INTRODUCTION

According to the World Health Organization (WHO), overweight and obesity (excess weight) are characterized by excessive accumulation of fatty tissue, resulting from an energy imbalance between the greater amount of calories consumed and the smaller amount of calories expended by the body ¹ . Both overweight and obesity are considered global epidemics, which generate real problems for public health in several countries ² .

People are becoming overweight at an increasingly premature age, with adolescence being the phase that has been most marked by this risk and causing multiple health complications. According to data from the Household Budget Survey (POF), conducted by the Brazilian Institute of Geography and Statistics (IBGE), there was a 9.2% increase in the prevalence of overweight between 1974-1975 and 2008-2009. In addition, the obesity rate in this population group also increased by 3.8% ^3,4 . In the Study of Cardiovascular Risks in Adolescents – ERICA, conducted in Brazil between 2013 and 2014, a prevalence of obesity of 25.5% was identified among schoolchildren aged 12 to 17 years ⁵ .

Excess weight in adolescence can trigger orthopedic, oncological, cardiometabolic (dyslipidemia, type 2 diabetes) and psychological problems, even compromising professional development in adulthood. There are multiple factors associated with excess weight at this stage, which can be of biological, behavioral and environmental origin. Examples include sedentary behavior, such as hours spent in front of the television, computer and video games, as well as unhealthy eating habits, physical inactivity, sociodemographic characteristics and even parents being overweight ^6,7 .

In 2015, the third edition of the National School Health Survey (PeNSE) presented a portrait of Brazilian adolescents, addressing several aspects related to the health of this population, including food consumption, physical activity and sedentary behavior ⁸ . In view of this, the present study aimed to evaluate the association between sociodemographic and behavioral factors and excess weight in Brazilian school adolescents investigated in PeNSE 2015.

METHODOLOGY

This is an analytical cross-sectional study, developed based on data from the third edition of PeNSE, carried out in 2015, through an agreement between the Brazilian Institute of Geography and Statistics (IBGE) and the Ministry of Health, with the support of the Ministry of Education ⁸ .

PeNSE 2015 was composed of two independent samples of students, one of students in the 9th grade of Elementary School (Sample 1), already investigated in previous editions of PeNSE (2019 and 2012), and another of students who attended the 6th grade of Elementary School to the 3rd grade of High School (Sample 2) in the daytime and evening shifts in public and private schools, located in urban and rural areas. Sample 2 was composed of students aged 13 to 17 years old, with the purpose of providing data that would allow comparison with international population-based studies such as the School-based Student Healthy Survey (GSHS), which use this age range ⁸ . For the analysis of this article, we used the data belonging to Sample 2, consisting of 10,926 Brazilian students, aged 13 to 17 years old.

Anthropometric data for classification of nutritional status were obtained by IBGE technicians by directly measuring the weight and height of students, following the WHO recommendations for assessing the nutritional profile of adolescents, by calculating the body mass index (weight in kilograms divided by the square of the height in meters): BMI-for-age, expressed as a Z score ⁸ . PeNSE classified adolescents into four categories regarding nutritional status: underweight, normal weight, overweight, and obesity ⁸ . For this study, excess weight was considered the dependent variable based on the recategorization of the nutritional status variable in relation to excess weight: yes (overweight and obesity) and no (underweight and normal weight).

The independent variables were grouped into: sociodemographic – geographic region, school administration, age group, sex, living with the mother and/or father, maternal education, and number of people living with; and behavioral – food consumption in the seven days prior to the survey (vegetables, fruits, soft drinks, sweets, fried snacks, industrialized/ultra-processed foods and fast food ); average time spent by students on physical activities in the seven days prior to the survey, considering the following domains: going to and from school, physical education classes and other extracurricular activities, in minutes; and time spent on a typical weekday sitting, watching television, using the computer, playing video games, talking to friends or doing other activities while sitting. The variable of how many days a week the student practiced at least one hour of physical activity in the seven days prior to the survey was also used to calculate the average.

For analysis purposes, some independent variables were categorized as follows:

Age: 13 to 15 and 16 to 17 years old;
How many people do you live with: ≤ 3 people and > 3 people;
Mother’s education: Did not study, Incomplete and complete elementary school, Incomplete and complete high school, and Incomplete and complete higher education;
Frequency of food consumption, by type of food, in the seven days prior to the survey: ≤ 4 days and > 4 days;
Average total time spent on physical activities: insufficient (up to 299 minutes) active (300 minutes more);
Time spent on a typical weekday sitting: ≤ 3 hours and > 3 hours;

