Prevalence of excessive body fat among adolescents of a south Brazilian metropolitan region and State capital, associated risk factors, and consequences

This is a cross-sectional study carried out from 2014 to 2016, with adolescents (11–18 years) whose parents gave permission to participate in the research by signing an informed consent form. For the sample size, a sampling error of 4.5% was specified at 91% confidence level of a universe of 82,414 individuals (Additional file 1). The study sample consisted of adolescents who met the following criteria: a) parents authorized their participation; b) did not make use of medicines containing calcium; c) did not undergo radiography/computed tomography in the seven days prior to the evaluation; d) did not suspect pregnancy. The subjects of the study were gathered from public and private schools, as well as sports training centers in the metropolitan region (composed of the city of Curitiba-PR, Brazil, and 29 other municipalities).

The body mass was assessed with a mechanical scale and the height was assessed using a stadiometer coupled to the scale (Filizola, São Paulo, Brazil). The waist circumference was measured in the smallest circumference of the abdomen with a flexible and inelastic tape measure (Cardiomed, Curitiba, Brazil). For classification purposes, the adopted cut-off point was ≥80 cm for boys and ≥90 cm for girls [14].

The evaluation of body fat and bone mineral density was performed using dual energy X-ray absorptiometry (DXA) technology using a Hologic Discovery A linear sweep fan-beam scanner (Hologic, Inc., Bedford, U.S.A.). All DXA procedures were performed by an expert in the equipment. For the classification of the excess fat, the cut-off point of Williams et al. was used, which establishes a percentage of fat (%BF) with a value equal to or greater than 25% for boys, and 30% for girls [15]. With regard to the BMI classification, it was made from the index calculation (kg/m²) using the Brazilian recommendation of Conde and Monteiro [16].

Serum glucose, total cholesterol (TC) and triglyceride levels were measured using a portable test-strip type dosing equipment (Accutrend Plus, Roche, Germany). For the classification of the individuals, the reference values of the Brazilian Society of Diabetes were used, being classified as individuals with high levels those with glucose levels ≥140 mg/dl, and triglycerides levels ≥150 mg/dl. For TC, 150 to 169 mg/dl were considered as limit values, and values ≥170 mg/dl were considered as high values.

The levels of physical activity and sedentary lifestyle were evaluated through the International Physical Activity Questionnaire (IPAQ) short version, validated in Brazil by Matsudo et al. [17]. The IPAQ is a questionnaire that allows estimating the weekly time spent in moderate and vigorous physical activities in different contexts of everyday life, such as work, transportation, domestic tasks and leisure, as well as the time spent in passive activities performed in the sitting position.

A classification of aggregated factors was proposed, in which the individuals were grouped into three categories: a) adequate physical activity and not sedentary (classified as “no factor”); b) either no physical activity or sedentary (classified as “1 factor”); c) low levels of physical activity and sedentary (classified as “2 factors”).

The socioeconomic profile was assessed through the questionnaire proposed by the Brazilian Association of Research Companies (ABEP in the Portuguese abbreviation). This instrument allows estimating the purchasing power of families and classifying them into classes: A1, A2, B1, B2, C1, C2, D and E. For this, the subjects were asked to answer nine questions about how many of these items/services they had at home: TV, refrigerator, monthly house cleaner, automobile, bathroom and DVD player. The classes were gathered into three categories: A1 and A2 (high purchasing power category), B1 and B2 (medium purchasing power category) and C1, C2, D and E (low purchasing power category) [18].

For the statistical analysis, the sample was characterized with position (mean) and dispersion (standard deviation) measurements, as well as frequency distribution (%). To investigate the association between excess fat and factors such as biochemical, socioeconomic, active transportation, physical activity level and sedentary lifestyle, the chi-square test was applied. The risk ratio [odds ratio (OR)] and its 95% confidence intervals (95% CI) were obtained through a specific procedure of the Statistical Package for the Social Sciences (SPSS), version 17.0 (SPSS Inc. Chicago, IL). The value of p < 0.05 was adopted as statistical significance.

The metropolitan area of Curitiba is the eighth most populous area in Brazil. The HDI (Human Development Index) of Curitiba is 0.823, considered to be very high, similar to some developed countries such as Poland (0.853). However, it is the sixth most violent capital in the country. As most of the southern Brazilian population, Curitiba is inhabited mostly by Brazilians who are descendant from European cultures, especially Polish, Ukrainian, Italian, German and Asian.

In our study, it was detected a critical stage in the prevalence of excess fat (40.4%), higher than the projection for the year 2025 [9]. In addition to it, this prevalence was much higher among female adolescents (92.1%) than males (18.2%). In concordance with our results, another study using DXA, performed in the southeast region of Brazil, with 215 adolescents aged 10 to 14 years, found a prevalence of 44.2% of excess fat, without stratification by sex [19]. The high prevalence of overweight and obesity, as high as 38.9%, was also indicated by a review study composed of 16 researches made in Brazil [7]. This review study, as well as ours, shows that Brazilian public policies focused on low weight might be equivocal. Thus, differently from previous decades when undernutrition was a chronic problem in our country, the new reality seems to point towards high prevalence of overweight and obesity. Therefore, this new information should be considered both when evaluating the maintenance of existing public policies and when creating new control policies for overweight and obesity in adolescence.

