe obesity

Figure 1  Nutritional status of children aged 6 to 10 years from a private school, according to gender (n = 331). Pelotas (RS),
2012.

4.50
A
4.00
Mean scores in the subscales of CEBQ

3.50

3.00

2.50 Food responsiveness (FR)

2.00 Enjoyment of food (EF)

Emotional overeating (EOE)

1.50
Desire to drink (DD)
1.00

0.50

0.00
Normal weight Overweight Obesity Severe obesity
BMI categories

B 4.50
Mean scores in the subscales of CEBQ

4.00

3.50

3.00
Satiety responsiveness (SR)
2.50
Slowness in eating (SE)
2.00
Food fussiness (FF)
1.50
Emotional undereating (EUE)
1.00

0.50

0.00
Normal weight Overweight Obesity Severe obesity

BMI categories

Figure 2  Mean scores for the subscales of “interest in food” (A) and “lack of interest in food” (B) of the CEBQ, according to
categories of body mass index of the children (n=331). Pelotas, RS, 2012.
Table 1  Mean ± standard deviation of the CEBQa subscales according to categories of body mass index for age (BMI/age),b gender and age of children.
46

Food Enjoyment Emotional Desire Satiety Slowness Food Emotional

Responsiveness of Food Overeating to Drink Responsiveness in Eating Fussiness Undereating
(FR) (EF) (EOE) (DD) (SR) (SE) (FF) (EUE)
BMI/age (n=331)
Normal weight (n=163) 1.87 (0.56) 2.82 (0.78) 1.72 (0.66) 2.48 (1.03) 3.07 (0.60) 2.79 (0.57) 2.98 (0.33) 2.63 (0.92)
Overweight (n=86) 2.10 (0.68) 3.03 (0.75) 1.97 (0.81) 2.52 (1.05) 2.88 (0.52) 2.62 (0.59) 2.98 (0.25) 2.68 (0.82)
Obesity (n=50) 2.80 (0.93) 3.55 (0.81) 2.68 (0.96) 2.76 (1.17) 2.93 (0.42) 2.69 (0.45) 3.06 (0.25) 2.63 (0.82)
Severe obesity (n=32) 3.33 (0.93) 3.94 (0.77) 2.74 (1.02) 3.38 (1.17) 2.61 (0.49) 2.43 (0.46) 2.94 (0.29) 2.45 (0.82)
p (ANOVA)c <0.001 <0.001 <0.001 <0.001 <0.001 0.003 0.254 0.637
p (Trend test)d <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.752 0.450
Gender (n=335)
Male (n=163) 2.26 (0.83) 3.10 (0.86) 2.12 (0.87) 2.80 (1.11) 2.97 (0.55) 2.72 (0.57) 2.99 (0.28) 2.70 (0.85)
Female (n=172) 2.17 (0.87) 3.08 (0.87) 1.95 (0.90) 2.47 (1.07) 2.94 (0.58) 2.68 (0.54) 2.99 (0.31) 2.58 (0.90)
p (ANOVA)c 0.377 0.857 0.070 0.005 0.643 0.490 0.894 0.199
Age (n=335)
<7 years (n=134) 2.15 (0.79) 3.04 (0.90) 1.94 (0.84) 2.56 (1.10) 3.00 (0.57) 2.81 (0.57) 2.97 (0.32) 2.68 (0.86)
7–7.9 years (n=119) 2.26 (0.87) 3.11 (0.90) 2.11 (0.90) 2.76 (1.08) 2.96 (0.54) 2.66 (0.56) 3.00 (0.27) 2.61 (0.86)
≥8 years (n=82) 2.25 (0.90) 3.15 (0.74) 2.08 (0.92) 2.58 (1.14) 2.88 (0.58) 2.55 (0.49) 3.00 (0.28) 2.60 (0.90)
p (ANOVA)c 0.529 0.602 0.279 0.319 0.313 0.002 0.717 0.751
a
CEBQ (Children’s Eating Behaviour): questionnaire to assess infant feeding behavior.
b
Classification according to cutoffs recommended by the World Health Organization (2007) for children older than 5 years.
c
p value by ANOVA.
d
p value by the linear trend test.

