Artigos

Conteúdo (article):

A scale to assess teachers’ dietary knowledge by the Brazilian Dietary Guidelines

Thanise Sabrina Souza Santos1
Geralda Kelen Fonseca2
Mariana Mendes Pereira3
Marise Fagundes Silveira4
Lucinéia de Pinho5

AFFILIATIONS:
1Núcleo de Pesquisas Epidemiológicas em Nutrição e Saúde. Universidade de São Paulo, Brasil. Orcid: https://orcid.org/0000-0003-4087-1815. Email: thanisesouza@gmail.com
2Faculdades Unidas do Norte de Minhas. Brasil. Orcid: https://orcid.org/0000-0003-1015-7688. Email: kelensme@gmail.com
3Universidade Estadual de Montes Claros. Brasil. Orcid: https://orcid.org/0000-0002-1847-1746. Email: mariana.mendes@edu.montesclaros.mg.gov.br
4Programa de Pós-graduação em Ciências da Saúde. Universidade Estadual de Montes Claros, Brasil. Orcid: https://orcid.org/0000-0002-8821-3160. Email: marise.silveira@unimontes.br
5Universidade Estadual de Montes Claros, Brasil. Orcid: https://orcid.org/0000-0002-2947-5806. Email: lucineiapinho@hotmail.com

ABSTRACT
This psychometric study adapted and validated a scale to assess teachers’ knowledge of healthy eating. After selecting an instrument (GAB1), its wording was adapted to the school environment and the clarity, importance and relevance of the 16 statements were assessed. Teachers’ responses to the adapted questionnaire were analysed using a three-parameter logistic model based on Item Response Theory. Assessment of parameters for retaining or excluding statements from the scale was followed by analysis of fit. Levels of interpretation of the scale were proposed and descriptive analyses were carried out to characterise the teachers. The statements presented satisfactory clarity, importance and relevance. After analysing the teachers’ responses, the final 10-item model returned adequate goodness-of-fit (RMSEA = 0.012; SRMSR = 0.042; TLI = 0.99 and CFI = 0.99). Three levels of knowledge were identified: food choice; from food to meals; and food and nutrition education. Teachers who worked in schools where healthy eating was included in programme content were more knowledgeable. The GAB1-professores scale contributes to planning and evaluating teacher qualification processes.

Keywords: Food Guide. Food and Nutrition Education. Surveys and Questionnaires. Psychometrics.

INTRODUCTION
Eating habits acquired in childhood can be associated with health, disease and diet in later years1. Given the relationship between unhealthy eating habits and disease1,2, early intervention during childhood and adolescence can prevent health problems.
During these periods, schools play an important role in forming eating habits and thus figure as a locus for the implementation of public policies to promote appropriate and healthy eating among children and adolescents3. In that regard, the UNICEF Nutrition Strategy 2020 to 2030 is designed to support action through policies and programmes in the school context to prevent poor nutrition by introducing Food and Nutrition Education (FNE) into school curricula4.
These interventions are to be aligned with current national recommendations for appropriate, healthy food. The first edition of the Brazilian Dietary Guidelines was published in 20065. Given the population’s changing epidemiological and nutritional profile and aligning with the World Health Organisation’s recommendations for periodically updating nutritional guidelines, the Ministry of Health issued a second edition of the Dietary Guidelines in 20146. The central recommendation includes prioritising fresh or minimally processed foods and culinary preparations and avoiding ultraprocessed foods6.
As regards school meals in Brazil, the national school meals programme (Programa Nacional de Alimentação Escolar, PNAE) operates to foster and form healthy eating habits by setting out guidelines for school meals, encouraging the inclusion of varied, regional foods and acknowledging the value of local dietary culture. Note that the PNAE guidelines take the Dietary Guidelines as their frame of reference and recommend that FNE measures be introduced into schools’ educational project, signalling the importance of classroom discussions on the subject7.
In that scenario, teacher training is fundamental to fostering teachers’ knowledge and skills in implementing FNE activities with their pupils8. In Brazil, teacher training occurs predominantly in courses framed by the “banking” approach and, in the classroom, teachers are observed to be limited in their handling of the subject of healthy eating9. Also, since the Dietary Guidelines were updated6, instruments with evidence of validity for assessing primary and secondary school teachers’ knowledge of healthy diet can be seen to be lacking in Brazil. The provision and validation of such an instrument can help assess how prepared teachers are to undertake FNE activities with children and adolescents in the school environment. In view of the foregoing, the GAB-1, an instrument used in the primary health care context, was adapted and validated to measure primary and secondary school teachers’ knowledge of healthy eating in keeping with the Dietary Guidelines.

