Skip to main content
PMC home page
Preventive Medicine Reports logoLink to Preventive Medicine Reports
. 2016 Jul 30;4:391–396. doi: 10.1016/j.pmedr.2016.07.010

Trends in diet quality among adolescents, adults and older adults: A population-based study☆☆☆★

Samantha Caesar de Andrade a,⁎,1, Ágatha Nogueira Previdelli a,2, Chester Luiz Galvão Cesar b,3, Dirce Maria Lobo Marchioni a,4, Regina Mara Fisberg a,5
PMCID: PMC4990660  PMID: 27547722

Abstract

This study aimed to monitor diet quality and associated factors in adolescents, adults and older adults from the city of São Paulo, Brazil. We conducted a cross-sectional population-based study involving 2376 individuals surveyed in 2003, and 1662 individuals in 2008 (Health Survey of São Paulo, ISA-Capital). Participants were of both sexes and aged 12 to 19 years old (adolescents), 20 to 59 years old (adults) and 60 years old or over (older adults). Food intake was assessed using the 24-h dietary recall method while diet quality was determined by the Brazilian Healthy Eating Index (BHEI-R). The prevalence of descriptive variables for 2003 and 2008 was compared adopting a confidence interval of 95%. The means of total BHEI-R score and its components for 2003 and 2008 were compared for each age group. Associations between the BHEI-R and independent variables were evaluated for each survey year using multiple linear regression analysis. Results showed that the mean BHEI-R increased (54.9 vs. 56.4 points) over the five-year period. However, the age group evaluation showed a deterioration in diet quality of adolescents, influenced by a decrease in scores for dark-green and orange vegetables and legumes, total grains, oils and SoFAAS (solid fat, alcohol and added sugar) components. In the 2008 survey, adults had a higher BHEI-R score, by 6.1 points on average, compared to adolescents. Compared to older adults, this difference was 10.7 points. The diet quality remains a concern, especially among adolescents, that had the worst results compared to the other age groups.

Keywords: Adolescents, Diet quality, Eating habits, Food intake

Highlights

  • •

    Over the five-year period studied, the diet quality remains worrying.

  • •

    The adolescents had the worst diet quality compared with the other age groups.

  • •

    The score for SoFAAS (solid fat, alcohol and added sugar) became even lower.

  • •

    Were found association between diet quality and life style variables.

  • •

    Head of household education showed an inverse association with diet quality.

1. Introduction

Dietary monitoring is part of the effort toward improving the population's nutritional health. A well-designed health-surveillance program can frame political initiatives by identifying adverse tendencies in the population's eating behavior (Ballard-Barbash, 2001).

Public policies, including raising the minimum wage and income transfer programs and expansion of the family health strategy have contributed to the decline of child malnutrition in Brazil (Monteiro et al., 2010). On the other hand, changes in eating patterns and physical activity have led to a new perspective for the population's health.

Analysis of Household Budget Surveys (HBS) in Brazil (2002/03–2008/09) revealed an insufficient amount of fruits and vegetables and an excess of calories from free sugars and saturated fats in the diet (Levy et al., 2012).

Studies of the Mediterranean population revealed unfavorable trends in diet quality, characterized by higher consumption of total and saturated fat and less consumption of fruits and vegetable in 2005 compared to 2000 (Valdés et al., 2009). Data from the National Health and Nutrition Examination Survey (NHANES) showed that the diet quality of Americans is far from optimal and, according to the Healthy Eating Index (HEI), did not improve overall between 2001/02 and 2007/08 (USDA, 2013).

Recently, dietary pattern analysis has emerged as an alternative and complementary approach to examining the relationship between diet and the risk of chronic diseases. Instead of looking at individual nutrients or foods, pattern analysis examines the effects of overall diet. There is a growing interest in using dietary quality indices to evaluate the adherence to a certain dietary pattern or current dietary guidelines (Hu, 2002).

Given all these changes in recent decades, we hypothesized that Brazilian diet quality has deteriorated over the years across all age groups. Therefore, the primary objective of this study was to monitor diet quality and associated factors in adolescents, adults and older adults from the city of São Paulo, Brazil, in 2003 and 2008, using the Brazilian Healthy Eating Index (BHEI-R).

2. Methods

2.1. Study design and population

Data were obtained from two cross-sectional, population-based surveys on health and living conditions in a representative sample of adolescents (12–19 years old), adults (20–59 years old) and older adults (60 years old or over) living in the city of São Paulo (ISA - Capital), one of the most important metropolitan cities in Brazil, conducted in 2003 and 2008.

The two surveys employed a similar sampling process. Sample selection was carried out by two-stage cluster sampling: census tracts and households. For the present study, inclusion criteria were individuals at least 12 years of age who answered the socioeconomic and dietary survey, giving a total of 2376 subjects (814 adolescents, 746 adults and 816 older adults) for the ISA-Capital 2003 and 1662 subjects (560 adolescents, 585 adults and 517 older adults) for the ISA-Capital 2008. Further details on the sampling can be found in another publication (Castro et al., 2009).

The Research Ethics Committee of the School of Public Health, University of São Paulo, approved the study protocol. All participants provided all the data required and signed the informed consent form.

2.2. Data collection

A structured questionnaire collecting demographic (sex, age and race), socioeconomic (head of household education and household income per capita), family (family household members), anthropometric (body weight and height) and lifestyle characteristics (smoking habits, alcohol consumption and leisure-time physical activity) was applied by trained interviewers at the participants' homes during the periods March to December 2003 and September 2008 to March 2009. All data are shown in Table 1.

