Scholarly article on topic 'Calcium intake by adolescents: a population-based health survey'

Calcium intake by adolescents: a population-based health survey Academic research paper on "Health sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Jornal de Pediatria
OECD Field of science
Keywords
{Adolescent / "Calcium dietary" / "Food consumption" / "Health surveys" / Adolescente / "Cálcio na dieta" / "Consumo de alimentos" / "Inquérito de saúde"}

Abstract of research paper on Health sciences, author of scientific article — Daniela de Assumpção, Marcia Regina Messaggi Gomes Dias, Marilisa Berti de Azevedo Barros, Regina Mara Fisberg, Antonio de Azevedo Barros Filho

Abstract Objective To analyze calcium intake in adolescents according to sociodemographic variables, health-related behaviors, morbidities, and body mass index. Methods This was a cross-sectional population-based study, with a two-stage cluster sampling that used data from a survey conducted in Campinas, São Paulo, Brazil, between 2008 and 2009. Food intake was assessed using a 24-hour dietary recall. The study included 913 adolescents aged 10–19 years. Results Average nutrient intake was significantly lower in the segment with lower education of the head of the family and lower per capita family income, in individuals from other cities or states, those who consumed fruit less than four times a week, those who did not drink milk daily, those who were smokers, and those who reported the occurrence of headaches and dizziness. Higher mean calcium intake was found in individuals that slept less than seven hours a day. The prevalence of calcium intake below the recommendation was 88.6% (95% CI: 85.4–91.2). Conclusion The results alert to an insufficient calcium intake and suggest that certain subgroups of adolescents need specific strategies to increase the intake of this nutrient.

Academic research paper on topic "Calcium intake by adolescents: a population-based health survey"

J Pediatr (Rio J). 2015;xxx(xx):xxx-xxx

Jornal de

Pediatría

www.jped.com.br

ORIGINAL ARTICLE

Calcium intake by adolescents: a population-based health survey^

Daniela de Assumpcäo3, Marcia Regina Messaggi Gomes Diasb, Marilisa Berti de Azevedo Barros3, Regina Mara Fisbergc, Antonio de Azevedo Barros Filhob*

a Department of Collective Health, Faculdade de Ciencias Médicas, Universidade Estadual de Campinas (UNICAMP), Campinas, SP, Brazil

b Department of Pediatrics, Faculdade de Ciencias Médicas, Universidade Estadual de Campinas (UNICAMP), Campinas, SP, Brazil c Department of Nutrition, Faculdade de Saúde Pública, Universidade de Sao Paulo (USP), Sao Paulo, SP, Brazil

Received 17 April 2015; accepted 25 September 2015

KEYWORDS

Adolescent; Calcium dietary; Food consumption; Health surveys

Abstract

Objective: To analyze calcium intake in adolescents according to sociodemographic variables, health-related behaviors, morbidities, and body mass index.

Methods: This was a cross-sectional population-based study, with a two-stage cluster sampling that used data from a survey conducted in Campinas, Sao Paulo, Brazil, between 2008 and 2009. Food intake was assessed using a 24-hour dietary recall. The study included 913 adolescents aged 10-19 years.

Results: Average nutrient intake was significantly lower in the segment with lower education of the head of the family and lower per capita family income, in individuals from other cities or states, those who consumed fruit less than four times a week, those who did not drink milk daily, those who were smokers, and those who reported the occurrence of headaches and dizziness. Higher mean calcium intake was found in individuals that slept less than seven hours a day. The prevalence of calcium intake below the recommendation was 88.6% (95% CI: 85.4-91.2). Conclusion: The results alert to an insufficient calcium intake and suggest that certain subgroups of adolescents need specific strategies to increase the intake of this nutrient. © 2015 Sociedade Brasileira de Pediatria. Published by Elsevier Editora Ltda. All rights reserved.

* Please cite this article as: de Assumpcâo D, Dias MRMG, de Azevedo Barros MB, Fisberg RM, de Azevedo Barros Filho A. Calcium intake by adolescents: a population-based health survey. J Pediatr (Rio J). 2015. http://dx.doi.org/10.1016/j.jped.2015.09.004

* Corresponding author.

E-mail: abarros@fcm.unicamp.br (A. de Azevedo Barros Filho).

http://dx.doi.org/10.1016/j.jped.2015.09.004

0021-7557/© 2015 Sociedade Brasileira de Pediatria. Published by Elsevier Editora Ltda. All rights reserved.

de Assumpçâo D et al.