METHODS
This psychometric study was conducted in three stages: selection and adaptation of the instrument (Stage 1); content validation (Stage 2); and internal structure validation (Stage 3).

Stage 1: instrument selection and adaptation
The GAB1 instrument considered here was developed originally to assess primary health care personnel’s knowledge of the Dietary Guidelines’ content and approach10. Given the instrument’s alignment with the current version of the Dietary Guidelines6 and with the authors’ permission, that instrument was selected for use in the study reported here to assess primary and secondary school teachers’ knowledge. The statements were adjusted to the educational environment. Words and/or sentences in the instrument were rewritten without altering the construct under investigation. The instrument’s original structure was maintained: 16 statements, each with three response options: true, false and I don’t know11. The GAB1 presents statements that implicitly address knowledge of the benefits and harm of foods at various stages of processing, as well as knowledge of commensality and setting in the act of eating, and their impacts on healthy eating habits. It comprises five theoretical dimensions: 1- a new paradigm of healthy eating; 2- how to choose foods; 3- how to assemble meals; 4- the act of eating and commensality: regularity and attention in eating, eating in appropriate settings and eating in company; and 5- how to overcome obstacles: information, supply, cost, culinary skills, time and publicity10.

Stage 2: content validation
Content validation of the version adapted in Stage 1 was performed by an online panel of judges to evaluate clarity, importance, relevance and representation of the dimensions of the theoretical frame of reference (the Dietary Guidelines). The judges included graduates in Nutrition who were active professionally in relation to school meals, and graduates in Education who had teaching experience. Fifteen judges were invited, by e-mail, to take part and received a manual with guidelines about the study and instructions on content validation. They were also sent the digital version of the Dietary Guidelines6 and asked to read it beforehand.
Clarity, importance and relevance were scored from 1 to 4, where scores of 1 and 2 indicated that the statements were not clear, important or relevant, while scores of 3 and 4 indicated that the statements met all the criteria satisfactorily. In order to assess the theoretical dimension, the experts were asked to classify each statement by theoretical dimension of the Dietary Guidelines. They were also asked to consider suggesting statements for inclusion or elimination. Their responses served to calculate proportions of agreement among the experts on the clarity, importance and relevance of each statement. The number of satisfactory scores (3 and 4) for each aspect were added and the total then divided by the number of judges. The values obtained for each aspect were added together in order then to calculate mean agreement, that is, the content validity index (CVI), for each statement. The mean of all CVIs gave the total index for the instrument (CVIt). CVI and CVIt were considered satisfactory at values > 0.811.

