Common mental disorders in Brazilian adolescents: association with school characteristics, consumption of ultra-processed foods and waist-to-height ratio

Transtornos mentais comuns em adolescentes brasileiros: associação com características escolares, consumo de alimentos ultraprocessados e razão cintura/estatura

Trastornos mentales comunes en adolescentes brasileños: asociación con características escolares, consumo de alimentos ultraprocesados y relación cintura/estatura

Lucia Helena Almeida Gratão Thales Philipe Rodrigues da Silva Luana Lara Rocha Mariana Zogbi Jardim Tatiana Resende Prado Rangel de Oliveira Cristiane de Freitas Cunha Larissa Loures Mendes About the authors

Abstract:

Half of all mental health problems diagnosed in adulthood have their onset before or during adolescence, especially common mental disorders (CMD). Thus, it is relevant to study the factors associated with these disorders. This study aimed to investigate the association of school characteristics, consumption of ultra-processed foods, and waist-to-height ratio with the presence of CMD in Brazilian adolescents. This is a school-based, cross-sectional study that analyzed data from 71,553 Brazilian adolescents aged 12-17 years. The prevalence of CMD in these adolescents was 17.1% (cut-off point 5 for the General Health Questionnaire-12). Associations were estimated using multilevel logistic models, with the presence of CMD as the dependent variable. The final model, adjusted for non-modifiable individual variables, modifiable individual variables and family characteristics, identified a positive association between private-funded schools (OR = 1.10; 95%CI: 1.07-1.14), advertisements for ultra-processed foods (OR = 1.13; 95%CI: 1.09-1.17), the second to fourth quartiles of ultra-processed food intake and waist-to-height ratio (OR = 2.26; 95%CI: 2.03-2.52). This study demonstrated that the private-funded schools , the presence of ultra-processed food advertisements, the consumption of ultra-processed food, and an increased waist-to-height ratio are risk factors for CMD in Brazilian adolescents.

Keywords:
Adolescent Health; Mental Health; Food and Beverages

Resumo:

Metade de todos os problemas de saúde mental diagnosticados na idade adulta têm seu início antes ou durante a adolescência, especialmente os transtornos mentais comuns (TMC). Desta maneira, é importante estudar os fatores associados a estes transtornos. Este estudo objetiva investigar a associação entre características escolares, consumo de alimentos ultraprocessados e razão cintura/estatura com a presença de TMC em adolescentes brasileiros. Trata-se de um estudo transversal de base escolar que analisou dados de 71.553 adolescentes brasileiros de 12-17 anos. A prevalência de TMC nesses adolescentes foi de 17,1% (ponto de corte 5 para o Questionário de Saúde Geral - GHQ-12). As associações foram estimadas por meio de modelos logísticos multiníveis, tendo como variável dependente a presença de TMC. O modelo final, ajustado para variáveis individuais não modificáveis, variáveis individuais modificáveis e características familiares, identificou uma associação positiva entre dependência administrativa privada (OR = 1,1; IC95%: 1,07-1,14), presença de propaganda de alimentos ultraprocessados (OR = 1,13; IC95%: 1,09-1,17), segundo a quarto quartis de consumo de alimentos ultraprocessados e razão cintura/estatura (OR = 2,26; IC95%: 2,03-2,52). Este estudo demonstrou que a dependência administrativa privada da escola, a presença de propagandas de alimentos ultraprocessados, o consumo de alimentos ultraprocessados e o aumento da razão cintura/estatura são fatores de risco para TMC em adolescentes brasileiros.

Palavras-chave:
Saúde do Adolescente; Saúde Mental; Alimentos e Bebidas

Resumen:

La mitad de todos los problemas de salud mental diagnosticados en la edad adulta empiezan antes o durante la adolescencia, sobre todo los trastornos mentales comunes (TMC). Así, es importante estudiar los factores asociados a estos trastornos. El objetivo de este estudio es investigar la asociación entre las características escolares, el consumo de alimentos ultraprocesados y la proporción cintura/estatura y la presencia de TMC en adolescentes brasileños. Se trata de un estudio transversal de base escolar que analizó datos de 71.553 adolescentes brasileños de 12 a 17 años. La prevalencia de TMC en estos adolescentes fue del 17,1% (punto de corte 5 para el Cuestionario General de Salud - GHQ-12). Se estimaron las asociaciones a través de modelos logísticos multinivel, con la presencia de TMC como variable dependiente. El modelo final, ajustado para variables individuales no modificables, variables individuales modificables y características familiares, identificó una asociación positiva entre la dependencia administrativa privada (OR = 1,10; IC95%: 1,07-1,14), la presencia de publicidad de alimentos ultraprocesados (OR = 1,13; IC95%: 1,09-1,17), segundo al cuarto cuartil de consumo de alimentos ultraprocesados y la proporción cintura/estatura (OR = 2,26; IC95%: 2,03-2,52). Este estudio demostró que la dependencia administrativa privada de la escuela, la presencia de publicidades de alimentos ultraprocesados, el consumo de alimentos ultraprocesados y el aumento de la proporción cintura/estatura son factores de riesgo para trastornos mentales comunes en adolescentes brasileños.

Palabras-clave:
Salud del Adolescente; Salud Mental; Alimentos y Bebidas

Introduction

Adolescence is a transitional period between childhood and adulthood, during which individuals undergo profound physical, social, cognitive, and psychological changes 11. Blakemore SJ, Burnett S, Dahl RE. The role of puberty in the developing adolescent brain. Hum Brain Mapp 2010; 31:926-33.. In this stage, individuals develop autonomy, self-control, interaction, and learning in society, which are important skills for strengthening mental health, both during this period and in subsequent stages of development 22. WHO Regional Office for Europe. Factsheet for World Mental Health Day 2018. Adolescent mental health in the European Region. Geneva: World Health Organization; 2018..

According to the World Health Organization 2, half of all mental health problems diagnosed in adulthood have their onset before or during adolescence, especially common mental disorders (CMD). The term CMD refers to two main categories of diagnoses: anxiety and depressive disorders, and non-specific and somatic complaints, which may or may not be associated 33. Goldberg DP, Huxley P. Common mental disorders: a bio-social model. New York: Tavistock/Routledge; 1992.,44. Goldberg D. A bio-social model for common mental disorders. Acta Psychiatr Scand Suppl 1994; 385:66-70.,55. Ferrari AJ. Charlson FJ, Norman RE, Patten SB, Freedman G, Murray CJL, et al. Burden of depressive disorders by country, sex, age, and year: findings from the Global Burden of Disease Study 2010. PLoS Med 2013; 10:e1001547.. These disorders are among the leading causes of illness in childhood and adolescence, increasing the risk of self-harm and suicide in this age group 66. Organização Pan-Americana da Saúde; Organização Mundial da Saúde. Saúde mental dos adolescentes. https://www.paho.org/pt/topicos/saude-mental-dos-adolescentes (accessed on 26/Jan/2022).
https://www.paho.org/pt/topicos/saude-me...
,77. Gili M, Castellví P, Vives M, Torre-Luque A, Almenara J, Blasco MJ, et al. Mental disorders as risk factors for suicidal behavior in young people: a meta-analysis and systematic review of longitudinal studies. J Affect Disord 2019; 245:152-62.. In this sense, recognizing factors that may be associated with CMD could help guide action plans to mitigate these disorders in adolescents.