Table 1.

Weighted percentage distribution of studied population according to sociodemographic, anthropometric and lifestyle characteristics. ISA 2003 and 2008. São Paulo, SP, 2003–08.

Variables ISA 2003
 ISA 2008
n %a (CI 95%) n %a (CI 95%)
Sex
 Male 1163 44.92 (42.21–47.62) 722 46.90 (44.36–49.44)
 Female 1213 55.08 (52.38–57.79) 940 53.10 (50.56–55.64)
Age group (years)
 12 a 19b 814 19.84 (17.66–22.01) 560 15.14 (12.93–17.34)
 20 a 59 746 68.42 (66.19–70.65) 585 71.00 (68.00–74.01)
 60 or older 816 11.74 (10.22–13.27) 517 13.86 (11.62–16.10)
Racec
 White 1509 61.22 (57.30–65.15) 963 58.46 (52.72–64.20)
 Non white 856 38.78 (34.85–42.70) 697 41.54 (35.80–47.28)
House head's educationc
 Up to 4 yearsb 1033 39.27 (35.44–43.09) 595 22.65 (18.82–26.48)
 5–8 years 481 23.65 (20.71–26.60) 407 22.10 (16.86–27.36)
 9–11 yearsb 426 20.55 (17.67–23.44) 418 35.44 (29.81–41.07)
 12 years or more 405 16.53 (12.96–20.09) 216 19.81 (13.94–25.66)
Nutritional statusc
 Underweight 187 5.48 (04.24–6.72) 96 4.05 (03.20–04.91)
 Normal weight 1224 58.57 (55.47–61.66) 899 54.02 (50.67–57.36)
 Overweight 670 35.95 (33.03–38.88) 590 41.93 (38.74–45.12)
Leisure-time physical activityc, d
 Activeb 177 07.03 (05.21–08.85) 253 14.41 (11.81–17.00)
 Not activeb 2018 92.97 (91.15–94.79) 1407 85.59 (83.00–88.19)
Smoking habitsc
 Never smoked 1641 67.53 (64.86–70.21) 1155 64.17 (60.45–67.87)
 Former smoker 363 14.21 (11.71–16.70) 274 16.04 (13.21–18.86)
 Smoker 334 18.26 (15.76–20.76) 232 19.79 (16.44–23.14)
Alcoholic beverage consumptionc
 No 1341 50.43 (46.78–54.07) 962 47.89 (44.69–51.08)
 Yes 1024 49.57 (45.93–53.22) 699 52.11 (48.92–55.31)
Open in a new tab

n - number of individuals in the unweighted sample.

CI 95% - Confidence Interval 95%.

a

Prevalence in the weighted sample.

b

Intervals do not overlap, so we have (at least) 95% confidence that the true values are not equal.

c

Exclusion of individuals with ignored information.

d

Active individual with a physical activity practice of at least 30 min of low or moderate intensity physical activity daily, five or more days of the week, or practice of at least 20 min of vigorous intensity physical activity daily, 3 or more days of the week.

The nutritional status was assessed by body mass index (BMI = weight/stature2) (Center for Disease Control and Prevention, 2012, World Health Organization (WHO), 1998, Lipschitz, 1994) and the leisure-time physical activity (active/not active) was defined as the practice of physical activity for at least 30 min per day, five days a week at moderate intensity, or at least 20 min a day three days a week in vigorous intensity, using the International Physical Activity Questionnaire (IPAQ), long version.

Food intake was measured by one 24-h dietary recall, considering all seasons of the year and days of the week, contributing to daily variation in dietary intake. Data were collected using the multiple-pass method, that employs five sequential passes in the interview to facilitate the accurate recall of dietary intake (Conway et al., 2004), and critically reviewed to identify any failures relating to descriptions of foods or preparations consumed and their apportioning and quantification, in accordance with Pinheiro et al. (2000) and Fisberg and Villar (2002). The nutritive values of foods and ingredients were checked. Total energy intake was also analyzed, and none of the 24HRs were excluded. Thus, the 24HR data was controlled for errors that might have occurred in estimating nutrient intake. Data were analyzed using the Nutrition Data System for Research (NDSR database, version 2007, NCC, University of Minnesota, Minneapolis). The primary source of nutritional food values was the US Department of Agriculture (USDA) nutritional composition table, and this was extended with information provided by food manufacturers, the scientific literature and the Brazilian Table of Food Composition (TACO) (NEPA-UNICAMP, 2011). A concordance rate between 80 and 120% for energy and macronutrients content was required to accept a food selection from local composition table (Fisberg and Marchioni, 2012). When regional foods did not have an exact equivalent or similar food available in the NDS-R database, one or more foods combined were inserted as a recipe that represents the same nutritional value as the original version (Kovalskys et al., 2015).

Overall diet quality was assessed using the revised version of the Healthy Eating Index 2005 for the Brazilian population (BHEI-R) by Previdelli et al. (2011) and evaluated according to the psychometric properties of Andrade et al. (2013). This index evaluates a combination of different types of foods, nutrients and other components of the diet regarding the daily intake recommendations and/or health outcomes.