Stage 3: internal structure validation
For this stage, teachers from the municipality of Montes Claros, Minas Gerais, Brazil, working actively in urban and rural areas were invited to take part. At the time of the study, the municipal education system comprised a total of 1,491 teachers working in daycare/preschool, primary, secondary and adult and youth education. At this stage of the study, a sample size of 500 participants was established so as to obtain very close estimates when working with larger samples12–14.
The Montes Claros municipal education department was contacted, and authorised the study. Then the heads of schools were informed of the study and invited to assist in persuading teachers to participate. Once assessed by the panel of judges, the version of the instrument was input to Google Forms, including also the statements on identification particulars (name, contact information, sex, age, self-reported skin colour and marital status), socioeconomic conditions (family income and number of persons living in the household), professional background and position (schooling, length of time working as a teacher, in which school system and participation in continued professional development) and addressing food and nutrition content in the classroom. Those who agreed to participate in the study received a link through a group in a social network in order to access the online questionnaire.
The teachers’ right and wrong answers as regards healthy eating were identified by reference to the response guide to the instrument statements. Responses were coded as: 1- ‘wrong’, 2- ‘right’ and P- ‘I don’t know’. Those whose answers were all wrong (n = 1); all right (n = 2); or all ‘I don’t know’ (n = 7) were excluded from subsequent analysis. On the basis of the one-dimensionality considered by the authors of GAB110, responses were analysed by applying confirmatory full-information factor analysis. For a statement to be retained for subsequent analysis, factor load of > 0.30 and communality > 0.20 were considered acceptable15.
Item response theory (IRT) was applied in order to estimate the discrimination, difficulty and guessing parameters in applying a three-parameter logistical model. That model is represented by the following equation:
P(U_ij = 1|θ_j ) = c_i+(1+c_i ) 1/(1+e^(〖-Da〗_i (θ_j-b_i ) ) )
Where:
θj – IRT score for individual j;
P(Uij = 1|θj) – probability that individual j with ability θj will respond correctly to statement i;
D – constant scale factor equal to 1.7;
ci – guessing parameter for statement i;
ai – discrimination parameter for statement i;
bi – difficulty parameter for statement i.

The discrimination parameter (ai) indicates the item’s ability to discriminate individuals with different levels of knowledge about healthy eating. The difficulty parameter (bi) identifies the point on the scale at which the probability of a correct answer is greater than or equal to (1+ ci)/216. The guessing parameter (ci) indicates the probability of giving a right answer by chance, with values from zero to one (0 ≤ ci ≤ 1). All parameters (ai, bi and ci) were estimated by marginal maximum likelihood with a maximum of twenty quadrature points, 500 expectation-maximisation cycles and convergence criterion of 0.001.
The estimates of these parameters were examined with the corresponding standard errors. Statements with an ai parameter of 0.70 or less were considered less discriminatory and were excluded from subsequent analysis17. Statements with a bi parameter with high standard errors compared with other statements18 were excluded from subsequent analyses. Assessment of ci parameter estimates considered the magnitude of the likelihood of answering correctly by chance.
The following indices were used as indicators of goodness of fit: root mean square error of approximation (RMSEA), standardised root mean square residual (SRMSR), comparative fit index (CFI) and Tucker-Lewis index (TLI). Fit was considered to be good when RMSEA and SRMSR < 0.08 and CFI and TLI > 0.9519.
After identifying the model with best fit, the teachers’ scores for knowledge of healthy eating were calculated by the expected a posteriori method, on a scale with a mean of zero and a standard deviation of one. Levels of knowledge were ascertained by calculating the probability of a right answer for each statement at different scores, using the three-parameter logistical method. Each statement was positioned at the score at which the probability of a right answer was equal to or greater than (1+ ci)/2. The levels of knowledge about healthy eating were specified according to this positioning and the technical similarity between the statements identified in discussions between authors TSSS and LP.
In order for the levels to be more easily interpreted, the scores estimated on a scale with a mean of zero and a standard deviation of one were subsequently transformed to a scale with a mean of 50 and standard deviation of 10.
The data were processed electronically using IBM SPSS Statistics for Windows, version 22 (IBM Corp., Armonk, N.Y., USA), with Amos. The Stage 3 and descriptive analyses of the sample were performed using R Studio software. The normality of distribution of the quantitative variables (age, per capita income and time as a teacher) was investigated by the Shapiro-Wilk test and the association with knowledge about healthy eating was assessed by Spearman correlation. The Fisher exact test was used to assess differences among percentages in the socioeconomic, professional and work process variables, by level of knowledge about healthy eating.
The ethical aspects of this study conformed with National Health Council Resolution No. 466 of 12 December 2012. The institution’s declaration of agreement was signed by the Montes Claros Municipal Education Secretary and all participating teachers signalled their agreement in a declaration of free and informed consent. This study was approved by the research ethics committee of the Universidade Estadual de Montes Claros as in Opinion No. 3.586.107.

RESULTS

Stage 1: instrument selection and adaptation
The 16 statements of the GAB1 were maintained in the version adapted to the school context. In statement 1, the expression “usuários de serviços de saúde” [health service users] was changed to “alunos” [pupils] and “profissional de saúde” [health professional], to “professores” [teachers]. No changes were required in the other statements, which were suited to the proposed study.

