0243/2024 - Padrões alimentares derivados de três índices de qualidade da dieta e fatores associados: resultados da linha de base do ELSA-Brasil
Dietary patterns derivedthree diet quality indices and associated factors: findingsthe baseline of ELSA-Brasil
Autor:
• Leandro Teixeira Cacau - Cacau, L. T. - <lcacau@usp.br>ORCID: https://orcid.org/0000-0003-1681-5960
Coautor(es):
• Paulo Andrade Lotufo - Lotufo, P. A. - <palotufo@usp.br>ORCID: https://orcid.org/0000-0002-4856-8450
• Isabela Martins Benseñor - Benseñor, I. M. - <isabensenor@gmail.com>
ORCID: https://orcid.org/0000-0001-6889-7334
• Dirce Maria Marchioni - Marchioni, D. M. - <marchioni@usp.br>
ORCID: https://orcid.org/0000-0002-6810-5779
Resumo:
O objetivo deste estudo foi descrever os padrões alimentares através de três índices de qualidade da dieta e os fatores associados entre 15.081 participantes da linha de base do Estudo Longitudinal de Saúde do Adulto (ELSA-Brasil). O consumo alimentar foi avaliado por meio de questionário de frequência alimentar. A partir disto, foram aplicados três índices de qualidade da dieta: o Índice de Dieta da Saúde Planetária (PHDI), Índice de Qualidade da Dieta para Saúde Cardiovascular (CHDI) e o Índice de Qualidade da Alimentação-2015 (HEI-2015). Modelos de regressão linear foram construídos para avaliar os fatores associados. A população apresentou pontuações que variaram de baixas a moderadas nos três índices avaliados. Observou-se que as mulheres, os idosos, pessoas com maior renda per capita, prática de atividade física moderada e vigorosa apresentaram, em média, maiores pontuações nos três índices de qualidade da dieta avaliados. Ao passo que, os fumantes e as pessoas com sobrepeso e obesidade apresentaram, em média, menores pontuações nos três índices de qualidade da dieta avaliados. O presente estudo observou que condições sociodemográficas e de estilo de vida estão associadas com a adesão a recomendações dietéticas saudáveis e sustentáveis.Palavras-chave:
Padrões alimentares; Qualidade da dieta; Nível socioeconômico; Estilo de vidaAbstract:
The aim of this study was to describe dietary patterns using three diet quality indices and the associated factors among 15,081 participantsthe baseline of the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil). Dietary intake was assessed through a food frequency questionnaire. Three diet quality indices were applied: the Planetary Health Diet Index (PHDI), the Cardiovascular Health Diet Index (CHDI), and the Healthy Eating Index-2015 (HEI-2015). Linear regression models were constructed to assess the associated factors. The population exhibited low to moderate scores on all three evaluated indices. It was observed that women, the elderly, individuals with higher per capita income, and those engaged in moderate and vigorous physical activity had, on average, higher scores on all three diet quality indices assessed. Conversely, smokers and individuals with overweight and obesity had, on average, lower scores on all three diet quality indices assessed. This study found that sociodemographic and lifestyle conditions are associated with adherence to healthy and sustainable dietary recommendations.Keywords:
Dietary patterns; diet quality; socioeconomic status; lifestyleConteúdo:
Acessar Revista no ScieloOutros idiomas:
Dietary patterns derivedthree diet quality indices and associated factors: findingsthe baseline of ELSA-Brasil
Resumo (abstract):
The aim of this study was to describe dietary patterns using three diet quality indices and the associated factors among 15,081 participantsthe baseline of the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil). Dietary intake was assessed through a food frequency questionnaire. Three diet quality indices were applied: the Planetary Health Diet Index (PHDI), the Cardiovascular Health Diet Index (CHDI), and the Healthy Eating Index-2015 (HEI-2015). Linear regression models were constructed to assess the associated factors. The population exhibited low to moderate scores on all three evaluated indices. It was observed that women, the elderly, individuals with higher per capita income, and those engaged in moderate and vigorous physical activity had, on average, higher scores on all three diet quality indices assessed. Conversely, smokers and individuals with overweight and obesity had, on average, lower scores on all three diet quality indices assessed. This study found that sociodemographic and lifestyle conditions are associated with adherence to healthy and sustainable dietary recommendations.Palavras-chave (keywords):
Dietary patterns; diet quality; socioeconomic status; lifestyleLer versão inglês (english version)
Conteúdo (article):
Dietary patterns derived from three diet quality indices and associated factors: ELSA-Brasil baseline resultsLeandro Teixeira Cacau1 – lcacau@usp.br – ORCID https://orcid.org/0000-0003-1681-5960
Paulo Andrade Lotufo2 – palotufo@usp.br – ORCID https://orcid.org/0000-0002-4856-8450
Isabela Martins Benseñor2 - isabensenor@gmail.com – ORCID https://orcid.org/0000-0001-6889-7334
Dirce Maria Marchioni1 – marchioni@usp.br – ORCID https://orcid.org/0000-0002-6810-5779
1Department of Nutrition, College of Public Health, Universidade de São Paulo
2Clinical and Epidemiological Research Center, University Hospital, Universidade de São Paulo
Corresponding author: Leandro Teixeira Cacau, lcacau@usp.br.
