Abstracts

Introduction

Diet quality has changed over the years, with a reduction in the consumption of fruits, vegetables, grains, and legumes, and an increase in the consumption of industrially processed foods and beverages and ready-to-eat food products ¹1. Monteiro CA, Moubarac JC, Levy RB, Canella DS, Louzada MLC, Cannon G. Household availability of ultra-processed foods and obesity in nineteen European countries. Public Health Nutr 2017; 21:18-26.. Analyzing these changes in dietary patterns, Monteiro et al. ²2. Monteiro CA, Levy RB, Claro RM, Castro IRR, Cannon G. A new classification of foods based on the extent and purpose of food processing. Cad Saúde Pública 2010; 26:2039-49. developed a food classification based on the level of processing and the nature, extent, and purpose of industrial processes to foods. The NOVA system is internationally recognized and has been widely used in epidemiological studies on individual food consumption, diet quality, and health conditions ³3. Beslay M, Srour B, Méjean C, Allès B, Fiolet B, Debras C, et al. Ultra-processed food intake in association with BMI change and risk of overweight and obesity: a prospective analysis of the French NutriNet-Santé cohort. PLoS Med 2020; 17:e1003256.^,44. Kim H, Hu EA, Rebholz CM. Ultra-processed food intake and mortality in the USA: results from the Third National Health and Nutrition Examination Survey (NHANES III, 1988-1994). Public Health Nutr 2019; 22:1777-85.^,55. Monteiro CA, Cannon G, Lawrence M, Louzada MLC, Machado PP. Ultra-processed foods, diet quality, and health using the NOVA classification system. Rome: Food and Agriculture Organization of the United Nations; 2019..

Ultra-processed foods are industrial formulations with little or no real food that are marketed for quick consumption ⁶6. Monteiro CA, Cannon G, Levy RB, Moubarac JC, Jaime P, Martins AP, et al. NOVA. The star shines bright. World Nutrition 2016; 7:28-40.^,77. Gomes CB, Malta MB, Benício MHD, Carvalhães MABL. Consumption of ultra-processed foods in the third gestational trimester and increased weight gain: a Brazilian cohort study. Public Health Nutr 2021; 24:3304-31.. They have a negative effect on diet quality due to their high levels of sodium, saturated fat, and sugar ⁸8. Moubarac JC, Martins AP, Claro RM, Levy RB, Cannon G, Monteiro CA. Consumption of ultra-processed foods and likely impact on human health. Evidence from Canada. Public Health Nutr 2013; 16:2240-8., which are important factors related to morbidity and mortality from noncommunicable diseases ⁹9. World Health Organization. Diet, nutrition and the prevention of chronic diseases. Geneva: World Health Organization; 2003. (Technical Report Series, 916).. Recent studies in adults have shown an association between high ultra-processed foods consumption and an increased risk of overweight/obesity, cancer, type 2 diabetes, cardiovascular diseases, and all-cause mortality ³3. Beslay M, Srour B, Méjean C, Allès B, Fiolet B, Debras C, et al. Ultra-processed food intake in association with BMI change and risk of overweight and obesity: a prospective analysis of the French NutriNet-Santé cohort. PLoS Med 2020; 17:e1003256.^,44. Kim H, Hu EA, Rebholz CM. Ultra-processed food intake and mortality in the USA: results from the Third National Health and Nutrition Examination Survey (NHANES III, 1988-1994). Public Health Nutr 2019; 22:1777-85.^,1010. Askari M, Heshmati J, Shahinfar H, Tripathi N, Daneshzad E. Ultraprocessed food and the risk of overweight and obesity: a systematic review and meta-analysis of observational studies. Int J Obes 2020; 44:2080-91..

Data from the literature have consistently shown that, during pregnancy, a diet based on healthy eating habits contributes to the health of the pregnant woman, the development of the fetus, and the prevention of complications related to the pregnancy and the postpartum period ¹¹11. Mirmiran P, Hosseinpour-Niazi S, Moghaddam-Banaem L, Lamyian M, Goshtasebi A, Azizi F. Inverse relation between fruit and vegetable intake and the risk of gestational diabetes mellitus. Int J Vitam Nutr Res 2019; 89:37-44.^,1212. Procter SB, Campbell CG. Position of the Academy of Nutrition and Dietetics: nutrition and lifestyle for a healthy pregnancy outcome. J Acad Nutr Diet 2014; 114:1099-103.^,1313. Oliveira PG, Sousa JM, Assunção DGF, Araujo EKS, Bezerra DS, Dametto JFDS, et al. Impacts of consumption of ultra-processed foods on the maternal-child health: a systematic review. Front Nutr 2022; 9:821657.. Recently, studies on ultra-processed foods observed an important relationship between high consumption and increased gestational weight gain, gestational diabetes, overweight/obesity, and depression and sadness ⁷7. Gomes CB, Malta MB, Benício MHD, Carvalhães MABL. Consumption of ultra-processed foods in the third gestational trimester and increased weight gain: a Brazilian cohort study. Public Health Nutr 2021; 24:3304-31.^,1414. Ferreira LB, Lobo CV, Carmo AS, Souza RCVE, Santos LC. Dietary patterns during pregnancy and their association with gestational weight gain and anthropometric measurements at birth. Matern Child Health J 2022; 26:1464-72.^,1515. Cummings JR, Lipsky LM, Schwedhelm C, Liu A, Nansel TR. Associations of ultra-processed food intake with maternal weight change and cardiometabolic health and infant growth. Int J Behav Nutr Phys Act 2022; 19:61.. Inadequate maternal weight gain favors the development of gestational and postpartum complications, besides influencing fetal health conditions, such as birth weight, mode of delivery, and duration of pregnancy ¹⁶16. Morais SS, Nascimento SL, Godoy-Miranda AC, Kasawara KT, Surita FG. Body mass index changes during pregnancy and perinatal outcomes: a cross-sectional study. Rev Bras Ginecol Obstet 2018; 40:11-9..