Stage 2: content validation
At this stage, the participants were 11 expert professionals, 10 of them nutritionists and one an education specialist, from the Northeast, South and Southeast macro regions of Brazil. Content validation returned CVIt = 0.94, which was considered satisfactory, thus approving the scale. No alterations were suggested to the wording of statements, confirming that the language was clear and appropriate, the content was important and relevant and also that the dimensions of the theoretical frame of reference were satisfactorily contemplated throughout the instrument.

Stage 3: internal structure validation
The results of factor analysis and IRT are described in Tables 1 and 2. All the statements returned appropriate factor loads. Statements 13, 15 and 3 were excluded, separately and in that order, because they returned low communality values. The models tested up to the exclusion of statement 3 returned poor fit (TLI and/or CFI ≤ 0.95). Statement 14 returned a bi parameter with high standard error in comparison with the other statements and was excluded. Statements 12 and 2 returned high estimates for the ci parameter and were excluded, separately and in that order. The final model with the 10 remaining statements (1, 4, 5, 6, 7, 8, 9, 10, 11 and 16) returned proper fit (RMSEA = 0.012; SRMSR = 0.042; TLI = 0.99 and CFI = 0.99) and the statements returned appropriate factor loads, communalities and ai, bi and ci parameters. The GAB1 adapted for teachers (GAB1-professores, in the original Portuguese) displayed greater measurement precision between scores zero and two. The final questionnaire used to assess the teachers’ knowledge of healthy eating is shown in Chart 1.

[INSERT TABLE 1]

[INSERT TABLE 2]

[INSERT CHART 1]

Statement 11, “As long as they are baked, chicken nuggets can replace chicken meat at lunch or dinner”, returned the best discrimination parameter (a11 = 1.607), while statement 8, “Bean stew should be avoided, because it is a preparation that combines fresh and minimally processed foods with processed cooking ingredients”, returned the lowest discrimination (a8 = 0.746).
As regards parameter bi, statements 1, “Food and nutrition education strategies that expand pupils’ autonomy are those that make them obey the nutrition guidance given by their teacher”, and 8, “Bean stew should be avoided, because it is a preparation that combines fresh and minimally processed foods with processed cooking ingredients”, were considered the most difficult (b1 = 2.971; b8 = 1.748, respectively). Meanwhile, statement 10, “The label on a bottle of passionfruit juice describes the following ingredients: 100% passionfruit juice, water, passionfruit juice concentrate, citric acid acidulant, aroma identical to natural passionfruit, xanthan gum stabiliser, sodium benzoate and sodium metabisulphite preservatives, natural carmine colouring and polydimethylsiloxane antifoaming agent. Examination of the label indicates that this is an ultraprocessed food” was considered the easiest (b10 = -1.088).
In order to facilitate interpretation, scores can be transformed to a scale with mean of 50 and standard deviation of 10 by means of the constant, θ(50,10) = 10 × θ(0,1) + 50. By positioning the statements, it was possible to identify three levels of knowledge about healthy eating: choosing foods; from foods to meals; and FNE, which are described in the Supplementary Material, Chart 1.
Level 1 comprised knowledge about choosing foods, including the statements about: how eating foods such as rice, beans and nuts relates to body weight (7), reading labels (10), replacing fresh meats (11) and reliable sources of information on healthy eating (16). Level 2 comprised knowledge about choosing and combining foods, including also the statements about: food prices (4) and combining foods in meals (5, 6 and 9) and in culinary preparations (8). Level 3 comprised knowledge about choosing and combining foods and the importance of FNE activities, including also the statement about the potential of activities to promote autonomy in choosing foods (1).
The characteristics of the teachers who participated in this stage of the study, considering the total sample and the sample stratified by level of knowledge about healthy eating, are shown in Tables 3 and 4. Of the total of 591 teachers participated in the study, most were female (92.4%), aged from 23 to 55 years (93.2%) and university graduates (64.5%). None displayed knowledge at Level 3 (FNE). Teachers with more schooling and income and who worked in schools where the subject of healthy eating formed part of programme content were at Level 2 on the scale (from foods to meals). Meanwhile, those teachers who reported addressing food sources in the classroom were at Level 1 on the scale (choosing foods) (p < 0.05).

