Breakfast eating among Brazilian adolescents : Analysis of the National Dietary Survey 2008-2009

Objective To characterize breakfast eating among Brazilian adolescents. Methods Food intake was estimated based on a 1-day food record of adolescents aged 10-19 years, according to the first National Diet Survey (2008-2009). Breakfast was considered as the first meal of the day eaten between 4 and 11 am.


I N T R O D U C T I O N
Eating habits have a significant influence on the growth, development, and health of individuals [1,2].High consumption of fat and sugar rich foods and low consumption of fruits and vegetables [3] has been observed among Brazilian adolescents, resulting in nutritionally inadequate diets [4].Meal skipping [5,6] and eating away from home have also been observed [7].Moreover, when compared to adults and older adults, Brazilian adolescents consume more soft drinks, cookies, and sandwiches and less beans, salads, and vegetables [8].
Breakfast is considered as the first and most important meal of the day [9].The quality BREAKFAST AMONG BRAZILIAN ADOLESCENTS 465 of food at breakfast has been identified as essential for children and adolescents to achieve or maintain adequate health conditions [10] since the consumption of cereal and fruit is important for the prevention of chronic non-communicable diseases [11].
In children and adolescents, breakfast has been associated with improvement in attention, memory, and mood; it has also been possibly associated with improvements in motivation, cognitive function, and academic achievement [12,13], as well as with higher intakes of vitamin D and calcium [14].However, irregular breakfast habits have been associated with unfavorable health outcomes and weight gain among adolescents [15][16][17].
In Brazil, studies carried out in different locations have evaluated breakfast eating among adolescents [9,17].However, there are few studies addressing breakfast patterns among adolescents according to population-based data.Therefore, the present study proposes to characterize Brazilian adolescents' breakfast eating based on the data from the first National Dietary Survey, conducted in 2008-2009.

M E T H O D S
The data used were obtained from the Inquérito Nacional de Alimentação ( The instrument used to collect food consumption data was the two-day food record collected on non-consecutive days.Participants were asked to record all foods and beverages they consumed over the course of the day, including food items (food or preparations), amount consumed (in household measures or volume/mass measures), time and place of consumption (home -which also included foods prepared at home for consumption elsewhere or away from home -foods purchased and consumed away from home).
There was an extra question about the use of sugar and sweetener in beverages and other preparations [8].For the participants who reported using only sugar to sweeten beverages, the proportion of sugar added to beverages was standardized at 10%; for those who reported using sugar and artificial sweetener, this proportion was 5% [8].
In the present study, the foods mentioned by the participants in the food record were divided into 58 groups, according to Pereira et al. [18].Briefly, this food grouping system, proposed by the Department of Nutrition of the University of North Carolina, considers nine major United States Departament of Agriculture food groups disaggregated into subgroups according to the nutritional characteristics of foods and beverages and usual consumption pattern.The nutritional composition of the foods reported in the food survey was estimated based on the table of nutritional composition of foods adopted by the INA [19].
Only food records of the first day were used in this study.This decision was made due to the fact that even when two data sets are used, during the analysis for correction for intraindividual variations, a large number of zero intakes for many foods/food groups is observed, which can lead to biased estimates.In addition, the literature shows that the use of a single day of food record can provide reliable estimates of a mean intake of large groups and that the first day of record is the one with higher quality of reported information [20].
Breakfast was considered as the first meal of the day eaten between 4 and 11 am.To characterize it, the 43 groups of foods mentioned as foods consumed at breakfast by at least 2% of the adolescents were considered.Chart 1 shows the contributions to the daily consumption of energy and nutrients considering the regions of the country, the income quartiles, the time of the meal, and the number of items consumed at breakfast.The per capita household income was estimated by the sum of the total income of all household residents divided by the number of residents.The monthly per capita household was classified into quartiles.
Breakfast habits were evaluated according to: -Macro-regions of Brazil (North, Northeast, Central-West, South, and Southeast); -Place of residence (urban or rural); -Day of consumption: reports made on Saturday and Sunday were considered as weekend consumption; reports made form Monday to Friday were considered as weekday consumption; -Location of breakfast consumption (home and away from home); -Time of breakfast consumption was divided into three categories (4-6 am, 7-8 am, and 9-11 am); -Number of food items consumed at breakfast (1 food item, 2 food items, 3 food items, ≥4 food items).
Sample weights and the sample design effect were considered in the analysis.Differences in energy and nutrient consumption (continuous variables) between the categories evaluated were analyzed by linear regression (Generalized Linear Models [GLMs]) using the Complex Sample module of the Statistical Package for the Social Sciences (SPSS, Chicago, Illinois, United States), version 19 with Bonferroni correction.They were compared according to household income class, and geographic regions, time of breakfast, and number of items consumed.
Chi-square test was used to evaluate the differences between strata of the categorical variables, considering 95% confidence interval.

