Influence of the consumatory behavior onto the risk of obesity and its complications in primary school children

Full Text

Introduction

As per the World Health Organization (WHO), the proportion of overweight and obese individuals has attained epidemic levels in the European region by the end of 2022. One in three school-age children is overweight, with 7.9 % of children under the age of five being overweight (29 % of males and 27 % of girls)¹.

The issue of childhood obesity is also pertinent in the territory of the Russian Federation. Thus, in 2021, overweight and obesity among children and adolescents from 6 to 17 years of age are registered in 18.2 and 6.9 % of cases in boys and in 15.9 and 5.3 % of cases in girls [1]. In the Tyumen Region, the incidence of overweight among boys and girls aged 8–11 years was 19.8 and 18.6 %, and obesity was 20.2 and 11.7 %, respectively [2].

The annual increase in childhood obesity has demonstrated that conservative therapy, which involves the correction of nutrition and physical activity, exhibits decreased effectiveness. Therefore, there is a greater necessity to investigate additional modifiable risk factors for obesity development and devise supplementary prevention strategies.

Childhood obesity is regarded as a chronic, non-communicable disease with a prominent social impact [3]. It is currently recognized that cardiovascular diseases (CVD) are the consequences of the gradual progression of atherosclerosis, which commences at an early age and culminates in clinical manifestations decades later [4; 5]. Thus, studies conducted globally have documented a substantial correlation between childhood obesity and the formation of various CVD risk factors in adulthood, including elevated serum lipids, elevated body mass index, and blood pressure [6; 7].

Globally, chronic diseases are the primary cause of mortality and disability, and they are influenced by social and behavioral risk factors, including eating habits, as per the American Heart Association and WHO²[8]. By directly influencing food preferences (as modifiable risk factors), an additional method of effective obesity control can be implemented.

Recently, there has been a growing emphasis on the evaluation of eating habits among children to develop additional methods for preventing and correcting childhood obesity. The Child Eating Behavior Questionnaire (CEBQ) is a tool that is specifically designed to evaluate a variety of nutritional characteristics in preschool and primary school age. It is employed to analyze eating behavior [9]. Certain studies, based on the CEBQ questionnaire, reveal that modern food products consist of high-calorie, flavorful, inexpensive, and nutrient-poor foods, which elevate the risk of future CVD development based on the food approach type eating behavior among children [10–12]. However, the currently available studies focus on a restricted metabolic profile and do not consider the patient’s sex.

Thus, the identification of children at an increased risk of obesity and the clinical and metabolic factors that contribute to CVD development based on preclinical markers of cardiovascular risk and CEBQ questionnaire results will facilitate the development of preventive measures that could mitigate disability and premature mortality of the able-bodied population from obesity-related disorders.

This study aimed to ascertain the relationship between eating behavior and the clinical and metabolic status of children aged 6–11 to devise effective approaches for preventing and treating obesity and its consequences.

Materials and methods

This cross-sectional, multi-sample study included 94 children aged 6–11 years. In accordance with the inclusion and exclusion criteria, two groups of participants were formed: Group I (n = 64), consisting of children with an established diagnosis of “exogenous-constitutional obesity” [based on the WHO classification (2007) and the classification of obesity in children by etiology (V.A. Peterkova et al., 2014)] and their parents (n = 115), and group II (n = 30), comprising the control group, which included children with normal body weight and their parents (n = 50) [13; 14]. According to WHO criteria (2007), overweight was diagnosed with SDS BMI ranging from +1 to +2; grade I obesity, with SDS BMI from +2 to +2.5; grade II obesity, SDS BMI 2.6–3.0; grade III obesity, SDS BMI 3.1–3.9; and SDS BMI ≥ 4.0 was regarded as morbid obesity. The normal body weight was defined when SDS BMI ranged between –1 and +1. The study was performed by a pediatric endocrinologist during 2020–2022 at the Federal State Budgetary Educational Institution of Higher Education “Tyumen State Medical University” of the Ministry of Health of Russia (Rector I.M. Petrov) and children’s polyclinic of the State Budgetary Healthcare Institution of the Tyumen Region, “Regional Clinical Hospital No. 2,” Tyumen (chief physician R.V. Paskov).