1Department of Nutrition, College of Public Health, Universidade de São Paulo
Av. Dr. Arnaldo Antunes 715 – Cerqueira Cesar, São Paulo, SP.
Abstract
The aim of this study was to describe dietary patterns using three dietary quality indices and the associated factors among 15,081 participants from the baseline of the Brazilian Longitudinal Study of Adult Health (ELSA-Brazil). Dietary intake was assessed through a food frequency questionnaire. Three diet quality indices were applied: the Planetary Health Diet Index (PHDI), the Cardiovascular Health Diet Index (CHDI), and the Healthy Eating Index-2015 (HEI-2015). Linear regression models were constructed to assess the associated factors. The population exhibited low to moderate scores on all three evaluated indexes. It was observed that women, the elderly, individuals with higher per capita income, and those engaged in moderate and vigorous physical activity had, on average, higher scores on all three dietary quality indexes assessed. Conversely, smokers and individuals with overweight and obesity had, on average, lower scores on all three dietary quality indices assessed. This study found that sociodemographic and lifestyle conditions are associated with adherence to healthy and sustainable dietary recommendations.
Keywords: Dietary patterns; diet quality; socioeconomic status; lifestyle
Introduction
The dietary pattern analysis has been used in the field of Nutritional Epidemiology to investigate consumption patterns, considering that we consume food in complex meals, and not just isolated nutrients 1,2. Within the analysis of dietary patterns, we have a i) a priori or hypothesis-oriented dietary patterns, which comprise diet quality indices, which are based on scientific evidence; ii) a posteriori or exploratory dietary patterns, which use statistical data reduction methods, such as factor analysis and the principal component analysis. In this method, data is reduced regardless of the outcome to be analyzed; and iii) empirical methods or outcome-dependent data, where concepts of a priori and a posteriori methods will be used, combining scientific evidence and the reduction of dietary data 3,4.
Diet quality indexes – also known as a priori dietary patterns – can be used for comparisons within and between populations, enabling, this way, the knowledge for specific dietary intervention needs. Moreover, the diet quality indexes are useful to look into if the dietary recommendations have beneficial effect in relation to health outcomes, as for example cardiovascular diseases. The advantages of using diet quality indexes are that they are easy to calculate and are easily reproducible and comparable 4–7.
Several diet quality indexes have been proposed, with emphasis on the Healthy Eating Index (HEI), an index based on dietary recommendations from the United States 8 and widely used in epidemiological studies. The HEI is updated every five years, and the HEI-2015 has 13 food components divided into adequacy components and moderation components 9.
Recently, the Cardiovascular Health Diet Index (CHDI) 10 has been proposed. The CHDI can be defined as a diet quality index that is based on the dietary recommendations for preventing cardiovascular diseases, from the American Heart Association (AHA) 11, but adapted to the Brazilian dietary culture. This index incorporates specific elements, such as the inclusion of red meat and a component dedicated to the group of beans, foods characteristic of the Brazilian dietary pattern. Besides, the CHDI stands out for being the first diet quality index to include a metric for the consumption for ultra-processed foods, recognizing their significant impact on human health 12, becoming aligned with the latest Dietary Guidelines for the Brazilian Population, launched in 2014. CHDI has 11 components and a total score that can vary from 0 to 110 points.
In addition to this, recently, the Planetary Health Diet Index (PHDI) 13 was proposed to evaluate adherence to the reference diet proposed by the EAT-Lancet Commission for healthy and sustainable diets 14 and uses targets dietary allowances at their cutoff points to ensure dietary quality and environmental sustainability. The PHDI has 16 components and a score that can vary from 0 to 150 points 13,15.
Considering that the CHDI and PHDI are recent indexes, for which there are still no studies on the associated factors, while the HEI-2015 is an index widely recognized in the literature as an index of diet quality 16, this study aims to evaluate the factors that influence adherence to these three indexes in the population participating in the Longitudinal Study of Adult Health, ELSA-Brasil.
Methods
Study population
This study is a cross-sectional analysis of baseline data from the ELSA-Brasil study, an ongoing multicenter cohort with 15,105 public servants, of both genders, active and retired, from six institutions (five public universities and one public research institute) located in six different Brazilian cities (São Paulo, Rio de Janeiro, Belo Horizonte, Vitória, Porto Alegre and Salvador) in three Brazilian regions (Northeast, Southeast and South). All active or retired employees of the aforementioned institutions, aged between 35 and 74 years, were eligible for the study. The ELSA-Brasil baseline data was collected by trained and certified personnel under strict quality protocol between August 2008 and December 2010 17–19.