To better understand the factors associated with adolescents’ mental health, it is essential to observe the environments in which they live, such as schools. Some studies have shown that the school environment may be associated with the development of mental disorders 88. Buttazzoni A, Doherty S, Minaker L. How do urban environments affect young people's mental health? A novel conceptual framework to bridge public health, planning, and neurourbanism. Public Health Rep 2022; 137:48-61.,99. Zhang L, Wu L. Community environment perception on depression: the mediating role of subjective social class. Int J Environ Res Public Health 2021; 18:8083.. The school environment is where adolescents spend at least 20 hours per week - if enrolled in part-time education - and up to 40 hours per week - if enrolled in full-time education 1010. Alves CCR, Fonseca MA, Carmo AP, Silveira LSS, Senra K. Documento orientador da Política de Educação Integral e Integrada e implementação das escolas polo de educação múltipla em Minas Gerais. https://acervodenoticias.educacao.mg.gov.br/images/documentos/Documento%20Pol%C3%ADtica%20de%20Educa%C3%A7%C3%A3o%20Integral%20e%20Integrada%20FINAL.pdf (accessed on 27/Nov/2022).
https://acervodenoticias.educacao.mg.gov...
. In these spaces, besides educational activities, students receive or purchase meals, or both. In some cases, these places also feature food advertisements that direct the purchase and consumption of food. According to some studies conducted in Brazil, private schools have a more obesogenic food environment compared to public schools, due to the high availability, access to and presence of advertisements and ultra-processed foods 1111. Institute of Medicine; Board on Children, Youth, and Families; Food and Nutrition Board; Committee on Food Marketing and the Diets of Children and Youth. Food marketing to children and youth: threat or opportunity? Washington DC: National Academies Press; 2006., which are associated with overweight and chronic noncommunicable diseases in children and adolescents 1111. Institute of Medicine; Board on Children, Youth, and Families; Food and Nutrition Board; Committee on Food Marketing and the Diets of Children and Youth. Food marketing to children and youth: threat or opportunity? Washington DC: National Academies Press; 2006.,1212. Werneck AO, Costa CS, Horta B, Wehrmeister FC, Gonçalves H, Menezes AMB, et al. Prospective association between ultra-processed food consumption and incidence of elevated symptoms of common mental disorders. J Affect Disord 2022; 312:78-85..

The scientific literature has explored the association between ultra-processed foods and mental health in adults 1212. Werneck AO, Costa CS, Horta B, Wehrmeister FC, Gonçalves H, Menezes AMB, et al. Prospective association between ultra-processed food consumption and incidence of elevated symptoms of common mental disorders. J Affect Disord 2022; 312:78-85.,1313. Adjibade M, Julia C, Allès B, Touvier M, Lemogne C, Srour B, et al. Prospective association between ultra-processed food consumption and incident depressive symptoms in the French NutriNet-Santé cohort. BMC Med 2019; 17:78.,1414. Lane MM, Lotfaliany M, Hodge AM, O'Neil A, Travica N, Jacka FN, et al. High ultra-processed food consumption is associated with elevated psychological distress as an indicator of depression in adults from the Melbourne Collaborative Cohort Study. J Affect Disord 2023; 335:57-66.,1515. Lane MM, Gamage E, Travica N, Dissanayaka T, Ashtree DN, Gauci S, et al. Ultra-processed food consumption and mental health: a systematic review and meta-analysis of observational studies. Nutrients 2022; 14:2568.,1616. Lee S, Choi M. Ultra-processed food intakes are associated with depression in the general population: the Korea National Health and Nutrition Examination Survey. Nutrients 2023; 15:2169.,1717. Arshad H, Head J, Jacka FN, Lane MM, Kivimaki M, Akbaraly T. Association between ultra-processed foods and recurrence of depressive symptoms: the Whitehall II cohort study. Nutr Neurosci 2024; 27:42-54. and adolescents 1818. Mesas AE, González AD, De Andrade SM, Martínez-Vizcaíno V, López-Gil JF, Jiménez-López E. Increased consumption of ultra-processed food is associated with poor mental health in a nationally representative sample of adolescent students in Brazil. Nutrients 2022; 14:5207.,1919. Faisal-Cury A, Leite MA, Escuder MML, Levy RB, Peres MFT. The relationship between ultra-processed food consumption and internalising symptoms among adolescents from São Paulo city, Southeast Brazil. Public Health Nutr 2022; 25:2498-506.. Studies have suggested that the consumption of ultra-processed foods may be associated with mental disorders 1313. Adjibade M, Julia C, Allès B, Touvier M, Lemogne C, Srour B, et al. Prospective association between ultra-processed food consumption and incident depressive symptoms in the French NutriNet-Santé cohort. BMC Med 2019; 17:78.,1414. Lane MM, Lotfaliany M, Hodge AM, O'Neil A, Travica N, Jacka FN, et al. High ultra-processed food consumption is associated with elevated psychological distress as an indicator of depression in adults from the Melbourne Collaborative Cohort Study. J Affect Disord 2023; 335:57-66.,1515. Lane MM, Gamage E, Travica N, Dissanayaka T, Ashtree DN, Gauci S, et al. Ultra-processed food consumption and mental health: a systematic review and meta-analysis of observational studies. Nutrients 2022; 14:2568.,1616. Lee S, Choi M. Ultra-processed food intakes are associated with depression in the general population: the Korea National Health and Nutrition Examination Survey. Nutrients 2023; 15:2169.,1717. Arshad H, Head J, Jacka FN, Lane MM, Kivimaki M, Akbaraly T. Association between ultra-processed foods and recurrence of depressive symptoms: the Whitehall II cohort study. Nutr Neurosci 2024; 27:42-54.,1818. Mesas AE, González AD, De Andrade SM, Martínez-Vizcaíno V, López-Gil JF, Jiménez-López E. Increased consumption of ultra-processed food is associated with poor mental health in a nationally representative sample of adolescent students in Brazil. Nutrients 2022; 14:5207.,1919. Faisal-Cury A, Leite MA, Escuder MML, Levy RB, Peres MFT. The relationship between ultra-processed food consumption and internalising symptoms among adolescents from São Paulo city, Southeast Brazil. Public Health Nutr 2022; 25:2498-506.. However, regardless of the design, these studies only considered the intake of ultra-processed foods, without considering other food environment factors. We believe that in addition to the consumption of ultra-processed foods, considered in isolation, some factors in the school environment, such as the type of school, the presence of advertising for ultra-processed foods and overweight, may be associated with CMD in Brazilian adolescents, as mental disorders have multifactorial causes 2020. Safaei M, Sundararajan EA, Driss M, Boulila W, Shapi'i A. A systematic literature review on obesity: understanding the causes & consequences of obesity and reviewing various machine learning approaches used to predict obesity. Comput Biol Med 2021; 136:104754., including environmental characteristics.

This study aimed to investigate the association between school characteristics (type of school funding and presence of ultra-processed foods advertisements) and individual characteristics (ultra-processed foods consumption and waist-to-height ratio - WHtR) related to food consumption and body adiposity with the presence of CMD in Brazilian adolescents.

Materials and methods

Design, sampling, and participants

The data used in this study were obtained from the Brazilian Study of Cardiovascular Risk in Adolescents (ERICA, acronym in Portuguese). ERICA was a cross-sectional, nationwide, school-based study with data collection carried out from March 2013 to December 2014. Its sample consisted of adolescents aged 12 to 17 years, of both sexes, enrolled in the last three years of middle school and the three years of high school, in public and private schools in Brazil. In addition to adolescents, school administrators were also interviewed 2121. Bloch KV, Szklo M, Kuschnir MCC, Abreu GA, Barufaldi LA, Klein CA, et al. The study of cardiovascular risk in adolescents - ERICA: rationale, design and sample characteristics of a national survey examining cardiovascular risk factor profile in Brazilian adolescents. BMC Public Health 2015; 15:94..

The ERICA study included 273 Brazilian cities. To determine the number of eligible cities, the sampled population was divided into 32 geographic strata: each capital of the 27 Federative Units and five strata of metropolises with more than 100,000 inhabitants in each of the five macroregions of Brazil. After this geographic stratification, a selection of schools and classes in the eligible municipalities was carried out 2121. Bloch KV, Szklo M, Kuschnir MCC, Abreu GA, Barufaldi LA, Klein CA, et al. The study of cardiovascular risk in adolescents - ERICA: rationale, design and sample characteristics of a national survey examining cardiovascular risk factor profile in Brazilian adolescents. BMC Public Health 2015; 15:94..