Ingestâo de cálcio por adolescentes: inquérito de saúde de base populacional Resumo

Objetivos: Analisar a ingestâo de cálcio em adolescentes segundo variáveis socio demográficas, de comportamentos relacionados à saúde, morbidades e índice de massa corporal. Métodos: Trata-se de estudo transversal de base populacional, com amostra por conglomerados, tomada em dois estágios e que utilizou dados de inquérito realizado em Campinas, Sâo Paulo, Brasil, em 2008/09. O consumo alimentar foi estimado pelo Recordatorio de 24 horas. Foram analisados 913 adolescentes de 10 a 19 anos.

Resultados: Médias significativamente inferiores de ingestâo do nutriente foram verificadas nos segmentos de menor escolaridade do chefe da familia, de menor renda familiar per capita, nos naturais de outros municipios ou Estados, nos que consomem frutas menos que quatro vezes na semana, nos que nâo bebem leite diariamente, nos fumantes e nos que referiram presenca de dor de cabeca e tontura. Média superior de ingestâo de cálcio foi encontrada nos individuos que dormem menos de sete horas por dia. A prevalência de ingestâo de cálcio inferior ao recomendado foi de 88,6% (IC95%: 85,4-91,2).

Conclusôes: Os resultados deste estudo alertam para o consumo insuficiente de cálcio e sugerem que determinados subgrupos de adolescentes necessitam de estratégias mais especificas para aumentar a ingestâo deste nutriente.

© 2015 Sociedade Brasileira de Pediatria. Publicado por Elsevier Editora Ltda. Todos os direitos reservados.

PALAVRAS-CHAVE

Adolescente; Cálcio na dieta; Consumo de alimentos; Inquérito de saúde

Introduction

Calcium is an essential nutrient for maintaining bone health, as it contributes to bone mineralization and rigidity, thus preventing problems such as osteoporosis and fractures in adulthood and old age.1,2 It also participates in fibrin formation during the blood coagulation process and regulation of muscle contraction, since the protein troponin, a regulator of actin and myosin contractility, depends on calcium.3

Adequate calcium intake is essential for the development and maintenance of bone mass peak during adolescence.1,2 From birth to adulthood, bone mass increases 40 times, and the peak is reached at the end of the second decade of life.2 Between 40% and 60% of bone mass increase occurs during adolescence.2

The nutritional recommendation of calcium intake for children and adolescents aged 9-18 years is of 1100 mg/day, according to the estimated average requirements (EAR), which should not exceed the tolerable upper intake level of 3000 mg/day.4 According to the 2008-2009 Household Budget Survey (Pesquisa de Orcamentos Familiares [POF]), the highest mean values of calcium intake were 565.7 mg in boys aged 14-18 years and 521.7mg in girls aged 10-13 years.5

Dairy products, dark green vegetables, certain types of fish, and nuts are important dietary sources of this mineral.6 Brazilian research has shown that the dietary patterns of adolescents is characterized by low consumption of certain foods, such as milk and other dairy products, fruit, and vegetables, and by the increased intake of high-calorie foods, saturated fats, sugars, and sodium, such as sugary drinks, sweets, and cookies.7"10 This dietary pattern may be harming the consumption of calcium-rich foods.

Considering the importance of the nutrient intake for good health and the results of studies that evidence a severely inadequate calcium intake by adolescents, this

study aimed to evaluate the epidemiological profile of calcium intake in adolescents in the city of Campinas, state of Sao Paulo, Brazil, and to identify which segments are more susceptible to calcium intake deficiency, according to demographic and socioeconomic variables, health-related behaviors, morbidities, and body mass index (BMI).

Methods

This was a cross-sectional, population-based study that included 929 non-institutionalized adolescents aged 10-19 years, living in the urban area of the city of Campinas. The study data were obtained from the Campinas Health Survey (Inquérito de Saúde no municipio de Campinas - ISACamp 2008/09), carried out between February of 2008 and March of 2009.

The study sample was determined by probabilistic sampling procedures by clusters, and in two stages: census sectors and households. In the first stage, 50 census sectors were selected by drawing lots, with probability proportional to size (number of households). In the second stage, the households were selected by drawing lots.

The sample size was calculated considering the estimated prevalence of 50% (corresponding to the maximum variability), with a confidence level of 95%, sampling error between 4% and 5%, and a second design effect, totaling 1000 adolescents (10-19 years). Expecting 20% of non-response, the sample size was corrected to 1250. To achieve this number of individuals, 2150 households were randomly selected for interviews with the adolescents.

Information was collected through a questionnaire structured in thematic blocks that were tested in a pilot study and applied by trained and supervised interviewers. Dietary intake was estimated using the 24-hour food recall (24HR).