The BHEI-R components were based on Dietary Guidelines for Brazilians Population (Ministério da Saúde, 2005) and expressed in terms of food density (1000 kcal/g). BHEI-R comprises 12 components, nine of which are food groups, two of which are nutrients, and the last of which represents the sum of the energy value contributed by solid fats, alcohol, and added sugar (the SoFAAS component). The maximum number of total points was assigned for intakes at the standard level: 5 points (total grains, whole grains, dark-green and orange vegetables and legumes, total vegetables, total fruit and whole fruit), 10 points (milk and dairy products, meat and beans, oils, saturated fat and sodium) and 20 points for SoFAAS. Thus, the final score of the BHEI-R ranged from zero to 100 points. Higher values of score indicates greater adherence to dietary guidelines.

For most components, higher intakes result in higher scores. It should be noted, however, that for three components: saturated fat, sodium and SoFAAS, lower intake levels result in higher scores.

2.3. Statistical analysis

The prevalence of descriptive variables for ISA 2003 and 2008 were compared using a confidence interval of 95% (95% CI).

Total BHEI-R mean and components scores from 2003 and 2008 were compared using the Wald test or one-way analysis of variance for each age group.

The relationship between the BHEI-R (continuous variable) and the associated factors was assessed by multiple linear regression analysis (stepwise forward selection), given that the BHEI-R had a normal distribution. Data modeling was performed for the two survey years (2003 and 2008). The dependent variable was overall diet quality (BHEI-R) whereas independent variables were age group (adolescent, adult or older adults), smoking habits (smoker, former smoker or non-smoker), head of household education (years), gender (male or female), nutritional status (overweight or not overweight), household income per capita, alcohol consumption (yes or no), ethnicity (white or non-white) and the energy intake (kcal) to adjust for potentially confounding effects. Interactions among the variables included in the final model were also investigated, but no significant values were found. A residual analysis was performed to test for homoscedasticity of errors in regression, and no bias was found.

All analyses were performed using the “survey” module of the STATA software package (version 12, Stata Corporation, College Station, TX, USA) and a significance level of 5%.

3. Results

The two representative samples from the city of São Paulo were similar for most variables, according to descriptive analysis (Table 1). The mean of schooling years of the family head was 7.3 years (SE = 0.41; 95% CI 6.51–8.13) in 2003 and 8.4 years (SE = 0.21; 95% CI 7.95–8.79) in 2008. The mean energy intake was 1864.15 kcal (SE = 26.94; 95% CI 1810.20–1918.10) in 2003 and 1949.58 kcal (SE = 42.00; 95% CI 1865.77–2033.40) in 2008.

Over the five-year period studied, diet quality remains worrying (BHEI-R = 54.99 vs. 56.42, p < 0.005) with an increase in scores for total and whole fruit, milk and dairy products, saturated fat and sodium and a decline in scores for total grains, oils and SoFAAS components (comprising total calories derived from solid fats, alcohol and added sugar) (Table 2).

Table 2.

Comparison of means of components and BHEI-R final score of population studied according to age group in 2003 and 2008. ISA 2003 and 2008. São Paulo, SP, 2003–08.

Component (maximum score) Total population
 Adolescents
 Adults
 Older adults
2003 (n = 2376)
Meana (SE)		 2008 (n = 1662)
Meana (SE)		 2003 (n = 814)
Meana (SE)		 2008 (n = 560)
Meana (SE)		 2003 (n = 746)
Meana (SE)		 2008 (n = 585)
Meana (SE)		 2003 (n = 816)
Meana (SE)		 2008 (n = 517)
Meana (SE)
Total fruit (5) 1.63 (0.09) 1.94 (0.08)⁎ 1.44 (0.11) 1.36 (0.08) 1.53 (0.09) 1.88 (0.11)⁎ 2.51 (0.10) 2.88 (0.14)⁎
Whole fruit (5) 1.33 (0.08) 1.92 (0.09)⁎ 1.06 (0.10) 1.22 (0.08) 1.21 (0.09) 1.86 (0.12)⁎ 2.45 (0.11) 3.01 (0.14)⁎
Total vegetable (5) 4.16 (0.05) 4.17 (0.06) 3.82 (0.06) 3.59 (0.12) 4.24 (0.07) 4.25 (0.07) 4.32 (0.06) 4.34 (0.07)
DGOV&Lb (5) 3.64 (0.07) 3.61 (0.09) 3.22 (0.08) 2.86 (0.11)⁎ 3.72 (0.10) 3.73 (0.11) 3.90 (0.05) 3.79 (0.11)
Total grains (5) 4.62 (0.03) 4.52 (0.04)⁎ 4.73 (0.03) 4.55 (0.04)⁎ 4.59 (0.37) 4.50 (0.05) 4.59 (0.04) 4.59 (0.04)
Whole grains (5) 0.38 (0.03) 0.41 (0.04) 0.32 (0.04) 0.29 (0.04) 0.38 (0.04) 0.39 (0.05) 0.49 (0.05) 0.62 (0.07)
Milk and dairy (10) 3.95 (0.13) 4.35 (0.14)⁎ 3.70 (0.17) 4.43 (0.18)⁎ 3.86 (0.16) 4.21 (0.18) 4.91 (0.16) 4.99 (0.18)
Meats, eggs and Legumes (10) 8.52 (0.09) 8.54 (0.09) 8.18 (0.19) 8.08 (0.12) 8.68 (0.12) 8.69 (0.11) 8.14 (0.10) 8.24 (0.15)
Oilsc (10) 9.33 (0.04) 9.14 (0.08)⁎ 9.29 (0.08) 8.80 (0.13)⁎ 9.33 (0.05) 9.20 (0.11) 9.41 (0.09) 9.21 (0.11)
Saturated fat (10) 5.93 (0.13) 6.62 (0.14)⁎ 5.95 (0.17) 6.34 (0.15) 5.85 (0.16) 6.63 (0.19)⁎ 6.33 (0.15) 6.88 (0.17)⁎
Sodium (10) 2.12 (0.09) 2.60 (0.10)⁎ 2.53 (0.15) 3.38 (0.16)⁎ 2.03 (0.11) 2.49 (0.14)⁎ 1.97 (0.09) 2.30 (0.13)⁎
SoFAASd (20) 9.40 (0.30) 8.61 (0.26)⁎ 7.83 (0.40) 5.44 (0.26)⁎ 9.46 (0.29) 8.64 (0.36) 11.73 (0.24) 11.93 (0.28)
BHEI-R total (100) 54.99 (0.42) 56.42 (0.50)⁎ 52.03 (0.59) 50.33 (0.42)⁎ 54.87 (0.46) 56.47 (0.66)⁎ 60.73 (0.51) 62.78 (0.57)⁎
Open in a new tab

