Major depressive episodes and diet quality among Brazilian adults and older adults: results from the 2013 and 2019 National Health Survey

doi:10.21203/rs.3.rs-4979618/v1

Objective: To evaluate the association between major depressive episodes and diet quality among Brazilian adults and older adults.

Methods: We conducted a cross-sectional, population-based study using data from the National Health Survey (PNS) from 2013 and 2019. Major depressive episodes were assessed using the Patient Health Questionnaire-9 (PHQ-9) and classified in two ways: based on a cutoff point of ≥9 points on the scale and according to the diagnostic criteria of the Statistical Manual of Mental Disorders. Diet quality was assessed using a diet quality score that considered the week consumption of seven food items. We used both crude and adjusted linear regression models to explore the association between major depressive episodes and food consumption.

Results: In both 2013 and 2019, individuals with major depressive episodes had a diet quality score 0.20 standard deviations higher compared to their peers, indicating a worse diet quality. This association remained consistent regardless of the criteria used to define exposure and the potential confounders. Sensitivity analysis considering each food item separately revealed a lower prevalence of regular consumption of beans, vegetables, fruits, and milk, as well as a higher prevalence of regular consumption of sugar-sweetened beverages and sweets among individuals with major depressive episodes.

Conclusion: Major depressive episodes appear to be linked to lower diet quality in the Brazilian population. Policymakers and public health programs should consider approaches to improve both mental health and dietary habits.

Mental health

Dietary intake

Epidemiological surveys

The burden of non-communicable chronic diseases (NCDs) is increasing and is one of the leading causes of death worldwide ¹. In Brazil, NCDs are responsible for the majority of deaths and are also associated with other health-related outcomes, such as loss of quality of life and increased disability ^2,3. Unhealthy eating habits are related to the rise in NCDs and the risk of mortality from all causes, with dietary risk factors being one of the largest contributors to the global burden of disease ¹.

Evidence shows that ultra-processed foods (UPF) account for 50% and 30% of total energy intake in high- and middle-income countries, respectively ⁴. In Brazil, it is estimated that approximately 20% of the average daily calorie intake comes from UPF ⁵. Furthermore, there is an early introduction of UPF consumption in Brazil, reaching 40% of the total energy intake of school-age children ⁶. According to Elizabeth et al., high consumption of these foods is observed mainly in younger age groups, with a decrease as age increases ⁷.

If there is evidence of high consumption of UPF in the Brazilian population, the consumption of healthy food items, such as fruits and vegetables, is below the recommended levels. According to the 2014 Surveillance System for Risk and Protection Factors for Chronic Diseases by Telephone Survey (VIGITEL), only 24.1% of the Brazilian population reached the recommended consumption of fruits and vegetables. On the other hand, 20.8% of the population reported regular consumption (5 or more days per week) of soft drinks and 18.1% of sweets, with the prevalence of both being higher among individuals aged 18 to 34 years ⁸.

It is known that unhealthy dietary intake is associated with NCDs such as obesity, hypertension and type 2 diabetes. However, poor nutrition can also be linked to mental health problems, such as depression and anxiety ⁹. Mental health disorders are characterized by neurological dysfunctions that can lead to mood swings, discouragement, stress, appetite and sleep disorders, and low self-esteem, among other issues ⁹.

Although most research analyzes the association between food consumption and mental health by treating mental health as the primary outcome, evidence suggests that this relationship can be bidirectional. Specifically, the presence of depressive symptoms can lead to increased consumption of unhealthy foods and decreased consumption of healthy ones ^10,11,12. However, this relationship has not yet been fully elucidated, and further studies are needed to explore the role of mental health on individuals' food consumption, especially in middle-income countries like Brazil, where the prevalence of depressive symptoms has increased in recent years ^13,14. Therefore, our aim was to evaluate the association between major depressive episodes and diet quality among adults and older adults in Brazil in 2013 and 2019.

Design and sampling

This is a cross-sectional, population-based study conducted using data from the National Health Survey (PNS in the Portuguese acronym)^15,16, conducted by the Brazilian Institute of Geography and Statistics in partnership with the Ministry of Health. The PNS aims to produce data on the health situation and lifestyles of the Brazilian population. In this study, data from the first and second editions of the PNS, conducted in 2013 and 2019, respectively, were used.

The sampling process was similar in both PNS editions, using a three-stage cluster sampling. The first stage was characterized by the stratification of primary sampling units (PSUs). The PSUs were formed by the census sectors of each municipality selected to take part in the study. The second-stage units were the households drawn from each of the selected census sectors. Finally, the third stage of sampling was slightly different in each PNS edition: in the 2013 edition, the third stage of sampling consisted of residents aged 18 or over living in each selected household. In the 2019 edition, the third stage of sample selection consisted of residents aged 15 or over.

In both PNS editions, data collection was carried by the Brazilian Institute of Geography and Statistics, which used a structured questionnaire containing 20 blocks in 2013 and 26 in 2019. The interviews were conducted in the 26 Brazilian states plus the Federal District and each state had a specific number of interviewees. By the end of data collection, 62,986 interviews were conducted for the 2013 PNS. In the 2019 PNS, the total number of interviews conducted was 90,846, of which 88,531 interviewees were 18 years of age or older. More information about the sampling process and data collection for the 2013 and 2019 editions of PNS can be found on the website: <https://www.ibge.gov.br/estatisticas/sociais/saude/9160-pesquisa-nacional-de-saude.html?=&t=downloads>.