+Model

Calcium intake by adolescents 3

116 During the fieldwork, the content of 24HR was checked to morbidity variables that had p< 0.20 in the bivariate analysis 172

117 identify and resolve the incorrect filling out of forms. were added to the model and those that remained at the 5% 173

118 A 24HR quantification was performed to convert the significance level were maintained. The model was adjusted 174

119 amounts of foods or preparations mentioned in household for total caloric intake, as recommended by Willett et al.16 175

120 measures into grams or milliliters. For this purpose, the The GLM allows for the assumption of distributions other 176

121 authors used information available in household measure than the normal for the response variable, and provides 177

122 tables,11,12 food labels, and customer care services. The greater flexibility for the functional association between 178

123 intake information was inserted into the Nutrition Data Sys- the mean of the response variable and the linear predic- 179

124 tem for Research software database (NDS-R, version 2007, tor. Distributions modeled by GLM belong to the exponential 180

125 University of Minnesota, USA). The analyses excluded ado- family, including distributions for both continuous and mild 181

126 lescents with caloric intake lower than 600kcal/day and response variables.17 The graphical analysis and statistical 182

127 greater than 6000kcal/day.13 hypothesis testing demonstrated that the gamma distribu- 183

tion was appropriate to model the calcium intake. 184

Study variables The interviews were entered into the database developed 185

128 using Epidata 3.1 (Epidata Assoc., Odense, Denmark) and the 186

statistical analyses were performed using Stata 11.0 (Stata 187

129 Depended variable: mean caldum mtake (mg/day) Corp., College Station, USA), svy module, which takes into 188

account the weights and the study sampling design. 189

130 The prevalence of inadequate nutrient intake was estimated The present study was approved by the Research Ethics 190

131 using the EAR as the cutoff point, which corresponds to Committee of Universidade Estadual de Campinas and 191

132 110°mg and 800mg for adolescents aged 10-18 years and the National Research Ethics Committee, under CAAE No. 192

133 19year^ respectively.4 37303414.4.0000.5404. For adolescents younger than 18 193

134 The following sets of independent variables were years, the consent form was signed by their parents or 194

135 selected: guardians. 195

136 • Demographic and socioeconomic information: gender,

137 age (years), ethnicity/skin color (self-reported), number

138 of household members, place of birth, educational level Results 196

139 of the head of the family (years), employment, private

140 health insurance, school attendance by the adolescent, Of the total of 929 interviews, five were excluded due to 197

141 number of appliances in the household, and per capita refusal in filling out the 24HR and 11 for having a daily 198

142 family income (in Brazilian minimum wages). caloric intake <600kcal or >6000kcal. Therefore, 913 ado- 199

143 • Health-related behaviors: smoking, frequency of weekly lescents were assessed, with a mean age of 14.1 years (95% 200

144 consumption of alcohol, fruit, raw vegetables, and milk CI: 13.8-14.4). 201

145 obtained by the food frequency questionnaire; physical Calcium intake was significantly lower in girls, in ado- 202

146 activity in the context of leisure, categorized as: active lescents belonging to the lower per capita income strata, 203

147 (adolescents who practiced at least 150 minutes a week, and in those whose head of the family had lower educa- 204

148 distributed in at least three days), insufficiently active tional level, those without private health insurance, and 205

149 (those who practiced less than 150 minutes a week or those who had fewer household appliances in the residence. 206

150 more, but in fewer than three days a week) and sedentary Compared to adolescents from private schools, those who 207

151 (those who did not practice any kind of leisure physi- did not attend school or who attended public schools had 208

152 cal activity on any day of the week)14; and number of lower mean calcium intake (Table 1). 209

153 hours/day intended for sleep and computer use. Table 2 shows that individuals who consumed fruit and 210

154 • Morbidities and BMI: presence of headache/migraine, milk less than four times a week had a lower intake of the 211

155 dizziness, number of chronic diseases, and number of nutrient, which was also significantly lower among those 212

156 health complaints among those included in the check- who did not consume raw vegetables daily, in smokers, 213

157 list. BMI was calculated from the self-reported weight those who consumed alcohol two or more times a week, 214

158 and height information. The cutoffs recommended for those who reported nine or more hours of sleep, and those 215

159 adolescents15 were used, with the following classifi- who did not use a computer. 216

160 cations: underweight, normal weight, overweight, and Lower mean calcium intake was found in adolescents who 217

161 obese. reported the presence of headache and dizziness (Table 3). 218

The results in Table 4 indicate a lower mean calcium 219

162 For the study analyses, the mean calcium intake was intake in individuals from other municipalities or states, 220

163 calculated according to the categories of the indepen- those with lower educational level of the head of the 221

164 dent variables. The means and 95% confidence intervals family, those with a family income below a minimum 222

165 were estimated using single and multiple generalized lin- wage, those who consumed fruit at a lower weekly fre- 223

166 ear regression models (GLM). The selection of variables for quency, those who did not drink milk daily, smokers, and 224

167 the model adjustment was performed in two stages. In the those who reported the presence of headaches and dizzi- 225

168 first stage, the demographic and socioeconomic variables ness. Conversely, a higher mean was found in adolescents 226

169 that showed a significance level <0.20 in the bivariate anal- who slept less than seven hours/day when compared with 227

170 ysis were entered, and those with p < 0.05 remained in the those who reported between seven and eight hours of 228

171 model. In the second stage, the health-related behavior and sleep. 229

+Model ARTICLE IN PRESS

4 de Assumpçâo D et al.

