A significant body of evidence indicates that problematic eating behaviours are an important factor related to the body mass index (BMI) in children and adolescents [1
]. The term “problematic eating behaviours” includes many types of eating behaviours that adversely affect somatic and mental health (including leading to an increase in BMI) [4
]. Among problematic eating behaviours, the literature distinguishes among the following [3
]: (a) the tendency to eat when experiencing negative emotions and stress; (b) eating without feeling in control and without taking into account the level of hunger and satiety; (c) high reactivity to external triggers under the influence of which the amount of consumed food increases; and (d) the tendency to restrict food intake.
At this point, it is worth noting that the results of previous studies on the relationship between problematic eating behaviours and BMI are not consistent. On the one hand, in studies with an increase in BMI, the level of restrictive eating increases and the level of emotional and external eating decreases [3
]. On the other hand, research by Baños et al. [1
] shows that the higher BMI is the greater emotional and restrained eating becomes. Moreover, similar inconsistencies occur in studies in which girls and boys as well as older and younger children are compared in terms of the intensity of problematic eating behaviours. In the studies of Nagl et al. [11
], women have a higher level of restrictive and emotional eating compared to men, and younger children eat in a more restrictive, uncontrolled and emotional way than the older group does [12
]. On the basis of other studies [1
], it has been concluded that girls and boys as well as older and younger children do not differ in the context of emotional overeating, food responsiveness and satiety responsiveness. The aforementioned inconsistencies require further research that takes into account the fact that their source may also be (a) using various questionnaires to measure problematic eating behaviours or (b) using different perspectives in the assessment of children’s problematic eating behaviours (perspective of a child vs. perspective of a parent).
There are many questionnaires used in the assessment of children’s eating behaviour. However, very often discussed in the literature are, among others, the Three-Factor Eating Questionnaire (TFEQ) [2
], the Dutch Eating Behaviour Questionnaire for Children (DEBQ-C) [1
] and the Child Eating Behaviour Questionnaire (CEBQ) [13
In summary, in many studies, the relationship between problematic eating behaviours and BMI has been analysed [1
]. However, these studies have not verified the following: (1) which of the currently used questionnaires assessing abnormal eating styles (TFEQ or DEBQ-C or CEBQ) are the most effective at predicting a child’s BMI; and (2) whether the perspective of a child or mother (in the assessment of problematic eating behaviours) is the most effective at predicting a child’s BMI. Therefore, the first aim is to compare the child’s perspective of problematic eating behaviours (two questionnaires: TFEQ vs. DEBQ-C) with the mother’s perspective (one questionnaire: CEBQ). Moreover, because of the inconsistency among the results from other studies (e.g., [1
]), the second aim of the study is to determine whether there are differences in the level of problematic eating behaviours between girls and boys in different age groups (young children: 8–11 years old vs. adolescents: 12–16 years old) using various questionnaires (TFEQ, DEBQ-C, CEBQ).
