Dietary patterns of Brazilian farmers and their relation with sociodemographic, labor, and lifestyle conditions
Nutrition Journal volume 19, Article number: 23 (2020)
The eating habits have changed in the last few decades, but few studies prioritize the food consumption of farmers and the rural population. Therefore, the objective of this study was explore the sociodemographic, occupational and lifestyle factors to the high adherence these dietary patterns.
This is a cross-sectional epidemiological study of 740 farmers (51.5%, n = 381 males; 48.5%, n = 359 females) from a municipality in Southeastern Brazil. Food intake data were obtained by applying multipass 24-h recall and dietary intake was presented in dietary patterns determined by Principal Component Analysis with varimax orthogonal rotation.
Three dietary patterns were identified. The first pattern, “local traditional”, was associated with sociodemographic and labor variables, being considered typical of the region’s farmer as white race/color (p = 0.003), not extra-physical activity (p = 0.014) and cultivating 5 or more crops (p = 0.005). The permanence of a “traditional Brazilian” pattern and the occurrence of an “industrialized” pattern were also observed. Farmers working in non-conventional agriculture were 54% less adhere to “traditional Brazilian” pattern (OR 0.46, 95% CI 0.25–0.86, p = 0.014). Individuals aged 50 and over years were 82% less likely (OR 0.18, 95% CI 0.10–0.30) to adhere to “industrialized” pattern. Still, individuals of lower socioeconomic class were 52% less likely to adhere to this pattern (OR 0.48, 95% CI 0.24–0.96). Farmers who spent R$ 200 or more per capita to buy food were more than twice as likely to adhere to this food pattern (OR 2.22, 95% CI 1.32–3.73), and who had the habit of frequently eating out were 1.62 as likely adhere to “industrialized” pattern (95% CI 1.11–2.36).
The findings indicate changes in dietary patterns in rural areas of the country, maintaining a traditional Brazilian pattern, as well as a local and an industrialized pattern. This last pattern demonstrates that the contemporary rural population also opts for a diet with ultra-processed products, being associated with the characteristic habits of a more urbanized rural region.
Agriculture employs 1.3 billion people worldwide, representing about 40% of the global workforce , reaching 75% of occupations in low-income countries . Agricultural productivity is essential to sustaining the nutritional and health status of billions of people , and small farmers are critical to the global food supply . However, agriculture is a dangerous profession  and with priority of attention , due to its occupational health risks and the presence of non-communicable chronic diseases and their risk factors [7,8,9,10,11].
With regards to dietary pattern, the eating habits of Brazilians have been undergoing changes in recent decades, especially the decrease in the consumption of legumes, roots and tubers and higher relative consumption of meat, milk, sugar and soft drinks and other ultra-processed foods , higher expenses with food away from home  and continued growth in the purchase of ready-to-eat products and significant reduction in the share of food and culinary ingredients . In Brazilian food surveys, some differences were identified for the rural area, presenting, in general, higher prevalence of micronutrient inadequacy , lower consumption of food away from home  and higher consumption of beans [16, 17].
Despite the finding of an increase in the morbidity and mortality profile of this population [7,8,9,10,11], representative research on the food consumption of farmers and the rural population is still scarce , especially studies developed with the purpose of studying the dietary patterns of these populations and their associated factors [4,5,6, 19,20,21,22,23,24,25,26,27,28,29].
In this scenario, the objective of this paper was explore the sociodemographic, occupational and lifestyle factors to the high adherence these dietary patterns.
This is an epidemiological study of cross-sectional and quantitative design developed in the municipality of Santa Maria de Jetibá, located in the highlands of the state of Espírito Santo, Southeastern Brazil. This study integrates, in larger scope, the study “Health condition and associated factors: a study in farmers of Espírito Santo - AgroSaúdES”.