Table 1 Mean calcium intake (mg) in adolescents aged 10-19 years, according to sociodemographic variables. Campinas Health Survey (Inquerito de Saude de Campinas [ISACamp]), 2008/09.

Variables and categories n Mean (95% CI) p-Valuea

Gender

Maleb 462 692.3 (636.9-747.7)

Female 451 540.7 (436.4 -645.0) 0.000

Total 913 618.2 (570.8-665.5)

Age range (in years)

10-14b 504 620.3 (567.0-673.5)

15 -19 409 615.5 (496.0-735.0) 0.886

Ethnicity

Whiteb 587 639.7 (595.6-683.8)

Non-white 323 579.0 (461.7-696.4) 0.102

Number of people in household

1- b 220 646.0 (574.7-717.4)

4-6 579 627.3 (489.6-764.9) 0.572

7 or more 114 515.6 (290.6-740.5) 0.094

Origin

Campinasb 692 633.8 (580.9 -686.6)

Another municipality or state 221 569.5 (445.6-693.3) 0.075

Head of family educational level (in years)

0-7 382 528.6 (379.8-677.4) 0.000

8-11 310 625.9 (476.2-775.5) 0.003

12 or moreb 210 759.8 (695.5-824.1)

Per capita income (in minimum wages)

<1 578 564.6 (399.4-729.9) 0.000

>1 to <2 193 646.7 (450.5-842.9) 0.021

>2b 142 785.1 (705.5 -864.8)

Occupational activity

Worksb 148 655.5 (557.8-753.2)

Does not work 754 610.4 (418.5-802.3) 0.341

Has health care insurance

Yesb 307 712.8 (654.6-771.0)

No 601 566.7 (440.7-692.6) 0.000

Attends school

No 143 596.4 (412.9-779.8) 0.019

Yes, public 607 589.7 (437.5-741.8) 0.001

Yes, privateb 162 735.0 (666.4 -803.5)

Number of appliances in household

0-10 420 538.6 (373.1-704.2) 0.000

11 -15 231 643.2 (486.6-799.8) 0.090

16 or moreb 261 718.7 (649.5-787.8)

n, number of individuals in the unweighted sample. a In bold, p value <0.05. b Reference category used for comparison.

The prevalence of inadequate calcium intake was 88.6% for the total population (95% CI: 85.4-91.2), reaching 85.1% in boys (95% CI: 80.6-88.7) and 92.2% in girls (95% CI: 89.1-94.5). In the segment with higher educational level of the head of the family (12 years or more), the values corresponded to 79.8% in the group of adolescents (95% CI: 71.1-86.0), 75.9% in males (95% CI: 65.0-84.4) and 84.5% in females (95% CI: 76.0-90.4) (data not shown).

Discussion 238

This study identified inadequate calcium intake in adoles- 239

cents. It was significantly lower in individuals from other 240

municipalities or states, in the strata of lower educational 241

level of the head of the family, in the lowest per capita 242

income segment, in those who consumed fruit less than 243

four times a week, in those who reported not drinking milk 244

+Model ARTICLE IN PRESS

Calcium intake by adolescents 5

Table 2 Mean calcium intake (mg) in adolescents aged 10-19 years, according to variables of health-related behaviors. Campinas Health Survey (Inquerito de Saude de Campinas [ISACamp]), 2008/09.

Variables and categories n Mean (95% CI) p-Valuea

Fruit consumption

7 times a weekb 246 675.6 (602.5-748.7)

4-6 times a week 189 671.0 (499.0- 842.9) 0.927

<3 times a week 478 567.4 (413.7-721.1) 0.010

Raw vegetable consumption

7 times a weekb 295 658.0 (597.5-718.6)

4-6 times a week 214 581.2 (443.0-719.3) 0.052

<3 times a week 404 608.2 (480.9-735.5) 0.140

Milk consumption

7 times a weekb 552 711.5 (663.8-759.2)

4-6 times a week 94 649.0 (502.5-795.4) 0.209

<3 times a week 267 413.6 (316.8- 510.4) 0.000

Soft drink consumption

7 times a week 207 625.4 (491.1-759.6) 0.581

4-6 times a week 145 668.7 (518.3- 819.2) 0.186

<3 times a weekb 561 602.4 (551.1- 653.8)

Smoking status

Nonsmokerb 873 624.3 (577.0- 671.5)