The percentage of fat mass found in our study was 24.5 ± 8.0% (20.9 ± 5.9% for boys and 33.0 ± 5.4% for girls). These are so high that come close to numbers found in the USA, as shown in a study conducted among USA children and adolescents, those being 25.4 ± 0.2% for boys and 33.3 ± 0.3% for girls [20]. On the other hand, a study conducted in Poland, a developed country with an HDI similar to the area of our study, found a body fat 15.6% in boys and 13.4% in girls, all aged between 14 and 18 years [21].

Nevertheless, in Asian populations lower values of fat mass were found, such as in a study among adolescents of Malay and Chinese origins (aged 12 to 19). For boys, values of 17.1 ± 10.0% and 18.8 ± 9.4% were found for Malays and Chinese, respectively. For girls, the values found were 31.7 ± 8.4% and 32.8 ± 7.1% for Malays and Chinese, respectively [22]. Although these values are smaller than ours, this study also points to greater fat mass values in girls.

Regarding the relationship between physical inactivity, gender and obesity, some studies have also found association between these variables. In our study, male adolescents were more active than females (66.3% and 52.2%, respectively). In sum, the inactive had 2.9 times more chance of being classified with excess fat (4.0 times more among boys). Female adolescents had more sedentary habits (79.3%) when compared to boys (75.2%). This could be a factor in explaining the higher prevalence of excess fat among girls, since the act of watching television has been related to the consumption of high-calorie food [7, 8].

In comparison to the behavior of boys, more frequent sedentary behavior among girls has been reported in other studies conducted in Brazil and in Asian countries. In Goiânia, Brazil, a study with adolescents aged 14 to 18 years found that 78% of the girls and 54.3% of the boys were sedentary [11]. A study in Malaysia with adolescents with ages ranging from 12 to 19 years (Malay and Chinese origin) showed a significant relationship between excess fat and sedentary behavior only among girls. Separating by ethnicity, girls of Malay origin had sedentary behavior on average of 3.1 h/day while those of Chinese origin reached 3.8 h/day [23].

Another Brazilian study carried out in São Paulo, with children and adolescents (aged 7 to 18), found that 28.3% of the boys and 20.4% of the girls were overweight/obese. Among the risk factors were: going to school by car (0.72) (when compared to going by bus 0.61 and walking 0.59), more than one hour of computer use per day (1.64) or more than two hours per day (1.94) [6].

In Brazil, full-time schools are rare. Therefore, the time available for performing sedentary activities (use of television, computer, and video game) is higher than in developed countries [7].

We did not find relation between the excess fat in adolescents with the income or the way of transportation to the school (active transportation). Active transportation is adopted by a small fraction of the adolescents in the area of the study [24]. Violence might be an impacting factor, for active transportation was more common among boys (19.4% than girls (6.4%). When investigating the causes for this low rate, 84% of the parents and/or guardians show concern as to attacks of strangers being a barrier to adopting active transportation.

A study carried out in seven African developing countries meets our results. It showed that the active transportation to school did not influence the risk of overweight/obesity in adolescents in Benin, Djibouti, Egypt, Malawi and Mauritania. However, it was found that in Morocco the active transportation appeared to reduce the odds (OR = 0.77; 95% CI: 0.61–0.97) and that the opposite occurred in Ghana, where the odds appeared to have increased (OR = 1.28; 95% CI: 1.01–1.62) [9].

Despite there is no concordance in literature, the majority of the studies have not found significant relations between excessive body fat and risk factors as socio-economic level of the families, education level of the parents, among others [7, 12].

In the southeastern region of Brazil, authors also reported not having found a significant association between the type of school, total annual family income, employment status or maternal or paternal schooling with obesity and levels of physical activity [12].

Conversely, a review study (with 16 articles published between 2004 and 2012 in Brazil) showed that only one study, conducted in the center-west region, indicated some relation between overweight/obesity and socioeconomic level [7].

The male adolescents who presented abdominal obesity had 13.5% higher risk of being classified with excess fat. This risk was not verified among the girls. Although girls have a higher percentage of body fat due to hormonal differences between the sexes, they tend to accumulate adipose tissue in the hip region, while boys tend to accumulate it in the waist region [8, 25].

A study conducted in São Paulo, Brazil, with adolescents aged 14 to 19, found a prevalence of abdominal obesity of 10.5% in males and 10.8% in females. The authors present sedentary behavior as the main lifestyle factor associated with abdominal obesity [8].