Definition of food consumption and diet quality

We used data about food consumption of adults and older adults who responded to the 2013 or 2019 PNS. Among the questions about food consumption, the weekly frequency of consumption of beans, greens, raw vegetables, cooked vegetables, red meat (beef, pork, goat), chicken, fish, fruit, natural fruit juice, soft drinks or artificial juice, milk, sweets (pieces of cake or pie, sweets, chocolates, candies, cookies or crackers), quick snacks (sandwiches, savory snacks or pizzas) and salt consumption (considering freshly prepared food and processed foods) was assessed. For each food item, the interviewees were asked “How many days a week do you usually eat/drink...?”

Diet quality, the main outcome of our study, was assessed using the food consumption score proposed by Francisco et al., 2019 ¹⁷. To construct the score, a point is assigned to eight food items, three of them indicating unhealthy food consumption (soft drinks or artificial juices, red meat and sweets) and five of them indicating healthy food consumption (beans, fruits, raw vegetables, cooked vegetables and milk). The points from each food item varies from zero to four points depending on the food and the frequency of consumption. For healthy food items, the higher the consumption, the lower the score (closer to zero). For unhealthy food items, the more frequent the consumption, the higher the score (closer to four). The total score on the original scale varies from zero (best) to 32 points (worst quality of diet).

However, we adapted the score in our study, since in the 2019 PNS edition, information on weekly consumption of raw and cooked vegetables was investigated in a single question (“On how many days of the week do you usually eat at least one type of vegetable (not counting potatoes, cassava, taro or yam) such as lettuce, tomatoes, cabbage, carrots, chayote, eggplant, zucchini?”). Thus, the food consumption score ranged from zero (best) to 28 (worst diet quality). Further details on the score used in our study can be seen in the supplementary material (supplementary table 1).

Screening for major depressive episodes

The main exposure of our study was major depressive episodes. In both PNS editions, the Patient Health Questionnaire (PHQ-9) was applied. PHQ-9 is a diagnostic tool to screen adults in a primary care setting for the presence and severity of depressive episodes. The PHQ-9 has been validated for the Brazilian population ¹⁸ and has been used in different epidemiological studies since them ^19,20,21.

The PHQ-9 assess the presence of depressive symptoms in the last two weeks based on nine questions that evaluate symptoms such as depressive behavior, inability to feel pleasure, trouble sleeping, fatigue, changes in appetite, feelings of guilt, agitation, problems concentrating, and suicidal behavior. The nine questions of the scale have scores ranging from zero to three, corresponding to the response options “Never”, “Less than once a week”, “Once a week or more” and “Almost every day”.

We used two different ways to define major depressive episodes: the sum of the scores from zero to three of the response options, with a result ≥ 9 being considered as the presence of major depressive episodes. This cutoff point, which was proposed by Santos et al.¹⁸ and has already been used in other epidemiological studies conducted in Brazil, has high sensitivity and specificity for detecting depressive symptoms in population-based studies.

However, major depressive episodes were also assessed using the diagnostic criteria proposed by the Statistical Manual of Mental Disorders (DSM-IV-TR). This manual classifies a test as positive screening for major depressive episodes when five or more depressive symptoms were present during the last two weeks, for at least seven days, one of which being depressive behavior or anhedonia. The aim of using two different criteria to classify major depressive episodes was to increase the robustness of the findings, considering the inherent limitations of the cutoff point of ≥ 9 proposed by Santos et al. ¹⁸, as well as the limitations of cross-sectional studies to analyze causal associations.

Covariables

Individuals’ socioeconomic and demographic characteristics were included as covariables of our study and as potential confounders of the association between major depressive episodes and diet quality. The following covariables were included: geographical region of residence (North, Northeast, Midwest, Southeast and South), household area (urban or rural), sex (male or female), age (collected in complete years and categorized into three groups: 20–39, 40–59 and 60 years or older), self reported skin color (white, brown or black), marital status (married, separated/divorced, widowed or single), number of residents in the household (1–2, 3–4 or 5 or more), per capita household income in quintiles, education (none, incomplete elementary school, complete elementary school, complete high school, complete higher education) and occupation/employment status (working or not working).

Statistical analyses

We presented the relative frequencies of all variables included in the study with their respective 95% confidence intervals (95%CI). The associations of the socioeconomic and demographic characteristics with major depressive episodes were assessed using Poisson regression, with p-value from the Wald test, with a significance level of 5% and a 95%CI. To assess the association between major depressive episodes and diet quality, crude and adjusted linear regression models were used. The adjusted models included geographical region of residence, household area, age, sex, skin color, marital status, number of residents in the household, per capita household income, education and occupation status. We adjusted the regression models for these factors to verify whether the associations analyzed were independent of the socioeconomic and demographic characteristics. Diet quality score was standardized to provide a more convenient scale, and results were presented as β with their respective 95% CI. All analyses were stratified by PNS edition, in order to allow the assessment of changes in the magnitude of the associations between 2013 and 2019.