Ex-smoker 18 591.9 (344.9- 838.8) 0.746

Smoker 22 396.1 (255.0- 537.2) 0.000

Alcohol consumption

Does not drink alcoholb 764 616.4 (564.5- 668.3)

1-4 times a week 118 669.2 (525.6- 812.7) 0.253

2 or more times a week 26 476.7 (333.5- 620.0) 0.004

Physical activity at leisure

Sedentary 278 566.9 (428.4-705.5) 0.013

Insufficiently active 305 610.0 (474.5-745.6) 0.132

Activeb 330 667.7 (607.7-727.7)

Sleep (hours/day)

<7 60 615.3 (443.0-787.5) 0.445

7- b 450 657.9 (597.0-718.8)

9 or more 393 572.9 (446.6 699.2) 0.012

Computer use (hours/day)

0b 437 563.8 (499.5- 628.1)

1-2 298 657.5 (522.0-792.8) 0.011

3 or more 172 688.3 (527.2- 849.4) 0.013

n, number of individuals in the unweighted sample. a In bold, p-value <0.05. b Reference category used for comparison.

daily, in smokers, and in those who reported the presence of headache and dizziness.

The limitations of this study include the application of only one 24HR, which does not reflect the usual intake of adolescents, as it does not detect consumption variability. However, the 24HR is considered an appropriate tool to assess the mean intake of foods and nutrients when applied in population-based samples and on different days of the week and months of the year, as in the ISACamp 2008/09.18 Moreover, the possibility of reverse causality in a cross-sectional study prevents interpreting the results as of cause and effect.

The Dietary Guideline for the Brazilian Population, published in 2014, proposed a comprehensive assessment of diet and its association with health, setting its guidelines based on nutrients, foods, combinations of foods, and preparations.19 The focus of this study was to assess calcium intake, as it is an essential nutrient for the development and maintenance of bone mass throughout life, as well as for the prevention of osteoporosis during childhood and adulthood, and fracture resulting from this condition.2

Among the adolescents, evidence on the low consumption of calcium-rich foods10,20 and on the high intake of sodium20 and soft drinks,8 together with the finding that a significant

260 261 262

6 de Assumpçâo D et al.

Table 3 Mean calcium intake (mg) in adolescents aged 10-19 years, according to morbidities and body mass index (BMI). Campinas Health Survey of Campinas (Inquerito de Saude de Campinas [ISACamp]), 2008/09.

Variables and categories n Mean (95% CI) p-valuea

Headache/migraine

Yes 227 543.4 (431.8-655.0) 0.003

Nob 686 642.7 (594.3-691.1)

Dizziness

Yes 55 490.8 (349.5-632.2) 0.005

Nob 858 626.2 (578.2-674.2)

Number of chronic diseases

0b 739 613.7 (561.5-665.9)

1 or more 170 620.5 (500.6-740.4) 0.840

Number of health complaints

0b 344 638.0 (572.9-703.2)

1 295 627.0 (494.5-759.4) 0.742

2 or more 274 583.9 (443.7-723.9) 0.153

BMI (kg/m2)

Low weight 30 468.2 (260.0-676.4) 0.055

Normal weightb 565 619.8 (566.6-673.0)

Overweight 133 668.5 (521.3-815.6) 0.303

Obesity 81 626.7 (438.3-815.2) 0.918

n, number of individuals in the unweighted sample. a In bold, p value <0.05. b Reference category used for comparison.

portion does not meet the recommended levels of physical activity,21 reinforce the need for performing epidemiological research on calcium intake.

In this study, the mean calcium intake was 692.3mg in boys and 540.7 mg in girls, higher than the values found in the 2008-2009 POF, which were 565.7mg in boys and 521.7mg in girls aged 10-18 years.5 However, it was lower than the mean obtained in a study of 507 students in the city of Ouro Preto, state of Minas Gerais, which observed a consumption of 730.6 mg and 679.4 mg in boys and girls, respectively.22

Calcium intake below the EAR was observed in 88.6% of the adolescents in the city of Campinas. Analyzing data from the National Food Survey (Inquerito Nacional de Alimentacao [INA]) 2008-2009, Veiga et al.20 found an inadequacy prevalence >95.0% in individuals aged between 10 and 18 years. This result was attributed to a low consumption of dairy products, which are the main food sources of the nutrient, as well as its replacement by soft drinks and other sugary

beverages.5,23

In the bivariate analysis, female gender was associated with lower calcium consumption, but it did not remain in the final model, due to the adjustment by caloric consumption (kcal). The caloric intake was significantly lower in girls, being, respectively, in boys was 2715.2 kcal (95% CI: 2522.8-2907.6) and 2277.1 (95% CI: 1977.0-2577.1) in girls. Nonetheless, the authors decided to maintain the gender variable to adjust the model.