In the first stage, schools in each geographic stratum were selected with probability proportional to their size, which was considered equal to the ratio between the number of students in the eligible classes and the distance from the state capital. The selection was made after classifying the school records by location (urban or rural areas) and type of school funding (private or public). In total, 1,251 schools in 124 municipalities were selected 2121. Bloch KV, Szklo M, Kuschnir MCC, Abreu GA, Barufaldi LA, Klein CA, et al. The study of cardiovascular risk in adolescents - ERICA: rationale, design and sample characteristics of a national survey examining cardiovascular risk factor profile in Brazilian adolescents. BMC Public Health 2015; 15:94..

In the second stage three classes from each school in the sample were selected with equal probability. Using Brazilian grade year as a proxy for age, 7th, 8th, and 9th grades of middle school and 10th, 11th and 12th grades of high school were considered eligible for selection. In each selected class, all students were invited to take part in the research, which consisted of interviews, anthropometric measurements and blood pressure measurements 2121. Bloch KV, Szklo M, Kuschnir MCC, Abreu GA, Barufaldi LA, Klein CA, et al. The study of cardiovascular risk in adolescents - ERICA: rationale, design and sample characteristics of a national survey examining cardiovascular risk factor profile in Brazilian adolescents. BMC Public Health 2015; 15:94..

Detailed information on the sampling process, research protocol, participant selection and data collection can be found in studies previously published by the ERICA Study Committee 2121. Bloch KV, Szklo M, Kuschnir MCC, Abreu GA, Barufaldi LA, Klein CA, et al. The study of cardiovascular risk in adolescents - ERICA: rationale, design and sample characteristics of a national survey examining cardiovascular risk factor profile in Brazilian adolescents. BMC Public Health 2015; 15:94.,2222. Silva TLN, Klein CH, Souza AM, Barufaldi LA, Abreu GA, Kuschnir MCC, et al. Response rate in the Study of Cardiovascular Risks in Adolescents - ERICA. Rev Saúde Pública 2016; 50 Suppl 1:3s.,2323. Vasconcellos MTL, Silva PLN, Szklo M, Kuschnir MCC, Klein CH, Abreu GA, et al. Sampling design for the Study of Cardiovascular Risks in Adolescents (ERICA). Cad Saúde Pública 2015; 31:921-30..

Instruments and data collection

ERICA included three questionnaires: one for adolescents, one for parents/educators and one for school administrators. For the analysis of this study, the questionnaire for adolescents was used, including a 24-hour dietary recall (24hR), anthropometric measurements (height and waist circumference) and the school questionnaire.

The questionnaire for adolescents consisted of 105 questions, with specific ones for each of the 11 thematic blocks, which consisted of sociodemographic characteristics, work and employment, physical activity, eating habits, smoking, alcohol consumption, reproductive health, oral health, reported morbidity, sleep duration, and mental health. The adolescents completed the questionnaire using the personal digital assistant (PDA, model LG GM750Q, https://www.lg.com/br/suporte/produto/lg-GM750Q.ABRATN) electronic device for data collection.

The school questionnaire consisted of 28 questions covering three thematic blocks: general school characteristics, physical structure and school food. The questionnaire was administered during an interview between the field researcher, using a PDA device, and the principal or another staff member of the school.

The 74,953 eligible adolescents took part in the ERICA study. Of these, 74,589 completed the questionnaire for adolescents, 71,553 responded to the 24hR, 73,787 had their anthropometric measurements taken and 73,637 underwent blood pressure testing. For this study, only adolescents who participated in all of the steps mentioned above were considered, that is, 71,553 adolescents (Figure 1). According to the sensitivity analysis, this generated no significant difference in the final sample 2121. Bloch KV, Szklo M, Kuschnir MCC, Abreu GA, Barufaldi LA, Klein CA, et al. The study of cardiovascular risk in adolescents - ERICA: rationale, design and sample characteristics of a national survey examining cardiovascular risk factor profile in Brazilian adolescents. BMC Public Health 2015; 15:94.. Adolescents with some degree of disability that could affect the anthropometric assessment or prevent them from completing the questionnaire were excluded from the sample, as well as pregnant adolescents and the ones who were out of the age range (Figure 1).

Figure 1
Flowchart of eligible adolescents and sample completeness regarding the information blocks and subsets. Brazilian Study of Cardiovascular Risk in Adolescents (ERICA), 2013-2014.

Dependent variable

To create the CMD variable, the General Health Questionnaire (GHQ-12), validated for use in adolescents 2424. French DJ, Tait RJ. Measurement invariance in the General Health Questionnaire-12 in young Australian adolescents. Eur Child Adolesc Psychiatry 2004; 13:1-7., was included in the questionnaire for adolescents. The GHQ-12 is a widely used self-administered instrument that is known to be a reliable measure of mental health and assists in screening for psychiatric disorders in community and non-psychiatric clinical settings, using an index generated from individuals’ responses 2525. Goldberg DP, Williams P. A user's guide to the General Health Questionnaire: GHQ. London: GL Assessment; 1988..

To screen for CMD in adolescents, the binary system with a shear point of 5 was considered - in other words, CMD was considered present if at least 5 of the 12 items were answered with one of the last two options of the questionnaire (“a little more than usual” or “much more than usual”). This cut-off point had a 86.7% sensitivity, a 88.9% specificity, a 71.2% positive predictive value and a 0.94 area under the receiver operating characteristics curve (ROC) 2626. Goldberg DP, Gater R, Sartorius N, Ustun TB, Piccinelli M, Gureje O, Rutter C. The validity of two versions of the GHQ in the WHO study of mental illness in general health care. Psychol Med 1997; 27:191-7..

Independent variables

For the selection of independent variables, those related to schools and those associated with the consumption of ultra-processed foods and increased WHtR were tested: type of school funding (public or private) and presence of ultra-processed foods advertising (yes or no), consumption of ultra-processed foods in the previous 24 hours (quartiles of kcal/day) and WHtR (numerical variable).

To create the variable “presence of advertisements for ultra-processed foods”, advertisements for sweets, candy, lollipops, chocolate, sweet cookies, soft drinks, natural guarana, mate tea, other iced teas, guarana, isotonic, ice cream, popsicles and other ultra-processed foods were aggregated from the school questionnaire. Ultra-processed foods advertising was considered present if the school had at least one advertisement for these foods.

Data from the 24hR were used to determine ultra-processed foods intake. The 24hR data were collected during face-to-face interviews carried out by trained researchers. The Brasil-Nutri software (http://nebin.com.br/novosite/conteudo.php?id=4) was used to record food consumption data directly on the netbooks. The interview technique used was the multiple-pass method, which consists of a guided interview in five stages, to reduce underreporting of food consumption. The food database used in the research was developed by Brazilian Institute of Geography and Statistics (IBGE, acronym in Portuguese) in 2008-2009 2727. Instituto Brasileiro de Geografia e Estatística. Pesquisa de Orçamentos Familiares 2002-2003: perfil das despesas no Brasil: indicadores selecionados. Rio de Janeiro: Instituto Brasileiro de Geografia e Estatística; 2007.,2828. Instituto Brasileiro de Geografia e Estatística. Pesquisa de Orçamentos Familiares 2008-2009: análise do consumo alimentar pessoal no Brasil. Rio de Janeiro: Instituto Brasileiro de Geografia e Estatística; 2011..

After food weight was converted to grams, the dataset was linked to a nutritional composition table to calculate each adolescent’s energy intake. Foods were classified based on the degree of processing, as indicated by the NOVA food classification system 2929. Monteiro CA, Cannon G, Levy R, Moubarac JC, Jaime P, Martins AP, et al. NOVA. The star shines bright. World Nutr 2016; 7:28-38.. This system classifies all foods into the following four groups according to the nature, extent and purpose of the industrial processes they undergo: unprocessed or minimally processed foods; culinary ingredients; processed foods; and ultra-processed foods. Foods were categorized by two independent researchers and, in the event of disagreement, evaluated by a third expert researcher. For each adolescent, the total daily energy intake (kcal/day) was quantified and calculated in quartiles.