Table 2  Linear regression analysis for BMI z-scores (reference category: normal weight) in CEBQ subscales (n = 331).

Subscales Crude b Coefficient (SE) pa Adjusted b Coefficientb (SE) pa

Overweight Obesity Severe obesity Overweight Obesity Severe obesity
Food Response (FR) 0.22 (0.09) 0.93 (0.11) 1.46 (0.13) <0.001 0.22 (0.10) 0.93 (0.12) 1.49 (0.14) <0.001
Enjoyment of food (EF) 0.21 (0.10) 0.73 (0.12) 1.11 (0.15) <0.001 0.21 (0.10) 0.76 (0.13) 1.16 (0.15) <0.001
Emotional Overeating (EOE) 0.25 (0.10) 0.96 (0.13) 1.02 (0.15) <0.001 0.21 (0.11) 0.94 (0.13) 1.00 (0.16) <0.001
Desire to drink (DD) 0.04 (0.14) 0.28 (0.17) 0.89 (0.21) <0.001 0.05 (0.14) 0.24 (0.18) 0.75 (0.21) <0.001
Satiety Response (SR) —0.18 (0.07) —0.14 (0.09) —0.46 (0.11) <0.001 —0.19 (0.07) —0.14 (0.09) —0.48 (0.11) 0.002
Slowness in eating (SE) —0.17 (0.07) —0.10 (0.09) —0.36 (0.11) 0.003 —0.15 (0.07) —0.09 (0.09) —0.36 (0.11) <0.001
Food fussiness (FF) —0.0009 (0.04) 0.08 (0.05) —0.04 (0.06) 0.254 —0.004 (0.04) 0.09 (0.05) —0.05 (0.06) 0.480
Emotional Undereating (EUE) 0.05 (0.12) —0.005 (0.14) —0.18 (0.17) 0.637 0.03 (0.12) —0.01 (0.14) —0.21 (0.17) 0.811
SE, standard error.
a
p value at Wald test.
b
Analysis adjusted for gender and age of children and parental level of schooling.
Rodrigues dos Passos D et al
Children’s eating behavior: comparison between normal and overweight children from a school in Pelotas 47