Sociodemographic, cultural, and behavioral factors can compromise maternal eating habits and thus lead to increased ultra-processed foods consumption. Identifying these factors, especially potentially modifiable factors, can support more effective nutritional guidance measures. Given the increase in ultra-processed foods consumption in medium-/low-income countries, this study aims to describe the factors associated with higher ultra-processed foods consumption, based on data from a cohort of pregnant women in two Family Health Strategy (FHS) health units in the city of Rio de Janeiro, Brazil.

Methodology

Study design and population

This study analyzed data from pregnant women who participated in the research project entitled Factors Associated with Pregestational Obesity and its Repercussions on Maternal and Neonatal Health, a prospective cohort study conducted from February 2016 to November 2019 in two FHS units. The health units are located in one of the most vulnerable regions of Rio de Janeiro city, with the fifth lowest human development index in the city ¹⁷17. Sistema Municipal de Informações Urbanas. Índice de Desenvolvimento Humano (IDH), por ordem decrescente, dos Municípios do Estado do Rio de Janeiro em 1991/2000/2010. http://www.data.rio/datasets/%C3%ADndice-de-desenvolvimento-humano-municipal-idh-por-ordem-de-idh-segundo-asregi%C3%B5es-administrativas-ra-em-1991-2000-2010 (accessed on 15/Aug/2019).
http://www.data.rio/datasets/%C3%ADndice... .

This sample size was estimated for a 5% prevalence of negative outcomes (gestational diabetes or hypertension), 95% confidence interval (95%CI), and 80% power, allowing the detection of a difference of ≥ 2 in relative risk, considering a ratio of about 3:1 (35% overweight/obesity) between exposed and unexposed. In total, 512 pregnant women with a low obstetric risk, gestational age < 20 weeks, and aged ≥ 18 years were included in the baseline study. The first interview was performed during prenatal visits, where pregnant women were recruited sequentially until the planned sample size was reached. Two more interviews were conducted: at 34 weeks gestational age and two months postpartum. Women who answered the three questionnaires were included in this analysis, totaling 393 women.

Outcome variable: consumption of ultra-processed foods

Food consumption during pregnancy was assessed using a food frequency questionnaire (FFQ) applied in the third interview. The questionnaire presented eight different options for consumption frequency that were converted into daily intake: “more than three times per day”, “two to three times per day”, “once per day”, “five to six times per week”, “two to four times per week”, “once per week”, “one to three times per month”, and “never/almost never”. The list of foods included 88 items and, for each item, standardized portions, as an option to assess the amount consumed ¹⁸18. Sichieri R, Everhart JE. Validity of a Brazilian food frequency questionnaire against dietary recalls on estimated energy intake. Nutr Res 1998; 18:1649-59.. The questionnaire was validated by Giacomello et al. ¹⁹19. Giacomello A, Schmidt MI, Nunes MAN, Duncan BB. Validation of a Food Frequency Questionnaire conducted among pregnant women attended by the Brazilian National Health Service, in two municipalities of the State of Rio Grande do Sul, Brazil. Rev Bras Saúde Mater Infant 2008; 8:445-54. among pregnant women who used public healthcare services in Brazil.

Food energy value was estimated by converting the daily intake, consulting a food consumption table and giving references in 100-g portions and household measurements ²⁰20. Núcleo de Estudos e Pesquisas em Alimentação, Universidade Estadual de Campinas. Tabela Brasileira de Composição de Alimentos - TACO, 4ª edição revisada e ampliada. Campinas: Núcleo de Estudos e Pesquisas em Alimentação, Universidade Estadual de Campinas; 2011.^,2121. Philippi ST. Tabela de composição de alimentos: suporte para decisão nutricional. Brasília: Manole; 2020.. ultra-processed foods were identified according to the NOVA classification proposed in the Dietary Guidelines for the Brazilian Population²²22. Primary Health Care Department, Secretariat of Health Care, Ministry of Health of Brazil. Dietary guidelines for the Brazilian population. Brasília: Ministry of Health of Brazil; 2015., which considers the following food groups: (1) natural or minimally processed foods; (2) oils, fats, salt, and sugar; (3) processed foods; and (4) ultra-processed foods. The variable ultra-processed foods did not follow a Gaussian distribution and, therefore, was analyzed by tertile distribution. The third tertile corresponded to the highest consumption and the first tertile was the reference category in the analysis with the second and third tertiles.