The inclusion criteria also included that written informed consent for medical intervention be provided by a legal representative. The exclusion criteria were: history of receiving medications that could affect body weight and metabolic parameters (e.g., anticonvulsants, antidepressants, glucocorticoids, insulin, antipsychotics); delayed physical development (growth SDS £ –2 for a given age and sex based on the WHO reference tables); established diagnosis of neurological and mental disorders; syndromic, monogenic, hypothalamic, and neuroendocrine forms of obesity.

The study included measurements of waist circumference (WC), height, weight, BMI, and SDS BMI. Additionally, the assessment of blood pressure based on the average value of three readings was conducted in accordance with the WHO classification of arterial hypertension in children (WHO, 1962, 1999).

The laboratory studies included quantitative determination of cholesterol (CH), low and high density lipoproteins (LDL and HDL), triglycerides (TG) in serum by spectrophotometric analysis, uric acid (UA), serum thyroid-stimulating hormone (TSH), and cortisol levels using enzyme immunoassay, as well as the evaluation of carbohydrate metabolism employing a standard oral glucose tolerance test.

The children’s eating behavior was evaluated using a validated CEBQ questionnaire. The questionnaire was completed by the parents of the children and included 35 questions. Each of which was assessed on a 5-point scale, ranging from “never” to “always.” Each question corresponded to one of eight scales that were categorized into two groups of eating behavior: “food approach” and “food avoidance” (Table 1).

 

Table 1

Child eating behavior questionnaire scales (CEBQ, J. Wardle, 2001)

The “food approach” scale	The “food avoidance” scale
Emotional over-eating, English emotional over-eating	Emotional under-eating, English emotional under-eating
Enjoyment of food, English enjoyment of food	Satiety responsibility, English satiety responsibility
Food responsiveness, English food responsiveness	Food fussiness English food fussiness
Desire to drink, English desire to drink	Slowness in eating, English slowness in eating

 

The study was approved by the Committee on Ethics of Scientific Research of the Federal State Budgetary Educational Institution of Higher Education “Tyumen State Medical University” of the Ministry of Health of the Russian Federation (minute o. 95 dated 20.12.2020).

Statistical processing was performed using the IBM SPSS Statistics 26.0.0 software. To assess the normality of distribution, the Kolmogorov–Smirnov statistic was calculated for all variation series. Median (Me) and interquartile range (Q₁; Q₃) are used to describe variables in nonparametric data distribution, while mean value (M) and standard deviation (SD) are used in parametric distribution. The bootstrap method was employed to determine the 95 % confidence interval for the Me quantitative indicator. The Mann – Whitney U test was employed to compare the groups in pairings for quantitative feature analysis. Correlation analysis was conducted to assess the associations between quantitative indicators and their verification. The Spearman correlation coefficient was estimated for nonparametric data distributions while the Pearson correlation coefficient was used for parametric data distributions. Results were considered statistically significant if the p-value was < 0.05.

Results and discussion

The mean age of the children studied was eight and was comparable in both groups [7–9] years, p = 0.369. The male-to-female ratio in group I was 2:1 and 1:1 in group II (see Figure).

 

Fig. Structure of the studied groups by sex

 

The average SDS BMI in the obese children group was 2.8 [2.5–3.8], while in the control group it was –0.6 [–0.9...–0.3], as indicated by the anthropometry data in Table 2 (p < 0.001). The incidence of grade I obesity was 31.3 % (n = 20), grade II obesity was 35.9 % (n = 23), grade III obesity was 23.4 % (n = 15), and morbid obesity was 9.4 % (n = 6). It was consistently determined that the average values of WC and diastolic blood pressure (DBP) in the obese children group exceeded those of the normal-weight children group: DBP was 10 mmHg higher (70 [64–70] mmHg), and WC was 1.5 times higher (86 [82–90] cm), p < 0.001. The average levels of TG and UA [1.1 ± 0.6 mmol/l and 253.9 ± 71.3 mmol/l (p = 0.001)] were discovered to be significantly higher in obese children than in children with normal body weight [0.7 ± 0.3 mmol/l and 217.8 ± 27.4 mmol/l (p = 0.002)] (Table 3). Furthermore, we detected hypertriglyceridemia in 6.3 % (n = 4) and hyperuricemia in 1.6 % (n = 1) of the obese children.