Note: Adolescents (12–19 years old), adults (20–59 years old) and older adults (60 years old or over).

SE – Standard Error.

⁎

Wald test, p < 0.05.

a

Weighted mean.

b

DGOV&L = Dark-green and orange vegetables and legumes.

c

Oils = Vegetable oils, seed oils and oily fish.

d

SoFAAS = total calories from solid fat, alcohol and added sugar.

Age group stratification was carried out to better understand dietary changes. The adolescents revealed deterioration in diet quality influenced by a decrease in scores for dark-green and orange vegetables and legumes (DGOV&L), total grain, oils and SoFAAS components. Milk and dairy products and sodium were the only components that showed improvement in this age group. For adults and older adults, diet quality showed improvement, influenced by an increase in scores for total and whole fruit, as well as saturated fat and sodium. Sodium was the only component that showed improvement across all age groups (Table 2).

In addition to age group, a difference in mean BHEI-R was observed for lifestyle characteristics such as smoking habits. Former smokers had better diet quality, followed by non-smokers, on the 2003 survey, as did non-consumers of alcoholic beverages (Table 3). The BHEI-R showed an inverse correlation with energy intake (r = − 0.207 in 2003 and r = − 0.260 in 2008) and with socioeconomic variables such as head of household education (r = − 0.104 in 2003 and r = − 0.132 in 2008). On the 2008 survey, the BHEI-R showed a weak positive correlation with per capita income (r = 0.045).

Table 3.

Descriptive statistics of the Brazilian Healthy Eating Index Revised (BHEI-R) according to sociodemographic, anthropometric and lifestyle characteristics. ISA 2003 and 2008. São Paulo, SP, 2003–08.

Variables ISA 2003
 ISA 2008
Meana (SE) pb Meana (SE) pb
Sex
 Male 54.25 (0.46) 0.051 55.59 (0.66) 0.059
 Female 55.60 (0.59) 57.14 (0.62)
Age group (years)
 12 to 19b 52.03 (0.59) < 0.001 50.33 (0.41) < 0.001
 20 to 59 54.87 (0.46) 56.47 (0.66)
 60 or older 60.73 (0.51) 62.78 (0.57)
Racec
 White 55.28 (0.44) 0.294 56.93 (0.61) 0.106
 Non white 54.56 (0.65) 55.70 (0.61)
Nutritional statusc
 Not overweight 54.49 (0.52) 0.312 55.89 (0.61) 0.090
 Overweight 55.27 (0.61) 57.31 (0.68)
Leisure-time physical activityc, d
 Activeb 55.23 (1.07) 0.870 56.57 (0.54) 0.436
 Not activeb 55.04 (0.45) 55.50 (1.29)
Smoking habitsc
 Never smoked 55.15 (0.46) < 0.001 55.77 (0.37) < 0.001
 Former smoker 57.39 (0.86) 61.16 (0.65)
 Smoker 52.70 (0.72) 54.69 (0.81)
Alcoholic beverage consumptionc
 No 55.65 (0.52) 0.029 57.01 (0.56) 0.193
 Yes 54.54 (0.51) 55.87 (0.75)
Open in a new tab

SE – Standard Error.

a

eighted mean.

b

Wald test or One-way analysis of variance.

c

Exclusion of individuals with ignored information.

d

Active individual practicing at least 30 min of low or moderate intensity physical activity daily, five or more days of the week, or practicing at least 20 min of vigorous intensity physical activity daily, 3 or more days of the week.

The modeling in the multiple linear regression analysis (Table 4) revealed that adults in 2003 were associated with a BHEI-R score that was 2.5 points higher compared to adolescents. This difference was even greater for older adults, whose mean score was 7.2 points higher on the index. On the other hand, smoker status was associated with an index which was lower, on average, by 3 points, regardless of energy intake, age, or head of household education. Comparing 2008, a greater difference was found in index scores for age, with an average increase of 6.1 and 10.7 points among adults and older adults, respectively, compared with adolescents.

Table 4.

Factors associated with the Brazilian Healthy Eating Index Revised identified by multiple regression analysis for participants of the ISA 2003 and 2008. São Paulo, SP, 2003–08.