We performed a sensitivity analysis evaluating the association between major depressive episodes and each food item that composed the diet quality score, to observe which specific food items major depressive episodes could be associated with (beans, fruits, vegetables, milk, red meat, soft drinks or artificial juices, and sweets). To do so, regular consumption of each food item was considered, assuming that regular consumption of each food item was at least five times a week ¹⁷.

Subsequently, crude and adjusted Poisson regression models were used to assess the association between major depressive episodes and each food item, adjusting for the same confounders used in the linear regression models. The results were presented as prevalence ratios (PR) with their respective 95% CI. All analyses were performed using Stata, version 16.1, taking into account the sampling weights and the complex design of the two editions of the PNS.

We included in the analyses 62,986 adults and older adults from the 2013 PNS and 88,531 adults and older adults from the 2019 PNS. In both PNS editions, almost half of the sample lived in the Southeast region and just over 85% lived in urban areas. The majority of sample was composed by women in both surveys. In addition, more than 50% of the sample self reported their skin color as black or brown, with the proportion of individuals who self-declared as being Afro-descendants increasing in 2019 compared to 2013 (Table 1).

Almost 20% of individuals were 60 years of age or older in 2013. In 2019, this percentage reached 22.5%, indicating a slight population aging between the two surveys. In both 2013 and 2019, most respondents were married and lived in households with three or four individuals. Regarding household income, in both editions of the PNS, approximately 10% of respondents were classified in the lowest income quintile. Furthermore, the proportion of individuals with higher education degree (University degree) increased from 13.7% in 2013 to 16.4% in 2019. Finally, approximately 60% of the sample reported having some type of occupation in both editions of the survey (Table 1).

Regarding mental health, considering the DSM-IV-TR criterion, the prevalence of major depressive episodes was 4.2% (95%CI 3.9; 4.5) in 2013, increasing to 5.6% (95%CI 5.3; 5.9) in 2019. When the cut-off point of ≥ 9 was used to define major depressive episodes, we observed that the prevalence more than doubled, being 9.7% (95%CI 9.3; 10.2) in 2013, increasing to 13.0% (95%CI 12.5; 13.4) in 2019. Independent of the criterion used to define major depressive episodes, this condition was significantly lower in the North region and in rural households in both editions of the survey (Table 2).

Table 2 presents the prevalence of major depressive episodes according to the socioeconomic and demographic characteristics of the sample. Women, those who were, at least, 40 years of age and separated/divorced or widowed had a significantly higher prevalence of major depressive episodes when compared to their counterparts. On the other hand, individuals who lived in households with three or four residents, had a higher per capita household income, higher education level and who had some type of occupation presented a lower prevalence of major depressive episodes when compared to their peers.

The average diet quality score was − 0.01 standard deviations (SD) in 2013 and increased to 0.03 SD in 2019, indicating a slight worsening in the quality of diet between the surveys periods. We also observed that the diet quality score was higher in the North region in 2013 and in the Midwest in 2019, indicating worse diet quality in these locations. Furthermore, in both editions of the PNS, the score was higher in rural areas, among men, among those who aged between 20 and 39 years, with black or brown skin color, single and who lived in households with five or more residents, evidencing worse diet quality in these specific subgroups. Finally, individuals in the lowest per capita income quintile, with complete high school education and who had some type of occupation also presented a higher average diet quality score (Table 3).

Analysis of the association between major depressive episodes and diet quality showed that major depressive episodes were associated with higher diet quality score, regardless of the classification method used (either the DSM-IV-TR criteria or a cutoff point of ≥ 9). In both edition of the PNS, individuals with major depressive episodes had a diet quality score approximately 0.20 standard deviations higher compared to those without depressive episodes, regardless of potential confounders included in the analyses (Table 4).

Sensitivity analysis showed the association of major depressive episodes with each of the seven food items present in the diet quality score. In this analysis, we observed that, in 2013, individuals with major depressive episodes had a prevalence of regular consumption of beans and vegetables 10% lower when compared with individuals without this condition. In 2019, along with the consumption of beans and vegetables, the consumption of fruits and milk was also approximately 10% lower in those who reported major depressive episodes, independent of the criterion used to define the exposure (Table S2). For food items linked to unhealthy diet, we observed results in the opposite direction. In both 2013 and 2019 surveys, individuals with major depressive episodes presented higher prevalence of regular consumption of soft drinks or artificial juices as well as of sweets, no matter the way we used to define major depressive episodes (Table S2).

Our study, which aimed to evaluate the association between major depressive episodes and diet quality in adults and older adults from Brazil in 2013 and 2019, found that major depressive episodes were associated with poorer diet quality. Individuals who experienced major depressive episodes averaged higher diet quality scores, regardless of the criterion used to define the exposure and the potential confounders included in the analysis.

Studies focusing on the association between mental health and food consumption have mostly treated food consumption as the exposure and mental health as the outcome. However, some investigations have evaluated the association between mental health and food consumption in the same direction as the analysis carried out here ^12,22,23. For example, Quehl et al., analyzing a sample of female university students in Canada, observed that depressive symptoms were associated with a diet of low nutritional quality ¹². Similarly, a study conducted by Cebrino and Portero de la Cruz, involving adults in Spain across three phases (2006, 2011/2012 and 2017), found that depressive symptoms were a risk factor for poor or moderate diet quality ²². On the other hand, Elstgeest et al. suggested a possible reverse causality in the association between depression and diet. They noted that the presence and history of depressive symptoms are associated with a less healthy diet, especially in men. However, according to the authors, further studies are needed to understand the gender differences in this relationship ²³.

