INTRODUCTION

Breakfast is considered one of the main meals consumed throughout the day and deserves special attention^1,2, especially in adolescence when eating behavior and lifestyle habits that may persist throughout adulthood are being formed^3,4.

Regular breakfast consumption is associated with a higher daily intake of micronutrients, such as water-soluble vitamins (thiamine, riboflavin, niacin, pyridoxine, folate, cobalamin, and pantothenic acid); fat-soluble vitamins (A and D); minerals (calcium, iron, sodium, zinc, iodine, magnesium, potassium, manganese, and copper)⁵; a better nutritional profile⁶; and a higher cognitive performance⁷ among children and adolescents.

However, studies have shown a growing tendency to skip breakfast, particularly among adolescents^2,6,8,9,10. The omission of this meal has been associated with a nutritionally unbalanced diet and unhealthy eating patterns⁶, changes in metabolism and hormone secretion due to the long period of fasting^11,12,13 and reduced postprandial energy expenditure¹³, which can lead to health problems. Consequently, adolescents are more susceptible to weight gain and its consequences as well as having an impact on this condition in adulthood^14,15.

The prevalence of overweight among adolescents has increased and has become one of the main public health concerns, since this condition impairs health and quality of life at this stage, and later, predisposes individuals to chronic noncommunicable diseases and early mortality^6,16.

A significant association has been observed between skipping breakfast and being overweight in adolescents, internationally^6,16,17,18, as well as in some recent studies on Brazilian adolescents^19,20,21. However, some studies did not observe a significant association^22,23,24.

Thus, studies that explore this topic among adolescents are necessary, since the habits acquired at this stage can persist into adulthood and cause health problems. This study aimed to evaluate the association between skipping breakfast and being overweight among Brazilian school-going adolescents using nationally representative data, contributing to the few studies in this country on this topic.

METHODS

This was a cross-sectional study using data from the National School Health Survey (Pesquisa Nacional de Saúde do Escolar, PeNSE), conducted from April to September 2015, with Brazilian students from public and private schools. Although this edition had two independent samples, only the data from sample 2 were used owing to the availability of anthropometric data (weight and height), which is necessary for the body mass index (BMI) classification. The sample comprised adolescents aged 11 to 19 years who were from the 6th to 9th grade of elementary school and 1st to 3rd grade of high school of selected schools in 179 municipalities throughout Brazil and their macroregions. A total of 16,608 questionnaires were collected. However, questionnaires were considered valid if adolescents reported their sex and age (n=16,556). More information about the sample design and the PeNSE data collection process is available in other publications^25,26.

Information on the omission of breakfast was obtained with the question, “Do you usually have breakfast?” considered by the frequency of less than 5 days in the week prior to the interview²⁷. The dependent variable was overweight (overweight and obesity), classified by BMI and obtained by measuring weight and height according to sex and age²⁸, with values greater than 1 z-score of the reference distribution.

The covariates were sociodemographic and economic characteristics related to school, lifestyle, and body image. The sociodemographic and economic characteristics considered were sex (male or female), age group (11-14 years and 15-19 years), socioeconomic status, skin color (white, black, yellow, brown, or indigenous), maternal schooling (no education, incomplete primary education, complete primary education/incomplete secondary education, complete secondary education, or higher education), paid work, and lives with parents (yes; only with mother; only with father or neither).

The socioeconomic strata of the students were estimated using the goods and services score (EBS) according to Levy et al.²⁹ and categorized into tertiles of the sample distribution (1st tertile: low; 2nd tertile: medium; or 3rd tertile: high). This is composed of items such as possession of television, refrigerator, stove, microwave, dishwasher, washing machine, landline or cell phones, DVD player, computer, car, bathroom inside the home, and using the services of a housekeeper 5 or more days a week. A weight was assigned to each item, representing the inverse of the amount of possession as well as the presence of the good in the total sample studied. The distribution of the score was determined by summing the respective weights of the goods evaluated in tertiles according to the structural information of the sample²⁹.

Information related to the school included the presence of a school cafeteria, administrative dependency (public or private), and school schedule (morning, afternoon, full-time, or evening). For the variables on adolescents’ lifestyle and body image, the following were considered: level of physical activity, consumption of alcoholic beverages, smoking, consumption of industrialized foods, sedentary behavior, school meals, frequency of having meals with parents, and self-perception of body image.

The level of physical activity was estimated based on questions about commuting to school, leisure activities, and participating in the physical education classes at school, and the duration and frequency of these activities were considered. Adolescents were classified into three categories based on the frequency of physical activity per week: active (≥300 min), insufficiently active (≤299 min), and inactive (those who did not perform physical activity)³⁰.