ing was carried out (Table 2). The results obtained in the CEBQ usually show that emotional overeating is positively
crude analysis for each of the subscales did not change associated with BMI, whereas emotional undereating is neg-
after the adjustment, and the association between all atively related to BMI.5,6
CEBQ subscales and the BMI z-score categories was main- The CEBQ subscales that reflect “lack of interest in food”
tained, except for the subscales “Food Fussiness” and (“Satiety Responsiveness”, “Slowness in Eating”, “Emotional
“Emotional Undereating”, as seen in Table 1. Undereating” and “Food Fussiness”) seem to better charac-
terize the eating behavior of underweight children.4-7,13 In
our study, even though there were no underweight chil-
Discussion dren, we could observe a significant difference in the
scores of the subscales “Satiety Responsiveness” and
The findings of the present study suggest that eating behav- “Slowness in Eating” between children with normal weight
ior was strongly associated with the child’s nutritional sta- and children with excess weight, although no difference
tus. Children with excess weight had higher scores at all was found for the subscales “Food Fussiness” and “Emotional
CEBQ subscales that reflect “interest in food”, and lower Undereating”.
scores at the subscales that reflect “lack of interest in Children with excess weight had lower scores at the sub-
food”, when compared to normal weight children. In gen- scale “Satiety Responsiveness” when compared to the normal
eral, there were no differences in eating behavior between weight ones, going for the idea that a decrease in response
boys and girls, or depending on age. to satiety makes children less capable of regulating food
The children with excess weight showed greater response intake, and thus contributes to excess weight gain.5,6,7,13,16,22
to food, pleasure in eating, increased food intake due to Other studies25-27 have indicated that the strategies used by
the emotional state and a greater desire for beverages, and parents to make their children eat a meal or try new foods
on the other hand, weaker response to satiety and a pat- can hinder the learning of appetite regulation capacity. In
tern of faster food intake when compared to normal weight this study, overweight children had lower scores at the sub-
children. Similar results were previously found in stud- scale “Slowness in Eating”, demonstrating a faster eating
ies4-7,13 that used CEBQ to compare the eating behavior in pattern. An experimental study14 evaluated the eating behav-
samples of English, Portuguese and Dutch children and ado- ior of 80 children aged between 8 and 12 years during a test
lescents. meal carried out in a laboratory, in the presence of the moth-
It was observed that children with higher BMI/older age ers, and demonstrated that overweight children ate faster
had higher scores at the subscales “Pleasure Eating” and and with greater bite size when compared to normal weight
“Response to Food”, in agreement with studies that show children. Another similar experimental study by Berkowitz et
that children who are overweight have increased interest in al15 found that the behavior of fast food intake, character-
food and a more pronounced response capacity to the influ- ized by a greater number of bites per minute during the
ence of external food attributes such as taste, color and meal, was crucial to excessive weight gain and changes in
smell.4,6,7,13,22 These two dimensions of eating behavior are BMI from 4 to 6 years of age, while factors such as calorie
also investigated in the Dutch Eating Behavior Questionnaire consumption rate and/or warnings received by parents during
for Children (DEBQ-C),10 in the subscale “External Eating”, the meal did not affect weight gain.
which indicated that excess weight children scored higher In general, boys and girls showed a very similar eating
for “external eating” than those with normal weight behavior in our study, with a difference in mean points only
(p=0.02), similar to the findings of the present study. for the subscale “Desire to Drink”, in which boys had high-
It was verified that children with higher BMI/older age er scores. This is in agreement with the Portuguese study,7
had higher scores at the subscale “Desire to Drink”, which which evaluated 249 young individuals aged between 3 and
reflects the desire of children to carry with them beverag- 13 years. This behavior has been pointed out among the
es with low nutritional value and high energy density (soft probable causes for the increase in childhood obesity.23,24 In
drinks and sweetened juices). Studies23,24 have demonstrat- this study, boys simultaneously had increased interest in
ed a positive association between consumption of sugary sugary drinks and higher prevalence of overweight and
beverages and BMI, suggesting that a decrease in soda con- severe obesity. Studies have shown no consensus regarding
sumption could result in a reduction in the number of over- the variation in eating behavior depending on the child’s
weight children. gender.4-6 In the CEBQ validation study,[4] the authors found
Children with excess weight had higher scores at the sub- no differences in behavior between the genders either, and
scale “Emotional Overeating” when compared to the ones they state that it is probably during adolescence that major
with normal weight, but for the subscale “Emotional differences between boys and girls can be observed, as it is
Undereating” no significant difference was found between during this phase that girls begin to worry about body
the groups, similar to the findings of the study by Webber self-image, which leads to food restriction attitudes and
et al.13 Our results add to the discussion about the associa- increased esthetic awareness.28,29
tion between emotional eating and nutritional status. The present study showed no differences regarding the
Tanofsky-Kraff et al12 developed a specific scale to assess eating behavior according to the children’s age range,
the behavior of emotional eating (Emotional Eating Scale except for the subscale “Slowness in Eating”, in which
adapted for children and adolescents – EES-C), and when scores decreased with increasing age, suggesting that older
studying a sample of young individuals with and without children tend to eat faster. A study5 that evaluated young
excess weight they found no association between emotion- individuals in a broader age range (3-13 years) was able to
al eating and body weight. On the other hand, studies using observe significant differences in all CEBQ subscales
48 Rodrigues dos Passos D et al