[INSERT TABLE 3]

[INSERT TABLE 4]

DISCUSSION
In order to adapt the GAB-1 so as to permit assessment of primary and secondary school teachers’ knowledge about healthy eating, the expression “health service users” had to be replaced by “pupils” and “healthcare personnel”, by “teachers”. The panel of judges approved the adapted version, confirming the clarity of the content and the importance and relevance of the instrument. During internal structure validation, some statements were found not to bring in information to the instrument or to return a low degree of precision. In that connection, of the 16 adapted statements, 10 were maintained as returning satisfactory fit (RMSEA = 0.012; SRMSR = 0.042; TLI = 0.99 and CFI = 0.99). A scale was then proposed for interpreting knowledge about healthy eating at three levels: choosing foods; from foods to meals; and FNE. Among the teachers who participated in this study, the highest level of knowledge was displayed by those with most schooling and income and who were working in an institution where healthy eating was included in programme content. However, the potential of FNE activities was not recognised, pointing to a need for a higher level of knowledge about healthy eating.
Level 1 comprised knowledge about choosing foods in agreement with the golden rule of the Dietary Guidelines, that is: fresh or minimally processed foods should be the basis of diets in Brazil6. Also at this first level of the scale, teachers were knowledgeable about reliable sources of information, reading food labels and the relationship between food choices and health. This knowledge is extremely important, given that ultraprocessed foods have become increasingly available to Brazilian households in recent decades and the relative participation of fresh or minimally processed foods and culinary ingredients can be seen to be declining20. By sharing this knowledge with their pupils, teachers can support them in recognising foods by level of processing and thus contribute to food choices aligned with the recommendations of the Dietary Guidelines and to reducing the availability of ultraprocessed foods in households.
Also at Level 1, recognition of the relationship between unhealthy eating and health problems is essential in order to promote more appropriate eating habits among the pupils. Studies that include children and adolescents have shown the relationship between increasing consumption of ultraprocessed foods and impacts on health, including overweight21,22, metabolic risk, wheezing22, oral health problems21 and impaired sleep quality and duration23. In a scenario of increasing consumption of ultraprocessed foods in Brazil’s population24,25, especially among black, indigenous and low-income groups24, it is urgent for coordinated inter-sector action to be taken to advocate for a food system that promotes appropriate, healthy eating. Public school teachers who are knowledgeable about healthy eating can contribute to more appropriate food environments and practices and thus leverage endeavours to discourage consumption of ultraprocessed foods among their pupils and to promote dialogue with other sectors, such as health, urban farming, supply and so on.
Amid the large amounts of information broadcast about foods, special attention should be directed to abusive practices that contribute to maintaining the hegemonic food system. Governments and the media construct different meanings about what is understood by healthy eating and the agents of the foods industry circulate information favourable to their interests, their products and the corporations themselves, thus influencing science, policy making and the information that is broadcast about nutrition26,27. Level 1 includes knowledge about reliable sources of information, which is strategic to a critical discussion, with no conflicts of interest, of the quality of foods in the school environment. To complement the work of the teachers, policies to limit the sale of ultraprocessed foods in schools can also contribute to a school environment that promotes appropriate, healthy eating, free from the abusive influence of large corporations28.
Level 2 of the scale elicits knowledge about choosing and combining foods. It is known that in order to understand and discuss food choices, it is necessary to understand their determinants, such as preference, health, convenience, tradition, price and socialisation29,30. In that context, by recognising the influence of prices on choices, teachers can work with their pupils on decision-making strategies and discussing factors such as the healthiness of foods and their costs and benefits within financial realities and identifying public facilities that promote appropriate, healthy eating. Burlandy et al.31 highlight the need to move away from a discourse centred on traditionalism, nutritional and biological rationality, nutritionism and the logic of individual accountability for food choices. It is thus fundamentally important to incorporate different dimensions of healthy eating, with reference to food systems, food and nutrition security and the human right to appropriate, healthy diet31. In that context, school vegetable gardens are one example of a teaching strategy with the potential to introduce food into the syllabus of disciplines by providing opportunities for interpersonal experience outside the classroom for the whole school community and overall contact with foods, from cultivation through to preparation32.