Covariables

First questionnaire: age (18-24; 25-34; ≥ 35 years old); years of schooling (< 7; 8-11; ≥ 12 years); ethnicity/skin-color (white, black, and mixed-race); paid work (“Do you currently have a job that you earn money with?”: yes/no); marital status (“Do you live with a spouse/partner?”: yes/no); parity (no previous childbirths; one previous childbirth; two or more previous childbirths); planned pregnancy (yes, if the woman “wanted to become pregnant” versus no, if the woman “wanted to wait longer” or “did not want to become pregnant”); satisfaction with weight before pregnancy (yes/no); leisure time physical activity before pregnancy based on women’s information (yes/no); social support (high; above the median on the scale developed in the Medical Outcomes Study) ²³23. Silva KS, Coutinho ES. Social support scale: test-retest reliability in pregnant women and structures of agreement and disagreement. Cad Saúde Pública 2005; 21:979-83..

Second questionnaire: smoking during pregnancy (at least one cigarette per day every day); alcohol abuse (2 on the TWEAK scale) ²⁴24. Moraes CL, Viellas EF, Reichenheim ME. Assessing alcohol misuse during pregnancy: evaluating psychometric properties of the CAGE, T-ACE and TWEAK in a Brazilian setting. J Stud Alcohol 2005; 66:165-73.; diabetes mellitus (diagnosis of gestational or pregestational diabetes); hypertension (diagnosis of gestational or pregestational hypertension); prenatal nutritional guidance (yes/no).

Third questionnaire: pregestational nutritional status (classified according to the body mass index [BMI], measured until the 13th gestational week and recorded in the woman’s prenatal booklet); leisure time physical activity during pregnancy (according to the Pregnancy Physical Activity Questionnaire [PPAQ] ²⁵25. Silva FT, Araujo Júnior E, Santana EF, Lima JWO, Cecchino GM, Costa FS. Translation and cross-cultural adaptation of the Pregnancy Physical Activity Questionnaire (PPAQ) to the Brazilian population. Ceska Gynekol 2015; 80:290-8., pregnant women are classified as active [≥ 150 minutes/week] or insufficiently active or inactive [< 150 minutes/week]); symptoms of depression (≥ 10 on the Edinburgh Postnatal Depression Scale [EPDS]) ²⁶26. Santos IS, Matijasevich A, Tavares BF, Barros AJD, Botelho IP, Lapolli C, et al. Validation of the Edinburgh Postnatal Depression Scale (EPDS) in a sample of mothers from the 2004 Pelotas Birth Cohort Study. Cad Saúde Pública 2004; 23:2577-88.; symptoms of anxiety (≥ 3 on the Patient Health Questionnaire-4 [PHQ4]).

Data analysis

To assess food consumption during pregnancy, a sectional analysis of the FFQ in the third wave was performed. Foods were initially quantified according to their energy value and percentage contribution to total daily energy intake, grouped according to the NOVA classification.

(1) Natural or minimally processed food: rice; pasta; beans and legumes (lentil); fruits (orange, banana, papaya, apple, watermelon, pineapple, mango, grape, pear, passion fruit, lemon, watermelon, avocado, and guava); root and tuber vegetables (potato, cassava, carrot, and beet); milk, chicken; red meat; juice; fish; eggs; giblets; flour (cassava or manioc flour, and polenta); peanuts; popcorn; and coffee.

(2) Oils, fats, salt, and sugar: sugar; butter.

(3) Processed foods: cheese; canned foods (maize, peas, and tuna/sardine); bacon.

(4) Ultra-processed foods: bread; cookies; soft drinks; cakes; chocolate bars; pizza; candies; caramels; ice cream; chocolate powder; margarine; mayonnaise; yogurt; processed meats (hamburger and sausage) and alcoholic beverages (beer and wine).

The analysis excluded women with implausible total energy intake (< 600 or > 6,000Kcal/day) ²⁷27. Alves-Santos NH, Eshriqui I, Franco-Sena AB, Cocate PG, Freitas-Vilela AA, Benaim C, et al. Dietary intake variations from pre-conception to gestational period according to the degree of industrial processing: a Brazilian cohort. Appetite 2016; 105:164-71.. A bivariate analysis of the independent variables in relation to ultra-processed foods consumption was performed using the chi-square test and 5% statistical significance. Factors associated with ultra-processed foods consumption were identified based on the literature on the topic among pregnant women, introducing a set of covariables into the analysis using a hierarchical multinomial logistic regression model. The underlying logic of the proposed model is that the hierarchically superior factors use the inferior factors to perform their actions (Figure 1).

Variables at level 1 that reached significance (p < 0.20) in the simple logistic regression remained in the multivariate regression model, adopting the backward procedure with gradual exclusion of the variables with the lowest statistical significance, until the final model at this level retained only variables with p < 0.05.