Data were analyzed using IBM SPSS ^r software , version 21.0, using procedures from the Complex Samples Module , suitable ^for analyzing data obtained through a complex sampling plan, where weight, stratum and primary sampling unit were considered. First, the prevalence of overweight was calculated according to sociodemographic and behavioral variables. The average number of days per week in which adolescents consumed various foods and the average number of days in which adolescents practiced physical activity for at least one hour were calculated. Bivariate analysis was then performed using simple logistic regression to obtain unadjusted Odds Ratio (OR) values. For variables that showed a significant association with the outcome (p < 0.20), the adjusted OR ( _adj OR ) was calculated using multivariate logistic regression, with its 95% confidence intervals (95%CI). All analyses were performed with a significance level of 5%. PeNSE 2015 complied with the recommendations of Resolution 466 of December 12, 2012 of the National Health Council and was approved by the National Research Ethics Committee, under opinion no. 1,006,467, of March 30, 2015. All information, both from the student and the school, was confidential and unidentified ⁸ . RESULTS The presence of excess weight (overweight and obesity) was identified in 23.7% (95%CI 22.6-24.9) of schoolchildren aged 13 to 17 years, being significantly higher among schoolchildren from the southern region (28.2%; 95%CI: 26.2 – 30.3), students from private schools (28.4%; 95%CI: 25.8 – 31.2), aged 13 to 15 years (25.1%; 95%CI: 23.7 – 26.6), who lived with their mother and/or father (23.8%; 95%CI: 22.7 – 25.0), who lived with up to three people (26.0%; 95%CI: 23.9 – 28.1) and who had mothers with higher education (incomplete/complete). (28.5%; 95% CI: 26.0 – 31.2). The prevalence of excess weight was similar between sexes (Table 1).

Regarding eating behaviors, excess weight was more prevalent among those who consumed vegetables (26.0%; 95% CI: 24.2 – 27.9) and fruits (24.5%; 95% CI: 22.6 – 26.6) more regularly (> 4 days a week), and among those who consumed unhealthy foods less regularly (≤ 4 days a week) such as soft drinks (24.5%; 95% CI: 23.2 – 25.9), sweets (25.6%; 95% CI: 24.3 – 27.0), fried snacks (24.4%; 95% CI: 23.2 – 25.6) and processed foods (24.1%; 95% CI: 22.8 – 25.5). Regarding physical activity, there was a slight prevalence of active schoolchildren (24.7%; 95%CI: 22, – 24.8) with the outcome studied in relation to those insufficiently active (23.3%; 95%CI: 22.9 – 26.5) (Table 1). The prevalence of excess weight was the same among those who sat on a normal weekday, during and for more than three hours.

The variables that showed a significant association with the outcome in the bivariate analysis were: region, school, age, number of people living with the child, mother’s education, vegetable consumption, soft drink consumption, sweets consumption and fried snacks consumption, as shown in Table 1.

Table 2 shows the multivariate analysis where the following remained associated with excess weight: geographic region (p=0.007; p=0.001), age group (p=0.003), maternal education (p=0.017), vegetable consumption (p=0.001) and sweets (p=0.006). The adolescents most likely to be overweight are those living in the Southeast (34%) and South (40%) regions, those aged 13 to 15 years (26%), with maternal education of incomplete/complete higher education (53%), who consumed vegetables more than four days a week (23%) and ate fewer sweets on four days (21%). Living with more than three people also showed a significant association (p=0.007), however it constituted a protective factor (ORaj=0.82; 95%CI: 0.72 – 0.94) for the outcome studied (Table 2).

DISCUSSION

The results of this study showed that 23.7% of Brazilian adolescents are overweight and that sociodemographic and behavioral factors influenced this outcome.

Adolescents from the South region had a higher prevalence of being overweight (28.2%), as did those from the Southeast region. These data were similar to ERICA, where the highest prevalence of excess weight (overweight and obesity) was identified in adolescents living in the South (29.8%) and Southeast (26.0%) regions ⁵ . Another national population-based survey indicated that the pattern of food availability in Brazilian households is directly related to the high and growing prevalence of overweight and obesity in the population, and it was observed that the excess fat content consumed tended to increase with family income, being more evident in the South and Southeast regions ⁴ .

Furthermore, the ERICA results also revealed that other aspects related to excess weight were more pronounced in these regions, with lower frequencies of healthy eating behaviors observed among adolescents in the Southeast region. This region, like the South, is among the most prevalent for physical inactivity and sedentary behavior (spending two or more hours a day using TV, computers, and video games) ^5,9 . Considering this information, it is assumed that adolescents in the South and Southeast regions are more susceptible to adopting lifestyle habits that predispose to excessive weight gain, thus there is a greater tendency for this finding in these regions.

Despite the higher prevalence observed among students in private schools (28.4%), the analysis of the data from PeNSE 2015 showed that there was no significant difference in excess weight between students in public and private schools. However, the study by Benedet et al. (2013) revealed that studying in public schools increases the chances of being overweight in female adolescents, while it decreases the chances for male adolescents ⁶ . On the other hand, Dias et al. (2014) highlight that adolescents from higher classes, as well as those from private schools, had a greater chance of having sedentary behaviors because they have greater access to technological means, such as the use of computers and video games ¹⁰ . Thus, because they have a sedentary behavior, it is likely that adolescents from private schools have a greater chance of developing overweight when compared to those from public schools.