Assessment of food consumption
Food consumption was assessed using a semi-quantitative food frequency questionnaire (FFQ) with 114 items, referring to the last 12 months, developed and validated for use in ELSA-Brasil 20,21. The FFQ estimated usual food consumption with questions structured into three sections: (1) food products/food preparations; (2) measures of products consumed; and (3) frequencies of consumption with eight response options (more than 3 times/day, 2 to 3 times/day, once a day, 5 to 6 times a week, 2 to 4 times a week, once a week, 1-3 times a month, and never/almost never). The daily consumption of each FFQ item (in g per day) was obtained by multiplying the portion size by the corresponding frequency. Food measurements were then converted to nutrient intakes using the Nutrition Data System for Research (NDSR) and Table of Food Composition (TACO). More information about the FFQ was previously described 20,21.
Planetary Health Diet Index
The Planetary Health Diet Index (PHDI) 13 is a 16-component dietary index that considers all food groups proposed in the sustainable diet proposed by the EAT-Lancet Commission and has a gradual scoring system, i.e., the components are scored according to the amount of consumption 15. The 16 components are divided into 4 categories: 1) adequacy components (nuts and peanuts, fruits, vegetables, legumes and whole grains); 2) optimal components (eggs, dairy products, fish and seafood, tubers and potatoes and vegetable oils); 3) proportion components (dark green vegetables / total vegetables and orange-red vegetables / total vegetables); and 4) moderation components (red meat, poultry and substitutes, animal fats and added sugars). The components and the total score have a continuous score. The adequacy, optimal and moderation components can score from 0 to 10 points, while the proportion components score from 0 to 5 points. The total PHDI score can range from 0 to 150 points, where a higher score means greater adherence to the sustainable diet. Table 1 presents the components, scoring criteria and cutoff points.
For the adequacy components, the maximum score (10 points) was assigned if consumption met or exceeded the recommendation. Otherwise, the score was proportional, being calculated as follows: (current intake ÷ recommended value * maximum score). For example, individuals who consumed 2% fruit originated energy of their total daily energy intake received 4 points = [(2 ÷ 5) * 10 = 4]. For the optimal components, the maximum score (10 points) has been assigned by reaching the superior limit of consumption. As consumption exceeds the upper limit, the score becomes inversely proportional to consumption. Finally, the moderation components received a minimum score (0 points) when consumption reached or exceeded the recommended value. When consumption was lower than the consumption limit, the calculation was as follows: [(1 – current intake ÷ recommended value) * maximum score]. For example, individuals who consumed 1.5% fruit originated energy of their total daily energy intake received 3.75 points [(1 – 1,5 ÷ 2,4) * 10 = 3,75]. Further details of the development process, component information, cutoff points, scoring criteria and validity results have been previously described 13.
Cardiovascular Health Diet Index
The Cardiovascular Health Diet Index (CHDI) is a dietary index that considers recommendations for a healthy diet proposed by the American Heart Association 11 with some adaptations to Brazilian food culture, such as the inclusion of beans and red meat, and also the inclusion of a metric for the consumption of ultra-processed foods. The CHDI has 11 components: fruits, vegetables, fish and seafood, sugar-sweetened beverages (SSBs), whole grains, nuts, legumes, processed meats, red meat, dairy products and ultra-processed foods. All components receive a continuous score from 0 to 10 points, totaling a final score that can vary from 0 to 110 points, where a higher score indicates higher dietary quality for cardiovascular health 10.
Table 2 describes the components, scoring criteria and cutoff points of the index. Scores are assigned according to the consumption in grams per day of the foods that make up the index. The components fruits, vegetables, fish and seafood, whole grains, nuts, beans and dairy products received the maximum score (10 points) if consumption met or exceeded the recommendation and received the minimum score (0 points) if there was no consumption. Otherwise, the score was proportional. For the components SSBs, red meat, processed meat and the ultra-processed food metric, the process was the opposite to the previous ones, that is, the minimum score (0 points) was assigned when consumption reached or exceeded the recommended value, while the score maximum (10 points) was awarded when there was no consumption. Otherwise, the score was proportional. More information about the development criteria, components, cutoff points, scoring system and validity are described here 10.
Healthy Eating Index – 2015
The HEI-2015 is an index based on US dietary recommendations, but widely used as an indicator of diet quality in several studies outside the US. The HEI-2015 has 13 components, divided into adequacy components (total fruits, whole fruits, total vegetables, vegetables and legumes, whole cereals, dairy products, total protein foods, fish and seafood, vegetable proteins and fatty acids) and in moderation components (refined cereals, sodium, added sugars and saturated fats). The HEI-2015 components are scored from 0 to 10 or from 0 to 5 points. The maximum total score is 100 and higher scores characterize high-quality diets. All components are based on energy density 9,16 (Table 3).
Social and lifestyle factors
The sociodemographic characteristics of sex, age, self-declared race/skin color, education, per capita income and marital status have been used. Smoking, alcohol consumption and physical activity were used as lifestyle characteristics. All information included in this analysis was self-reported by means of standardized questionnaires 18,19,22.
Sociodemographic characteristics have been used: sex (men or women), age (adults aged 34 to 59 years or elderly people ≥60 years), self-declared race/skin color (white, brown, black, and yellow/indigenous), marital status (living with or without a partner), per capita family income (based on self-report, it was calculated as the total monthly family income divided by the number of family members and then divided into tertiles: low, medium and high) and the level of education (<8 years, from 9 to 11 years and >12 years of study).