For each variable at level 2, the adjusted odds ratio (OR) was estimated for the variables retained at the level 1. Variables that reached statistical significance (p < 0.20) were included in the multivariate regression model at this level, along with the variables retained at the previous level. A new backward procedure with gradual exclusion of the variables with the lowest statistical significance was performed until the model retained only variables with p < 0.05. Finally, the same procedures were performed at level 3.

The final model for the hierarchical multivariate logistic regression analysis retained only variables with p < 0.05 at each level. Statistical analyses were performed using SPSS version 22 (https://www.ibm.com/).

Ethical aspects

The study was approved according to the recommendations of Resolution n. 466/2012 of the Brazilian National Health Council, which defines the procedures for research in human subjects, and has been filed with the Ethics Research Committee of the Sergio Arouca National School of Public Health, Oswaldo Cruz Foundation (ENSP/Fiocruz, CAAE 21982613.6.0000.5240).

Results

Of the 520 pregnant women recruited for the first interview, 393 completed the three questionnaires. Losses (n = 120; 23.4%) occurred mainly due to address changes (n = 63; 12.3%), refusal to participate (n = 25; 4.9%), miscarriage or stillbirth (n = 17, 3.3%), and not located (n = 15; 2.9%). Years of schooling (p = 0.011) was the only variable with a significant difference between respondents and nonrespondents. We found no significant difference in age (p = 0.272), ethnicity/skin-color (p = 0.650), paid work (p = 0.388), or marital status (p = 0.958) (Table 1).

After excluding implausible total energy intake (< 600 or > 6,000Kcal/day), we analyzed data from 344 pregnant women. Mean daily energy intake during pregnancy was 3,335Kcal (standard deviation - SD ±1,147.57Kcal), of which 52.5% (1,752.2Kcal) came from unprocessed or minimally processed foods, 4.8% (161.4Kcal) from cooking ingredients, 7.3% (242.7Kcal) from processed foods, and 35.3% (1,178.6Kcal) from ultra-processed foods (Table 2). The group 1 foods that contributed the most were fruits (8.4%), beans and legumes (7.8%), root and tuber vegetables (5.9%), rice (5.6%), milk (5%), chicken (4%), red meat (3.9%), juice (2.2%), fish (1.8%), pasta (1.7%), and eggs (1.4%). In the cooking ingredients group, sugar and butter contributed 4.5% and 0.4%, respectively. In group 3, cheese (1%), canned foods (0.5%), and bacon (0.3%) contributed the highest percentage of calories. The most widely consumed ultra-processed foods were breads (9.8%), cookies (5.8%), soft drinks (3.2%), crackers (2.1%), cakes (2%), chocolate bars (1.5%), pizza (1.4%), processed meat (1.3%), ice cream (1.1%), and chocolate powder (1%).

The mean age of respondents was 26.7 years (SD ±6.0), ranging from 18 to 44 years. They reported a mean of 10.1 (SD ±2.8) years of schooling and 56.4% had eight to eleven years of schooling. Table 3 shows the characteristics of pregnant women according to their energy intake from ultra-processed foods. Compared with the 1st tertile of consumption (lowest consumption), women in the 3rd tertile (highest consumption) were younger, had fewer years of schooling, reported less physical activity before pregnancy, and were more likely to report alcohol abuse. Obstetric variables, pregestational nutritional status, and psychological variables, such as depression and anxiety during pregnancy, did not show statistically different proportions between tertiles of ultra-processed foods consumption.

Tables 4 and 5 present the results of the analysis of the three hierarchical levels, comparing the 2nd and 3rd tertiles with the 1st tertile of ultra-processed foods consumption (reference). In the comparative analysis between the 2nd with the 1st tertiles, two variables at level 1 (age and years of schooling) showed a crude association with ultra-processed foods consumption (Table 4). However, after multivariate analysis, no variable remained in the final model with a significance level < 0.05 (Table 5). At level 2, only parity and leisure time physical activity before pregnancy were associated with the 2nd tertile of ultra-processed foods consumption with a significance level < 0.20 (Table 4). After adjusting for the significant variables at this level and the previous level, only parity remained in the final model (p < 0.05) (Table 5). At level 3, the variables associated with the second tertile of ultra-processed foods consumption (p < 0.20) were alcohol abuse, diabetes mellitus, and anxiety disorder during pregnancy. All variables lost significance in the multivariate model, therefore, we did not include them in the final model.

When comparing the 3rd tertile with the reference category, the variables at level 1 that showed an association with p < 0.20 were age, years of schooling, and ethnicity/skin-color (Table 4). In the multivariate model, the variable ethnicity/skin-color lost statistical significance (Table 5). At level 2, for parity, leisure time physical activity before pregnancy, and pregestational weight, p < 0.20 (Table 4). The only variables that remained in the final model were parity and leisure time physical activity before pregnancy (Table 5). No variable at level 3 showed statistical significance among women in the highest tertile of consumption, either in the analysis adjusted for variables retained at levels 1 and 2 or in the multivariate model among variables at the same level (Table 4). Thus, no variables at level 3 were included in the final model for high ultra-processed foods consumption among pregnant women in this cohort study (Table 5).