Level 2 of the scale included the statement with the highest degree of difficulty, which considered the composition of bean stew, a dish highly representative of Brazilian cuisine. The bean stew that has resulted from ethnic and cultural mixing and the social and economic scenario comprises black beans plus fresh, processed and ultra-processed foods. A typical food is culturally representative of a region, a people or a nation and these aspects demonstrate that diet is sociocultural and can be described as part of a system mediated by socialisation33. However, despite its symbolic importance to Brazilian culinary culture, bean stew is a preparation also connected with ‘forbidden’ and ‘unhealthy’ things, because of its calorific value and the presence of ultraprocessed foods. In that connection, nutritionism rooted in a conception of appropriate, healthy food offers important justification of the need for a higher level of knowledge in order to answer the statement correctly.
It was also found that no teacher displayed Level-3 knowledge, which would have included recognition of the reach of Food and Nutrition Education activities. The FNE field involves actions and strategies to promote knowledge, skills and attitudes relating to healthy eating. For that purpose, the actions should not concentrate on transmitting content and changing behaviour; rather, individuals should be encouraged to engage in critical thinking in order to develop their autonomy in food choices34. Given that schools are ideal spaces for promoting healthy eating habits, it is planned to conduct FNE activities as part of the national school meals programme, PNAE7. However, although this policy is designed to ensure more long-lasting food and nutrition security – and despite the fact that the policy reference framework for the food and nutrition education (Marco de Referência de Educação Alimentar e Nutricional para Políticas Públicas34) – was published more than 10 years ago, other findings of this study reinforce the lack of recognition for the potential of FNE: approximately 55% of the teachers in this study did not participate in capacity-building in healthy eating and more than 80% had no connection with the activities of the Health at School Programme (Programa Saúde na Escola, PSE). In a scenario, as in this study, where participants vary in how long they have been teaching, they need support in conducting FNE activities more systematically, transversely and in articulation with primary health care teams and the school community35,36. Strategies proposed include introducing the subjects of food and nutrition into teaching books and educational materials37 and preparing technical material with suggestions for educational activities coordinated with the school curriculum and pupils’ psychomotor, cognitive and social development36. However, their gains and limitations need to be evaluated to identify their impact on FNE in schools, as well as on schoolchildren’s dietary practice.
Certain limitations of the study need to be taken into account. The online format of the panel of judges precluded interaction among them or any shared construction in assessing the instrument and presenting their suggestions. Nonetheless, participation by judges from different fields broadened the scope of contributions, which enhanced the process of adapting the instrument for use in the educational environment. Importantly, content validation assessed only the level of agreement among the judges, although the space left for them to write suggestions did elicit valuable contributions as to the content of the statements, permitting adaptations to the instrument to suit it better to the school context.
On the other hand, the potential of the findings is to be emphasised. The use of item response theory with estimation of a ci parameter helped refine the instrument by broadening the assessment of each statement through modelling the probability of a right answer achieved by guessing rather than knowing. The GAB1-teachers scale makes it possible to ascertain and characterise different levels of knowledge, which helped to describe variations in teachers’ knowledge with their socioeconomic and professional characteristics. Assessment using the GAB1-professores scale can help to identify points to be worked on in training, so as to direct efforts towards improving competences that are still deficient, as well as identifying whether, after participating in such training, new skills have been acquired. Also, an instrument aligned with what is already being used in primary health care can leverage collaboration among professions and promote a more collaborative and integrated approach, as foreseen by the PSE. Note that the instrument is short and uses clear language, facilitating comprehension by all potential respondent professionals.