When compared with adolescents born in Campinas, those from other municipalities or states had lower calcium intake. Regarding individuals aged 10-18 years, the 2008 -2009 POF showed changes in mean calcium intake

among the major regions of Brazil: the Northeast region presented the lowest values and the Southeast, the highest.5 The findings of this study show a significant increase in calcium intake with the improvement in the educational level of the head of the family. Data from the 2009 National Schoolchildren's Health Survey (Pesquisa Nacional de Saude do Escolar [PeNSE]) demonstrated a positive association between regular consumption of milk (five days a week or more) with maternal educational level.24 Studies that analyzed the quality of diet in adolescents found a higher intake of dairy products and vegetables in the strata with higher educational level of the head of the family.25,26

Lower calcium intake was observed in individuals who lived in households with a monthly per capita income of less than Brazilian one minimum wage. The PeNSE results showed a low proportion of regular milk consumption in individuals from lower socioeconomic classes.24 Based on the 2008-2009 POF, Levy et al.27 identified an increase in the participation of milk and dairy products in the diet associated with an increase in household income. In Pelotas, researchers assessed 2209 adolescents and observed a significant reduction in the frequency of daily consumption of milk with worsening in the socioeconomic level.7

Fruit consumption less than four times a week was associated with lower calcium intake. A population-based study carried out in Sao Paulo found that, of the 812 interviewed adolescents, only 6.5% met the minimum daily recommendation of 400 g of fruit and vegetables.9 According to data from the 2009 PeNSE, 31.5% (95% CI: 30.8-32.2) of the students reported consuming fresh fruit regularly, with no differences between genders and the administrative location of schools.21 The increase in the consumption of fruit

+Model ARTICLE IN PRESS

Calcium intake by adolescents 7

Table 4 Two-step generalized linear model. Campinas Health Survey (Inquerito de Saude de Campinas [ISACamp]), 2008/09.

Variables and categories First stepa (95% CI) p-Valueb Second stepc (95% CI) p-Valueb

Gender Maled Female 328.6 (213.2-443.8) 297.5 (148.7-446.5) 0.068 408.3 399.8 (281.3-535.3) (229.6-570.0) 0.694

Origin Campinasd Another municipality or state 328.6 (213.2-443.8) 279.7 (123.6-435.8) 0.020 408.3 372.4 (281.3-535.3) (216.1-528.7) 0.017

Educational level of the head of the family (in years) 0-7 173.4 (9.25-337.6) 8-11 245.7 (68.2-423.2) 12 or mored 328.6 (213.2-443.8) 0.000 0.010 300.3 344.3 408.3 (125.6-475.0) (158.3-530.2) (281.3-535.3) 0.000 0.034

Per capita income (in minimum wages) <1 > 1 to <2 >2d 241.2 (52.0-430.5) 240.3 (44.7-435.8) 328.6 (213.2-443.8) 0.022 0.032 321.7 355.6 408.3 (117.5-526.0) (145.5-565.8) (281.3-535.3) 0.029 0.209

Fruit consumption 7 times a weekd 4-6 times a week <3 times a week 408.3 392.8 355.9 (281.3-535.3) (210.7-575.0) (183.4-527.4) 0.576 0.025

Milk consumption 7 times a weekd 4-6 times a week <3 times a week 408.3 328.6 215.4 (281.3-535.3) (155.3-502.1) (51.0-379.9) 0.001 0.000

Smoking status Nonsmokerd Ex-smoker Smoker 408.3 453.4 336.1 (281.3-535.3) (211.4-695.3) (151.2-521.0) 0.434 0.016

Sleep (hours/day) <7 7- d 9 or more 506.7 408.3 413.6 (297.2-716.2) (281.3-535.3) (255.7-571.5) 0.020 0.732

Headache/migraine Yes Nod 371.5 408.3 (218.0-525.0) (281.3-535.3) 0.007

Dizziness Yes Nod 332.3 408.3 (153.0-511.6) (281.3-535.3) 0.005

a Adjusted for calories and for demographic and socioeconomic variables. b In bold, p-value <0.05.

c Adjusted for calories and for all the variables of the table. d Reference category used for comparison.

and vegetables is one of the national targets intended to halt the increase of non-communicable chronic diseases, which involves actions aimed at promoting healthy eating in the School Feeding Program, lower prices, and greater supply and production of these foods.28

Individuals who reported drinking milk daily had higher mean levels of the nutrient. Among the participants of PeNSE, 53.6% reported consuming milk five or more days per week.24 Of the 20 foods most often mentioned by the adolescents included in the 2008-2009 POF, the item ''juices and artificial juices'' was in the sixth position (43.5%), while whole milk was in the 18th (12.9%).29 National data show that, between 1974 and 2003, the household availability

of dairy products increased by 36% and of soft drinks, by 400%.23 A cross-sectional study conducted in Canada with 610 children aged 8-10 years observed a decrease in the consumption of sugary drinks with the increased consumption of dairy products, as well as a significant reduction in systolic blood pressure in the segment that had two or more daily servings of dairy products.30