The final hierarchical model (Table 5) included the variables age and years of schooling (level 1), and parity and regular leisure time physical activity before pregnancy (level 2). Pregnant women with up to seven years of schooling were more than five times more likely to belong to the 3rd tertile of ultra-processed foods consumption (OR = 5.58; 95%CI: 1.62-19.23) compared with pregnant women with 12 or more years of schooling (reference). Pregnant women with two or more previous childbirths were more than four times (OR = 4.11; 95%CI: 1.72-9.80) more likely to belong to the 2nd tertile of consumption and seven times (OR = 7.53; 95%CI: 3.02-18.70) more likely to belong to the 3rd tertile compared with pregnant women with no previous childbirth (reference). Moreover, the lack of leisure time physical activity before pregnancy increased by twice the odds of high ultra-processed foods consumption (OR = 2.40; 95%CI: 1.31-4.38). Meanwhile, women aged ≥ 30 years showed lower odds of high consumption of these foods (OR = 0.33; 95%CI: 0.15-0.71).

Discussion

This study identified a set of factors associated with high ultra-processed foods consumption. Pregnant women with fewer years of schooling, higher parity, and no regular physical activity before pregnancy reported higher ultra-processed foods consumption. Moreover, our data showed a protective effect of age, as older pregnant women were less likely to consume ultra-processed foods.

According to Brazilian ²⁷27. Alves-Santos NH, Eshriqui I, Franco-Sena AB, Cocate PG, Freitas-Vilela AA, Benaim C, et al. Dietary intake variations from pre-conception to gestational period according to the degree of industrial processing: a Brazilian cohort. Appetite 2016; 105:164-71.^,2828. Pereira MT, Cattafesta M, Santos Neto ET, Salaroli LB. Maternal and sociodemographic factors influence the consumption of ultraprocessed and minimally-processed foods in pregnant women. Rev Bras Ginecol Obstet 2020; 42:380-9. and international studies ²⁹29. Wall C, Gammon C, Bandara D, Grant CC, Carr PEA, Morton SMB. Dietary patterns in pregnancy in New Zealand - influence of maternal socio-demographic, health and lifestyle factors. Nutrients 2016; 8:300., age seems to has an important effect on eating behavior. Unhealthy eating habits, including replacing regular meals with snacks, eating while watching TV, and consuming high energy-dense beverages, are behaviors related to younger individuals ²2. Monteiro CA, Levy RB, Claro RM, Castro IRR, Cannon G. A new classification of foods based on the extent and purpose of food processing. Cad Saúde Pública 2010; 26:2039-49., who tend to be more susceptible to marketing appeals ³⁰30. Vasconcellos AB, Goulart D, Gentil PC, Oliveira TP. A saúde pública e a regulamentação da publicidade de alimentos. http://189.28.128.100/nutricao/docs/geral/regulamentaPublicidadeAlimentos.pdf (accessed on 15/Jul/2021).
http://189.28.128.100/nutricao/docs/gera... . On the other hand, older pregnant women tend to adhere to a “healthy awareness” pattern consisting mainly of a higher consumption of whole wheat bread, fruits, vegetables, skim milk, and white meat, among other healthy foods ³¹31. McGowan CA, McAuliffe FM. Maternal dietary patterns and associated nutrient intakes during each trimester of pregnancy. Public Health Nutr 2013; 16:97-107..

As in the general population, women’s diet and lifestyle before and during pregnancy are strongly influenced by their sociodemographic characteristics. Evidence consistently suggests a social gradient by which older women with more years of schooling and higher income, or other markers of wealth, adopt a “healthier” dietary pattern, scoring higher on nutritional quality scales ³²32. Doyle IM, Borrmann B, Grosser A, Razum O, Spallek J. Determinants of dietary patterns and diet quality during pregnancy: a systematic review with narrative synthesis. Public Health Nutr 2017; 20:1009-28..

A study on pregestational food consumption in a cohort of 454 Brazilian pregnant women found an independent association between dietary pattern and age and years of schooling. Women who adhered to “lentils, whole grains, and soups” dietary patterns were older and had more schooling than women with low adherence to this pattern. Women who adhered more to “snacks, sandwiches, sweets, and soft drinks” dietary patterns were younger and had less schooling ³³33. Teixeira JA, Castro TG, Grant CC, Wall CR, Castro ALS, Francisco RPV, et al. Dietary patterns are influenced by socio-demographic conditions of women in childbearing age: a cohort study of pregnant women. BMC Public Health 2018; 18:301.. Similarly, a cohort of 5,664 pregnant women in New Zealand showed that the “junk food” pattern was positively associated with younger maternal age and fewer years of schooling. Moreover, pregnant women adhered less to the Ministry of Health’s Food and Nutrition Guidelines ²⁹29. Wall C, Gammon C, Bandara D, Grant CC, Carr PEA, Morton SMB. Dietary patterns in pregnancy in New Zealand - influence of maternal socio-demographic, health and lifestyle factors. Nutrients 2016; 8:300..