Lifestyle characteristics were smoking habit (never smoked and ex-smoker or current smoker), excessive alcohol consumption (amount ingested per week [≥ 210 g for men and ≥ 140 g for women]) and then dichotomized into yes or not, and leisure-time physical activity classified as low, moderate or vigorous according to the International Physical Activity Questionnaire – IPAQ long version – (≥150 min/week of moderate activity or ≥75 min/week of vigorous activity) 23.
Nutritional status was assessed using anthropometric measurements of weight and height, which were obtained using international techniques and criteria 18. Body weight was measured with the subject barefoot, fasting, and wearing a standard uniform over underwear. An electronic scale (Toledo®, model 2096PP) was used, with a capacity of 200 kg and an accuracy of 50 g. Height was measured with a wall stadiometer (Seca®, Hamburg, BRD) with a precision of 1 mm, fixed to the wall, with the individual barefoot, resting the head, buttocks and heels against the wall and looking at the horizontal plane. Body Mass Index (BMI) was calculated as weight (kg) divided by square height (m2). BMI values <25 kg/m2 were considered adequate, >25 km/m2 as overweight 24.
Statistical analysis
The normality of the variables was tested through descriptive analysis and visually using histograms. Descriptive analyses for each of the three indexes (PHDI, CHDI and HEI-2015) were carried out using means with their respective 95% confidence intervals (95% CI) for continuous variables and frequencies (n) with percentages for the categorical variables.
The associations among diet quality indices (HEI-2015, PHDI and CHDI – dependent variable) and sociodemographic, lifestyle and health status (predictors) were tested using multiple linear regression models. The sociodemographic, lifestyle and health status factors included in the model were sex (reference: men), age (reference: adults), education (reference: lower), per capita income (reference: low), race/skin color self-declared (reference: white), marital status (reference: lives with a partner), smoking habit (reference: non-smoker), alcohol consumption (reference: no), level of physical activity (reference: low), and nutritional status (reference: ideal weight).
All statistical analyzes were performed using the STATA® program (Statistical Software for Professionals, College Station, Texas, USA), version 14.2, and the p value < 0.05 was considered statistically significant.
Results
The descriptive results and averages of the PHDI, CHDI and HEI-2015 total scores according to the sociodemographic, health and lifestyle variables are presented in Table 4. In general, those with the highest total average scores in the three diet quality indexes are women, elderly people, individuals with higher education or postgraduate studies, with greater per capita family income, living without a partner, self-declared white and indigenous/Asian people, non-smokers, without alcohol consumption, with a moderate to vigorous level of physical activity, and ideal weight.
The results of multiple linear regression analyses between diet quality indices and associated factors are described in Table 5. The results indicated positive associations between the total PHDI score and the following characteristics: women, elderly people, individuals with medium and high per capita income, moderate and vigorous physical activity level. On the other hand, negative associations were observed for people who self-declared as brown and black, current smokers and overweight or obese.
In the analyses, with the total CHDI score, positive associations were found with women, elderly people, individuals with medium and high per capita income, self-declared black people, and those with moderate and vigorous levels of physical activity. Negative associations were found with current smokers, alcohol consumption, and being overweight or obese.
Finally, the results of the multiple linear regression indicated positive associations between the total HEI-2015 score and the following characteristics: women, elderly people, individuals with medium and high per capita income, living without a partner, people who self-identify as black, those with moderate and vigorous levels of physical activity. Negative associations were found between the HEI-2015 total score and higher or postgraduate education, current smoker and overweight or obesity.
Discussion
In this present study, the a priori dietary pattern of ELSA-Brasil participants was described using three different dietary evaluation indexes. It was observed that participants have low to moderate adherence to healthy and sustainable dietary recommendations. As associated factors, it was observed that women, the elderly, people with greater per capita income, practicing moderate and vigorous physical activity presented, on average, higher scores in the three diet quality indexes evaluated. Meanwhile, smokers and overweight and obese people had, on average, lower scores in the three diet quality indices assessed.
Despite having different components, scoring criteria and weights, the average of the three diet quality indices assessed was similar, which suggests that the quality of the population\'s diet is low to moderate, even when assessed by different indexes. This result is corroborated by other studies along with the ELSA-Brasil population, which, through other techniques for evaluating dietary patterns (i.e., factor analysis, cluster analysis, reduced rank regression and treelet transform), demonstrate that, in general, the population has a dietary pattern commonly called Western, which is characterized by greater consumption of red meat, eggs, and ultra-processed foods 25,26.
The ELSA-Brasil population reached around 40% of the total PHDI score, a result similar to that of the Brazilian population, evaluated by Marchioni and collaborators 27, who used data from the National Food Survey, a component module of the Family Budget Survey (INA/ POF) from 2017-2018 carried out by the Brazilian Institute of Geography and Statistics (IBGE). The authors observed an average of 45.9 points on the PHDI, representing around 30% of the total score, indicating low adherence to the sustainable dietary recommendations proposed by the EAT-Lancet Commission by the Brazilian population. To date, the CHDI has not yet been applied in other researches, but the results over the ELSA-Brasil population demonstrate adherence of around 52% of the total score.