The WHtR was calculated by dividing the adolescents’ waist circumference by their height 3030. Yoo EG. Waist-to-height ratio as a screening tool for obesity and cardiometabolic risk. Korean J Pediatr 2016; 59:425.. Waist circumference was obtained using an anthropometric fiberglass tape (Sanny, https://www.sanny.com.br/) with a resolution in millimeters and a length of 1.50 meters. Height was obtained by averaging two consecutive measurements obtained with a portable and detachable stadiometer (Alturexata, http://www.alturexata.com.br/) with millimeter and field resolution of use up to 213 centimeters. Standardized procedures and training were used to ensure the quality of the information to be obtained both from the questionnaire and from direct measurements.

Adjustment variables

The age of the adolescents was divided into the two following categories: 12-14 years and 15-17 years, in accordance with the classification used in other articles published with data from ERICA 3131. Lopes CS, Abreu GA, Santos DF, Menezes PR, Carvalho KMB, Cunha CF, et al. ERICA: prevalence of common mental disorders in Brazilian adolescents. Rev Saúde Pública 2016; 50 Suppl 1:14s.. For gender, the alternatives in the questionnaire for students were: female and male. The variable region of Brazil referred to five Brazilian macroregions: North; South; Central-West; Northeast; and Southeast.

The time of practice of weekly physical activity was categorized according to the cut points proposed by the Brazilian National Survey of School Health (PeNSE, acronym in Portuguese), in which students who did not practice physical activity in the reference period were considered inactive; those who practiced physical activity for 1-149 minutes were placed into the category “insufficiently active 1”; those who practiced physical activity for 150-299 minutes were placed into the category “insufficiently active 2”; and those who practiced for 300 minutes or more were categorized as active 3232. Instituto Brasileiro de Geografia e Estatística. Pesquisa Nacional de Saúde do Escolar, 2015. Rio de Janeiro: Instituto Brasileiro de Geografia e Estatística; 2016..

The variable “living with parents” referred to the following two categories: lives with both parents or only with the mother/father and does not live with either parent. The variable “work” was constructed using two variables based on the questions “did the student work without pay in the last year?” and “did the student work with pay in the last year?”, that is, the performance of paid and unpaid activities was considered work. Therefore, the categories of the variable considered for this study were: “does not work” and “works”.

To obtain the variable “mean sleep time”, the weighted mean between the usual duration, in hours, of sleep on weekdays and weekend days was calculated separately. Those who reported sleeping less than four hours and more than 14 hours were excluded for not meeting the usual parameters of sleep for adolescents.

To establish the socioeconomic status of the adolescents, it was decided to calculate a pattern of socioeconomic indicators (Supplementary Material. Table S1: https://cadernos.ensp.fiocruz.br/static//arquivo/suppl-e00068423_7936.pdf) by principal component analysis (PCA), consisting of the variables identified in the study by Ribeiro et al. 3333. Ribeiro IBS, Correa MM, Oliveira G, Cade NV. Common mental disorders and socioeconomic status in adolescents of ERICA. Rev Saúde Pública 2020; 54:4., namely: the presence of employees in the residence; the number of residents per room; the number of bathrooms in the residence; and the number of refrigerators in the residence. The pattern of socioeconomic indicators generated by the PCA identified a single main component, with a contribution of 36.22% of the explained accumulated variation. The pattern was characterized by the presence of employees, fewer residents per room, more bathrooms, and more refrigerators in the residence (Supplementary Material. Table S1: https://cadernos.ensp.fiocruz.br/static//arquivo/suppl-e00068423_7936.pdf).

Statistical analysis

Descriptive analysis included the calculation of absolute and relative frequencies for categorical variables, in addition to measures of central tendency. The chi-square test and t-test were used to compare proportions between variables.

The association between school characteristics, individual characteristics and the presence of CMD was estimated using multilevel logistic models, with the presence of CMD as the dependent variable. The inclusion of adjustment variables followed a hierarchical pattern (including only the independent variables, then the adjustment variables that were non-modifiable characteristics, and, lastly, the modifiable characteristics).

Thus, four models were proposed: (i) null model (M0), estimating the random effect of the intercept; (ii) model 1 (M1), containing the independent variables (type of school funding [public/private], ultra-processed foods advertising at school [no/yes], ultra-processed foods consumption [quartile of total kcal of ultra-processed foods consumed per day] and WHtR [numerical variable]), and non-modifiable adjustment variables (sex [female/male], age [12-14 and 15-17] and race/color [white/black/mixed-race/yellow/Indigenous]); (iii) model 2 (M2), containing the variables of M1 in addition to modifiable individual variables (work activities by the adolescent, total daily energy intake [numerical variable], mean sleep time [numerical variable], physical activity [inactive, insufficiently active 1, insufficiently active 2, active]); and (iv) model 3 (M3), containing the variables of M2 in addition to adjustment variables related to family characteristics (living with parents [both parents or only the mother/father] or not), the pattern of socioeconomic characteristics (terciles of pattern) and the region of residence (North, South, Central-West, Southeast, and Northeast).

The variance partition coefficient (VPC) was quantified to verify the proportion of total variance attributed to the schools. The assessment of the models was done by comparing the values of Akaike’s information criterion (AIC), in which a decrease in the AIC value indicates a better fit of the model to the response variable. At the end of the modeling, the variance reduction was calculated to verify the final fit.

The statistical software for professionals Stata, version 14.0 for Mac (https://www.stata.com), package was used. For the multilevel models, the “gllamm” command was used, allowing for non-independent data and multilevel analyses with the inclusion of sample weights for complex samples. The aggregation unit used was the adolescents’ school. A 5% significance level was used for all analyses.

Ethical aspects

This report was approved by the Institutional Review Board of the Institute of Collective Health Studies of the Federal University of Rio de Janeiro (IESC/UFRJ, acronym in Portuguese), which is part of the central coordination of the report (approval n. 45/2008), and of each Brazil’s Federative Units. Written informed consent was obtained from all subjects, their parents and legal guardian(s) in two copies, one of which remained in the possession of the research subjects. The adolescents also signed a written assent form.

Results

Sample characteristics

Data from 71,553 Brazilian adolescents were evaluated. The prevalence of CMD in these adolescents was 17.1% (cut-off point 5 for the GHQ-12). Table 1 shows the characterization of the adolescents studied, most of whom were male (50.21%); aged 12 to 13 years (35.1%); mixed-race (48.83%); belonged to the first tercile of the pattern “socioeconomic status”, which corresponds to those with better socioeconomic conditions (46.16%); resided in the Southeast Region (50.78%); and studied in public schools (83.61%) located in the urban area (96.1%). Of the adolescents studied, 26.07% performed some work activity, 36.85% lived with their parents or only with their mother, and 5.88% did not live with either parent.

Table 1
Characteristics of adolescents enrolled in schools in Brazilian capitals stratified by the presence of common mental disorders (CMD). Brazilian Study of Cardiovascular Risk in Adolescents (ERICA), 2013-2014, (n = 71,553).

The presence of CMD was more frequent and statistically significant among female adolescents (67.98%), aged 14-15 years (36.05%), who did not work (69.38%), had mothers with an education level up to complete middle school (33.55%) and who lived with both parents (51.46%) (Table 1). It was also observed that 26.34% of the adolescents with CMD had a WHtR of 0.44±0.001. The mean energy intake was 2,352.72±26.10 kcal/day (Table 1), a higher value than that observed in adolescents without CMD.

Association between school characteristics and CMD

Table 2 shows the multilevel logistic regression models, with the presence of CMD as the outcome variable and the type of school funding, ultra-processed foods consumption, presence of ultra-processed foods advertising at school and WHtR as independent variables.

Table 2
Multilevel logistic regression models for ultra-processed foods consumption, waist-to-height ratio, and school environment variables associated with common mental disorders (CMD) in adolescents enrolled in Brazilian schools. Brazilian Study of Cardiovascular Risk in Adolescents (ERICA), 2013-2014 (n = 71,553).