Another important point of this study was regular physical activity before pregnancy. Women classified as sedentary before becoming pregnant, according to the new World Health Organization (WHO) guidelines ³⁴34. World Health Organization. WHO guidelines on physical activity and sedentary behaviour. https://www.who.int/publications/i/item/9789240015128 (accessed on 01/Mar/2021).
https://www.who.int/publications/i/item/... , reported higher ultra-processed foods consumption. Regular physical activity before pregnancy is a strong predictor of physical activity during pregnancy, and its benefits are widely reported in the literature ³⁵35. Gaston A, Cramp A. Exercise during pregnancy: a review of patterns and determinants. J Sci Med Sport 2011; 14:299-305.. However, common problems in pregnancy, such as nausea, pain, and fatigue, may also interfere with women’s adherence to physical activity, contributing to a combination of unhealthy habits that pose potential risks of negative pregnancy outcomes ³⁶36. Cannon S, Lastella M, Vincze L, Vandelanotte C, Hayman M. A review of pregnancy information on nutrition, physical activity and sleep websites. Women Birth 2020; 33:35-40.. Health-related behavioral changes involve great complexity and, to be successful, they require involvement, motivation, and support. Thus, during prenatal care, it is important to understand and address the barriers involved in this process with nutritional information and encouragement of healthy habits, such as physical activity. Studies suggest that diet during pregnancy tends to reflect other health-related behaviors before and during pregnancy ³⁷37. Barker M, Dombrowski SU, Colbourn T, Fall CHD, Kriznik MN, Lawrence WT, et al. Intervention strategies to improve nutrition and health behaviours before conception. Lancet 2018; 391:1853-64..

Parity is the most frequently assessed obstetric variable in studies on dietary patterns. Our results show a direct association with ultra-processed foods consumption in the adjusted model. However, this association has shown conflicting results in the literature, sometimes with positive, sometimes with negative effects. A systematic review of dietary patterns and diet quality in pregnant women confirmed this finding. Among the 10 studies that addressed this variable, five found an inverse association between parity and healthy eating; in four studies, the association was positive, and one found no association ³²32. Doyle IM, Borrmann B, Grosser A, Razum O, Spallek J. Determinants of dietary patterns and diet quality during pregnancy: a systematic review with narrative synthesis. Public Health Nutr 2017; 20:1009-28..

Based on the Dietary Guidelines for the Brazilian Population²²22. Primary Health Care Department, Secretariat of Health Care, Ministry of Health of Brazil. Dietary guidelines for the Brazilian population. Brasília: Ministry of Health of Brazil; 2015., the diet should be based on a wide variety of unprocessed or minimally processed foods, cooking ingredients and processed foods should be used sparingly, and ultra-processed foods consumption should be avoided. The latter group is particularly critical, since ultra-processed foods contains high calories, low nutritional value, and additives in their composition ²²22. Primary Health Care Department, Secretariat of Health Care, Ministry of Health of Brazil. Dietary guidelines for the Brazilian population. Brasília: Ministry of Health of Brazil; 2015.. During pregnancy, ultra-processed foods consumption may jeopardize placental and fetal growth and development ³⁸38. Teixeira JA, Hoffman DJ, Castro TG, Saldiva SRDM, Francisco RPV, Vieira SE, et al. Pre-pregnancy dietary pattern is associated with newborn size: results from ProcriAr study. Br J Nutr 2021; 126:903-12. and increase the risk of gestational diabetes, hypertensive syndromes, and gestational weight gain, compromising the health of fetuses and mothers in the medium and long term ⁷7. Gomes CB, Malta MB, Benício MHD, Carvalhães MABL. Consumption of ultra-processed foods in the third gestational trimester and increased weight gain: a Brazilian cohort study. Public Health Nutr 2021; 24:3304-31.^,3939. Silva CFM, Saunders C, Peres W, Folino B, Kamel T, Santos MS, et al. Effect of ultra-processed foods consumption on glycemic control and gestational weight gain in pregnant with pregestational diabetes mellitus using carbohydrate counting. PeerJ 2021; 9:e10514..