Smoking and alcohol consumption habits were considered regular if practiced more than 1 day during the previous 30 days²⁵. Sedentary behavior was classified according to the time adolescents watched television, used the computer, or practiced other sitting activities, except on weekends, holidays, and time spent sitting at school; accordingly, more than 4 hours a day in front of these screens was considered sedentary behavior³¹.

Eating meals with parents and consuming industrialized foods (e.g. hamburger, ham, bologna, salami, sausage, instant noodles, packaged snacks, and crackers) were considered regular when the frequency was greater than 5 days³², these types of food are considered industrial formulations rich in sugars, fats and sodium and low in micronutrients, bioactive compounds and fibers¹; school meals were considered regular if taken for more than 3 days³³ during the previous week.

The self-perception of body image was evaluated based on the question, “As for your body, do you consider yourself:” with the following response options: very thin, thin, normal, fat or very fat and categorizing students as thin/very thin; normal; or fat/very fat³⁴.

The prevalence and their respective 95% confidence interval values (95% CI) were estimated for all variables using the chi-square test with second-order Rao-Scott correction. Poisson regression was performed to analyze the association between skipping breakfast and being overweight with adjustment for covariates.

Multiple models were built using the backward elimination method with all variables that presented a p value <0.20 in the bivariate analysis, with only the variables considered significant (p <0.05) were retained in the multiple models. Since the models were similar after stratifying for sex, the model was retained for the general population of the study.

Statistical analyses were performed using the Statistical Software for Professionals (Stata), version 16.0, and the survey module (svy) was used to consider the sample weight and research design. A significance level of 5% was considered statistically significant.

PeNSE was approved by the National Research Ethics Commission of the National Health Council through Conep Opinion. 1,006,467, of 03/30/2015, and the adolescents who provided informed consent participated in the study.

The data underlying the results presented in the study are publicly available from the IBGE repository (https://downloads.ibge.gov.br/downloads_estatisticas.htm).

RESULTS

In this study, 16,556 adolescents aged between 11 and 19 years were evaluated (mean age 14.1 years [95% CI:14.6; 14.2]); 50.8% were male, 46.1% were aged 11 to 14 years, 86.6% were from public schools, and 30.2% regularly consumed industrialized foods.

The prevalence of skipping breakfast and being overweight were 33.8% (95% CI:32.1; 35.4) and 25.2% (95% CI:24.2; 26.3), respectively. As for the sociodemographic and economic characteristics, the prevalence of being overweight was the highest among the following: youngest group (11–14 years) regardless of sex; boys who were in the 1st and 2nd tertiles of the socioeconomic stratum when compared to the 3rd tertile; those enrolled in private schools regardless of sex; boys who studied in the morning versus the evening shift; girls who studied in the morning versus the full shift; white-skinned boys compared to those with black skin color; boys whose mothers had completed higher education in contrast to adolescents whose mothers had incomplete elementary education and lacked education; and girls who did not live with their fathers in relation to the other categories (Table 1).

For the lifestyle characteristics, the prevalence of overweight was higher among school children who skipped breakfast (30.8%; 95% CI:29.0; 32.6) than among those who ate breakfast (22.4%; 95% CI:21.2; 23.7), regardless of sex. The prevalence of being overweight was higher among the following: girls who practiced physical activity than among those who were insufficiently active; girls who irregularly consumed industrialized foods, school meals, and meals with their parents; boys who studied in schools with canteens; and students who reported feeling fat/very fat regarding body concern, regardless of sex (Table 2).

After controlling for confounding variables (sex, age group, socioeconomic status, administrative dependence of school, consumption of industrialized foods, and body image), skipping breakfast remained significantly associated with overweight, with 2% more chance of being overweight among skippers (OR:1.02 [95% CI:1.01; 1.04]) (Table 3).

DISCUSSION

In a 2015 PeNSE study, with nationally representative data from Brazilian school adolescents aged between 11 and 19 years, the prevalence of skipping breakfast was high and was significantly associated with the prevalence of overweight, even after adjusting for potential confounding factors. The results of this study corroborate those found in other studies in which significant associations were observed between skipping breakfast and being overweight among adolescents^6,10,17,18. Monzani et al.6, in their systematic review of studies carried out between 2008 and 2018, observed that 94.7% of the studies evaluated in 33 different countries noted an association between skipping breakfast and being overweight during adolescence.