To better understand the relationship between mental health and diet quality, we analyzed the association of major depressive episodes with the consumption of specific food items used to create the diet quality score. This score is composed of food items linked to unhealthy diets (red meat, sugar-sweetened beverages and sweets) and healthy diets (beans, fruits, vegetables and milk). Regarding the food items linked to an unhealthy diet, we observed that individuals with major depressive episodes had a higher prevalence of regular consumption of sugar-sweetened beverages and sweets. These foods can be classified as comfort foods, which are high-calorie foods that evoke a sense of nostalgia. This nostalgic feeling can be tied to an individual’s personal memories (for example, condensed milk pudding that reminds them of a family member or an emotional relationship with another person) or a broader cultural connection (for example, a Brazilian living abroad who seeks Brazilian foods to emotionally connect with their homeland) ²⁴.

Emotions are directly related to eating behavior. According to Oliveira et al., “the emotional state is responsible for the process of choosing foods, as well as the frequency and quantity of food that is consumed”. Poor mental health influences food consumption since food can be seen as a form of reward or a source of affection. In the case of people with anxiety, for example, obsessively eating high-calorie foods can represent an escape from unpleasant situations and this practice can lead to the emergence of NCDs, such as obesity ²⁴.

Another explanation for the association between major depressive episodes and regular consumption of sweets and sugar-sweetened beverages observed in our study may be linked to the release of serotonin, commonly known as the “happiness hormone”. The consumption of high-calorie foods, particularly those rich in carbohydrates, sugar, salt, and fat, triggers the release of this hormone, creating an immediate sense of comfort and pleasure, and reducing stress. However, this effect is short-lived, which increases the likelihood of individuals repeatedly consuming high-calorie foods ²⁶. Additionally, our results showed a negative association between major depressive episodes and the consumption of food items linked to a healthy diet in both 2013 and 2019. According to Cebrino and Portero de la Cruz, the lack of energy or enthusiasm to prepare or enjoy food, as well as a loss of appetite, can negatively influence diet quality among individuals with depressive symptoms ²².

Our study showed other important results that go beyond the relationship between mental health and food consumption. One of them was the increase in the prevalence of major depressive episodes between 2013 and 2019, independent of the criteria used to define such condition, a fact that may indicate an increase in the presence of depressive symptoms in the general population. It is important to emphasize that the study analyzed a pre-pandemic context and future studies will be important to verify how much the scenario of the COVID-19 pandemic may have contributed to changing the magnitude of the relationship between major depressive episodes and diet quality.

The study has strengths, such as the use of data from the PNS, a population-based survey with national coverage, representative of the Brazilian population, whose data were collected by trained and standardized personnel. Furthermore, two different criteria were used to define major depressive episodes and, independent the way we used to define this condition, it remained associated with diet quality, indicating a certain robustness in the association analyzed here.

On the other hand, our study has limitations. One of them is the cross-sectional design, which does not allow establishing a cause-and-effect relationship. Another limitation is the presence of possible recall bias in the questionnaire responses, given that both exposure and outcome were self-reported. Finally, in the case of secondary data analyses, the association may be biased by residual confounding caused by some important variables possibly associated with major depressive episodes and diet quality, not available in the PNS database. One of these potential variables is food insecurity, associated with both depressive symptoms and diet quality, not available in both editions of the survey.

In conclusion, major depressive episodes was associated with worse diet quality, especially due to the increased consumption of unhealthy foods, such as sugar-sweetened beverages and sweets, and the decreased consumption of foods considered healthy, such as beans, vegetables, fruits and milk. Considering that poor diet quality is an important risk factor for NCDs and mortality, public health measures are needed to prevent and help reducing the prevalence of depressive symptoms and other mental health disorders in the Brazilian population. In addition, longitudinal studies are needed to better elucidate the association between mental health problems and food consumption.

Author Contribution

J.C.V.S.S. Schiatti contributed to the preparation and writing of the manuscript, statistical analysis and interpretation of results. L.C.Muniz participated in the preparation of the project and final review of the manuscript. L.P. Santos collaborated in the preparation of the database, statistical analysis, interpretation of results and preliminary and final review of the manuscript. All authors approve the final version of the manuscript to be published.