R E S U L T S
The sample analyzed consisted of 52.0% of males; 61.0% of the adolescents were aged between 10 and 15 years (mean age of 14.5 years), and 32.0% were from families in the first income quartile.It was observed that 93.1% of the adolescents reported having eaten breakfast between 4 and 11 am, and there were no differences according to gender.As for the geographic regions, a significant difference was observed only for the Southern region, where 90.3% of the adolescents reported breakfast eating within this period of time, and this proportion was significantly lower than that in the other regions of the country (Table 1).The first meal of the day was taken between 7-8 am by 59.0% of the adolescents, and only 10.7% had breakfast away from home (Table 1).
On average, breakfast contributed to 17.7% of the daily energy intake, and no significant difference was observed according to gender.In the highest income quartile, the contribution of breakfast to daily energy intake was lower than that in the other quartiles.The contribution of breakfast to daily energy intake was higher on weekends than on weekdays (19.6 versus 17.5%, p=0.02).It was also found that the contribution of breakfast to daily energy intake was greater when it was taken between 9-11 am (20.8%) than between 4-6 am (16.0%) and 7-8 am (19.2%).The higher the number of food items consumed at breakfast, the greater the breakfast contribution to the daily intake of energy and consequently of other nutrients, except for sugar, which remained stable around 30.0% regardless of the number of items consumed in that meal.Iron was another exception since it was about 4 times higher when the number of food items consumed at breakfast increased from 4 or more items (Table 2).
The contribution of breakfast to daily calcium intake among adolescents living in the urban area was higher than that of the adolescents living in rural areas (24.0 versus 19.2%, p<0.01).Breakfast contribution to the daily intake of added sugar among those living in the rural areas was higher than that of those in urban areas (42.1 versus 31.4%,p<0.01).The contribution of breakfast to the daily intake of carbohydrates in the Northeastern region was higher than that in the other regions.In the Northern and Northeastern regions, breakfast contribution to the daily intake of added sugar was higher than that in the Southeastern, Southern, and Central-Western regions.The contribution of breakfast to the daily intake of added sugar among the adolescents from families in the lowest income quartile income was higher than that of adolescents from families in the other income quartiles income (Table 2).
The contribution of breakfast consumed on weekends to the daily intake of carbohydrates, lipids, and fibers was higher than that consumed on weekdays.Breakfast eaten at home had a more significant contribution to the intake of lipids, saturated fat, and trans fat than that eaten away from home.It was also observed an increase in the contribution of breakfast to the daily intake of carbohydrates, proteins, total lipids, and fiber when breakfast was eaten between 9 and 11 am compared to that of the breakfast consumed earlier (Table 2).
The adolescents who ate breakfast away from home showed higher intake of deep fried or baked snacks (9 versus 2%, p<0.01), fruit juice (8 versus 4%, p<0.01), sugar-added beverages (5 versus 2%, p<0.01), and sweets and chocolates (5 versus 1%, p<0.01) than those who ate breakfast at home.Compared with the urban area, in the rural area, the consumption of the following food groups at breakfast was significantly higher: coffee and tea (74 versus 54%, p<0.01), packaged salty snacks (13 versus 7%, p<0.01), corn-based dishes (13 versus 3%, Compared with the adolescents from families in other income categories, adolescents from families in the first income quartile reported higher consumption of: coffee and tea, packaged salty snacks, corn-based dishes, and roots and tubers, and lower consumption of milk.As for the other quartiles of household income, the adolescents from families in the last income quartile reported significantly higher consumption of milk, fruit juice, cheese, and sugar-added beverages (Figure 1).
In the Northern region, there was higher consumption of coffee and tea (72%) and roots and tubers (7%), but milk consumption was lower (11%) than that in the other regions.In the Northeastern region, there was higher consumption of con-based dishes (13%) and eggs (6%) at breakfast.On the other hand, in the Southern region, there was higher consumption of processed meats (7%) and fruits (6%) (Data not shown).
The consumption of fruits (4%) was higher among the adolescents who ate breakfast between 4-6 am when compared to that of the adolescents who ate breakfast late in the morning.For those who ate breakfast between 7-8 am, there was higher consumption of breads (53%), butter/margarine (33%), and corn-based dishes (7%) than in the other breakfast times investigated.On the other hand, the adolescents who ate breakfast between 9-11 am showed a higher consumption of deep fried or baked snacks (6%), sugar-added beverages (5%), and sweets and chocolates (3%) (Data not shown).