Table 2 shows the M0. The intercept variation (0.20; 95%CI: 0.205-0.209) of M0 showed that the presence of CMD in adolescents differed between schools (p < 0.001). The variance partition coefficient (VPC) was 0.0518, that is, approximately 5.18% of the total variance was attributed to the characteristics of the schools of the adolescents.

M1 was adjusted only for sex, age, and race/color, which are factors that cannot be modified by the individual. It is possible to observe that there was a positive association between the private administration of the school (OR = 1.11; 95%CI: 1.08-1.14) and the presence of ultra-processed foods advertising in the school environment (OR = 1.16; 95%CI 1.13-1.20) with the presence of CMD.

In M2, with the inclusion of modifiable factors such as work activities by adolescents, total kilocalories consumed in the previous 24 hours, mean sleep time, and physical activity, a reduction in the odds ratio values for the school environment variables was observed. However, the type of funding of the school (OR = 1.12; 95%CI: 1.09-1.15) and the presence of ultra-processed foods advertising in the school (OR = 1.12; 95%CI: 1.09-1.16) remained directly associated with the presence of CMD.

M3, additionally adjusted for family characteristics, such as living with parents, the pattern of socioeconomic indicators and region of residence, maintained the behavior observed in M2, reducing the odds ratio of school environment variables, but still maintaining their association with the presence of CMD. There was a positive association between the private-funded schools (OR = 1.11; 95%CI: 1.08-1.15) and the presence of ultra-processed foods advertising in the school (OR = 1.10; 95%CI: 1.06-1.14).

Association between ultra-processed foods consumption, WHtR and CMD

Comparing M1, M2 and M3, it can be observed that there was an increase in ultra-processed foods consumption in the fourth quartile, that is, among those who consumed more kilocalories from ultra-processed foods (M1: OR = 1.17, 95%CI: 1.15-1.19; M2: OR = 1.19, 95%CI: 1.17-1.22; M3: OR = 1.20, 95%CI: 1.18-1.22).

At the same time, the WHtR decreased in magnitude, as observed by the reduced odds ratio in M2 and M3 in relation to M1. However, in all of them, there was still a positive association with the outcome (M1: OR = 2.88, 95%CI: 2.61-3.18; M2: OR = 2.32, 95%CI: 2.08-2.58; and M3: OR = 2.16, 95%CI: 1.94-2.41).

Discussion

This is the first study to identify an association between ultra-processed foods consumption, body adiposity, and characteristics of the school environment with the presence of CMD in adolescents. A positive association of the private administration of schools, ultra-processed foods advertising in schools, ultra-processed foods consumption and increased WHtR with the presence of CMD in Brazilian adolescents was found.

In addition, Hecht et al. 3434. Hecht EM, Rabil A, Steele EM, Abrams GA, Ware D, Landy DC, et al. Cross-sectional examination of ultra-processed food consumption and adverse mental health symptoms. Public Health Nutr 2022; 25:3225-34. set out to investigate whether adults aged over 18 years who consumed ultra-processed foods had more symptoms related to mental health. They found that individuals with high ultra-processed foods consumption were significantly more likely to report depression and anxiety and to have worse mental health. Other studies have found a similar association 3535. Coletro HN, Mendonça RD, Meireles AL, Machado-Coelho GLL, Menezes MC. Multiple health risk behaviors, including high consumption of ultra-processed foods and their implications for mental health during the COVID-19 pandemic. Front Nutr 2022; 9:1042425.,3636. Contreras-Rodriguez O, Reales-Moreno M, Fernández-Barrès S, Cimpean A, Arnoriaga-Rodríguez M, Puig J, et al. Consumption of ultra-processed foods is associated with depression, mesocorticolimbic volume, and inflammation. J Affect Disord 2023; 335:340-8.,3737. Chen H, Hou Y, Yang H, Wang X, Xu C. The associations of dietary patterns with depressive and anxiety symptoms: a prospective study. BMC Med 2023; 21:307.. The physiological mechanisms associated with these events are not yet known; however, it has been hypothesized that industrial additives used for preservation, odorization, and coloring can modify the neuronal mitochondrial function by various metabolic pathways 3838. Boulangé CL, Neves AL, Chilloux J, Nicholson JK, Dumas ME. Impact of the gut microbiota on inflammation, obesity, and metabolic disease. Genome Med 2016; 8:42.. The consumption of unhealthy foods has also been linked to inflammatory processes, nutrient and neurotransmitter defficiencies 3939. Godos J, Currenti W, Angelino D, Mena P, Castellano S, Caraci F, et al. Diet and mental health: review of the recent updates on molecular mechanisms. Antioxidants (Basel) 2020; 9:346. and increased likelihood of central nervous system demyelination 4040. Mannino A, Daly A, Dunlop E, Probst Y, Ponsonby AL, Mei IAF, et al. Higher consumption of ultra-processed foods and increased likelihood of central nervous system demyelination in a case-control study of Australian adults. Eur J Clin Nutr 2023; 77:611-4., as well as changes in the gut-brain axis, leading to changes in the production of neurotransmitters 4141. Cenit MC, Sanz Y, Codoñer-Franch P. Influence of gut microbiota on neuropsychiatric disorders. World J Gastroenterol 2017; 23:5486-98..

However, in the case of our study, the possibility that the presence of CMD can lead to a worsening of the quality of food choices cannot be ruled out, as there is a possibility that emotions regulate eating, just as eating can regulate emotions 4242. Macht M. How emotions affect eating: a five-way model. Appetite 2008; 50:1-11.. Keck et al. 4343. Keck MM, Vivier H, Cassisi JE, Dvorak RD, Dunn ME, Neer SM, et al. Examining the role of anxiety and depression in dietary choices among college students. Nutrients 2020; 12:2061., for example, in a study with 225 college students, observed that symptoms of depression were a greater risk factor for poor nutrition.

The association between parameters of body adiposity and the presence of CMD in adolescents has been found in this study and by other researchers. Our results showed that an increased WHtR may be associated with the presence of CMD in adolescents, as the WHtR is strongly correlated with visceral fat 4444. Ashwell M, Cole TJ, Dixon AK. Ratio of waist circumference to height is strong predictor of intra-abdominal fat. BMJ 1996; 313:559-60.. Scott et al. 4545. Scott KM, McGee MA, Wells JE, Oakley Browne MA. Obesity and mental disorders in the adult general population. J Psychosom Res 2008; 64:97-105., in a cross-sectional study of people aged over 16 years in New Zealand, found an association between obesity, depressive disorder and anxiety disorder. Lewis-de-Los-Angelis & Richard 4646. Lewis-de los Angeles WW, Liu RT. History of depression, elevated body mass index, and waist-to-height ratio in preadolescent children. Psychosom Med 2021; 83:1075-81. found that a history of depression was associated with a higher WHtR in individuals from the United States aged 9 and 10 years WHtR. In addition, girls with a history of depression were found to be more likely to have an elevated WHtR.

Therefore, an increased WHtR is associated with CMD in adolescents. In another study with adults aged from 20-89 years old 4747. Rivenes AC, Harvey SB, Mykletun A. The relationship between abdominal fat, obesity, and common mental disorders: results from the HUNT Study. J Psychosom Res 2009; 66:269-75., the authors found that an increase in the waist-hip ratio was associated with an increase in the prevalence of anxiety and depression. Notably, there is a possibility of reverse causality when referring to the association between WHtR and CMD, given the study design, and also considering that the literature has shown a bidirectional association between body adiposity and mental health outcomes.