When compared with non-smokers, smokers had lower calcium intake. In a population-based survey with adolescents aged 12-19 years, Bigio et al.9 found a decrease in the consumption of fruit and vegetables among smokers. In a sample of 2375 individuals aged 12 years or more, Andrade et al.31 observed that smokers had a worse overall diet

quality, reflected by a lower consumption of fruit and dairy products, and higher consumption of meat and sodium. By observing the association with diet, tobacco control can contribute to the effectiveness of strategies aimed at promoting healthy behaviors, including eating habits.

Adolescents who reported having headaches/migraine and dizziness showed significantly lower calcium intake. To the best of the authors' knowledge, no study in the literature has analyzed the association between calcium intake and the occurrence of headache or dizziness. Using the same sample assessed in this research, Braz et al.32 found a prevalence of health problems in 61.5% of the sample, and frequent headache/migraine was reported by 24.8% of the adolescents.

A higher calcium intake was observed in individuals that mentioned sleeping less than seven hours a day. Leal et al.10 found that 21.0% of adolescents did not have breakfast and 22.0% did not eat a mid-afternoon snack, meals that usually consist of milk other dairy products. Therefore, the fact that adolescents had longer sleep duration may have contributed to the skipping of breakfast and, consequently, to a lower intake of foods with high calcium content.

A calcium-poor diet was observed concurrently with other unhealthy behaviors, such as smoking and lower frequency of fruit consumption. This result highlights the importance of actions aimed at the group of behaviors that influence the adolescents' overall health, including the promotion of a healthy diet.

The present study provided population-level information on calcium intake in individuals aged 10-19 years living in the city of Campinas. It was observed that there is a deficiency in the intake of this nutrient in all assessed segments, demonstrating that those in the lower socioeconomic status strata and those with other inappropriate behaviors show even more unfavorable results. These findings suggest that strategies promoting healthy eating habits should take into account the specificities of each population subgroup.

Conflicts of interest

The authors declare to have no conflicts of interest. Acknowledgements

The authors would like to thank Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, process n. 409747/2006-8) for funding this research and for the productivity grants given to M.B.A. Barros and R.M. Fisberg; the Municipal Health Secretariat of Campinas and the Health Surveillance Secretariat of the Ministry of Health, for the financial support to perform the field research of ISACAMP 2008; and CAPES, for the Ph.D. scholarship received by D. Assumpcao.

References

1. Bueno AL, Czepielewski MA. The importance for growth of

dietary intake of calcium and vitamin D. J Pediatr (Rio J).

2008;84:386-94.

de Assumpçâo D et al

2. Golden NH, Abrams SA. Committee on nutrition optimizing bone health in children and adolescents. Pediatrics. 2014;134:e1229-43.

3. Silva AG, Cozzolino SM. Cálcio. In: Cozzolino SM, editor. org. Biodisponibilidade de nutrients. Sao Paulo: Manole; 2005. p. 421 -46.

4. Institute of Medicine. Dietary reference intakes for calcium and vitamin D. Washington, DC: The National Academies Press; 2011.

5. Instituto BrasileirodeGeografiaeEstatística(IBGE). Pesquisa de ornamentos familiares 2008-2009: análise do consumo alimentar pessoal no Brasil. Rio de Janeiro: IBGE; 2011.

6. Brasil, Ministério da Saúde Secretaria de Atencao a Saúde. Departamento de Atencao Básica. Coordenacao-Geral da Política de Alimentacao e Nutricao. Guia alimentar para a populacao brasileira: promovendo a alimentado saudável. Brasília: Ministério da Saúde; 2006.

7. Neutzling MB, Assuncao MC, Malcon MC, Hallal PC, Menezes AM. Hábitos alimentares de escolares adolescentes de Pelotas, Brasil. RevNutr. 2010;23:379-88.

8. Malta DC, de Andreazzi MA, Oliveira-Campos M, Andrade SS, de Sá NN, deMoura L, et al. Trend of the risk and protective factors of chronic diseases in adolescents National Adolescent School-based Health Survey (PeNSE 2009 e 2012). Rev Bras Epidemiol. 2014;17:77-91.

9. Bigio RS, Verly Junior E, Castro MA, César CL, Fisberg RM, Marchioni DM. Determinants of fruit and vegetable intake in adolescents using quantile regression. Rev Saude Publica. 2011;45:448-56.

10. Leal GV, Philippi ST, Matsudo SM, Toassa EC. Food intake and meal patterns of adolescents, Sao Paulo, Brazil. Rev Bras Epidemiol. 2010;13:457-67.