With regard to the first objective, the present study provides support for the perspective of the child based on the DEBQ-C as being most effective at predicting children’s BMI because this perspective explains 29% of the variance in children’s BMI (while the TFEQ-R13 explained 19% of the variance and the CEBQ explained 16% of the variance). Moreover, the best predictor of BMI is restricted eating from the DEBQ-C, followed by restrictive eating from the TFEQ-R13. This may indicate the particular importance of this style of eating for BMI in children and adolescents. From the children’s perspective, other statistically significant predictors (except for food restrictions) are uncontrolled eating (TFEQ-R13) and external eating (DEBQ-C). From the mothers’ perspective, the important predictors are emotional eating and correct reaction on satiety. Interestingly, emotional eating is not significant from the children’s perspective. In summary, the results related to the first objective of the study indicate that the next stage of this project and future research should take into account the perspectives of both children and mothers in the assessment of problematic eating behaviours (with consideration given to the following: (a) BMI will most accurately predict the DEBQ-C questionnaire, and (b) the parents’ perspective may be helpful in the assessment of the relationship between emotional eating and BMI). Moreover, for all significant predictors (except for the “satiety responsiveness” subscale, which is interpreted inversely), the higher their levels are, the higher the BMI is. The current findings are partly consistent with those of previous research showing that the more restrictive eating habits are, the higher the BMI is [1
]. However, our research did not confirm that the relationship between BMI and external and emotional food is negative [1
In relation to the second aim of this study, it is observed that girls and boys in two age groups (young children: 8–11 years old vs. adolescents: 12–16 years old) differ in the context of problematic eating behaviours. Both the TFEQ-R13 and DEBQ-C show that participants differ in the level of restrictive eating. The most important findings in this context are summarised as follows: (a) the highest level is observed in older girls, followed by younger girls; and (b) younger and older boys do not differ from each other. However, more significant differences among all analysed groups can be found using the TFEQ-R13. This finding shows, therefore, that restrictive eating is more intense among girls than among boys, and its intensity increases significantly with age. Interestingly, our findings confirm earlier results in relation to gender-related comparisons [11
] but contradict the results regarding the comparisons of age groups [12
Only the TFEQ-R13 and CEBQ show significant differences between groups in the level of emotional (over-)eating. Again, both questionnaires coherently indicate that the highest level of this type of problematic eating behaviour is characteristic of older girls. However, only on the TFEQ-R13 is the result obtained by them significantly higher than those obtained by all others (the CEBQ indicates a significant difference between younger and older girls). Our results are confirmed by the study of Nagl et al. [11
], which indicated that women have a higher level of emotional eating. However, our results negate other findings [12
] that indicate that younger children show higher levels of emotional eating.
One of the possible explanations of the results regarding restrictive and emotional eating is the development of emotional awareness and self-control in children and adolescents [27
]. These two characteristics develop with age, which can be connected, on the one hand, with the fact that emotions become triggers to eat. On the other hand, the frequency of introducing eating restrictions also increases (to achieve weight reduction). Such a pattern of problematic eating behaviours may lead to the development of eating disorders, which, as other studies show, are the most disseminated among adolescent girls [29
Finally, it should be mentioned that the highest level of external eating (DEBQ-C) is observed among younger boys and the highest level of a correct reaction to the feeling of fullness is observed among younger girls (the results of both groups differ significantly from those of the other three groups). These results confirm the above assumption (regarding eating disorders) that for girls over time, weight control through restrictive eating begins to have greater significance than physiological signals related to hunger and satiety and external triggers. The lack of significant between-group differences for emotional eating (DEBQ-C), uncontrolled eating (TFEQ-R13) and food responsiveness (CEBQ) is consistent with the findings of some previous research [1
These studies are valuable because there are few studies in Poland regarding the assessment of the eating behaviours of children and adolescents and BMI from a child–mother dyadic perspective (e.g., [31
]). However, a few limitations of the current study should be mentioned, which should be addressed in subsequent studies. First, a more objective and detailed assessment of body weight should be used (e.g., body composition by use of bioelectrical impedance analysis) among children and mothers. Second, the relationship between problematic eating behaviours and BMI should be verified in longitudinal studies. Third, based on other studies [1
], in the future (next to the division into sex and age), body weight should be categorized as either normal or excessive to compare the analyses carried out in these two groups. Fourth, apart from the perspective of mothers, it is necessary to consider the perspective of fathers. Fifth, in future research it may be interesting to speculate on the possibility to collect information on food choices of children. Sixth, it would be interesting to examine how the mothers’ BMI influences the BMI of children and their eating behaviours. All these indications will be included by the authors of this article in the further stages of the project.
In conclusion, this study showed certain findings that are consistent with those of many studies. These findings include a positive relationship between restrictive eating and BMI and a higher level of this type of problematic eating behaviour among girls than among boys. This finding may also indicate that restrictive eating is a characteristic of children and youth regardless of the cultural context. Other discoveries are divergent, which suggests that intercultural research in this context may prove extremely valuable in the investigation of the mechanism underlying these results.
It is well-known that eating behaviours established in childhood persist, with implications such as fussiness and poor dietary variety or high responsiveness to food cues and increased obesity risk [33
]. It is worth pointing out that nutrition policies and obesity prevention in Poland should include the need to educate children and adolescents in the context of eating behaviours, especially in young female adolescents and those with restrictive eating.