This leads us to believe that the neural mechanisms associated with the consumption of ultra-processed foods 3434. Hecht EM, Rabil A, Steele EM, Abrams GA, Ware D, Landy DC, et al. Cross-sectional examination of ultra-processed food consumption and adverse mental health symptoms. Public Health Nutr 2022; 25:3225-34.,3535. Coletro HN, Mendonça RD, Meireles AL, Machado-Coelho GLL, Menezes MC. Multiple health risk behaviors, including high consumption of ultra-processed foods and their implications for mental health during the COVID-19 pandemic. Front Nutr 2022; 9:1042425. and increased WHtR 4747. Rivenes AC, Harvey SB, Mykletun A. The relationship between abdominal fat, obesity, and common mental disorders: results from the HUNT Study. J Psychosom Res 2009; 66:269-75. may lead to adolescents having a greater chance of developing CMD, even during adolescence. For us, based on our results and on those of published studies, the advertising and sale of ultra-processed foods in private schools, favoring greater consumption of these foods 4848. Ministério da Saúde. Eu quero me alimentar melhor. A influência da publicidade nas escolhas alimentares. https://www.gov.br/saude/pt-br/assuntos/saude-brasil/eu-quero-me-alimentar-melhor/noticias/2023/a-influencia-da-publicidade-nas-escolhas-alimentares (accessed on 22/Mar/2024).
https://www.gov.br/saude/pt-br/assuntos/...
,4949. Mattos MC, Nascimento PCBD, Almeida SS, Costa TMB. Influência de propagandas de alimentos nas escolhas alimentares de crianças e adolescentes. Psicol Teor Prat 2010; 12:34-51. and increased body adiposity, increase the risk of CMD in adolescents.

In this regard, Carmo et al. 5050. Carmo AS, Assis MM, Cunha CF, Oliveira TRPR, Mendes LL. The food environment of Brazilian public and private schools. Cad Saúde Pública 2018; 34:e00014918. also found crucial data in a cross-sectional study with 1,427 public and private schools in Brazil, reporting that at least 76.1% of private schools marketed some ultra-processed foods. It is also known that the presence of unhealthy foods in the school environment is associated with higher consumption of these items by students 5151. Rocha LL, Pessoa MC, Gratão LHA, Carmo AS, Cordeiro NG, Cunha CF, et al. Characteristics of the school food environment affect the consumption of sugar-sweetened beverages among adolescents. Front Nutr 2021; 8:742744.,5252. Azeredo CM, Rezende LFM, Canella DS, Claro RM, Peres MFT, Luiz OC, et al. Food environments in schools and in the immediate vicinity are associated with unhealthy food consumption among Brazilian adolescents. Prev Med 2016; 88:73-9.,5353. Zimmerman FJ, Shimoga SV. The effects of food advertising and cognitive load on food choices. BMC Public Health 2014; 14:342.,5454. Shiv B, Fedorikhin A. Heart and mind in conflict: the interplay of affect and cognition in consumer decision making. J Consum Res 1999; 26:278-92.. Rocha et al. 5151. Rocha LL, Pessoa MC, Gratão LHA, Carmo AS, Cordeiro NG, Cunha CF, et al. Characteristics of the school food environment affect the consumption of sugar-sweetened beverages among adolescents. Front Nutr 2021; 8:742744. found that the caloric contribution of ultra-processed foods to the total kilocalories consumed by adolescents was significantly higher in those who studied in private schools.

Unlike public schools, private schools are profit-driven institutions, regulated by the Brazilian Ministry of Education only in terms of educational features 5555. Ministério da Educação. Educação básica teve 47,3 milhões de matrículas em 2020. https://www.gov.br/pt-br/noticias/educacao-e-pesquisa/2021/01/educacao-basica-teve-47-3-milhoes-de-matriculas-em-2020 (accessed on 26/Jan/2022).
https://www.gov.br/pt-br/noticias/educac...
. This means that these institutions do not have nationwide regulations on the advertising and marketing of foods and beverages on their premises, and it is up to them to determine how these items will be made available in the school environment. The food environment of private schools is characterized by the sale of ultra-processed foods and beverages inside and around their premises, in addition to the presence of food advertising 5050. Carmo AS, Assis MM, Cunha CF, Oliveira TRPR, Mendes LL. The food environment of Brazilian public and private schools. Cad Saúde Pública 2018; 34:e00014918.,5656. Wognski ACP, Ponchek VL, Schueda Dibas EE, Orso MR, Vieira LP, Ferreira BGCS, et al. Comercialização de alimentos em cantinas no âmbito escolar. Braz J Food Technol 2019; 22:e2018198.,5757. Browne S, Staines A, Barron C, Lambert V, Susta D, Sweeney MR. School lunches in the Republic of Ireland: a comparison of the nutritional quality of adolescents' lunches sourced from home or purchased at school or "out" at local food outlets. Public Health Nutr 2017; 20:504-14.,5858. Callaghan M, Molcho M, Nic Gabhainn S, Kelly C. Food for thought: analysing the internal and external school food environment. Health Educ 2015; 115:152-70.,5959. Dia OEW, Løvhaug AL, Rukundo PM, Torheim LE. Mapping of outdoor food and beverage advertising around primary and secondary schools in Kampala city, Uganda. BMC Public Health 2021; 21:707..

A school environment is a privileged place for health and nutrition interventions, but when it is characterized as an obesogenic environment, there is a risk of immediate and long-term negative effects on the health of children and adolescents, especially concerning healthy habits and behaviors 22. WHO Regional Office for Europe. Factsheet for World Mental Health Day 2018. Adolescent mental health in the European Region. Geneva: World Health Organization; 2018.,6060. Sorhaindo A, Feinstein L. What is the relationship between child nutrition and school outcomes? London: Centre for Research on the Wider Benefits of Learning Institute of Education; 2006. (Wider Benefits of Learning Research Report, 18).. In this sense, it may contribute to an increase in the prevalence of obesity 6161. Potvin Kent M, Velazquez CE, Pauzé E, Cheng-Boivin O, Berfeld N. Food and beverage marketing in primary and secondary schools in Canada. BMC Public Health 2019; 19:114. and is also a risk factor for CMD in adolescents.

This study has some limitations, such as the use of a 24hR of only one day to construct the variable identifying ultra-processed foods consumption, which may imply a consumption that does not correspond to that of the adolescents evaluated and the possibility of recall bias and individual attrition. To ensure that the data from this recall would be collected in the best possible way, the multiple-pass interview technique was used 6262. Conway JM, Ingwersen LA, Vinyard BT, Moshfegh AJ. Effectiveness of the US Department of Agriculture 5-step multiple-pass method in assessing food intake in obese and nonobese women. Am J Clin Nutr 2003; 77:1171-8.. The GHQ-12 6363. Goldberg DP. The detection of psychiatric illness by questionnaire: a technique for the identification and assessment of non-psychotic psychiatric illness. Oxford: Oxford University Press; 1972., although validated for use in adolescents by French & Tait 2424. French DJ, Tait RJ. Measurement invariance in the General Health Questionnaire-12 in young Australian adolescents. Eur Child Adolesc Psychiatry 2004; 13:1-7., may be subject to recall bias and divergent responses, in addition to the possibility of underestimating the cases of adolescents who are treated for mental illness with medication that reduces the symptoms of anxiety and depression, in which case this adolescent, even if diagnosed, may not be identified by the GHQ. Lastly, it is not possible to state that the ultra-processed foods consumption came from meals eaten in the school environment. We emphasize that there is no way to infer causality in this study, because it is a cross-sectional study. This design measures everything at the same time, with no way to define the temporality between risk factors and development. It is important to clarify that the factors studied can only be risk markers.

Despite its limitations, this investigation used the ERICA Study database, which was carefully constructed, as well as all the stages of the study, from sampling to data collection, with the more than 71,000 adolescents evaluated, representative of the adolescent population.

Conclusions

In this study, it was possible to observe that the type of school funding, the presence of ultra-processed foods advertising, ultra-processed foods consumption and increased WHtR are risk factors for CMD in Brazilian adolescents.

This study showed the importance of the school environment as a health-promoting place and how characteristics of this environment can contribute to the presence of CMD in adolescents. Due to the study design, it was not possible to determine causal relationships, leaving gaps in how the food environment of private schools could exert this relationship with mental health in this age group.

Acknowledgments

We thank the research committee of ERICA for making the data available and Brazilia Coordination for the Improvent of Higher Education Personnel/Brazilian National Research Council (CAPES/CNPq) for the PhD scholarships of Lucia Helena Almeida Gratão, Luana Lara Rocha, Mariana Zogbi Jardim, and Thales Philipe Rodrigues da Silva (MCT/FINEP/MS/SCTIE/DECIT - CT/SAÚDE e FNS - síndrome metabólica - 01/2008).