A study conducted in the same region as our study, with adolescents aged 11 to 19, identified a prevalence of general abdominal obesity of 12.2%, which was also higher in boys (15.1%) than in girls (12.2%) [26].

Comparing the prevalent values between developed and developing countries, a review study found that the prevalence of abdominal obesity in developing countries varied from 3.8% to 51.7%. In developed countries, however, rates ranged from 9.3% to 33.2%. Many of these discrepancies may have occurred due to the different criteria applied [27].

The adoption of public policies to restrain the excessive body fat is important because one of its consequences might be the alteration of biochemical parameters. In our study, the prevalence of high total cholesterol was 22.2% in boys and 42.3% in girls. Considering those with excess fat, boys were 2.2 times more likely to have increased triglycerides and 2.67 times more likely to have high total cholesterol. In girls, obesity seems to have affected only total cholesterol. Girls with excess fat had 8.0 times more chance of having high total cholesterol.

Risks for cardiovascular health and metabolic syndrome of the increased rates of cholesterol and triglycerides are well documented. However, some researches are also concerned about their impact on bone health. It seems this relation should be further investigated.

Our study found no direct relationship between excess fat and Bone Mineral Density (BMD). However, BMD values were lower (p = 0.046) in boys with two aggregate factors (physical inactivity and sedentary lifestyle). This is worrisome because adolescence is considered a critical period for the gain of bone mass and an impaired bone acquisition during this phase may increase the risk of osteopenia/osteoporosis and fractures in old age [28].

A possible explanation for the effect of HDL cholesterol on the skeleton would be the inhibition of osteogenic tissue, suggesting that it regulates the differentiation of osteoblasts. Therefore, the authors describe a study that detected a negative correlation between hypercholesterolemia with BMD (lumbar spine) and BMC (Bone Mineral Content) and increased total cholesterol in overweight/obese girls, but not in boys. The authors further suggested that the risk of cardiovascular disease and osteoporosis later in life would be greater in these girls [28, 29].

The authors describe another study that negatively correlated the increase in triglyceride levels and BMD in adolescents. It was observed that triglycerides were positively correlated with fat in the bone marrow, possibly because this type of fat stimulates osteoclasts (which degrade bone tissue) and it was also noted that high levels of bone marrow fat increased the risk of fracture [27, 30].

A review study that included 27 studies described that overweight or obese children have a significantly higher BMD than normal-weight children. That sensitivity analysis showed that the association was stronger in girls [31]. However, the authors found only one longitudinal study that investigated the long-term consequences of obesity in BMD. This longitudinal study found a higher trabecular density at the tibia in adult women and a lower cortical density at the tibia in adult men who were obese in childhood. In addition to these findings, the authors describe that the results of few studies that compared sexes indicate significant differences between normal weight, overweight or obesity are found more often in girls than in boys. This can occur due to the difference in hormonal development in girls and boys. Considering this result and the lack of evidence on the long-term consequences of childhood obesity in BMD, further prospective research is recommended [31].

It is worth noticing that bone quality in obesity is not as good as that obtained by physical activity. Some authors have related resistance to leptin, frequently found in obese children, due to poor bone health. They suggest that leptin has an indirect effect on bone formation influencing other hormones that affect bone density such as growth hormone, androgens, and cortisol. Thus, the microstructure of the bone would be altered and this would lead to a greater propensity to fractures, especially of the forearm, in obese children [31, 32].

Among the limitations of this research is the transversal design, offering no basis for the study of probabilities. Another limitation was the use of questionnaires to collect data regarding sedentary behavior, active transportation and level of physical activity. Although validated, the instruments may contain a bias of estimation and memory. Moreover, the analyzed adolescents are from a single region of the country, not providing a nationally representative sample.

More than 3000 adolescents were contacted, but we still had great difficulty in sensitizing them to participate in the research. Therefore, we analyzed data from adolescents who either had not filled any questionnaire or not agreed to take any specific collection (such as blood testing). This way, another limitation in this study was a lot of missing values, for example, socioeconomic data. The use of a single body fat cut-off to define obesity (25% in boys and 30% in girls) is also mentioned as limitation. The use of age-specific cut-offs may be more suitable to define pediatric obesity.

Another limitation was not being able to investigate the adolescents’ diets. Besides physical activity, diet is one of the most important determinants of adiposity. In this sense, high prevalence of excessive body fat might have been influenced by an hypercaloric and unhealthy diet.

Despite these limitations, this study is important because it uses DXA-based %BF to identify overweight and obese adolescents. In addition, it was possible to obtain objective and consistent measures of several risk factors that may help in the elaboration of local public policies, since Brazil is a continental country with great ethnic and cultural variety among its regions. Thus, it is possible to establish local strategies to help fight overweight/obesity among adolescents in this particular region.

Moreover, the use of DXA provided us with important information about BMD, justifying future studies investigating the relationships between BMD and excess fat, levels of physical activity, and rates of cholesterol and triglycerides.