The representative sample of male and female farmers exposed here met the following inclusion criteria: adults 18–59 years old, non-pregnant, who had agriculture as their main source of income and were in full employment for at least 6 months. To identify eligible farmers, we used data available in the records of individuals and families conducted by the Family Health Strategy teams, responsible for covering 100% of the eleven health regions of the municipality.
Seven thousand two hundred eighty-seven farmers were identified from a total of 4018 families. About this population universe, we calculated a minimum sample of 708 farmers, considering a sampling error of 3.5%, a 95% level of significance and prevalence of 50% to maximize sample . In order to reach the minimum sample and considering possible losses, recruitment included 806 individuals.
To define the sample universe one list was built with the survey of the registration of individuals and families by the Community Health Agents, through the data available in the family register used by the Family Health Strategy teams. This register covers 100% of the eleven health regions of the municipality. At the time of the sample (September/2016), 4018 families were enrolled, with 7287 farmers. The participants were selected by stratified draw, proportionality the number of families per health region, in order to respect proportionality among the regions. In families with more than one eligible, only one individual was drawn, avoiding thus the interdependence of information. In cases of refusal of participation or non-attendance in data collection, a new participant in the waiting list of the lottery was called, respecting the sex and region of origin of the dropout.
Data collection took place between December 2016 and April 2017 in the facilities of the municipal health units. A semi-structured questionnaire containing questions about socioeconomic, occupational, lifestyle and food consumption characteristics was applied.
Food consumption analysis
Food intake data were obtained by applying three 24-h recalls (R24h) during the interview, 2 days of the week and 1 day of the weekend within 15 days after the first R24h in the return interviews. In order to obtain greater accuracy of the portions consumed, photo albums were used to facilitate the identification and quantification of the consumed items. Data were collected from 790 farmers, but 27 individuals were excluded since they underwent only one R24h and 23 since they underwent only two (6.3% loss), resulting in a final sample of 740 farmers. As such, the total lied above the minimum sample of 708 farmers (Additional file 1).
The nutritional composition of the R24h was performed using the software AvaNutri 4.1, in which the Brazilian Table of Food Composition  was selected for extraction of nutritional information. Information from the manufacturer or from standard recipes was used for the registration of typical regional foods that were not available at a table and possible dietary supplements were registered. After the registration of the food and acquiring the calories, no exclusion was performed due to extremes in energy consumption .
Then, 355 different food items reported in the R24h were listed in order to identify eating patterns through Principal Component Analysis (PCA). From these items, 65 foods were removed for not constituting the eating habits of the analyzed population (consumption < 5% of the sample) [33, 34]. The remaining foods were allocated to 25 groups (Table 1), according to their nutritional characteristics and Pearson’s correlation between their food items [30, 34].
The applicability of the PCA method was evaluated by the Kaiser-Meyer-Olkin coefficient (KMO) and Bartlett’s test of sphericity (BTS). In this study, the KMO value was 0.609 and the BTS test p-value was p < 0.001, which indicates the adequacy of the data to the factor analysis and recommends its application to the group of farmers studied .
The number of factors retained was defined according to the following criteria: components with eigenvalues greater than 1.0; Cattel chart; and conceptual meaning of the identified patterns. After Cattel graph analysis, three factors were extracted during analysis based on the line’s inflection point on the graph . Factor analysis was subsequently applied to the 25 food groups, selecting varimax rotation to obtain uncorrelated factors .
Foods or food groups whose factor saturation charges were above 0.3 were evaluated as having a strong association with the component, providing better information for the description of a dietary pattern [30, 34]. The patterns were named in accordance with the interpretability and characteristics of the items retained in each pattern, and the items with the highest factor loadings were the ones that most influenced the interpretation and denomination of factors [30, 34, 35].