The proportions of calories from ultra-processed foods and unprocessed or minimally processed foods were similar to the proportions found in other Brazilian studies with pregnant women, ranging from 48.8% to 55% for natural foods and 32% to 43% for ultra-processed foods ²⁷27. Alves-Santos NH, Eshriqui I, Franco-Sena AB, Cocate PG, Freitas-Vilela AA, Benaim C, et al. Dietary intake variations from pre-conception to gestational period according to the degree of industrial processing: a Brazilian cohort. Appetite 2016; 105:164-71.^,4040. Sartorelli DS, Crivellenti LC, Zuccolotto DCC, Franco LJ. Relationship between minimally and ultra-processed food intake during pregnancy with obesity and gestational diabetes mellitus. Cad Saúde Pública 2019; 35:e00049318.. Some studies have shown slightly lower proportions of ultra-processed foods in the diet, ranging from 22.2% to 24.8% ⁷7. Gomes CB, Malta MB, Benício MHD, Carvalhães MABL. Consumption of ultra-processed foods in the third gestational trimester and increased weight gain: a Brazilian cohort study. Public Health Nutr 2021; 24:3304-31.^,4141. Graciliano NG, Silveira JAC, Oliveira ACM. The consumption of ultra-processed foods reduces overall quality of diet in pregnant women. Cad Saúde Pública 2021; 37:e00030120.. However, the consumption of this food group has increased ⁷7. Gomes CB, Malta MB, Benício MHD, Carvalhães MABL. Consumption of ultra-processed foods in the third gestational trimester and increased weight gain: a Brazilian cohort study. Public Health Nutr 2021; 24:3304-31.^,2727. Alves-Santos NH, Eshriqui I, Franco-Sena AB, Cocate PG, Freitas-Vilela AA, Benaim C, et al. Dietary intake variations from pre-conception to gestational period according to the degree of industrial processing: a Brazilian cohort. Appetite 2016; 105:164-71., with a large portion coming from bread, cookies, cold cuts, and soft drinks ²2. Monteiro CA, Levy RB, Claro RM, Castro IRR, Cannon G. A new classification of foods based on the extent and purpose of food processing. Cad Saúde Pública 2010; 26:2039-49.. Studies with population samples, especially in high-income countries, have shown that ultra-processed foods consumption represents more than half of the total daily energy intake ⁴²42. Juul F, Martinez-Steele E, Parekh N, Monteiro CA, Chang VW. Ultra-processed food consumption and excess weight among US adults. Br J Nutr 2018; 120:90-100.^,4343. Rauber F, Louzada MLC, Steele EM, Millet C, Monteiro CA, Levy RB. Ultra-processed food consumption and chronic non-communicable diseases-related dietary nutrient profile in the UK (2008-2014). Nutrients 2018; 10:587., and this is also a reality among pregnant women ⁴⁴44. Rohatgi KW, Tinius RA, Cade WT, Steele EM, Cahill AG, Parra DC. Relationships between consumption of ultra-processed foods, gestational weight gain and neonatal outcomes in a sample of US pregnant women. PeerJ 2017; 5:e4091.. ultra-processed foods dominate the food supply in high-income countries and their consumption has increased rapidly in middle-income countries ⁴⁵45. Monteiro CA, Moubarac JC, Cannon G, Ng SW, Popkin B. Ultra-processed products are becoming dominant in the global food system. Obes Rev 2013; 14 Suppl 2:21-8.. The differences in the proportions of food groups between studies can be partly explained by the use of different dietary data collection tools, such as the 24-hour dietary recall, food frequency questionnaires, and even the classification of foods to differentiate the groups.

Although our study assessed a low-income population living in a social vulnerable area, our data showed, even in this scenario, differences in ultra-processed foods consumption. We found that less educated pregnant women were more likely to report higher ultra-processed foods consumption. The choices that constitutes the basis of healthier eating may be related to a lack of access to information and an understanding of the importance of good eating habits. Reinforcing guidance on the consumption of minimally processed foods, such as grains, legumes, fruits, greens and vegetables, unprocessed meat, and other foods, rather than ultra-processed products, such as sausages, cold cuts, and ready-to-eat dishes, can restore traditional cultural eating patterns.

Despite the methodological care in this study, the external validity of our results is limited to women with low socioeconomic status who receive prenatal and obstetric care in public healthcare services. Another key point is that dietary assessment is complex and involves recording and analyzing numerous foods and beverages consumed daily in varying amounts. The food frequency questionnaire is subject to recall bias and may underestimate or overestimate the consumption of certain food groups, thus influencing the resulting estimates. Moreover, the tool used in this study was developed in the 1990s and was not designed specifically to classify foods according to the degree of processing. Some foods were difficult to classify, since the food frequency questionnaire does not discriminate between homemade and ready-to-eat dishes. To maintain comparability with another Brazilian study on ultra-processed foods consumption during pregnancy, we used the same food grouping method ²⁷27. Alves-Santos NH, Eshriqui I, Franco-Sena AB, Cocate PG, Freitas-Vilela AA, Benaim C, et al. Dietary intake variations from pre-conception to gestational period according to the degree of industrial processing: a Brazilian cohort. Appetite 2016; 105:164-71..

Strengths of this study include the cohort design with follow-up of pregnant women from early pregnancy to the postpartum period. The hierarchical analysis model consisted of sociodemographic, obstetric, psychological, and health variables in pregnancy that are rarely addressed in observational studies on dietary patterns and diet quality, allowing both the evaluation of the crude effect of the variables and the control for confounding in the multivariate model, identifying the variables that best explained ultra-processed foods consumption.