Accordingly, the average HEI-2015 in the ELSA-Brasil population was 60.2 points, representing moderate diet quality, corroborating other studies accomplished in Brazil. Through the analysis of INA/POF data from 2008-2009, Souza and collaborators 28 found an average of 45.7 points in the HEI-2015. Pires and collaborators 29, when using the Brazilian Diet Quality Index-Revised (IQD-R), in ELSA-Brasil participants, observed an average of 72.6 points. However, after adaptations proposed by the authors in the IQD-R components, the average changed to 64.3 points. In contrast, population-based studies in the municipalities of Campinas 30 and São Paulo 31 which used the IQD-R observed averages of 52.7 points and 58.4 points, respectively.
Women and the elderly had higher averages in the three diet quality indexes evaluated, when compared to men and adults, results that corroborate data from national surveys, such as the INA/POF 32 and the National Health Survey (PNS) 33 and the Surveillance System for Risk and Protective Factors for Chronic Diseases by Telephone Survey (VIGITEL) 34, in which women have a greater diet quality than men, while elderly people have a higher diet quality than adults.
The hypothesis of a greater diet quality in the elderly is centered on the nutritional transition, since the formation of the elderly\'s eating habits was carried out at an earlier stage of the transition, where the consumption of in natura and minimally processed foods, with culinary preparations, it was the norm. In younger generations, more affected by the intense intake of ultra-processed foods and eating outside home, there is a worse diet quality 35. In results from the health survey in the cities of Campinas (ISA-Campinas) 30 and São Paulo (ISA-Capital) 31, the quality of the diet assessed by the IQD-R increased progressively as the age of the participants was increasing. Hiza and collaborators 36, using data from NHANES 2003-2004 (National Health and Nutrition Examination Survey), observed higher HEI-2005 averages in elderly people aged between 64 and 74 years compared to adults, and, in addition, elderly people aged equal to or above 75 years old had the highest averages in relation to all age groups.
In relation to gender, the data also corroborate findings from national surveys. Women, in general, have greater concerns about their health, attend to more health services and, consequently, receive more health information, including information on healthy eating, which can influence the higher quality of their diet compared to men 37. In addition, women are also more concerned about the quality of food and check nutritional information more frequently 38. According to data from VIGITEL, women have a higher consumption of foods such as fruits, vegetables and whole grains, and a lower consumption of ultra-processed foods, when compared to men 34. According to data from ISA-Capital 31 and ISA-Campinas (Institutes of Food Solidarity of the Capital and of Campinas City) 30, women have a higher quality diet than men.
Income is also a characteristic related to higher diet quality. Among ELSA-Brasil participants, those with higher per capita income also had higher scores on the diet quality indices assessed. Likewise, to the results obtained in this present study, Souza and collaborators 28, evaluating data from the INA/POF from 2008-2009, observed that individuals with higher monthly income had higher scores on the HEI-2015. Marchioni and collaborators 27 observed that individuals with higher per capita income have a higher average on the PHDI, and, when stratifying by gender and age, they observed that women had higher scores in all income categories compared to men. Yet, Mello and collaborators observed a higher average IQD-R in São Paulo residents with greater per capita income 31.
Observing the averages of the indexes in relation to the self-declared race/skin color categories, we can observe that participants who self-declare as white have higher scores on the PHDI in relation to brown and black participants. The opposite happens for the CHDI, where blacks have higher averages compared to whites and browns. There was no difference in the means for the HEI. The PHDI is based on the EAT-Lancet reference diet and consists mainly of a plant-based diet, with reduced or almost no consumption of foods of animal origin, with low recommendation values for red meat, for example 13. Conversely, the CHDI is based on the AHA recommendations, with adaptations, considering consumption for the red meat and beans groups, besides including ultra-processed foods 10.
Another of the main differences between the indexes is also their components. The CHDI considers SSBs, processed meats, and ultra-processed foods among their components. 10. The HEI-2015 also considers some components common to ultra-processed foods, such as sodium, saturated fat, added sugar and refined cereals 9. According to Claro and collaborators, on a study with the POF 2008-2009, at the same period of data collection of ELSA-Brasil baseline, the ultra-processed food cost was greater than of foods such as rice and beans, traditional food of the Brazilian diet 39. While fruits and vegetables are more expensive 39, which may affect access and choice of food, and, consequently, the score of the components of the evaluated indexes. Furthermore, the recommendation used for the cutting point of fruits and vegetables is greater in the CHDI than in the PHDI, for example. And, as observed in previous studies, the average fruit score in the PHDI is 9.7 points 13, while 7.7 in the CHDI 10, for the entire ELSA-Brasil population. The same occurs with the vegetables group.
These differences in consumption are determined by many factors, of which race/skin color is the most visible sign. For example: race/skin color is associated with social position, income, culture and all of these factors play a role in food choices. The indexes tackled have characteristics that may express in a higher or lower level, choices and access to foods. Previous studies with the Brazilian population evidence this. According to Verly Jr and collaborators, an increase in the consumption of fruits and vegetables would imply an increase in the cost of the diet 40.