CONCLUSION
The adapted instrument is valid for use in the school environment, including the key statements to assess primary and secondary school teachers’ knowledge of healthy eating. The description of levels in the GAB1-professores suggests its potential for recognising points that should be worked on with teachers, so as to build and change their pupils’ eating habits in line with the Dietary Guidelines and foster a critical, thoughtful attitude with greater autonomy in food choices. Accordingly, use of the scale can contribute to guiding teacher training processes away from the locus of nutritionism and individual accountability towards an approach attentive to the complexity of the food system and the human right to suitable healthy food.

FUNDING
Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Universidade Estadual de Montes Claros (UNIMONTES).

REFERENCES
1. Elizabeth L, Machado P, Zinöcker M, Baker P, Lawrence M. Ultra-Processed Foods and Health Outcomes: A Narrative Review. Nutrients 2020;12(7):1955.
2. Monzani A, Ricotti R, Caputo M, et al. A Systematic Review of the Association of Skipping Breakfast with Weight and Cardiometabolic Risk Factors in Children and Adolescents. What Should We Better Investigate in the Future? Nutrients 2019;11(2):387.
3. Chaudhary A, Sudzina F, Mikkelsen BE. Promoting Healthy Eating among Young People—A Review of the Evidence of the Impact of School-Based Interventions. Nutrients 2020;12(9):2894.
4. Fundo das Nações Unidas para a Infância (UNICEF). Para cada criança, nutrição - Estratégia de Nutrição 2020–2030 do UNICEF: Resumo. Published online 2020. https://www.unicef.org/media/109816/file/Nutrition%20Strategy%202020-2030%20(Brief)%20-%20Portuguese.pdf
5. Brasil. Guia Alimentar Para a População Brasileira: Promovendo a Alimentação Saudável. Ministério da Saúde; 2006.
6. Brasil. Guia alimentar para população brasileira. Ministério da Saúde; 2014.
7. BRASIL. Resolução No 06, de 08 de Maio de 2020.; 2020.
8. Katsagoni CN, Apostolou A, Georgoulis M, et al. Schoolteachers’ Nutrition Knowledge, Beliefs, and Attitudes Before and After an E-Learning Program. J Nutr Educ Behav 2019;51(9):1088-1098.
9. Almeida RMMP, Pereira MM, Cunha AC, et al. Conhecimento e percepção de autoeficácia e eficácia coletiva de professores sobre o Guia Alimentar. DEMETRA Aliment Nutr Saúde 2023;18:e71030.
10. Reis LCD, Jaime PC. Pool of items to measure Primary Health Care workers’ knowledge on healthy eating. Rev Saúde Pública 2021;55:55.
11. Rubio DM, Berg-Weger M, Tebb SS, Lee ES, Rauch S. Objectifying content validity: Conducting a content validity study in social work research. Soc Work Res 2003;27(2):94-104.
12. Nunes CHS da S, Primi R. Impacto do tamanho da amostra na calibração de itens e estimativa de escores por teoria de resposta ao item. Aval psicol 2005;4(2).
13. Embretson SE, Reise SP. Item Response Theory for Psychologists. Lawrence Erlbaum Associates Publishers; 2000.
14. Muñiz J. Teoría de Respuesta a Los Ítems: Un Nuevo Enfoque En La Evolución Psicológica y Educativa. Ediciones Pirámide; 1990.
15. Hair JF, Black WC, Babin BJ, Anderson RE, Tatham RL. Análise fatorial. In: Análise Multivariada Dos Dados. Bookman; 2009.
16. Andrade DF, Tavares HR. Teoria da Resposta ao Item: Conceitos e Aplicações. Associação Brasileira de Estatística; 2000.
17. Bortolotti SLV, Tezza R, De Andrade DF, Bornia AC, De Sousa Júnior AF. Relevance and advantages of using the item response theory. Qual Quant 2013;47(4):2341-2360.
18. Chernyshenko OS, Stark S, Chan KY, Drasgow F, Williams B. Fitting Item Response Theory Models to Two Personality Inventories: Issues and Insights. Multivar Behav Res 2001;36(4):523-562.