The independent variables of this study were subdivided into sociodemographic, labor, and lifestyle variables. Among the sociodemographic variables were evaluated sex, age group (“up to 29 years”, “30 to 39 years”, “40 to 49 years” and “50 years or more”), marital status (“single”, “married/living with a partner” and “divorced/separated/widowed”), race/color (“white” and “non-white”), land bond (“owner” and “non-owner”), schooling (“less than 4 years”, “4 to 8 years” and “more than 8 years”), transport used most frequently (“own vehicle” and “on foot, by bicycle or bus”), nearby places for physical activity (“there is no proper place” and “around the house”), and socioeconomic class (“A or B”, “C” and “D or E”) . This classification is used in national studies and estimates socioeconomic classes according to the purchasing power of individuals and families, and also allows estimating the average monthly gross family income (A: approximately R$ 11,037.00; B: approximately R$ 6006.00; C: approximately R$ 1865.00; D/E: approximately R$ 895.00) .
Labor variables were investigated by questioning working time as a farmer (“under 10 years”, “from 10 to 29 years” and “30 years or more”), the current type of production (“conventional” and “non-conventional”), the number of worked crops (“up to 4 crops” and “5 or more crops”), the type of worked crops categorized into “temporary only”, “permanent only” and “temporary and permanent”, according to criteria of the Brazilian Institute of Geography and Statistics , the workload (hours/week) (“less than or equal to 40 hours” and “more than 40 hours”) and contact with pesticides (“direct contact” and “indirect contact, organic or agroecological”) .
Lifestyle variables included alcohol consumption, categorized as “non-drinking” and “drinking”; smoking, assessed according to the Smoker Approach and Treatment Consensus and categorized as “non-smoker” and “current and past smoker”; practice of physical activity extra-field (“yes” or “no”); screen time obtained by the sum of daily activities for television, video game and computer/cell phone, divided by the days of the week, classified as “no sedentary leisure” when < 2 h/day and “with sedentary leisure” when ≥ 2 h/day . Also evaluated were the number of places where they usually buy food (“2 places or less” and “3 places or more”), the frequency of food purchases (“twice/month or more” and “once/month or less”), travel time to purchase food (“up to 15 minutes”, “16 to 29 minutes” and “more than 30 minutes”), monthly per capita expenditure on food purchases (“R$ 100 or less”, “> R$ 100 to < R$ 200” and “R$ 200 or more”), the habit of eating away from home (“no or rarely” and “yes, often”) and the place where they usually meal (“at a table” and “under a different setting”).
The normality of the variables was assessed by the Shapiro-Wilk test. To describe the study variables, the median (50p) was used as a measure of central tendency, and the interquartile range (IQR) as a dispersion measure for continuous variables, and absolute and percentage values for categorical variables. Regarding the association tests between the independent variables and the outcome for the qualitative variables, Pearson’s Chi-square test was used. When the expected values in the table cells were less than five or when the sum of the column value was less than twenty, Fisher’s exact test was used. To analyze the association between a quantitative and a qualitative variable, due to the abnormality of the variables, the Mann-Whitney U test was used. When the qualitative variable had three or more categories, the Kruskal-Wallis test and the Mann-Whitney U test were performed two by two to identify the differences. Missing data were maintained due to low data loss, different number of individuals in each variable were reported in the table captions.
The binary logistic regression model was applied to assess the association between independent variables and adherence to dietary patterns (adherence below the median versus adherence above the median). Variables that were statistically significant with dietary patterns of up to 20% in the association analyses were tested in multiple models and adjusted for sex. We used the backward variable selection method with likelihood ratio test, adopting the model with the highest adjustment according to the Hosmer-Lemeshow test (p > 0.05, closer to 1.0). We also tested the assumptions of absence of multicollinearity (tolerance > 0.1 and variance inflation factor < 10), minimum sample size for the number of model variables (> 20 individuals per model variable and > 5 cases in each category of variables) and absence of outliers (absence of standardized residues > ± 3 standard deviations; up to 1% of standardized residues between ±2.5 and 3 standard deviations; and up to 5% of standardized residues between ±2.0 and 2.5 standard deviations, Cook’s distance < 1, and DFBeta < 1).