Moreover, in a longitudinal study of Brazilian adolescents, Hassan et al.²⁰ found that at baseline, the prevalence of overweight was higher among girls who skipped breakfast (i.e., a consumption frequency of less than 5 days a week) than among those who did not skip a meal. When evaluating the change in meal skipping time from baseline to the third year of follow-up, the authors noted a 40% increase in the odds of skipping a meal among overweight boys. In their longitudinal study of 10 years of follow-up among Japanese children and adolescents, Yaguchi-Tanaka et al.¹⁸ also found a significant association between male adolescents who skipped breakfast at 2.5 years of age and being overweight at 10 and 13 years of age, as well as among girls at 10 years of age.

Monzani et al.⁶ revealed that skipping a meal could be considered a potential marker of lifestyle behaviors not yet elucidated in the literature, which can favor the emergence of overweight and metabolic diseases among adolescents of both sexes. De Souza et al.²⁰ demonstrated significant associations of skipping breakfast and being overweight with other measures of body adiposity. In our study, there was no difference in the association between skipping breakfast and overweight in the models when stratified by sex, thus, we presented results for general population. Monzani et al.⁶, and Souza et al.³⁵ similarly identified few studies that reported associations only for girls.

However, some studies have not found a significant association between skipping breakfast and being overweight^22,23,24. Dialektakou et al.³⁶ observed that depending on the methodology adopted, such as analyzing BMI or overweight/obesity, controlling or not controlling for confounding factors, and the different definitions of breakfast skipping made comparisons between studies difficult. In the present study, the adjustment variables were selected based on reported literatures of potential confounding factors in the association between skipping breakfast and being overweight among adolescents. Thus, we considered sociodemographic, economic, lifestyle, and body self-image as the variables to be used^6,16,17.

Among the possible explanations for the association between skipping breakfast and being overweight among adolescents is the growing practice of fasting for long periods or even restricting food consumption to control or reduce weight, which, in turn, results in skipping certain meals, especially breakfast¹⁷. In addition, appetite regulation mechanisms may explain this association since skipping breakfast can lower postprandial energy expenditure and contribute to changes in lipid and glucose metabolism¹¹. Skipping breakfast prolongs the time on an empty stomach, hence increasing the secretion of ghrelin, and consequently, causing increased appetite and hyperphagia throughout the day. This eventually can lead to weight gain and accumulation of body fat^12,13.

In addition, diet quality has also been highlighted in the literature. Adolescents who skip breakfast have a less healthy eating pattern than those who eat meals regularly, with higher consumption of snacks, soft drinks, packaged fruit juices, and fast food, and lower consumption of fruits, vegetables, and milk^27,37, which may contribute to weight gain. The first Brazilian study to address breakfast eating patterns using a nationally representative data on adolescents was recently published by Hassan et al.³⁸. There were three patterns and their study revealed greater adherence to the pattern consisting of cereal, protein, fruit drinks, and the north/northeast pattern. In the present study, the association between skipping breakfast and being overweight remained significant even after considering the frequency of consuming industrialized foods as part of the adjustment variables of the model.

This study has several strengths and limitations. A strong point is that the data came from a school-based probability sample with a national representation. In addition, the sample in question provided data that assessed the weight status as anthropometric data. In the last edition of PeNSE (held in 2019), anthropometric data were not evaluated and hence made the investigation unfeasible. In this study, we adopted the most complete model to observe the independent association between skipping breakfast and being overweight in adolescents according to the literature.

Among the limitations, we highlighted the lack of the information and, consequently, adjustment for pubertal development, given the influence on body composition and its relationship with obesity, suggesting a bidirectional relationship³⁹. Furthermore, Souza et al.¹⁹ analyzing nationally representative data from Brazilian adolescents participating in the ERICA study, observed that, even after adjusting for the stage of sexual maturation or regardless of the stage of sexual maturation, skipping breakfast remained associated with excess body weight, assessed from the BMI, as well as having a high waist circumference and a higher waist-to-height ratio.

Moreover, the classification of breakfast omission was performed through the frequency response of the meal in the previous 7 days without further information about the composition of the meal. Furthermore, the lack of standardization of the definition of breakfast skipping and its classification in the literature made it difficult to make comparisons with other studies among adolescents.

Therefore, skipping breakfast is associated with being overweight among Brazilian school adolescents, even after controlling for several confounding variables. This finding is relevant since the data were analyzed through a model that considered numerous adjustment variables, and despite this, the association remained significant. However, there is a need for more studies to investigate this association, and standardize the definition of breakfast skipping, classification of breakfast skipping, and the cutoff points for classifying overweight adolescents.

Since these habits acquired in adolescence can persist throughout life, it is necessary to adopt preventive measures aimed at reducing the prevalence of skipping breakfast. This is a possible strategy to reduce the prevalence of overweight among Brazilian adolescents.