11. Fisberg RM, Villar BS. Manual de receitas e medidas caseiras para cálculo de inquéritos alimentares. Sao Paulo: Signus; 2002.

12. Pinheiro AB, Lacerda EM, Benzecry EH, Gomes MC, Costa VM. Tabela para avaliacao de consumo alimentar em medidas caseiras. 5th ed Sao Paulo: Atheneu; 2004.

13. Willett WC. Nutritional epidemiology. 2nd ed. New York: Oxford University Press; 1998.

14. World Health Organization (WHO). Global recommendations on physical activity for health. Geneva: WHO; 2010.

15. de OnisM, OnyangoAW, Borghi E, Siyam A, Nishida C, Siekmann J. Development of a WHO growth reference for school-aged children and adolescents. Bull World Health Organ. 2007;85:660-7.

16. Willett WC, Howe GR, Kushi LH. Adjustment for total energy intake in epidemiologic studies. Am J Clin Nutr. 1997;65:1220S-8S.

17. Nelder JA, Wedderburn RW. Generalized linear models. J R Statist SocA. 1972;135:370-84.

18. Breslow RA, Guenther PM, Juan W, Graubard BI. Alcoholic beverage consumption, nutrient intakes, and diet quality in the US adult population, 1999-2006. J Am Diet Assoc. 2010;110:551-62.

19. Brasil, Ministério da Saúde. Secretaria de Atencao a Saúde. Departamento de Atencao Básica. Guia alimentar para a populacao brasileira. Brasilia: Ministério da Saúde; 2014.

20. Veiga GV, Costa RS, Araújo MC, Souza A, de M, Bezerra IN, et al. Inadequate nutrient intake in Brazilian adolescents. Rev Saude Publica. 2013;47:212S-21S.

21. Malta DC, Sardinha LM, Mendes I, Barreto SM, Giatti L, Castro IR, et al. Prevalence of risk health behavior among adolescents: results from the 2009 National Adolescent School-based Health Survey (PeNSE). Cien Saude Colet. 2010;15:3009-19.

22. Santos LC, Martini LA, Freitas SN, Cintra IP. Ingestao de cálcio e indicadores antropométricos entre adolescentes. Rev Nutr. 2007;20:275-83.

23. Levy-Costa RB, Sichieri R, Pontes N, dos S, Monteiro CA. Household food availability in Brazil: distribution and trends (1974-2003). Rev Saude Publica. 2005;39:530-40.

Calcium intake by adolescents 9

480 24. Levy RB, Castro IR, Cardoso L, de O, Tavares LF, Sardinha LM, estratégicas para o enfrentamento das doencas crónicas nao 496

481 et al. Food consumption and eating behavior among Brazilian 497

482 adolescents: National Adolescent School-based Health Survey da Saude; 201 1. 498

483 (PeNSE), 2009. Cien Saude Colet. 2010;15:3085-97. 29. Souza Ade M, Pereira RA, Yokoo EM, Levy RB, Sichieri R. Most 499

484 2 5. Godoy FC, Andrade SC, Morimoto JM, Carandina L, Goldbaum consumed foods in Brazil: National Dietary Survey 2008-2009. 500

485 M, Barros MB, et al. Índice de qualidade da dieta de ado- Rev Saude Publica. 2013;47:190S-9S. 501

486 lescentes residentes no distrito do Butanta, municipio de Sao 30. Yuan WL, Kakinami L, Gray-Donald K, CzernichowS, Lambert M, 502

487 Paulo, Brasil. Rev Nutr. 2006;19:663-71. ParadisG. Influence of dairy product consumption on children's 503

488 26. de Assumpcao D, Barros MB, Fisberg RM, Carandina L, blood pressure: results from the QUALITY cohort. J Acad Nutr 504

489 Goldbaum M, Cesar CL. Diet quality among adolescents: a Diet. 2013;113:936-41. 505

490 population-based study in Campinas, Brazil. Rev Bras Epidemiol. 31. Andrade SC, PrevidelliÁN, Marchioni DM, Fisberg RM. Evaluation 506

491 2012;15:605-16. of the reliability and validity of the Brazilian Healthy Eating 507

492 27. Levy RB, Claro RM, Mondini L, Sichieri R, Monteiro CA. Regional Index Revised. Rev Saude Publica. 2013;47:675-83. 508

493 and socioeconomic distribution of household food availability in 32. Braz M, Barros Filho AA, Barros MB. Adolescent health: a 509

494 Brazil, in 2008 -2009. Rev Saude Publica. 2012;46:6 -15. population-based study in Campinas, Sao Paulo State, Brazil. 510

495 28. Brasil, Ministério da Saude Secretaria de Vigilancia em Saude. Cad Saude Publica. 2013;29:1877-88. 511

Departamento de Análise de Situacao de Saude. Plano de acöes