according to age, with the exception of the subscales 2. Popkin BM, Adair LS, Ng SW. Global nutrition transition and the
“Emotional Undereating” and “Desire to Drink”. Some stud- pandemic of obesity in developing countries. Nutr Rev.
ies4,5,30 have argued that as children grow up and have the 2012;70:3-21.
3. Brasil – Instituto Brasileiro de Geografia e Estatísticas
autonomy to choose what they want to eat and how much,
[homepage on the Internet]. Pesquisa de orçamentos familiares
their eating behavior tends to undergo changes, among
2008-2009 [accessed 14 February 2013]. Available from: http://
them an increase in the velocity of ingestion. www.ibge.gov.br.
The relevance of the present study is due to the fact that 4. Wardle J, Guthrie CA, Sanderson S, Rapoport L. Development
it was one of the first studies in our country to investigate of the children’s eating behaviour questionnaire. J Child
the psychobiological aspects of child eating behavior through Psychol Psychiatry. 2001;42:963-70.
an internationally validated questionnaire (CEBQ). Different 5. Viana V, Sinde S. O comportamento alimentar em crianças:
from other studies that sent the questionnaires to be filled estudo de validação de um questionário numa amostra
out by the parents, our study achieved a high response rate portuguesa (CEBQ). Ana Psicologica. 2008;1:111-20.
(93%). However, this study has some limitations that should 6. Sleddens EF, Kremers SP, Thijs C. The children’s eating
behaviour questionnaire: factorial validity and association with
be considered. The main limitation concerns the conve-
body mass index in Dutch children aged 6-7. Int J Behav Nutr
nience sample used, which was restricted to children attend-
Phys Act. 2008;5:49.
ing a single private school in the municipality of Pelotas. 7. Viana V, Sinde S, Saxton JC. Children’s eating behaviour
Underweight children and lower income groups are not rep- questionnaire: associations with BMI in Portuguese children. Br
resented in this study, and thus caution is required when J Nut. 2008;100:445-50.
generalizing the data found here to other populations. It is 8. Schacht M, Richter-Appelt H, Schulte-Markwort M, Hebebrand
worth mentioning that the children’s behavior was assessed J, Schimmelmann BG. Eating pattern inventory for children: a
by the parents’ subjective perception, which was used as a new self-rating questionnaire for preadolescents. J Clin
“proxy” of the eating behavior, through a questionnaire val- Psychol. 2006;62:1259-73.
idated in a sample of Portuguese children, not Brazilian 9. Braet C, Claus L, Goossens L, Moens E, van Vlierberghe L,
Soetens B. Differences in eating style between overweight and
ones. Furthermore, the cross-sectional design carries a lim-
normal-weight youngsters. J Health Psychol. 2008;13:733-43.
itation regarding the causal inference, as it is not possible to
10. Baños RM, Cebolla A, Etchemendy E, Felipe S, Rasal P, Botella
verify whether the assessed eating behaviors were determi- C. Validation of the Dutch eating behavior questionnaire for
nants or consequence of excess weight. children (DEBQ-C) for use with Spanish children. Nutr Hosp.
Studies currently investigating factors associated with 2011;26:890-8.
childhood excess weight still focus on the biological deter- 11. Tanofsky-Kraff M, Ranzenhofer LM, Yanovski SZ, Schvey NA,
minants and lifestyle, without considering aspects of eating Faith M, Gustafson J, et al. Psychometric properties of a new
behavior that may be involved in this process. Our results questionnaire to assess eating in the absence of hunger in
indicate the existence of several behavioral differences children and adolescents. Appetite. 2008;51:148-55.
between children with and without overweight, although 12. Tanofsky-Kraff M, Theim KR, Yanovski SZ, Bassett AM, Burns NP,
Ranzenhofer LM, et al. Validation of the emotional eating scale
no differences were found in eating behavior between gen-
adapted for use in children and adolescents (EES-C). Int J Eat
ders or depending on the age. It is worth mentioning that
Disord. 2007;40:232-40.
longitudinal studies are necessary to strengthen the evi- 13. Webber L, Hill C, Saxton J, Van Jaarsveld CH, Wardle J. Eating
dence base on the role of eating behaviors in the etiology behaviour and weight in children. Int J Obes (Lond). 2009;33:21-8.
of obesity and to understand the possible behavioral differ- 14. Laessle RG, Uhl H, Lindel B, Müller A. Parental influences on
ences according to the child’s gender and age range. The laboratory eating behavior in obese and non-obese children. Int
study findings may help the development of effective nutri- J Obes Relat Metab Disord. 2001;25 Suppl 1:S60-2.
tional interventions to promote healthy eating behaviors 15. Berkowitz RI, Moore RH, Faith MS, Stallings VA, Kral TV,
among children and reduce childhood excess weight in our Stunkard AJ. Identification of an obese eating style in 4-year-
country. old children born at high and low risk for obesity. Obesity
(Silver Spring). 2010;18:505-12.
16. Carnell S, Wardle J. Appetite and adiposity in children:
evidence for a behavioral susceptibility theory of obesity. Am J
Funding Clin Nutr. 2008;88:22-9.
17. Lazarou C, Kalavana T, Matalas AL. The influence of parents’
Master’s Degree grant from Conselho Nacional de dietary beliefs and behaviours on children’s dietary beliefs and
Desenvolvimento Científico e Tecnológico (CNPq), Brazil behaviours. The CYKIDS study. Appetite. 2008;51:690-6.
process 558620/2010-8. 18. Quaioti TC, Almeida SS. Determinantes psicobiológicos do
comportamento alimentar: uma ênfase em fatores ambientais
que contribuem para a obesidade. Psicol USP. 2006;17:193-211.
Conflicts of interest 19. Mikkilä V, Räsänen L, Raitakari OT, Pietinen P, Viikari J.
Longitudinal changes in diet from childhood into adulthood with
respect to risk of cardiovascular diseases: the cardiovascular risk
The authors declare no conflicts of interest.
in young finns study. Eur J Clin Nutr. 2004;58:1038-45.
20. Zambon MP, RGM Antonio MA, Mendes RT, Barros Filho AA.
Obese children and adolescents: two years of interdisciplinary
References follow-up. Rev Paul Pediatr. 2008;26:130-5.
21. World Health Organization [homepage on the Internet]. Growth
1. Rossi A, Moreira EA, Rauen MS. Determinants of eating behavior: reference 5-19 years [accessed 15 February 2013]. Available
a review focusing on the family. Rev Nutr. 2008;21:739-48. from: http://www.who.int/growthref/en/.
Children’s eating behavior: comparison between normal and overweight children from a school in Pelotas 49