Acknowledgement

During the preparation of this work, the authors used ChatGPT (chat.openai.com) in order to review the English style of our manuscript. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

PAGLIAI, G. et al. Consumption of ultra-processed foods and health status: a systematic review and meta-analysis. The British Journal of Nutrition, v. 125, p. 308-318, 2021.
SOUZA, A. M. et al. Alimentos mais consumidos no Brasil: inquérito nacional de alimentação 2008-2009. Revista de Saúde Pública, v. 47, p. 190-99, 2013.
RAUBER, F. et al. Ultra-processed Food Consumption and Chronic Non-Communicable Diseases-Related Dietary Nutrient Profile in the UK (2008-2014). Nutrients, v. 10, 5 587, 2018.
CHEN, X. et al. Consumption of ultra-processed foods and health outcomes: a systematic review of epidemiological studies. Nutrition Journal, v. 19, n.1, p. 1-10, 2020.
LOUZADA, M. L. C. et al. Ultra-processed foods and the nutritional dietary profile in Brazil. Revista de Saúde Pública, v. 49, 2015.
BIELEMANN, R. M. et al. Early feeding practices and consumption of ultraprocessed foods at 6y of age: Findings from the 2004 Pelotas (Brazil) Birth Cohort Study. Nutrition, v. 47, p. 27-32, 2018.
ELIZABETH, L. et al. Ultra-Processed Foods and Health Outcomes: A Narrative Review. Nutrients, v. 12, n. 7, p. 1955, 2020.
MALTA, D. C. et al. Fatores de risco e proteção para doenças crônicas por inquérito telefônico nas capitais brasileiras, Vigitel 2014. Revista Brasileira de Epidemiologia, v.18, p. 238-255, 2015.
JORGE, A. K. B.; PACHECO, A. F. C.; MOREIRA, R. V. Psiquiatria nutricional: a influência da alimentação na saúde mental. In: LÓSS, J. C. S. et al. (Orgs.). Principais transtornos psíquicos na contemporaneidade, p. 103, 2019.
KUCZMARSKI, M. F. et al. Higher Healthy Eating Index-2005 scores associated with reduced symptoms of depression in an urban population: findings from the Healthy Aging in Neighborhoods of Diversity Across the Life Span (HANDLS) study. Journal of the American Dietetic Association, v. 110, p. 383-389, 2010.
PULOKA, I. et al. Dietary behaviours and the mental well-being of New Zealand adolescents. Journal of Paediatrics and Child Health, v. 53, p. 657-662, 2017.
QUEHL, R. et al. Food and Mood: Diet Quality is Inversely Associated with Depressive Symptoms in Female University Students. Canadian Journal of Dietetic Practice and Research, v. 78, p. 124-128, 2017.
LOPES, C. S. Como está a saúde mental dos brasileiros? A importância das coortes de nascimento para melhor compreensão do problema. Cad. Saúde Pública, v. 36, n. 2, 2020.
BONADIMAN, C. S. C. et al. Depressive disorders in Brazil: results from the Global Burden of Disease Study 2017. Population Health Metrics, v. 18, n. 1, p. 1-13, 2020.
BRASIL. Instituto Brasileiro de Geografia e Estatística-IBGE. Pesquisa nacional de saúde, 2013: acesso e utilização dos serviços de saúde, acidentes e violências: Brasil, grandes regiões e unidades da Federação. 2015.
BRASIL. Instituto Brasileiro de Geografia e Estatística-IBGE. Pesquisa nacional de saúde, 2019: informações sobre domicílios, acesso e utilização dos serviços de saúde: Brasil, grandes regiões e unidades da federação. 2020.
FRANCISCO, P. M. S. B. et al. Prevalência e coocorrência de fatores de risco modificáveis em adultos e idosos. Rev Saude Publica, v. 53, p. 53-86, 2019.
SANTOS, I. S. et al. Sensibilidade e especificidade do Patient Health Questionnaire-9 (PHQ-9) entre adultos da população geral. Cadernos de Saúde Pública, v. 29, p. 1533-1543, 2013.
MUNHOZ, T. N.; SANTOS, I. S.; MATIJASEVICH, A. Depression among Brazilian adolescents: a cross-sectional population-based study. Journal of Affective Disorders, v. 175, p. 281-286, 2015.
SILVA, M. T. et al. Prevalence and correlates of depressive symptoms among adults living in the Amazon, Brazil: A population-based study. Journal of Affective Disorders, v. 222, p. 162-168, 2017.
SANTOS, L. P. et al. Association between food insecurity and major depressive episodes amid Covid-19 pandemic: results of four consecutive epidemiological surveys from southern Brazil. Public Health Nutrition, v. 25, p. 944-953, 2022.
CEBRINO, J.; PORTERO DE LA CRUZ, S. et al. Diet Quality According to Mental Status and Associated Factors during Adulthood in Spain. Nutrients, v. 13, n. 5, p. 1727, 2021.
ELSTGEEST, L. E. et al. Associations of depressive symptoms and history with three a priori diet quality indices in middle-aged and older adults. Journal of Affective Disorders, v. 249, p. 394-403, 2019.
SILVA, J. D. M. Distúrbio da ansiedade e impacto nutricional: obesidade e compulsividade alimentar. Revista Eletrônica Acervo Saúde, v. 15, p. 1-7, 2022.
OLIVEIRA, M.F.C. et al. Prevalência de obesidade em adolescentes e jovens. Revista Brasileira de Obesidade, Nutrição e Emagrecimento, v. 14, p. 811-820, 2020.
LOPES, K. C. S. P. e SANTOS, W. L. Transtorno de ansiedade. Revista de Iniciação Científica e Extensão, v. 1, p. 45-50, 2018.

Table 1. Socioeconomic and demographic characteristics of individuals from the 2013 and 2019 Brazilian National Health Survey (N = 62.986 in 2013 and 88.531 in 2019).