Independent variable 2003
 2008
β p r2 β p r2
Energy intake (Kcal) − 0.0021 < 0.001 0.089 − 0.0013 0.002 0.099
Age group Adultsa 2.4469 < 0.001 6.1154 < 0.001
Older adultsa 7.1968 < 0.001 10.7202 < 0.001
Smoking habits Smokerb − 2.9523 < 0.001 − 3.0910 0.003
Former smokerb 1.0748 0.241 0.5755 0.610
Head of household education (years) − 0.0664 0.032 − 0.1550 0.155
Open in a new tab

Note: Adolescents (12–19 years old), adults (20–59 years old) and older adults (60 years old or over).

a

In relation to adolescents.

b

In relation to never smoked.

4. Discussion

The study population showed increased scores on the BHEI-R but these remain worrying, falling short of scores obtained by other countries with high morbidity and mortality rates due to diseases related to inadequate eating habits, such as the United States (USDA, 2009). This concern is greatest among adolescents, who had the worst diet quality compared with the other age groups, a gap that has been widening.

Among the groups assessed, only the adolescent group showed a decline in scores on BHEI-R components. Despite a decline in scores for total grains and oils, scores remained close to expected values. The score for SoFAAS components, already relatively low (7.8) in 2003, became even lower (5.4), given the maximum score for this component is 20 points.

The points attributed to SoFAAS components are for the energy value derived from solid fat, saturated and trans fats, alcohol and added sugar intake. There are several mechanisms linking the consumption of these items to health problems, including: increased concentration of triglycerides and decreased concentration of high density lipoproteins (Fried and Rao, 2003).

Changes in dietary structure, associated with economic, social and demographic shifts and their repercussions on the population's health, have been observed in a number of developing countries (Popkin, 2001), including Brazil, and resemble the eating patterns seen in developed countries. Data from the 2003 NHANES, highlight that about 40% of the total energy intake of American children and adolescents from 2 to 18 years old was in the form of “discretionary calories”, and half of these calories were from soft drinks, fruit juice, dairy desserts, pizzas and whole milk (Reedy and Krebs-smith, 2010).

In the city of São Paulo, 70% of adolescents consume > 10% of their diet total energy value (TEV) in the form of added sugar, with soft drinks contributing most to this percentage (Colucci et al., 2012). Data from the HBS performed by the Brazilian Institute of Geography and Statistics (IBGE) confirms a tendency of increased availability of soft drinks within households, which has risen by 400% in the metropolitan areas of Brazil over a 30-year period (1974–2003) (Levy-Costa et al., 2005).

In the present study, half of the sample had excess body weight and, although the number of physically active had doubled, exercise prevalence was only 14% in 2008 (Table 1).

Several studies have shown the association of lifestyle variables with healthy eating (Lin et al., 2013, Andrade et al., 2010). Both of the ISA surveys revealed that former smokers had better diet quality (Table 3), possibly due to their commitment to improve lifestyle habits. Furthermore, smoker status was associated with a 3-point lower BHEI-R in both survey years, regardless of age and energy intake (Table 4).

The consumption of alcoholic beverages is also a growing concern, and studies have shown some tendencies in alcohol consumption at earlier ages (Gigliotti and Bessa, 2004, Galduroz et al., 2004). Alcohol consumption contributes to decrease SoFAAS component scores. Almost half of the interviewed adolescents have reported drinking alcohol, but an increase consumption along the five analyzed years was not observed (39.7%; 95% CI 34.57–45.06 in 2003 and 34.1%; 95% CI 28.97–39.53 in 2008) Besides the harmful effects of alcohol on health, data from NHANES of 2003–2008 identified poorer diet quality among males and females on days when they consumed alcoholic beverages, highlighting a higher consumption of total fat and lower intake of fruit, milk and dairy products (Breslow et al., 2013). This difference in diet quality among consumers and non-consumers of alcoholic beverages was also found in the population of São Paulo on the ISA 2003 (Table 3).

Monitoring the intake of milk and dairy products is essential, not only because of the protein value but also because it provides calcium (Miller et al., 2001). Martini et al. (2013) showed a prevalence of inadequate calcium intake of over 70% in the ISA 2008 sample. Despite improved milk and dairy product scores for the adolescent group, values were under half those recommended by the BHEI-R and well below the score reported for the American adolescent population (7.7 points) (USDA, 2009).

Akin to milk and dairy products, other components of the BHEI-R are of concern due to their insufficient intake. The consumption of fruit and vegetables has a protective effect on risk of obesity, diabetes mellitus type 2, cardiovascular diseases and some types of cancer. The World Health Organization (WHO) reported that an estimated 2.7 million of the deaths registered worldwide in 2000 could have been prevented with adequate consumption of fruit and vegetables (Lock et al., 2005). Table 2 shows that scores for total fruit and whole fruit increased among adults and older adults but not in adolescents.

Studies with the HEI indicate lower scores for the vegetable component registered by the Americans when compared to the Brazilian population (3.2 vs. 3.8 points) (NHANES 2001–04/ISA 2003). The same is true when we compare adolescents (2.4 vs. 3.8 points) but the scores are similar for the older adults (4.3 points).

Authors reported unfavorable trends in consumption of vegetables by Spanish (Valdés et al., 2009) but the vegetable consumption increased in Portuguese population; this finding is in agreement with other studies conducted in Sweden (Eiben et al., 2004) and New Zealand (Laugesen and Swinburn, 2000).

The control of the deficiencies of vitamins and minerals is a priority in public health, for this reason, the Dietary Guidelines for Brazilians Population recommends varying the fruits and vegetables, increasing the supply of fiber and vitamins, such as carotenoids, emphasizing the consumption of dark-green and orange vegetables (Ministério da Saúde, 2005). The score to dark-green and orange vegetables and legumes component also declined in adolescents group (3.2 vs. 2.9 points).