D I S C U S S I O N
The results obtained indicate that coffee or tea, bread and butter or margarine are the most typical breakfast foods among Brazilian adolescents.However, there were differences according to income and the macro-regions of the country.The contribution of breakfast to daily energy intake was slightly lower than those reported in similar studies [21,22].Milk was the fourth most commonly consumed food all over the country, and its consumption was three times higher in the Southeastern and Central-Western regions than in the Northern region.Energy, lipid, and trans fat intake was higher on weekends than on weekdays.
The most commonly consumed foods at breakfast found in the present study are similar to those described by Trancoso et al. [23], who carried out a review study characterizing breakfast among Brazilians in terms of the consumption of coffee with milk and bread and butter or margarine.This basic composition may be related to convenience and practicality, which has contributed to a change in the consumption of regional foods, as pointed out by Maluf [24] and Pinheiro [25].Nevertheless, significant regional differences were observed in the consumption of milk, corn-based dishes, roots and tubers, eggs, fruits, and processed meat.
Based on food frequency questionnaire data and on a population-based study of adolescents from two different regions in Spain, Monteagudo et al. [22] found that the contribution of breakfast to daily energy intake ranged from 21.14% (girls aged 14 to 17 years) to 25.81% (boys aged 10 to 13 years), which shows slightly higher proportions than those found in the present study.However, it is worth highlighting the different methods used in these two studies to obtain food consumption data.
It was found that the contribution of breakfast to daily calcium intake increased with income.Moreover, this contribution was more significant in the urban area than in the rural area.These findings are consistent with those of Levy et al. [26], who reported that the contribution of milk and dairy products in terms of dietary energy availability with the increase in income.
The large amount of saturated fat, trans fat, and added sugar in the foods consumed by Brazilian adolescents at breakfast found in this study corroborates the findings about food consumption characteristics among adolescents in international studies [27] and in data from the Pesquisa Nacional de Saúde Escolar (PeNSE, National School Health Survey) [28].Pereira et al. [18] evaluated the presence of foods high in saturated fat, trans fat, and added sugar consumed by the Brazilian population.The author found that the contribution of these foods to total energy intake was higher among adolescents (54%) than among adults (47%) and older adults (46%).
Good quality breakfast has been considered essential for children and adolescents to achieve and maintain adequate health conditions [10], and breakfast consumption has been associated with intake adequacy of micronutrients, cereals, and fruits [11].The quality of breakfast consumed by Brazilian adolescents may be compromised since it was observed reduced fruit consumption and the presence of unhealthy foods, such as deep fried or baked snacks, sugar-added beverages, packaged snacks, sweets and chocolates, and processed meats.An important aspect to highlight is the significant contribution of breakfast to the daily intake of added sugar, since it provided an average of 17.7% of the daily energy, and 33.4% of the added sugar consumed per day came from foods eaten at breakfast.
In the present study, the proportion of adolescents who skipped breakfast was much lower than that found by other authors [6,29,30], who reported breakfast skipping rates between 29 and 51%.These differences can be attributed to the methods used to evaluate breakfast eating habits.In this study, we analyzed the proportion of participants who reported having consumed breakfast on a 1-day food record; in the other studies, food frequency questionnaires were used to evaluate how often breakfast was usually consumed.
Food record is a widely used method in population studies, and it has the advantage of not relying on respondent's memory and providing more accurate information regarding dietary intake.However, one disadvantage stems from conscious or unconscious changes in eating habits that may occur while the person is keeping the record [31,32].Therefore, the high proportion of reports of breakfast eating observed in the present study may be related to the method used to evaluate food consumption and does not necessarily indicate the prevalence of breakfast consumption or breakfast skipping.Accordingly, it is worth highlighting the different breakfast definitions that have been provided.Aranceta et al. [33] define breakfast based on the time of consumption, considering breakfast as the intake of solid or liquid foods between 6 and 10 am on weekdays and between 6 and 11 am on weekends and holidays.Similarly, Affenito et al. [34] define breakfast according to the time of consumption.The authors characterize breakfast as the consumption of any food between 5 and 10 am on weekdays and between 5 and 11 am on the weekend.On the other hand, Matthys et al. [35] define breakfast as the first meal of the day.
The present study showed some limitations.Despite the great effort to obtain reliable data about the nutritional composition of foods, the estimates of added sugar intake may be biased since table sugar consumption data were not directly obtained.
This study describes the nutritional characteristics and identifies the most commonly consumed foods at breakfast among Brazilian adolescents, according to the regions of the country, urban-rural residence, and income.The results showed that breakfast contribution to daily energy intake is low and that its nutritional quality improves with income.In Brazil, information about breakfast eating habits of adolescents is still scarce although it is considered an important meal for health and healthy eating promotion, according to the Guia Alimentar para População Brasileira (Dietary Guidelines for the Brazilian Population) [36].This study offers a significant contribution to support the development of more effective intervention programs aimed at promoting health in Brazilian adolescents.

Figure 2 .
Figure 2. Most commonly consumed food groups at breakfast among adolescents, according to the number of food items included in the meal.Brazil, National Dietary Survey 2008-2009.Note: Chi-square test; * p-value<0.05.

A
C K N O W L E D G M E N T S For the financial support provided by the Ministry of Health, Conselho Nacional de Desenvolvimento Científico e Tecnológico, and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior.C O N T R I B U T I O N SLS MONTEIRO, AM SOUZA, BK HASSAN e CCP ESTIMA contributed to data analysis and interpretation and manuscript writing.R SICHIERI contributed to data interpretation and manuscript revision.RA PEREIRA contributed to the conception and design of this study, data analysis and interpretation, and manuscript writing and critical revision.

Chart 1 .
The most-frequently reported food groups consumed at breakfast among Brazilian adolescents -Inquérito Nacional de Alimentação (INA, National Dietary Survey).

Table 1 .
Sample characterization and prevalence of breakfast consumption among adolescents.Brazil, National Dietary Survey 2008-2009.

Table 2 .
Percentage contribution of breakfast to daily energy and nutrient intake among adolescents.Brazil, National Dietary Survey 2008-2009.