	2013	2019
Sociodemographic characteristics	% (IC95%)	% (IC95%)
Geographical region
North	7,3 (7,1; 7,6)	10,7 (10,4; 11,0)
Northeast	26,3 (25,8; 26,9)	23,4 (22,9; 23,9)
Midwest	7,4 (7,2; 7,6)	7,6 (7,3; 7,8)
Southeast	44,1 (43,4; 44,8)	43,5 (42,7; 44,3)
South	14,8 (14,4; 15,3)	14,8 (14,4; 15,2)

Household área
Urban	86,2 (85,7; 86,6)	86,2 (85,8; 86,6)
Rural	13,8 (13,4; 14,3)	13,8 (13,4; 14,2)

Sex
Male	46,9 (46,1; 47,7)	46,8 (46,1; 47,4)
Female	53,1 (52,3; 53,9)	53,2 (52,6; 53,9)

Age groups
20-39	45,0 (44,3; 45,8)	40,8 (40,1; 41,4)
40-59	36,0 (35,3; 36,7)	36,7 (36,1; 37,3)
60 or older	19,0 (18,4; 19,6)	22,5 (21,9; 23,1)

Skin color
White	48,5 (47,7; 49,4)	44,1 (43,4; 44,9)
Brown	42,2 (41,4; 43,0)	44,2 (43,6; 44,9)
Black	9,3 (8,8; 9,8)	11,6 (11,2; 12,1)

Marital status
Married	46,5 (45,7; 47,3)	45,5 (44,8; 46,2)
Separated/divorced	6,7 (6,4; 7,1)	7,4 (7,1; 7,7)
Widowed	7,0 (6,7; 7,3)	7,1 (6,8; 7,4)
Single	39,8 (39,0;40,5)	40,0 (39,3; 40,7)

Nº of residentes in the household
1-2	28,2 (27,6; 28,8)	32,3 (31,8; 32,8)
3-4	49,7 (49,0; 50,4)	50,6 (50,0; 51,2)
5 or more	22,1 (21,4; 22,8)	17,1 (16,5; 17,6)

Per capita household income (quintiles)
1° (lowest)	10,0 (9,6; 10,4)	12,5 (12,2; 12,9)
2°	20,5 (20,0; 21,1)	19,5 (19,0; 20,0)
3°	22,8 (22,2; 23,4)	17,1 (16,6; 17,5)
4°	23,7 (23,1; 24,4)	25,8 (25,2; 26,4)
5° (highest)	23,0 (22,0; 24,0)	25,2 (24,4; 26,0)

Formal education
None	7,2 (6,8; 7,6)	5,0 (4,8; 5,3)
Elementary (incomplete)	24,9 (24,1; 25,7)	25,5 (24,9; 26,1)
Elementary (complete)	19,2 (18,6; 19,8)	17,3 (16,8; 17,7)
High school	35,0 (34,3; 35,8)	35,8 (35,1; 36,4)
University degree	13,7 (12,9; 14,5)	16,4 (15,7; 17,1)

Ocupation status
Not working	38,0 (37,2; 38,7)	38,2 (37,6; 38,8)
Working	62,0 (61,3; 62,8)	61,8 (61,2; 62,4)
Total	100,0	100,0

Table 2. Prevalence of major depressive episodes according to socioeconomic and demographic characteristics of individuals from the 2013 and 2019 Brazilian National Health Survey (N = 62.986 in 2013 and 88.531 in 2019).

	MDE using the DSM-IV-TR criterion		MDE using the cutoff of ≥9
Sociodemographic characteristics	2013	2019	2013		2019
	% (IC95%)	% (IC95%)	% (IC95%)		% (IC95%)
Geographical region	<0,001	<0,001		<0,001		<0,001
North	2,9 (2,5; 3,4)	4,0 (3,6; 4,4)		7,8 (7,0; 8,7)		10,1 (9,5; 10,8)
Northeast	4,0 (3,5; 4,5)	5,4 (5,0; 5,9)		10,2 (9,5; 11,0)		13,3 (12,6; 13,9)
Midwest	3,9 (3,4; 4,5)	6,2 (5,5; 6,9)		10,0 (9,0; 11,0)		13,7 (12,6; 14,7)
Southeast	4,3 (3,8; 4,9)	6,0 (5,4; 6,6)		9,2 (8,4; 10,0)		13,5 (12,7; 14,3)
South	4,9 (4,2; 5,9)	5,7 (5,1; 6,3)		11,0 (9,8; 12,2)		12,5 (11,5; 13,5)

Household área	0,011	<0,001		0,004		<0,001
Urban	4,3 (4,0; 4,6)	5,9 (5,6; 6,3)		9,9 (9,4; 10,4)		13,5 (13,0; 14,0)
Rural	3,5 (2,9; 4,0)	3,7 (3,3; 4,1)		8,4 (7,5; 9,2)		9,4 (8,7; 10,2)

Sex	<0,001	<0,001		<0,001		<0,001
Male	2,5 (2,1; 2,8)	3,1 (2,8; 3,4)		6,0 (5,5; 6,5)		7,5 (7,0; 7,9)
Female	5,7 (5,3; 6,1)	7,9 (7,4; 8,4)		13,0 (12,3; 13,7)		17,8 (17,1; 18,4)