The increase in sodium component scores was observed in all age groups. Nevertheless, the values remain very low, and represent half of the scores registered by the American population (2003 NHANES) (2.12 vs. 4.1 points) (USDA, 2008).

According to the Institute for Applied Economic Research - IPEA (2009), in the period between 2003 and 2008 there was an average increase of 28.2% in Brazilian average income, influencing the availability of different food groups. According to data from the 2008 HBS (IBGE, 2011), food groups whose presence in the diet rose uniformly, along with the level of household income included: meats, milk and dairy products, fruits and vegetables. On the other hand, some negative markers of diet quality have been observed among higher income categories, including sweet products, soft drinks, pizza and baked pastries. The presence of soft drinks in the diet was five times greater in the higher income than in the lower income classes.

Data from the 2003 survey showed that the head of the household's education (proxy for income) was inversely associated with diet quality, a relationship that did not remain in 2008. In a study of trends in diet quality from 1999 to 2010 among the US adult population, using the Healthy Eating Index 2010, Wang et al. (2014) showed that family income and education level were positively associated with the index, and the gap between low and high socioeconomic status had widened over time, from 3.9 points in 1999 to 7.8 points in 2000.

In the present study, an increase in diet quality score was observed among adults and older adults, mainly due to increased fruit intake and decreases in saturated fats and sodium consumption. However, consumption for all food groups was inadequate and remains a concern, with low scores in relation to total score. This concern is greatest among the adolescent group, which had the worst diet quality and exhibited a tendency for further declines in this score, as evidenced in the final model of multiple linear regression analysis.

Adolescents are making poorer food choices compared to other age groups. Studies show that adolescents prefer processed foods with high sugar and fat content, affecting health during the development phase (USDA, 2009, Andrade et al., 2010). The latest 2008 HBS (IBGE, 2010) confirms the significant increase in the proportion of Brazilian adolescents overweight: in 1974–75, 3.7% of boys and 7.6% of girls aged between 10 and 19 years were overweight; in 2008–09, these rates were 21.7% and 19.4%, respectively.

This study provides evidence of the urgent need for actions aimed at improving diet quality across all income classes, including families whose head of household has a high educational level, and with a special focus on adolescents. These actions will determine eating behaviors in adult life, preventing the individual from developing future non-communicable chronic diseases.

The limitation of the BMI should be considered. BMI is a surrogate measure of body fatness because it is a measure of excess weight rather than excess body fat. Factor such as muscle mass can influence the BMI and change the nutritional status. However, the use of anthropometric measurements has become, albeit with limitations, the most practical and lowest cost means of analyzing individuals and populations, whether for clinical reasons, for screening projects or for monitoring tendencies (World Health Organization (WHO), 1998, Conde and Monteiro, 2006).

Other limitation of this study was evaluating consumption based on one day only. Food consumption may vary from one day to the next, and the evaluation of a single day may not reflect the individual's usual intake. Nevertheless, the 24HR provides details on all types of food consumed and their values, serving as a valuable tool for population-based surveys, such as the present study. Willett (2013) claims that a single 24HR may be adequate to estimate the mean value of consumption when the size and characteristics of the sample are estimated for this purpose.

These data may support the devising of actions to promote consumption of specific food groups, such as fruit and vegetables, milk and dairy products, and whole grains, while also encourage reductions in the use of added sugar and sodium by food manufacturers and in food preparation. In addition, the results can help raise awareness of the need for public actions promoting better lifestyle habits, such as programs encouraging the practice of regular physical activities, and anti-smoking and anti-alcohol campaigns.

Possibilities for further research include updating of the BHEI-R to reflect current dietary guidance and to continue the monitoring of the diet quality in future population-based studies. Research about frequency of eating out, changes in portion sizes, changes in dietary patterns and changes in the intake of energy-dense foods of low nutritional value may shed light on the observed trend.

Financial support

São Paulo Research Foundation (FAPESP - processes 2007/51488–2 and 2009/15831-0) and The Brazilian National Council for Scientific and Technological Development (CNPq - processes 502948/2003-5 and 481176/2008-0). The funding bodies had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Footnotes

☆

We note that there are no prior publications or submissions with any overlapping information, including studies and patients. The manuscript has not been and will not be submitted to any other journal while it is under consideration by The Preventive Medicine Reports and the authors declare no conflict of interest.

☆☆

The author Samantha C. de Andrade was the person who wrote the first draft of the manuscript. No payment was given to anyone to produce the manuscript. Each author's contributions were detailed below. All authors read and approved the final manuscript.

★

This work had financial support from: São Paulo Research Foundation (FAPESP - procedural 2007/51488-2 and 2009/15831-0) and National Counsel of Technological and Scientific Development (CNPq - procedural 502948/2003-5 and 481176/2008-0). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