 

Table 2

Indicators of anthropometry and blood pressure in the study groups

Indicator	Group I, n = 64	Group II, n = 30	p-value
Height, cm, Ме [Q1–Q3]	139.0 [131.8–141.8]	130.0 [126.5–135.0]	0.237
Weight, kg, Mе [Q1–Q3]	47.5 [40.4–53.3]	25.7 [23.5–29.0]	0.016*
BMI, kg/m2, Ме [Q1–Q3]	24.3 [22.6–26.1]	15.2 [14.7–16.2]	< 0.001*
SDS BMI, Ме [Q1–Q3]	2.8 [2.5–3.8]	–0.6 [–0.9 – (–0.3)]	< 0.001*
WC, cm, Ме [Q1–Q3]	86 [82–90]	58 [56–60]	< 0.001*
SBP, mmHg, Ме [Q1–Q3]	95 [95–100]	95 [95–100]	0.344
DBP, mmHg, Ме [Q1–Q3]	70 [64–70]	60 [60–65]	< 0.001*

Note: * – differences in indicators are statistically significant (p < 0.05).

 

Table 3

Comparative characteristics of the biochemical and hormonal data of primary school children with obesity and control groups

Indicator	Group I, n = 64		Group II, n = 30		p-value
	M ± SD	95 % СI	M ± SD	95 % СI
CH, mmol/l	4.2 ± 0.7	4.0–4.4	4.4 ± 0.7	4.1–4.6	0.418
TG, mmol/l	1.1 ± 0.6	0.9–1.2	0.7 ± 0.3	0.6–0.8	0.001*
LDL, mmol/l	2.6 ± 0.6	2.4–2.7	2.7 ± 0.7	2.4–3.0	0.382
HDL, mmol/l	1.2 ± 0.3	1.1–1.3	1.3 ± 0.3	1.2–1.4	0.075
UA, mmol/l	253.9 ± 71.3	234.0–273.7	217.8 ± 27.4	207.6–228.0	0.002*
TSH, mIU/l	3.0 ± 1.4	2.6–3.4	2.0 ± 0.8	1.7–2.3	< 0.001*
Cortisol, nmol/l	346.2 ± 93.0	321.5–370.9	333.2 ± 84.3	301.7–364.7	0.524
FG, mmol/l	4.5 ± 0.6	3.0–6.0	4.1 ± 0.5	2.8–5.2	< 0.001*

Note: * – differences in indicators are statistically significant (p < 0.05).

 