References

  • 1
    Blakemore SJ, Burnett S, Dahl RE. The role of puberty in the developing adolescent brain. Hum Brain Mapp 2010; 31:926-33.
  • 2
    WHO Regional Office for Europe. Factsheet for World Mental Health Day 2018. Adolescent mental health in the European Region. Geneva: World Health Organization; 2018.
  • 3
    Goldberg DP, Huxley P. Common mental disorders: a bio-social model. New York: Tavistock/Routledge; 1992.
  • 4
    Goldberg D. A bio-social model for common mental disorders. Acta Psychiatr Scand Suppl 1994; 385:66-70.
  • 5
    Ferrari AJ. Charlson FJ, Norman RE, Patten SB, Freedman G, Murray CJL, et al. Burden of depressive disorders by country, sex, age, and year: findings from the Global Burden of Disease Study 2010. PLoS Med 2013; 10:e1001547.
  • 6
    Organização Pan-Americana da Saúde; Organização Mundial da Saúde. Saúde mental dos adolescentes. https://www.paho.org/pt/topicos/saude-mental-dos-adolescentes (accessed on 26/Jan/2022).
    » https://www.paho.org/pt/topicos/saude-mental-dos-adolescentes
  • 7
    Gili M, Castellví P, Vives M, Torre-Luque A, Almenara J, Blasco MJ, et al. Mental disorders as risk factors for suicidal behavior in young people: a meta-analysis and systematic review of longitudinal studies. J Affect Disord 2019; 245:152-62.
  • 8
    Buttazzoni A, Doherty S, Minaker L. How do urban environments affect young people's mental health? A novel conceptual framework to bridge public health, planning, and neurourbanism. Public Health Rep 2022; 137:48-61.
  • 9
    Zhang L, Wu L. Community environment perception on depression: the mediating role of subjective social class. Int J Environ Res Public Health 2021; 18:8083.
  • 10
    Alves CCR, Fonseca MA, Carmo AP, Silveira LSS, Senra K. Documento orientador da Política de Educação Integral e Integrada e implementação das escolas polo de educação múltipla em Minas Gerais. https://acervodenoticias.educacao.mg.gov.br/images/documentos/Documento%20Pol%C3%ADtica%20de%20Educa%C3%A7%C3%A3o%20Integral%20e%20Integrada%20FINAL.pdf (accessed on 27/Nov/2022).
    » https://acervodenoticias.educacao.mg.gov.br/images/documentos/Documento%20Pol%C3%ADtica%20de%20Educa%C3%A7%C3%A3o%20Integral%20e%20Integrada%20FINAL.pdf
  • 11
    Institute of Medicine; Board on Children, Youth, and Families; Food and Nutrition Board; Committee on Food Marketing and the Diets of Children and Youth. Food marketing to children and youth: threat or opportunity? Washington DC: National Academies Press; 2006.
  • 12
    Werneck AO, Costa CS, Horta B, Wehrmeister FC, Gonçalves H, Menezes AMB, et al. Prospective association between ultra-processed food consumption and incidence of elevated symptoms of common mental disorders. J Affect Disord 2022; 312:78-85.
  • 13
    Adjibade M, Julia C, Allès B, Touvier M, Lemogne C, Srour B, et al. Prospective association between ultra-processed food consumption and incident depressive symptoms in the French NutriNet-Santé cohort. BMC Med 2019; 17:78.
  • 14
    Lane MM, Lotfaliany M, Hodge AM, O'Neil A, Travica N, Jacka FN, et al. High ultra-processed food consumption is associated with elevated psychological distress as an indicator of depression in adults from the Melbourne Collaborative Cohort Study. J Affect Disord 2023; 335:57-66.
  • 15
    Lane MM, Gamage E, Travica N, Dissanayaka T, Ashtree DN, Gauci S, et al. Ultra-processed food consumption and mental health: a systematic review and meta-analysis of observational studies. Nutrients 2022; 14:2568.
  • 16
    Lee S, Choi M. Ultra-processed food intakes are associated with depression in the general population: the Korea National Health and Nutrition Examination Survey. Nutrients 2023; 15:2169.
  • 17
    Arshad H, Head J, Jacka FN, Lane MM, Kivimaki M, Akbaraly T. Association between ultra-processed foods and recurrence of depressive symptoms: the Whitehall II cohort study. Nutr Neurosci 2024; 27:42-54.
  • 18
    Mesas AE, González AD, De Andrade SM, Martínez-Vizcaíno V, López-Gil JF, Jiménez-López E. Increased consumption of ultra-processed food is associated with poor mental health in a nationally representative sample of adolescent students in Brazil. Nutrients 2022; 14:5207.
  • 19
    Faisal-Cury A, Leite MA, Escuder MML, Levy RB, Peres MFT. The relationship between ultra-processed food consumption and internalising symptoms among adolescents from São Paulo city, Southeast Brazil. Public Health Nutr 2022; 25:2498-506.
  • 20
    Safaei M, Sundararajan EA, Driss M, Boulila W, Shapi'i A. A systematic literature review on obesity: understanding the causes & consequences of obesity and reviewing various machine learning approaches used to predict obesity. Comput Biol Med 2021; 136:104754.
  • 21
    Bloch KV, Szklo M, Kuschnir MCC, Abreu GA, Barufaldi LA, Klein CA, et al. The study of cardiovascular risk in adolescents - ERICA: rationale, design and sample characteristics of a national survey examining cardiovascular risk factor profile in Brazilian adolescents. BMC Public Health 2015; 15:94.
  • 22
    Silva TLN, Klein CH, Souza AM, Barufaldi LA, Abreu GA, Kuschnir MCC, et al. Response rate in the Study of Cardiovascular Risks in Adolescents - ERICA. Rev Saúde Pública 2016; 50 Suppl 1:3s.
  • 23
    Vasconcellos MTL, Silva PLN, Szklo M, Kuschnir MCC, Klein CH, Abreu GA, et al. Sampling design for the Study of Cardiovascular Risks in Adolescents (ERICA). Cad Saúde Pública 2015; 31:921-30.
  • 24
    French DJ, Tait RJ. Measurement invariance in the General Health Questionnaire-12 in young Australian adolescents. Eur Child Adolesc Psychiatry 2004; 13:1-7.
  • 25
    Goldberg DP, Williams P. A user's guide to the General Health Questionnaire: GHQ. London: GL Assessment; 1988.
  • 26
    Goldberg DP, Gater R, Sartorius N, Ustun TB, Piccinelli M, Gureje O, Rutter C. The validity of two versions of the GHQ in the WHO study of mental illness in general health care. Psychol Med 1997; 27:191-7.
  • 27
    Instituto Brasileiro de Geografia e Estatística. Pesquisa de Orçamentos Familiares 2002-2003: perfil das despesas no Brasil: indicadores selecionados. Rio de Janeiro: Instituto Brasileiro de Geografia e Estatística; 2007.
  • 28
    Instituto Brasileiro de Geografia e Estatística. Pesquisa de Orçamentos Familiares 2008-2009: análise do consumo alimentar pessoal no Brasil. Rio de Janeiro: Instituto Brasileiro de Geografia e Estatística; 2011.
  • 29
    Monteiro CA, Cannon G, Levy R, Moubarac JC, Jaime P, Martins AP, et al. NOVA. The star shines bright. World Nutr 2016; 7:28-38.
  • 30
    Yoo EG. Waist-to-height ratio as a screening tool for obesity and cardiometabolic risk. Korean J Pediatr 2016; 59:425.
  • 31
    Lopes CS, Abreu GA, Santos DF, Menezes PR, Carvalho KMB, Cunha CF, et al. ERICA: prevalence of common mental disorders in Brazilian adolescents. Rev Saúde Pública 2016; 50 Suppl 1:14s.
  • 32