The results suggest that potentially modifiable factors, such as physical activity (besides age, parity, and years of schooling), can guide nutritional guidance actions. The promotion of healthy eating practices even before pregnancy, with accessible guidance, including mainly minimally processed or unprocessed foods in the diet and a significant reduction in ultra-processed foods, will contribute to adequate gestational weight gain, an important indicator of pregnancy progress. During pregnancy, women and their families are more likely to follow guidelines that will benefit both the mother and the fetus. Studies on dietary interventions and encouragement of physical activity during pregnancy have shown a reduction in gestational weight gain and beneficial effects on women’s health ⁴⁶46. International Weight Management in Pregnancy (i-WIP) Collaborative Group. Effect of diet and physical activity based interventions in pregnancy on gestational weight gain and pregnancy outcomes: meta-analysis of individual participant data from randomised trials. BMJ 2017; 358:j3119.. However, realistic policies and actions to control or reduce ultra-processed foods consumption should extended beyond education and information programs in health services. These policies and actions should focus on government programs for the entire society. In various health areas, such as tobacco and alcohol control, the combination of health education, public information campaigns, product labeling, and government guidelines has proven effective in reducing ultra-processed foods consumption, with a positive effect on noncommunicable diseases and population’s health ⁴⁷47. Lustig RH. Ultraprocessed food: addictive, toxic, and ready for regulation. Nutrients 2020; 12:3401..

Some strategies have been implemented aimed at raising awareness of healthy food consumption among the population, such as the publication of dietary guidelines and the application of warning labels. In particular, the Dietary Guidelines for the Brazilian Population²²22. Primary Health Care Department, Secretariat of Health Care, Ministry of Health of Brazil. Dietary guidelines for the Brazilian population. Brasília: Ministry of Health of Brazil; 2015., published in 2014, which introduced the NOVA classification, values the context of food consumption and the sociocultural importance of eating ⁴⁸48. Davies VF, Moubarac JC, Medeiros KJ, Jaime PC. Applying a food processing-based classification system to a food guide: a qualitative analysis of the Brazilian experience. Public Health Nutr 2018; 21:218-29., clearly highlighting the principles and recommendations of adequate and healthy eating. This approach has been increasingly used for the classification of food groups in international studies in adults ³3. Beslay M, Srour B, Méjean C, Allès B, Fiolet B, Debras C, et al. Ultra-processed food intake in association with BMI change and risk of overweight and obesity: a prospective analysis of the French NutriNet-Santé cohort. PLoS Med 2020; 17:e1003256.^,4949. Rauber F, Steele EM, Louzada MLC, Millett C, Monteiro CA, Levy RB. Ultra-processed food consumption and indicators of obesity in the United Kingdom population (2008-2016). PLoS One 2020; 15:e0232676., pregnant women ⁴⁴44. Rohatgi KW, Tinius RA, Cade WT, Steele EM, Cahill AG, Parra DC. Relationships between consumption of ultra-processed foods, gestational weight gain and neonatal outcomes in a sample of US pregnant women. PeerJ 2017; 5:e4091., and official Pan-American Health Organization (PAHO) reports ⁵⁰50. Pan-American Health Organization. Ultra-processed food and drink products in latin america: sales, sources, nutrient profiles, and policy implications. Washington DC: Pan-American Health Organization; 2019..

The use of food warning labels is another important strategy. In Chile, one year after the implementation of this policy, a study showed that its participants (mothers of children aged two to 14 years with different socioeconomic levels) understood that labeling regulation had been implemented to combat childhood obesity in the country, causing changes in the eating habits of the Chilean population ⁵¹51. Correa T, Fierro C, Reyes M, Carpentier FRD, Taillie LS, Corvalan C. Responses to the Chilean law of food labeling and advertising: exploring knowledge, perceptions and behaviors of mothers of young children. Int J Behav Nutr Phys Act 2019; 16:21.. In 2020, following in the footsteps of different Latin American countries, such as Uruguay, Peru, Ecuador, and Bolivia, Brazil passed new legislation on the nutritional labeling of packaged foods, aiming to clarify the nutritional information on food labels and help consumers make more conscious choices ⁵²52. Agência Nacional de Vigilância Sanitária. Resolução de Diretoria Colegiada - RDC nº 429, de 8 de outubro de 2020. Dispõe sobre a rotulagem nutricional dos alimentos embalados. Diário Oficial da União 2020; 9 out.. These standards came into effect in October 2022. Thus, considering the multiple determinants of food practices and the complexity and challenges involved in shaping current food systems, these strategies aim to contribute to the promotion and fulfilment of the human right to adequate food.

Conclusion

Sociodemographic factors are important risk factors associated with ultra-processed foods consumption in the general population, particularly in pregnant women. Moreover, leisure time physical activity before and during pregnancy is a potentially modifiable factor associated with lower consumption of these foods. An unhealthy diet before and during pregnancy can have negative consequences for the health of both mother and fetus. The identification of risk and protective factors allows for the establishment of control measures and the encouragement of healthy practices aimed at the most vulnerable population. However, the greatest benefits come from intervention strategies during prenatal care associated with public policies that reach the entire population.

References

Additional information

Publication Dates

History