The highest prevalence and higher chances of being overweight in the 13- to 15-year-old age group were similar to other studies conducted internationally, such as in Iraq, with 1,656 students aged 13 and 17, with a higher prevalence of obesity and overweight (11.3% and 22.6%, respectively) also observed in the 13- to 15-year-old age group ¹¹ . These data are also corroborated by the 2008-2009 POF, given the similar behavior of the reduction in the prevalence of overweight with advancing age among adolescents ⁴ .

A systematic analysis involving global, regional and national surveys of overweight and obesity conducted with children and adults showed an increase in prevalence from 1980 to 2013 in all age groups, with a substantial increase among children and adolescents in both developed and developing countries, such as Brazil ¹² . Although the analyses of the present study do not establish a temporal trend in the prevalence of overweight with the age range of the adolescents studied, their findings suggest that this health problem manifests itself increasingly earlier in this population, which requires an early approach to health promotion, prevention and control.

The present study observed a significant association between mothers with higher education and overweight in adolescents. However, in the study by D’Avila et al. (2016), which aimed to verify the association between food consumption, physical activity, sociodemographic factors and body fat percentage in schoolchildren, the authors noted that parental education was not an influencing factor, although the outcome showed a higher prevalence of excess body fat among participants with higher maternal education ¹³ .

Although it is expected that mothers with higher levels of education have more access to information that would influence their decision to choose healthier foods, it is also possible to observe that this educational context is related to a greater probability of maternal insertion in the labor market and a more favorable family socioeconomic condition. This results in increased purchasing power, greater access to different foods and their availability to their children. However, it is worth noting that these aspects can also have a negative impact on the eating habits of adolescents, given that mothers who work outside the home tend to spend less time preparing food, leading to a preference for semi-prepared products such as industrialized/ultra-processed foods ^14,15 . This fact could be related to the greater chance of being overweight in these adolescents.

The behavioral aspects related to the food consumption of adolescents found in this study contradicted the expected results, where higher consumption of vegetables and fruits and lower consumption of sweets were associated with excess weight. Furthermore, no statistically significant associations were found between the time adolescents spend on physical activities and sitting and excess weight.

However, other national and international studies also indicate an association between higher weekly intake of vegetables and excess weight in adolescents ^16,17 . Reverse causality may be responsible for this finding, as concluded by a study carried out with Brazilian adolescents living in Florianópolis ¹⁶ . Given that the results of the present study indicate that adolescents with excess weight have a significantly higher average consumption of vegetables than adolescents without excess weight (Figure 1),

Figure 1. Average food consumption, physical activity and sitting time among schoolchildren with and without excess weight. Brazil, 2015.
* In days per week; ** In hours per day
Source: National School Health Survey, 2015.

Analyzing the results observed regarding the consumption of sweets, we can also infer that the findings point to reverse causality, since lower weekly consumption was associated with excess weight and the highest average consumption was observed in adolescents without excess weight.

The motivation of overweight adolescents to acquire healthy habits has been investigated in some studies ^18,19 . A study conducted with adolescents in Los Angeles found that those who were overweight and obese showed greater motivation to acquire healthier eating habits, compared to adolescents with normal weight. The researchers concluded that obese adolescents are more vulnerable to external influences from family, friends and health professionals, which results in changes that would lead to weight loss ¹⁸ .

It is also worth noting that, according to the literature researched, cross-sectional studies that evaluate the effect of the consumption of healthy foods, especially fruits and vegetables, on the body weight of adolescents, are still quite controversial. This suggests that the significant effects of this more frequent consumption may take time to be evidenced in terms of their influence on nutritional status, and therefore the importance of conducting longitudinal studies for this purpose is recognized ^16,20 .

CONCLUSION

It is noteworthy that excess weight was significantly associated with the younger age group of schoolchildren, which may show the greater susceptibility of schoolchildren in the early years of adolescence to external influences, since they are the target audience for media coverage of the consumption of ultra-processed products. Greater and more effective investments in health promotion and education are recommended, as well as monitoring for those who are already overweight.

It is necessary to alert educators, health professionals, managers and the community in general that these individuals are in the process of biopsychosocial formation and that the behavioral and physical foundations established in these years will be reflected in the years to come. The greatest challenge is to ensure that the years that have been added to the life expectancy of future generations are also of quality.

Large national studies are important for assessing the health status of the population so that intervention measures can be taken, changed or continued. The inclusion of a second sampling plan in PeNSE brought visibility to a broader group of students who can more accurately reflect the Brazilian adolescent population.