Costa and collaborators observed that there is a difference in the food consumption of the Brazilian population according to race/skin color: the brown and black population has, on average, a higher consumption of rice and beans compared to white people, while white people have higher consumption of fruits, vegetables, and also ultra-processed foods in relation to brown and black people 41. These differences in consumption can impact the final average of the evaluated indexes, since the indexes are tools that evaluate the diet quality through the sum of the points of the components, which can oftentimes express different dietary patterns 5. Another relevant point is the cost of the diet: the PHDI is an index that assesses adherence to a sustainable diet and, according to Hirvonen and collaborators, the EAT-Lancet sustainable diet is not accessible in many countries 42 and, as assessed by Verly Jr, in Brazil, the cost of a healthy and sustainable diet is high 43.
Mello and collaborators 31 have observed that individuals included under the “non-white” category, which relates to black, brown, yellow and indigenous people, have small differences in the IQD-R score, while white individuals having, on average, higher score. Beydoun and Wang observed that white individuals had higher scores on the HEI-2005 and the adapted Mediterranean Diet Score (aMDS), compared to African Americans 44. In consonance with, Gao and collaborators observed that white people have had higher scores in the HEI-2005, comparing to African-Americans, Latinos and Chinese 45.
Individuals who are non-smokers, who do not drink alcohol frequently and who practice physical activity have higher averages in the diet quality indexes assessed, corroborating findings in the literature 31,46. In contrast, the participants with overweight and obesity have shown less diet quality, a result that also underpins with the literature 31. Regular physical activity, not drinking alcohol, not smoking and keeping an ideal weight, combined with a healthy eating pattern, are considered healthy practices and are associated with a lower risk of chronic diseases, especially cardiovascular diseases 11.
This study demonstrates several strengths. Firstly, it employs two dietary assessment indices developed and validated using data from ELSA-Brasil. One index is based on recommendations for a healthy diet for cardiovascular health, includes cultural adaptations to the Brazilian context, and is the first diet assessment index to incorporate a metric for ultra-processed foods, aligning with the Dietary Guidelines for the Brazilian Population 47. The other index is based on recommendations for a sustainable diet from the EAT-Lancet Commission, which has driven discussions on incorporating sustainability into healthy diets. Additionally, this study utilizes data from ELSA-Brasil, a multicenter cohort that tracks individuals from six cities across three regions of Brazil. For the first time, this study identifies factors associated with the two developed and validated indices.
Some limitations can also be highlighted. The FFQ has a limitation related to the participant\'s memory, which must be for a relatively long period, which can cause measurement errors, as participants may tend to report recent or current eating habits instead of the previous period that the FFQ should cover 48,49. Another limitation is related to the accuracy of the quantity consumed, since in the FFQ the participant must report intake based on a preestablished preparation unit 48,49. Despite the biases attributed to the FFQ, it continues to be one of the most used methods in epidemiological studies, including large cohort studies, such as the Nurses\' Health Study 50 and the Study of Health Professionals 51. And, as established, all food consumption collection instruments are subject to measurement errors, even those considered the gold standard, such as R24h 48,49.
In conclusion, the present study has identified that social and lifestyle characteristics are independently associated with diet quality. The results still help for the evidence that, some populational groups must receive greater attention to foster a healthy and sustainable diet, although the whole population needs improvement in their dietary quality.
Besides, the results add to the findings in the literature and provide evidence that it is important to support the formulation and consolidation of public policies that promote the availability and access to fresh, nutritious and healthy foods; promote food environments that are conducive to healthy and sustainable eating and protect consumers from aggressive marketing techniques, for example, linked to commercial determinants of health.
Future studies may use social and lifestyle characteristics associated with diet quality indexes assessed in health-disease relationship studies.
References
1. Hu FB. Dietary pattern analysis: a new direction in nutritional epidemiology. Curr Opin Lipidol. 2002;13(1):3-9.
2. Mozaffarian D, Rosenberg I, Uauy R. History of modern nutrition science—implications for current research, dietary guidelines, and food policy. BMJ. Published online June 13, 2018:k2392.
3. Krebs-Smith SM, Subar AF, Reedy J. Examining Dietary Patterns in Relation to Chronic Disease. Circulation. 2015;132(9):790-793.
4. Ocké MC. Evaluation of methodologies for assessing the overall diet: dietary quality scores and dietary pattern analysis. Proc Nutr Soc. 2013;72(2):191-199.
5. Fransen HP, Ocké MC. Indices of diet quality. Curr Opin Clin Nutr Metab Care. 2008;11(5):559-565.
6. Waijers PMCM, Feskens EJM, Ocké MC. A critical review of predefined diet quality scores. Br J Nutr. 2007;97(2):219-231.
7. Burggraf C, Teuber R, Brosig S, Meier T. Review of a priori dietary quality indices in relation to their construction criteria. Nutr Rev. 2018;76(10):747-764.