22. Jansen PW, Roza SJ, Jaddoe VW, Mackenbach JD, Raat H, 26. Joyce JL, Zimmer-Gembeck MJ. Parent feeding restriction and
Hofman A, et al. Children’s eating behavior, feeding practices child weight. The mediating role of child disinhibited eating
of parents and weight problems in early childhood: results from and the moderating role of the parenting context. Appetite.
the population-based Generation R Study. Int J Behav Nutr Phys 2009;52:726-34.
Act. 2012;9:130. 27. Rodgers RF, Paxton SJ, Massey R, Campbell KJ, Wertheim EH,
23. Collison KS, Zaidi MZ, Subhani SN, Al-Rubeaan K, Shoukri M, Al- Skouteris H, et al. Maternal feeding practices predict weight
Mohanna FA. Sugar-sweetened carbonated beverage gain and obesogenic eating behaviors in young children: a
consumption correlates with BMI, waist circumference, and prospective study. Int J Behav Nutr Phys Act. 2013;10:24.
poor dietary choices in school children. BMC Public Health. 28. Finato S, Rech RR, Migon P, Gavineski IC, de Toni V, Halpern R.
2010;10:234. Body image insatisfaction in students from the sixth grade of
24. Pérez-Morales E, Bacardí-Gascón M, Jiménez-Cruz A. Sugar- public schools in Caxias do Sul, Southern Brazil. Rev Paul
sweetened beverage intake before 6 years of age and weight or Pediatr. 2013;31:65-70.
BMI status among older children; systematic review of 29. Leme AC, Philippi ST. Teasing and weight-control behaviors in
prospective studies. Nutr Hosp. 2013;28:47-51. adolescent girls. Rev Paul Pediatr. 2013;31:431-6.
25. Blissett J, Haycraft E, Farrow C. Inducing preschool children’s 30. Salvy SJ, Elmo A, Nitecki LA, Kluczynski MA, Roemmich JN.
emotional eating: relations with parental feeding practices. Influence of parents and friends on children’s and adolescents’s
Am J Clin Nutr. 2010;92:359-65. food intake and food selection. Am J Clin Nutr. 2011;93:87-92.