Age groups	<0,001	<0,001		<0,001		<0,001
20-39	3,2 (2,9; 3,6)	4,8 (4,3; 5,2)		8,0 (7,5; 8,6)		12,0 (11,3; 12,7)
40-59	4,9 (4,4; 5,4)	6,4 (6,0; 6,9)		11,1 (10,4; 11,9)		14,0 (13,3; 14,6)
60 or older	5,2 (4,4; 6,0)	5,9 (5,3; 6,5)		11,0 (10,0; 12,0)		13,1 (12,3; 13,9)

Skin color	0,473	0,091		0,036		0,110
White	4,0 (3,6; 4,4)	5,6 (5,1; 6,1)		9,2 (8,6; 9,8)		12,7 (12,0; 13,4)
Brown	4,4 (3,9; 4,8)	5,5 (5,1; 5,9)		10,1 (9,5; 10,8)		13,0 (12,4; 13,6)
Black	4,3 (3,4; 5,1)	6,5 (5,6; 7,3)		10,6 (9,2; 12,0)		14,1 (13,0;15,0)

Marital status	<0,001	<0,001		<0,001		<0,001
Married	4,2 (3,7; 4,6)	5,0 (4,6; 5,5)		9,2 (8,6; 9,9)		11,6 (10,9; 12,2)
Separated/divorced	5,4 (4,3; 4,6)	8,5 (7,4; 9,7)		12,2 (10,7; 13,8)		16,9 (15,4; 18,4)
Widowed	6,0 (4,9; 7,2)	7,9 (6,9; 8,9)		13,6 (12,0; 15,3)		16,4 (15,0; 17,7)
Single	3,7 (3,3; 4,1)	5,4 (5,0; 5,8)		9,2 (8,5; 9,8)		13,2 (12,5; 13,9)

Nº of residentes in the household	0,249	<0,001		0,009		<0,001
1-2	4,4 (4,0; 4,9)	6,2 (5,8; 6,7)		10,1 (9,5; 10,1)		13,5 (12,9; 14,1)
3-4	4,0 (3,6; 4,4)	5,1 (4,7; 5,5)		9,1 (8,5; 9,7)		12,1 (11,5; 12,7)
5 or more	4,4 (3,7; 5,1)	6,0 (5,2; 6,8)		10,6 (9,6; 11,7)		14,5 (13,3; 15,7)

Per capita household income (quintiles)	<0,001	<0,001		<0,001		<0,001
1° (lowest)	6,5 (5,5; 7,5)	7,4 (6,7; 8,0)		14,1 (12,7; 15,5)		15,9 (15,0;16,9)
2°	4,6 (4,1; 5,2)	6,6 (6,0; 7,2)		10,6 (9,8; 11,5)		14,6 (13,8; 15,5)
3°	4,3 (3,7; 4,9)	5,3 (4,7; 6,0)		10,0 (9,1; 10,9)		12,7 (11,7; 13,7)
4°	4,0 (3,4; 4,6)	5,6 (5,0; 6,2)		9,5 (8,6; 10,3)		12,2 (11,4; 13,1)
5° (highest)	2,9 (2,3; 3,5)	4,2 (3,7; 4,8)		6,9 (6,1; 7,7)		11,1 (10,2; 12,0)

Formal education	<0,001	<0,001		<0,001		<0,001
None	6,9 (5,6; 8,3)	7,1 (5,9; 8,3)		14,8 (13,0; 16,6)		16,4 (14,7; 18,1)
Elementary (incomplete)	5,1 (4,5; 5,7)	6,7 (6,1; 7,3)		11,8 (10,8; 12,8)		14,8 (14,0; 15,6)
Elementary (complete)	5,3 (4,5; 6,1)	6,2 (5,5; 7,0)		11,1 (10,0; 12,2)		13,3 (12,3; 14,3)
High school	2,8 (2,4; 3,1)	4,8 (4,3; 5,2)		7,5 (6,8; 8,1)		11,8 (11,1; 12,5)
University degree	2,9 (2,2; 3,6)	4,4 (3,6; 5,2)		6,8 (5,8; 7,8)		11,0 (9,9; 12,0)

Ocupation status	<0,001	<0,001		<0,001		<0,001
Not working	6,4 (5,8; 7,0)	8,0 (7,5; 8,6)		13,3 (12,5; 14,1)		16,8 (16,1; 17,5)
Working	2,8 (2,6; 3,1)	4,2 (3,8; 4,5)		7,5 (7,1; 8,0)		10,6 (10,1; 11,1)
Total	4,2 (3,9; 4,5)	5,6 (5,3; 5,9)		9,7 (9,3; 10,2)		13,0 (12,5; 13,4)

Table 3. Mean of the diet quality score according to socioeconomic and demographic characteristics of the adults and older adults from the 2013 and 2019 Brazilian National Health Survey (N = 62.986 in 2013 and 88.531 in 2019).