8. Kennedy ET, Ohls J, Carlson S, Fleming K. The Healthy Eating Index: design and applications. J Am Diet Assoc. 1995;95(10):1103-1108.
9. Krebs-Smith SM, Pannucci TE, Subar AF, et al. Update of the Healthy Eating Index: HEI-2015. J Acad Nutr Diet. 2018;118(9):1591-1602.
10. Cacau LT, Marcadenti A, Bersch-Ferreira AC, et al. The AHA Recommendations for a Healthy Diet and Ultra-Processed Foods: Building a New Diet Quality Index. Front Nutr. 2022;9:804121.
11. Benjamin EJ, Virani SS, Callaway CW, et al. Heart Disease and Stroke Statistics—2018 Update: A Report From the American Heart Association. Circulation. 2018;137(12).
12. Lichtenstein AH, Appel LJ, Vadiveloo M, et al. 2021 Dietary Guidance to Improve Cardiovascular Health: A Scientific Statement From the American Heart Association. Circulation. 2021;144(23).
13. Cacau LT, De Carli E, de Carvalho AM, et al. Development and validation of an index based on EAT-Lancet recommendations: the Planetary Health Diet Index. Nutrients. 2021;13(5):1698.
14. Willett W, Rockström J, Loken B, et al. Food in the Anthropocene: the EAT–Lancet Commission on healthy diets from sustainable food systems. Lancet. 2019;393(10170):447-492.
15. Cacau LT, Marchioni DM. The Planetary Health Diet Index Scores Proportionally and Considers the Intermediate Values of the EAT-Lancet Reference Diet. Am J Clin Nutr. 2022;115(4):1237.
16. Reedy J, Lerman JL, Krebs-Smith SM, et al. Evaluation of the Healthy Eating Index-2015. J Acad Nutr Diet. 2018;118(9):1622-1633.
17. Bensenor IM, Griep RH, Pinto KA, et al. Routines for organizing exams and interviews at the ELSA-Brasil research center. Rev. Saúde Pública (Public Health Journal). 2013;47(suppl 2):37-47.
18. Aquino EML, Barreto SM, Bensenor IM, et al. Brazilian Longitudinal Study of Adult Health (ELSA-Brasil): Objectives and Design. Am J Epidemiol. 2012;175(4):315-324.
19. Schmidt MI, Duncan BB, Mill JG, et al. Cohort Profile: Longitudinal Study of Adult Health (ELSA-Brasil). Int J Epidemiol. 2015;44(1):68-75.
20. Molina M del CB, Faria CP de, Cardoso L de O, et al. Diet assessment in the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil): Development of a food frequency questionnaire. Rev Nutr. 2013;26(2):167-176.
21. Molina M del CB, Benseñor IM, Cardoso L de O, et al. Reproducibility and relative validity of the Food Frequency Questionnaire used in the ELSA-Brasil. Cad Saúde Pública. (Public Health Register???) 2013;29(2):379-389.
22. Mill JG, Pinto K, Griep RH, et al. Measurements and clinical examinations carried out on ELSA-Brasil participants. Rev. Saúde Pública (Public Health Journal). 2013;47:54-62.
23. Craig CL, Marshall AL, Sjöström M, et al. International Physical Activity Questionnaire: 12-Country Reliability and Validity. Med Sci Sports Exerc. 2003;35(8):1381-1395.
24. World Health Organization. Physical Status: The Use and Interpretation of Anthropometry.; 1995. Accessed November 15, 2021. https://apps.who.int/iris/bitstream/handle/10665/37003/WHO_TRS_854.pdf?sequence=1&isAllowed=y
25. Alves M de A, Molina M del CB, Fonseca M de JM, Lotufo PA, Benseñor IM, Marchioni DML. Different statistical methods identify similar population-specific dietary patterns: an analysis of Longitudinal Study of Adult Health (ELSA-Brasil). Br J Nutr. 2022;128(11):2249-2257.
26. Silva VC, Gorgulho B, Marchioni DM, et al. Clustering analysis and machine learning algorithms in the prediction of dietary patterns: Cross‐sectional results of the Brazilian Longitudinal Study of Adult Health (ELSA‐Brasil). J Hum Nutr Diet. 2022;35(5):883-894.
27. Marchioni DM, Cacau LT, De Carli E, Carvalho AM de, Rulli MC. Low Adherence to the EAT-Lancet Sustainable Reference Diet in the Brazilian Population: Findings from the National Dietary Survey 2017–2018. Nutrients. 2022;14(6):1187.
28. de Paula Matos Souza J, Magela de Lima M, Martins Horta P. Diet Quality among the Brazilian Population and Associated Socioeconomic and Demographic Factors: Analysis from the National Dietary Survey 2008-2009. J Acad Nutr Diet. 2019;119(11):1866-1874.
29. Pires RK, Luft VC, Araújo MC, et al. Critical analysis of the revised diet quality index for the Brazilian population (DQI-R): its application in ELSA-Brasil. Cien Saude Colet. 2020;25(2):703-713.
30. Assumpção D de, Domene SMÁ, Fisberg RM, Barros MB de A. Social and demographic inequalities in dietary quality in a population-based study. Rev Nutr. 2016;29(2):151-162.