    Instituto Brasileiro de Geografia e Estatística. Pesquisa Nacional de Saúde do Escolar, 2015. Rio de Janeiro: Instituto Brasileiro de Geografia e Estatística; 2016.
  • 33
    Ribeiro IBS, Correa MM, Oliveira G, Cade NV. Common mental disorders and socioeconomic status in adolescents of ERICA. Rev Saúde Pública 2020; 54:4.
  • 34
    Hecht EM, Rabil A, Steele EM, Abrams GA, Ware D, Landy DC, et al. Cross-sectional examination of ultra-processed food consumption and adverse mental health symptoms. Public Health Nutr 2022; 25:3225-34.
  • 35
    Coletro HN, Mendonça RD, Meireles AL, Machado-Coelho GLL, Menezes MC. Multiple health risk behaviors, including high consumption of ultra-processed foods and their implications for mental health during the COVID-19 pandemic. Front Nutr 2022; 9:1042425.
  • 36
    Contreras-Rodriguez O, Reales-Moreno M, Fernández-Barrès S, Cimpean A, Arnoriaga-Rodríguez M, Puig J, et al. Consumption of ultra-processed foods is associated with depression, mesocorticolimbic volume, and inflammation. J Affect Disord 2023; 335:340-8.
  • 37
    Chen H, Hou Y, Yang H, Wang X, Xu C. The associations of dietary patterns with depressive and anxiety symptoms: a prospective study. BMC Med 2023; 21:307.
  • 38
    Boulangé CL, Neves AL, Chilloux J, Nicholson JK, Dumas ME. Impact of the gut microbiota on inflammation, obesity, and metabolic disease. Genome Med 2016; 8:42.
  • 39
    Godos J, Currenti W, Angelino D, Mena P, Castellano S, Caraci F, et al. Diet and mental health: review of the recent updates on molecular mechanisms. Antioxidants (Basel) 2020; 9:346.
  • 40
    Mannino A, Daly A, Dunlop E, Probst Y, Ponsonby AL, Mei IAF, et al. Higher consumption of ultra-processed foods and increased likelihood of central nervous system demyelination in a case-control study of Australian adults. Eur J Clin Nutr 2023; 77:611-4.
  • 41
    Cenit MC, Sanz Y, Codoñer-Franch P. Influence of gut microbiota on neuropsychiatric disorders. World J Gastroenterol 2017; 23:5486-98.
  • 42
    Macht M. How emotions affect eating: a five-way model. Appetite 2008; 50:1-11.
  • 43
    Keck MM, Vivier H, Cassisi JE, Dvorak RD, Dunn ME, Neer SM, et al. Examining the role of anxiety and depression in dietary choices among college students. Nutrients 2020; 12:2061.
  • 44
    Ashwell M, Cole TJ, Dixon AK. Ratio of waist circumference to height is strong predictor of intra-abdominal fat. BMJ 1996; 313:559-60.
  • 45
    Scott KM, McGee MA, Wells JE, Oakley Browne MA. Obesity and mental disorders in the adult general population. J Psychosom Res 2008; 64:97-105.
  • 46
    Lewis-de los Angeles WW, Liu RT. History of depression, elevated body mass index, and waist-to-height ratio in preadolescent children. Psychosom Med 2021; 83:1075-81.
  • 47
    Rivenes AC, Harvey SB, Mykletun A. The relationship between abdominal fat, obesity, and common mental disorders: results from the HUNT Study. J Psychosom Res 2009; 66:269-75.
  • 48
    Ministério da Saúde. Eu quero me alimentar melhor. A influência da publicidade nas escolhas alimentares. https://www.gov.br/saude/pt-br/assuntos/saude-brasil/eu-quero-me-alimentar-melhor/noticias/2023/a-influencia-da-publicidade-nas-escolhas-alimentares (accessed on 22/Mar/2024).
    » https://www.gov.br/saude/pt-br/assuntos/saude-brasil/eu-quero-me-alimentar-melhor/noticias/2023/a-influencia-da-publicidade-nas-escolhas-alimentares
  • 49
    Mattos MC, Nascimento PCBD, Almeida SS, Costa TMB. Influência de propagandas de alimentos nas escolhas alimentares de crianças e adolescentes. Psicol Teor Prat 2010; 12:34-51.
  • 50
    Carmo AS, Assis MM, Cunha CF, Oliveira TRPR, Mendes LL. The food environment of Brazilian public and private schools. Cad Saúde Pública 2018; 34:e00014918.
  • 51
    Rocha LL, Pessoa MC, Gratão LHA, Carmo AS, Cordeiro NG, Cunha CF, et al. Characteristics of the school food environment affect the consumption of sugar-sweetened beverages among adolescents. Front Nutr 2021; 8:742744.
  • 52
    Azeredo CM, Rezende LFM, Canella DS, Claro RM, Peres MFT, Luiz OC, et al. Food environments in schools and in the immediate vicinity are associated with unhealthy food consumption among Brazilian adolescents. Prev Med 2016; 88:73-9.
  • 53
    Zimmerman FJ, Shimoga SV. The effects of food advertising and cognitive load on food choices. BMC Public Health 2014; 14:342.
  • 54
    Shiv B, Fedorikhin A. Heart and mind in conflict: the interplay of affect and cognition in consumer decision making. J Consum Res 1999; 26:278-92.
  • 55
    Ministério da Educação. Educação básica teve 47,3 milhões de matrículas em 2020. https://www.gov.br/pt-br/noticias/educacao-e-pesquisa/2021/01/educacao-basica-teve-47-3-milhoes-de-matriculas-em-2020 (accessed on 26/Jan/2022).
    » https://www.gov.br/pt-br/noticias/educacao-e-pesquisa/2021/01/educacao-basica-teve-47-3-milhoes-de-matriculas-em-2020
  • 56
    Wognski ACP, Ponchek VL, Schueda Dibas EE, Orso MR, Vieira LP, Ferreira BGCS, et al. Comercialização de alimentos em cantinas no âmbito escolar. Braz J Food Technol 2019; 22:e2018198.
  • 57
    Browne S, Staines A, Barron C, Lambert V, Susta D, Sweeney MR. School lunches in the Republic of Ireland: a comparison of the nutritional quality of adolescents' lunches sourced from home or purchased at school or "out" at local food outlets. Public Health Nutr 2017; 20:504-14.
  • 58
    Callaghan M, Molcho M, Nic Gabhainn S, Kelly C. Food for thought: analysing the internal and external school food environment. Health Educ 2015; 115:152-70.
  • 59
    Dia OEW, Løvhaug AL, Rukundo PM, Torheim LE. Mapping of outdoor food and beverage advertising around primary and secondary schools in Kampala city, Uganda. BMC Public Health 2021; 21:707.
  • 60
    Sorhaindo A, Feinstein L. What is the relationship between child nutrition and school outcomes? London: Centre for Research on the Wider Benefits of Learning Institute of Education; 2006. (Wider Benefits of Learning Research Report, 18).
  • 61
    Potvin Kent M, Velazquez CE, Pauzé E, Cheng-Boivin O, Berfeld N. Food and beverage marketing in primary and secondary schools in Canada. BMC Public Health 2019; 19:114.
  • 62
    Conway JM, Ingwersen LA, Vinyard BT, Moshfegh AJ. Effectiveness of the US Department of Agriculture 5-step multiple-pass method in assessing food intake in obese and nonobese women. Am J Clin Nutr 2003; 77:1171-8.
  • 63
    Goldberg DP. The detection of psychiatric illness by questionnaire: a technique for the identification and assessment of non-psychotic psychiatric illness. Oxford: Oxford University Press; 1972.

Publication Dates

  • Publication in this collection
    17 May 2024
  • Date of issue
    2024

History

  • Received
    10 Apr 2023
  • Reviewed
    22 Dec 2023
  • Accepted
    17 Jan 2024
Escola Nacional de Saúde Pública Sergio Arouca, Fundação Oswaldo Cruz Rio de Janeiro - RJ - Brazil
E-mail: cadernos@ensp.fiocruz.br