	2013	2019
Socioeconomic and demographic characteristics	Mean (IC95%)†	Mean (IC95%)†
Geographical region	<0.001	<0.001
North	0.25 (0.21; 0.29)	0.11 (0.08; 0.13)
Northeast	0.00 (-0.03; 0.04)	-0.08 (-0.10; -0.05)
Midwest	0.07 (0.03; 0.10)	0.19 (0.13; 0.20)
Southeast	-0.09 (0.12; -0.06)	-0.00 (-0.03; 0.02)
South	0.04 (-0.00; 0.09)	0.16 (0.13; 0.20)

Household área	<0.001	0.033
Urban	-0.02 (-0.04; -0.01)	0.03 (0.01; 0.04)
Rural	0.09 (0.05; 0.13)	0.06 (0.03; 0.09)

Sex	<0.001	<0.001
Male	0.16 (0.13; 0.18)	0.16 (0.14; 0.18)
Female	-0.15 (-0.18; -0.13)	-0.08 (-0.10; -0.06)

Age groups	<0.001	<0.001
20-39	0.26 (0.23; 0.28)	0.36 (0.34; 0.38)
40-59	-0.11 (-0.14; -0.09)	-0.08 (-0.10; -0.06)
60 or older	-0.44 (-0.47; -0.41)	-0.39 (-0.41; -0.37)

Skin color	<0.001	<0.001
White	-0.05 (-0.08; -0.02)	-0.00 (-0.02; 0.02)
Brown	0.03 (0.01; 0.06)	0.05 (0.03; 0.07)
Black	0.02 (-0.02; 0.07)	0.07 (0.04; 0.11)

Marital status	<0.001	<0.001
Married	-0.12 (-0.14; -0.09)	-0.10 (-0.11; -0.07)
Separated/divorced	-0.13 (-0.18; -0.07)	-0.07 (-0.11; -0.03)
Widowed	-0.43 (-0.048; -0.38)	-0.40 (-0.43; -0.36)
Single	0.21 (0.18; 0.24)	0.26 (0.24; 0.28)

Nº of residentes in the household	<0.001	<0.001
1-2	-0.13 (-0.16; -0.11)	-0.09 (-0.11; -0.07)
3-4	0.01 (-0.02; 0.03)	0.05 (0.03; 0.07)
5 or more	0.11 (0.07; 0.15)	0.20 (0.17; 0.24)

Per capita household income (quintiles)	<0.001	<0.001
1° (lowest)	0.04 (0.00; 0.08)	0.11 (0.08; 0.13)
2°	0.06 (0.03; 0.09)	0.05 (0.03; 0.08)
3°	0.03 (0.00; 0.06)	0.02 (-0.01; 0.06)
4°	-0.04 (-0.07; -0.01)	0.02 (-0.01; 0.05)
5° (highest)	-0.09 (-0.13; -0.05)	-0.01 (-0.04; 0.02)

Formal education	<0.001	<0.001
None	-0.07 (-0.13; -0.01)	-0.07 (-0.11; -0.02)
Elementary (incomplete)	0.03 (-0.01; 0.07)	0.01 (-0.01; 0.03)
Elementary (complete)	-0.07 (-0.10; -0.03)	0.02 (-0.01; 0.05)
High school	0.05 (0.02; 0.08)	0.11 (0.08; 0.13)
University degree	-0.08 (-0.13; -0.04)	-0.03 (-0.07; -0.01)

Ocupation status	<0.001	<0,001
Not working	-0.18 (-0.20; -0.15)	-0.13 (-0.15; -0.11)
Working	0.09 (0.07; 0.12)	0.13 (0.11; 0.14)
Total	-0.01 (-0.03; 0.01)	0.03 (0.02; 0.04)

†Diet quality score was standardized to provide a more convenient scale and results are presented in z-score.

Table 4. Crude and adjusted associations between major depressive episodes and diet quality among adults and older adults from the 2013 and 2019 Brazilian National Health Survey.

MDE using the DSM-IV-TR criterion	Diet quality score
	2013	2019
	β (IC95%)	β (IC95%)
	Crude analysis
p-value	0.072	0.003
No	0.00	0.00
Yes	0.07 (-0.01; 0.14)	0.09 (0.03; 0.14)

	Adjusted analysis†
p-value	<0.001	<0.001
No	0.00	0.00
Yes	0.19 (0.12; 0.26)	0.16 (0.10; 0.22)

MDE using the cutoff of ≥9	2013	2019
MDE using the cutoff of ≥9	β (IC95%)	β (IC95%)
	Crude analysis
p-value	0.012	<0,001
No	0.00	0.00
Yes	0.06 (0.01; 0.11)	0.14 (0.10; 0.18)

	Adjusted analysis†
p-value	<0.001	<0.001
No	0.00	0.00
Yes	0.16 (0.12; 0.21)	0.20 (0.16; 0.24)

MDE – Major depressive episodes

† Adjusted for geographical region, household area, sex, age, skin color, marital status, number of residents in the household, per capita household income, formal education, and occupation

No competing interests reported.

Supplementarytable.docx

Major depressive episodes and diet quality among Brazilian adults and older adults: results from the 2013 and 2019 National Health Survey

Status:

Version 1

Abstract

Introduction

Methods

Design and sampling

Definition of food consumption and diet quality

Screening for major depressive episodes

Covariables

Statistical analyses

Results

Discussion

Declarations

Author Contribution

Acknowledgement

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1