31. Mello AV de, Pereira JL, Leme ACB, Goldbaum M, Cesar CLG, Fisberg RM. Social determinants, lifestyle and dietary quality: a population-based study from the 2015 Health Survey of São Paulo, Brazil. Public Health Nutr. 2020;23(10):1766-1777.
32. Instituto Brasileiro de Geografia e Estatística (IBGE) (Brazilian Institute of Geography and Statistics). Pesquisa de Orçamentos Familiares 2017-2018 (Household Revenue Survey 2017-2018): Personal Food Consumption Analysis in Brazil: (IBGE, ed.).; 2020.
33. Santin F, Gabe KT, Levy RB, Jaime PC. Food consumption markers and associated factors in Brazil: distribution and evolution, Brazilian National Health Survey, 2013 and 2019. Cad Saúde Pública. (Public Health Register???) 2022;38(suppl 1).
34. Brazil. Ministry of Health. Health Surveillance Secretariat. Department of Health Analysis and Noncommunicable Disease Surveillance. Vigitel Brazil 2019: Surveillance of Risk and Protective Factors for Chronic Diseases by Telephone Survey: Estimates on the Frequency and Sociodemographic Distribution of Risk and Protective Factors for Chronic Diseases in the Capitals of the 26 Brazilian States and the Federal District in 2019. Ministry of Health; 2020.
35. Popkin BM. The Nutrition Transition: An Overview of World Patterns of Change. Nutr Rev. 2004;62:S140-S143.
36. Hiza HAB, Casavale KO, Guenther PM, Davis CA. Diet Quality of Americans Differs by Age, Sex, Race/Ethnicity, Income, and Education Level. J Acad Nutr Diet. 2013;113(2):297-306.
37. Assumpção D de, Domene SMÁ, Fisberg RM, Canesqui AM, Barros MB de A. Differences between men and women in diet quality: population-based study in Campinas, São Paulo. Ciênc Saúde Colet. 2017;22(2):347-358.
38. Stran KA, Knol LL. Determinants of Food Label Use Differ by Sex. J Acad Nutr Diet. 2013;113(5):673-679.
39. Claro RM, Maia EG, Costa BV de L, Diniz DP. Food prices in Brazil: prefer culinary preparations to ultra-processed foods. Cad Saúde Pública. (Public Health Register???) 2016;32(8).
40. Verly Jr E, Oliveira DCRS de, Sichieri R. Cost of healthy and culturally acceptable diets in Brazil in 2009 and 2018. Rev Saude Publica. (Public Health Review). 2021;55(Supl.1):1-11.
41. Costa JC, Jesus AC da S de, Jesus JGL de, Madruga MF, Souza TN, Louzada ML da C. Differences in food consumption of the Brazilian population by race/skin color in 2017–2018. Rev. Saúde Pública (Public Health Journal). 2023;57(1):4.
42. Hirvonen K, Bai Y, Headey D, Masters WA. Affordability of the EAT–Lancet reference diet: a global analysis. Lancet Glob Health. 2020;8(1):e59-e66.
43. Verly-Jr E, de Carvalho AM, Marchioni DML, Darmon N. The cost of eating more sustainable diets: A nutritional and environmental diet optimization study. Glob Public Health. Published online March 15, 2021:1-14.
44. Beydoun MA, Wang Y. How do socio-economic status, perceived economic barriers and nutritional benefits affect quality of dietary intake among US adults? Eur J Clin Nutr. 2008;62(3):303-313.
45. Gao SK, Beresford SA, Frank LL, Schreiner PJ, Burke GL, Fitzpatrick AL. Modifications to the Healthy Eating Index and its ability to predict obesity: the Multi-Ethnic Study of Atherosclerosis. Am J Clin Nutr. 2008;88(1):64-69.
46. Breslow RA, Guenther PM, Juan W, Graubard BI. Alcoholic Beverage Consumption, Nutrient Intakes, and Diet Quality in the US Adult Population, 1999-2006. J Am Diet Assoc. 2010;110(4):551-562.
47. Brazil. Ministry of Health. Department of Health Care. Department of Primary Care. Dietary Guidelines for the Brazilian Population. 2nd ed.; 2014.
48. Thompson FE, Subar AF. Dietary Assessment Methodology. In: Coulston A, Boushey C, eds. Nutrition in the Prevention and Treatment of Disease. 3rd ed. Elsevier; 2013:5-46.
49. Willett W. Nutritional Epidemiology. 3rd ed. Oxford University Press; 2012.
50. Hu FB, Satija A, Rimm EB, et al. Diet Assessment Methods in the Nurses\' Health Studies and Contribution to Evidence-Based Nutritional Policies and Guidelines. Am J Public Health. 2016;106(9):1567-1572.
51. Rimm EB, Giovannucci EL, Stampfer MJ, Colditz GA, Litin LB, Willett WC. Reproducibility and Validity of an Expanded Self-Administered Semiquantitative Food Frequency Questionnaire among Male Health Professionals. Am J Epidemiol. 1992;135(10):1114-1126.