When evaluating carbohydrate metabolism, it was demonstrated that fasting glycemia (FG) was statistically significantly higher among group I participants (4.5 ± 0.6 mmol/l), in contrast to those group II participants (4.1 ± 0.5 mmol/l) (p < 0.001). Two patients with obesity (3.6 %) were diagnosed with impaired fasting glycemia. In group II participants, the TG, UA, and FG levels corresponded to the reference values. A comparison of serum TSH levels revealed that the average TSH level was 1.8 times higher in the group I participants (3.0 ± 1.4 mIU/l) than in the control group (2.0 ± 0.8 IU/l) (p < 0.001). Additionally, the TSH value in both groups did not exceed the reference data. The family is the primary environment in which children develop their eating patterns. The child acquires their initial knowledge through communication with their parents, who are the carriers of socio-cultural norms and values. The presence of eating disorders in parents elevates the sensitivity of the children to exogenous factors contributing to obesity, including environmental conditions, socio-economic status, physical activity, and eating behavior. However, recent studies have disproven the impact of heredity on food preferences in children and have emphasized the need for modifying the eating behavior of parents and methods of nutrition education for children [15]. Thus, 62 families of obese children and 26 families of children with normal body weight participated in the study. Two families included two children with obesity; four families included two children with normal body weight. The socio-economic factor of obesity development was excluded by the fact that the family composition, age, and educational level of the parents of the studied groups were comparable (p = 0.322, p = 0.327, p = 0.164, p = 0.218, p = 0.422, respectively). However, the prevalence of overweight and obesity was more prevalent in parents of group I participants – 64.5 % (n = 40) of mothers and 75.5 % (n = 40) of fathers – than in parents of group II participants [26.9 % (n = 7) and 33.3 % (n = 8), respectively, p = 0.004, p = 0.028]. Thus, children with obesity are more susceptible to exogenous factors such as their parents’ eating habits and food education style, which contribute to the development of their own eating behavior and, consequently, obesity. Table 4 presents the results of the CEBQ questionnaire assessment of children’s eating behavior, which reliably established that obese children exhibit a greater interest in food and use it as a source of comfort. Consequently, the average score of the “Food responsiveness” scale was significantly higher in group I than in group II, with values of 2.5 [1.9–3.5] and 1.8 [1.5–2.5], respectively, p = 0.016. Similarly, the "Emotional over-eating" measures yielded scores of 2.3 [1.6–3.0] and 1.0 [1.0–2.3], p = 0.004, respectively. Upon comparing the results of the CEBQ questionnaire by sex, depending on the study group, it was reliably established that females with obesity were more frequently characterized by high indicators of the scales “Food responsiveness,” “Emotional over-eating” and “Enjoyment of food,” as well as a low indicator of the scale "Satiety responsibility," in contrast to females from the control group (p = 0.014, p = 0.012, p = 0.003, and p = 0.007). Males with obesity exhibited high scores for the “Emotional under-eating” scale, p = 0.042.

 

Table 4

Comparative characteristics of the CEBQ questionnaire results among group I and II participants

Scales of eating behavior		Group		p-value
		Group I, n = 64, Me [Q1; Q3]	Group II, n = 30, Me [Q1; Q3]
The food approach scale	Food responsiveness	2.5 [1.9–3.5]	1.8 [1.5–2.5]	0.016*
	Emotional over-eating	2.3 [1.6–3.0]	1.0 [1.0–2.3]	0.004*
	Enjoyment of food	3.5 [2.8–4.3]	2.8 [2.3–4.0]	0.108
	Desire to drink	3.0 [2.0–3.9]	3.0 [2.3–4.3]	0.918
The food avoidance scale	Satiety responsibility	3.0 [2.3–3.4]	3.2 [2.8–3.6]	0.078
	Slowness in eating	2.3 [1.8–3.1]	2.3 [1.8–2.9]	0.807
	Emotional under-eating	2.7 [2.3–3.4]	2.0 [2.0–3.0]	0.160
	Food fussiness	3.2 [2.8–3.7]	2.8 [2.4–3.7]	0.263

Note: * – differences in indicators are statistically significant (p < 0.05).

 

Notably, when comparing laboratory parameters and the CEBQ questionnaire results among the group I participants based on the severity of obesity, no statistically significant results were discernible.

When examining the impact of eating behavior on the clinical and metabolic parameters, statistically significant direct correlations of the results of the “Enjoyment of food” scale with SDS BMI, WC, TSH, and FG levels were established in females (r_xy = 0.566; p = 0.004, r_xy = 0.433; p = 0.030, r_xy = 0.451; p = 0.027, and r_xy = 0.708; p < 0.001). The reverse was observed with the TG levels in boys (r_xy = –0.421; p = 0.016). The SDS BMI index and the level of FG in females were inversely correlated with the "Satiety Responsibility" scale (rxy = –0.592; p = 0.002 and rxy = –0.471; p = 0.020). There were direct moderate correlations between the results of the "Emotional under-eating" scale and the level of UA in males (rxy = 0.395; p = 0.023 and rxy = 0.458; p = 0.008). Table 5 displays the results of the correlation analysis between the clinical and metabolic parameters of primary school students and the results of the CEBQ eating behavior questionnaire.