  1. Andrade S.C., Barros M.B.A., Carandina L. Dietary quality index and factors among adolescents of the state of Sao Paulo, Brazil. J. Pediatr. 2010;156(3):456–460. doi: 10.1016/j.jpeds.2009.09.066. [DOI] [PubMed] [Google Scholar]
  2. Andrade S.C., Previdelli A.N., Marchioni D.M.L. Avaliação da confiabilidade e validade do Índice de Qualidade da Dieta Revisado. Rev. Saúde Pública. 2013;47(4):675–683. doi: 10.1590/S0034-8910.2013047004267. [DOI] [PubMed] [Google Scholar]
  3. Ballard-Barbash R. Designing surveillance systems to address emerging issues in diet and health. J. Nutr. 2001;131:S437–S439. doi: 10.1093/jn/131.2.437S. [DOI] [PubMed] [Google Scholar]
  4. Brasil. Ministério da Saúde. Secretaria de Atenção à Saúde . Ministério da Saúde; Brasília: 2005. Coordenação-Geral da Política de Alimentação e Nutrição. Guia alimentar para a população brasileira: promovendo a alimentação saudável. 236 pp. – (Série A. Normas e Manuais Técnicos) [Google Scholar]
  5. Breslow R.A., Chen C.M., Graubard B.I. Diets of drinkers on drinking and nondrinking days: NHANES 2003–2008. Am. J. Clin. Nutr. 2013;97(5):1068–1075. doi: 10.3945/ajcn.112.050161. May. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Castro M.A.d., Barros R.R., Bueno M.B. César CLG, Fisberg RM. Trans fatty acid intake among the population of the city of São Paulo, Southeasthern Brazil. Rev Saúde Pública. 2009;43:991–997. doi: 10.1590/s0034-89102009005000084. [DOI] [PubMed] [Google Scholar]
  7. Center for Disease Control and Prevention CDC growth charts. 2012. http://www.cdc.gov/growthcharts/ Available at: (Accessed on: May 24)
  8. Colucci A.C.A., Cesar C.L.G., Marchioni D.M.L. Factors associated with added sugars intake among adolescents living in São Paulo, Brazil. J. Am. Coll. Nutr. 2012;31(4):259–267. doi: 10.1080/07315724.2012.10720429. [DOI] [PubMed] [Google Scholar]
  9. Conde W.L., Monteiro C.A. Valores críticos do índice de massa corporal para classificação do estado nutricional de crianças e adolescentes brasileiros. J. Pediatr. 2006;82(4):266–272. [Google Scholar]
  10. Conway J.M., Ingwersen L.A., Moshfegh A.J. Accuracy of dietary recall using the USDA five-step multiple-pass method in men: an observational validation study. J. Am. Diet. Assoc. 2004;104:595–603. doi: 10.1016/j.jada.2004.01.007. [DOI] [PubMed] [Google Scholar]
  11. Eiben G., Andersson C.S., Rothenberg E., Sundh V., Steen B., Lissner L. Secular trends in diet among elderly Swedes – cohort comparisons over three decades. Public Health Nutr. 2004;7:637–644. doi: 10.1079/phn2003576. [DOI] [PubMed] [Google Scholar]
  12. Fisberg R.M., Marchioni D.M.L. Faculdade de Saúde Pública da USP; São Paulo, Brazil: 2012. Manual de Avaliação de Consumo Alimentar em Estudos Populacionais: A Experiência do Inquérito de Saúde em São Paulo (ISA) Available online: http://www.gac-usp.com.br/resources/manual%20isa%20biblioteca%20usp.pdf (accessed on 15 Apr 2016) [Google Scholar]
  13. Fisberg R.M., Villar B.S. Signus; São Paulo: 2002. Manual de receitas e medidas caseiras para cálculo de inquéritos alimentares: manual elaborado para auxiliar o processamento de inquéritos alimentares. [Google Scholar]
  14. Fried S.K., Rao S.P. Sugars, hypertriglyceridemia, and cardiovascular disease. Am. J. Clin. Nutr. 2003;78:S873–S880. doi: 10.1093/ajcn/78.4.873S. [DOI] [PubMed] [Google Scholar]
  15. Galduroz J.C., Noto A.R., Nappo S.A. Trends in drug use among students in Brazil: analysis of four surveys in 1987, 1989, 1993 and 1997. Braz. J. Med. Biol. Res. 2004;37:523–531. doi: 10.1590/s0100-879x2004000400009. [DOI] [PubMed] [Google Scholar]
  16. Gigliotti A., Bessa M.A. Síndrome de dependência do álcool: critérios diagnósticos. Rev. Bras. Psiquiatr. 2004;26(Suppl. I):S11–S13. [PubMed] [Google Scholar]
  17. Hu F.B. Dietary pattern analysis: a new direction in nutritional epidemiology. Curr. Opin. Lipidol. 2002;13(1):3–9. doi: 10.1097/00041433-200202000-00002. [DOI] [PubMed] [Google Scholar]
  18. Instituto Brasileiro de Geografia e Estatística (IBGE) IBGE; Rio de Janeiro: 2010. Coordenação de Trabalho e Rendimento. Pesquisa de orçamentos familiares 2008–2009: antropometria e estado nutricional de crianças, adolescentes e adultos no Brasil. 130 pp. [Google Scholar]
  19. Instituto Brasileiro de Geografia e Estatística (IBGE) IBGE; Rio de Janeiro: 2011. Coordenação de Trabalho e Rendimento. Pesquisa de orçamentos familiares 2008–2009: análise do consumo alimentar pessoal no Brasil. 150 pp. [Google Scholar]
  20. Instituto de Pesquisa Econômica Aplicada . Instituto de Pesquisa Econômica Aplicada; Rio de Janeiro: 2009. IPEA DATA. Available at http://www.ipeadata.gov.br. [Google Scholar]