19. Brown T. Confirmatory Factor Analysis for Applied Research. The Guilford Press; 2015.
20. Levy RB, Andrade GC, Cruz GLD, et al. Três décadas da disponibilidade domiciliar de alimentos segundo a NOVA – Brasil, 1987–2018. Rev Saúde Pública 2022;56:75.
21. Mescoloto SB, Pongiluppi G, Domene SMÁ. Ultra-processed food consumption and children and adolescents’ health. J Pediatr (Rio J) 2023:S0021755723001213.
22. Louzada MLDC, Costa CDS, Souza TN, Cruz GLD, Levy RB, Monteiro CA. Impacto do consumo de alimentos ultraprocessados na saúde de crianças, adolescentes e adultos: revisão de escopo. Cad Saúde Pública 2021;37(suppl 1):e00323020.
23. Delpino FM, Figueiredo LM, Flores TR, et al. Intake of ultra-processed foods and sleep-related outcomes: A systematic review and meta-analysis. Nutrition 2023;106:111908.
24. Louzada MLDC, Cruz GLD, Silva KAAN, et al. Consumo de alimentos ultraprocessados no Brasil: distribuição e evolução temporal 2008–2018. Rev Saúde Pública 2023;57(1):12.
25. Rodrigues RM, Souza ADM, Bezerra IN, Pereira RA, Yokoo EM, Sichieri R. Evolução dos alimentos mais consumidos no Brasil entre 2008-2009 e 2017-2018. Rev Saúde Pública 2021;55(Supl.1):1-10.
26. Mialon M, Cediel G, Jaime PC, Scagliusi FB. “A consistent stakeholder management process can guarantee the ‘social license to operate’”: mapping the political strategies of the food industry in Brazil. Cad Saúde Pública 2021;37(suppl 1):e00085220.
27. Barlow P, Thow AM. Neoliberal discourse, actor power, and the politics of nutrition policy: A qualitative analysis of informal challenges to nutrition labelling regulations at the World Trade Organization, 2007–2019. Soc Sci Med 2021;273:113761.
28. Silva AG da, Teixeira RA, Prates EJS, Malta DC. Monitoramento e projeções das metas de fatores de risco e proteção para o enfrentamento das doenças crônicas não transmissíveis nas capitais brasileiras. Ciênc Saúde Coletiva 2021;26(4):1193-1206.
29. Santos TSS, Julián C, Vincenzi SL, et al. A new measure of health motivation influencing food choices and its association with food intakes and nutritional biomarkers in European adolescents. Public Health Nutr 2021;24(4):685-695.
30. Moraes JMM, Alvarenga MDS. Adaptação transcultural e validade aparente e de conteúdo da versão reduzida da The Eating Motivation Survey (TEMS) para o Português do Brasil. Cad Saúde Pública 2017;33(10).
31. Burlandy L, Castro IRRD, Recine E, Carvalho CMPD, Peres J. Reflexões sobre ideias e disputas no contexto da promoção da alimentação saudável. Cad Saúde Pública 2021;37(suppl 1):e00195520.
32. Coelho DEP, Bógus CM. Vivências em hortas escolares: a construção de uma estratégia pedagógica para alimentação adequada e saudável. Campo Abierto 2018;37(1):19-32.
33. Bitelli FM, Jurema MLM. FEIJOADA: origem e considerações acerca de um patrimônio cultural imaterial. Contextos da Alimentação – Revista de Comportamento, Cultura e Sociedade 2019;7(1).
34. Brasil. Marco de Referência de educação alimentar e nutricional para as políticas públicas. Published online 2012.
35. DallaCosta M, Rodrigues RM, Schütz G, Conterno S. Programa Saúde na Escola: desafios e possibilidades para promover saúde na perspectiva da alimentação saudável. Saúde Em Debate 2022;46(spe3):244-260.
36. Maldonado L, Farias SC, Damião JDJ, Castro LMC, Silva ACFD, Castro IRRD. Proposta de educação alimentar e nutricional integrada ao currículo de Educação Infantil e Ensino Fundamental. Cad Saúde Pública 2021;37(suppl 1):e00152320.
37. Castro SFDF, Melgaço MB, Souza VMGD, Santos KSD. Convergência de políticas públicas educacionais na promoção da alimentação adequada e saudável. Rev Panam Salud Pública 2017:1-9.