 

Table 5

The results of correlation analysis between clinical and metabolic parameters of primary school children and the CEBQ results

Clinical and metabolic parameter	Indicator of the CEBQ questionnaire	Group of children
		total		depending on the sex
				females		males
		rxy	p-value	rxy	p-value	rxy	p-value
SDS BMI	Enjoyment of food	0.195	0.146	0.566	0.004*	–0.048	0.792
	Satiety responsibility	–0.201	0.134	–0.592	0.002*	–0.048	0.792
WC, cm	Enjoyment of food	0.168	0.210	0.443	0.030*	0.084	0.641
	Emotional under-eating	0.164	0.222	–0.055	0.799	0.395	0.023*
TG, mmol/l	Enjoyment of food	–0.117	0.389	0.367	0.078	–0.421	0.016*
UA, mmol/l	Emotional under-eating	0.246	0.068	0.180	0.400	0.458	0.008*
TSH, mIU/l	Enjoyment of food	0.082	0.543	0.451	0.027*	–0.194	0.280
FG, mmol/l	Enjoyment of food	0.192	0.153	0.708	< 0.001*	–0.123	0.494
	Satiety responsibility	–0.162	0.228	–0.471	0.020*	–0.044	0.810

Note: * – differences in indicators are statistically significant (p < 0.05).

 

The link we ascertained between the markers of chronic non-communicable diseases, such as the TG, UA, and FG levels, and the nature of eating behavior in primary school children are partially comparable with the results of a similar 2020 Portuguese study. S. Warkentin et al. investigated the impact of eating behavior of children aged 7–10 years based on the CEBQ questionnaire on cardiometabolic risk parameters. They discovered that the scales “Food responsiveness” and “Emotional over-eating” have significant positive relationships with TG level, HOMA index, and waist circumference [12].

The relationship between eating behavior, obesity, and elevated serum TSH levels within the reference values is a controversial issue that has yet to be thoroughly investigated. Nevertheless, a study has verified a comparable observation in overweight children between the ages of three and eighteen. V. Lundbäck et al. (2021) reported that high TSH levels within the reference values are linked to higher SDS BMI in children (p < 0.001) and levels of insulin, total cholesterol, TG, and HOMA index (p = 0.004, p = 0.001, p < 0.001, and p = 0.002) [16].

Hyperuricemia is a risk factor for developing metabolic syndrome and is significantly associated with childhood dyslipidemia and insulin resistance [17; 18]. Nonetheless, the literature does not contain any materials regarding the impact of preschool and elementary school children’s nutritional behavior on the metabolism of purines and thyroid. Our results complement the available scientific knowledge for “childhood obesity – eating habits” and require further corroboration. Thus, the study findings revealed that laboratory markers of atherosclerosis, disorders of carbohydrate metabolism, metabolic syndrome, and thyroid status directly correlate with the nature of eating behavior in children and are determined by sex. Individual modifications to eating patterns can prevent and treat childhood obesity and associated conditions, thereby preventing the development of metabolic disorders in primary school children.

Conclusions

The results obtained enable us to broaden our understanding of the potential causes of metabolic disorders and obesity in children.

The nature of eating behavior in children has been demonstrated to correlate with their metabolic parameters. The necessity of incorporating psychological techniques into the program for the diagnosis of childhood obesity on a par with clinical and laboratory studies is underscored by the fact that eating disorders are the primary cause of the cascade of clinical and metabolic abnormalities. Additionally, the findings of the assessment of eating behavior in childhood enable the enhancement of the existing principles of prevention and conservative therapy of obesity by incorporating targeted correction of identified food preferences into the development of individual psychological programs.

 

¹ WHO European Regional Obesity Report 2022. World Health Organization. Copenhagen 2022; 220.

² Hearts: technical package for cardiovascular disease management in primary health care. World Health Organization. Geneva, Switzerland 2016; 76.

About the authors

A. Yu. Surgutskaya

Tyumen State Medical University

Author for correspondence.
Email: kuchkina94@rambler.ru
ORCID iD: 0000-0002-1973-2434

Assistant of the Department of Childhood Diseases

Russian Federation, Tyumen

S. A. Smetanina

Tyumen State Medical University

Email: kuchkina94@rambler.ru
ORCID iD: 0000-0003-3525-9891

DSc (Medicine), Associate Professor, Head of the Department of Childhood Diseases

Russian Federation, Tyumen

References

Supplementary files