  21. Kovalskys I., Fisberg M., Gómez G. Standardization of the food composition database used in the Latin American Nutrition and Health Study (ELANS) Nutrients. 2015;7:7914–7924. doi: 10.3390/nu7095373. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Laugesen M., Swinburn B. The New Zealand food supply and diet-trends 1961–1995 and comparison with other OECD countries. Organisation for economic co-operation and development. N. Z. Med. J. 2000;113:311–315. [PubMed] [Google Scholar]
  23. Levy R.B., Claro R.M., Mondini L. Regional and socioeconomic distribution of household food availability in Brazil, in 2008–2009. Rev. Saúde Pública São Paulo. 2012;46(1) doi: 10.1590/s0034-89102011005000088. [DOI] [PubMed] [Google Scholar]
  24. Levy-Costa R.B., Sichieri R., Pontes N.D.S. Disponibilidade domiciliar de alimentos no Brasil: distribuição e evolução (1974–2003) Rev. Saúde Pública São Paulo. 2005;39(4) doi: 10.1590/s0034-89102005000400003. [DOI] [PubMed] [Google Scholar]
  25. Lin C.T., Gao Z., Lee J.Y. Associations between self-reported weight management methods with diet quality as measured by the Healthy Eating Index–2005. Prev. Med. 2013 doi: 10.1016/j.ypmed.2013.05.026. Jun. Available at. [DOI] [PubMed] [Google Scholar]
  26. Lipschitz D.A. Screening for nutritional status in the elderly. Prim. Care. 1994;21(1):55–67. [PubMed] [Google Scholar]
  27. Lock K., Pomerleau J., Altmann D.R. The global burden of disease attributable to low consumption of fruit and vegetables: implications for the global strategy on diet. Bull. World Health Organ. 2005;83(2):100–108. [PMC free article] [PubMed] [Google Scholar]
  28. Martini L.A., Verly E., Marchioni D.M.L. Prevalence and correlates of calcium and vitamin D status adequacy in adolescents, adults, and elderly from the health survey São Paulo. Nutrition. 2013;29(6):845–850. doi: 10.1016/j.nut.2012.12.009. Jun. [DOI] [PubMed] [Google Scholar]
  29. Miller G.D., Jarvis J.K., McBean The importance of meeting calcium needs with foods. J. Am. Coll. Nutr. 2001;20(2):S168–S185. doi: 10.1080/07315724.2001.10719029. [DOI] [PubMed] [Google Scholar]
  30. Monteiro C.A., Benicio M.H.D.'’.A., Conde W.L. Narrowing socioeconomic inequality in child stunting: the Brazilian experience, 1974–2007. Bull. World Health Organ. Genebra. 2010;88(4) doi: 10.2471/BLT.09.069195. Apr. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. NEPA-UNICAMP . Revisada e ampliada. 4. ed. NEPA-UNICAMP; Campinas: 2011. Tabela brasileira de composição de alimentos/Núcleo de Estudos e Pesquisa em Alimentação – Universidade Estadual de Campinas; p. 161. [Google Scholar]
  32. Pinheiro A.B.V., Lacerda E.M.A., Benzecry E.H. 4.ed. Atheneu; São Paulo: 2000. Tabela para avaliação de consumo alimentar em medidas caseiras. [Google Scholar]
  33. Popkin B.M. The nutrition transition and obesity in the developing world. J. Nutr. 2001;131:871–873. doi: 10.1093/jn/131.3.871S. [DOI] [PubMed] [Google Scholar]
  34. Previdelli A.N., Andrade S.C., Pires M.M. Índice de Qualidade da Dieta Revisado para população brasileira. Rev. Saúde Pública São Paulo. 2011;45(4) Aug. [Google Scholar]
  35. Reedy J., Krebs-smith S.M. Dietary sources of energy, solid fats, and added sugars among children and adolescents in the United States. J. Am. Diet. Assoc. 2010;110:1477–1484. doi: 10.1016/j.jada.2010.07.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. USDA . Vol. 37. Center for Nutrition Policy and Promotion. U.S. Department of Agriculture; 2008. Diet quality of Americans in 1994–96 and 2001–02 as measured by the Healthy Eating Index-2005. (Nutrition Insight). Available at http://www.cnpp.usda.gov. [Google Scholar]
  37. USDA . Vol. 43. Center for Nutrition Policy and Promotion. U.S. Department of Agriculture; 2009. The quality of children's diets in 2003–04 as measured by the Healthy Eating Index-2005. (Nutrition Insight). Available at http://www.cnpp.usda.gov. [Google Scholar]
  38. USDA . Vol. 51. Center for Nutrition Policy and Promotion. U.S. Department of Agriculture; 2013. diet quality of Americans in 2001–02 and 2007–08 as measured by the Healthy Eating Index 2010. (Nutrition Insight). Available at http://www.cnpp.usda.gov. [Google Scholar]
  39. Valdés J., Grau M., Subirana I. Secular trends in energy intake and diet quality in a Mediterranean population. Ann. Nutr. Metab. 2009;54:177–183. doi: 10.1159/000217814. [DOI] [PubMed] [Google Scholar]
  40. Wang D.D., Leung C.W., Li Y. Trends in dietary quality among adults in the United States, 1999 through 2010. JAMA Intern. Med. 2014;174(10):1587–1595. doi: 10.1001/jamainternmed.2014.3422. Oct. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Willett W.C. third ed. Oxford University Press; New York: 2013. Nutritional Epidemiology. [Google Scholar]
  42. World Health Organization (WHO) Vol. 894. WHO; Geneva: 1998. Obesity: preventing and managing the global epidemic. (Technical Report Series). [PubMed] [Google Scholar]

Articles from Preventive Medicine Reports are provided here courtesy of Elsevier

RESOURCES