For all analyses, the level of significance adopted was α < 5% and these were performed using the statistical software IBM SPSS Statistics for Windows, version 22.0 (Armonk, NY: IBM Corp).
The study was approved by the Research Ethics Committee of the Health Sciences Center of the Federal University of Espírito Santo (Ufes), under number 1,856,331 (CAAE 52839116.3.0000.5060), and followed the precepts of the Declaration of Helsinki. All respondents signed the Informed Consent Form.
General characteristics of study population
Most of the evaluated farmers were married or living with a partner, were in socioeconomic class C, had low schooling (67.7% with less than 4 years of schooling), 66.4% (n = 491) reported not having adequate place for physical activity and 93.5% (n = 690) used their own vehicle as means of transportation (Table 2). Almost 80% (78.1%, n = 578) owned their land and 90% (n = 666) worked in conventional agriculture. Half of the farmers (50.8%, n = 375) worked from 10 to 29 years in the field, 56.8% (n = 420) cultivated five or more crops (larger in men, p = 0.004), worked in temporary crop (43.2%, n = 320) and had a high workload (80.1% with more than 40 working hours per week, higher in men with p < 0.001). Almost 70% of farmers had direct contact with pesticides, this contact being greater in men (p < 0.001).
Regarding lifestyle (Table 2), 43.4% (n = 321) reported consuming alcohol and 15% (n = 111) reported being a current or former smoker, both more prevalent in males (p < 0.001). More than 80% didn’t practice any physical activity extra-field and 45.3% (n = 335) presented sedentary leisure assessed by screen time. Regarding eating habits, 54.5% (n = 403) purchased food from two or less different locations and purchased low-frequency food (72% once/month or less, n = 533). The amount spent per capita on food purchases in almost half of the farmers (44.9%, n = 319) ranged from over R$ 100 up to R$ 200 per month. The habit of eating out was frequently reported by 33.2% of the individuals (n = 246), with this practice being more common among men (p < 0.001) and eating at a table was present in 74.1% (n = 548) of farmers.
Dietary sources and nutritional characteristics
The most consumed items by this population were rice and coffee, beans, poultry meat, sugar, butter and margarine, homemade bread, oils, pasta, tomato, potato, and others typically foods (Table 3). Tomato was the only vegetable consumed in more than half of the population (64.9%) and 88.6% of the farmers consume at least some vegetables. The other vegetables consumed were green condiment, lettuce, cabbage, cucumber, carrot, and others. Fruit consumption was even lower, only 48.9% of farmers consume some kind of fruit. The most consumed fruit was banana, followed by lemon, apple, guava, mango, grape, watermelon and peach.
After rotational factor analysis, three dietary patterns were obtained (Table 4), namely: “pattern 1 – local traditional”: sugar; coffee; butter and margarine; homemade bread, brote, cakes and cookies; juice and sugary beverages; potatoes, yams and cassava; and pasta; “pattern 2 – traditional Brazilian”: beans; rice; vegetables (tomato, green condiment, lettuce, cabbage, cucumber, carrot, chayote, pepper, okra, green leafy vegetables, scarlet eggplant, pumpkin and pod); flour (flour and farofa); and oils and fats; and “pattern 3 – industrialized”: soda; snacks, fried food, hamburger, hot dog, garlic bread and trooper’s beans; red meat (beef and pork); sausage, canned food, industrialized condiment and sauce; alcoholic drinks (distilled beverages, bear and wine); and industrialized breads, cookies, toasts and threads. The component “homemade breads, brote, cakes and cookies” presented high negative factor load in the “industrialized” group, which demonstrates that individuals of this first dietary pattern have very low consumption of this type of food.
The total variance explained by the factors was 23.8%. Foods with a low factorial load in one component were considered to be of low correlation and did not saturated in the any dietary pattern, which makes it possible to consider them as foods of homogeneous consumption among individuals. They are: white meat (poutries and fishes), candies (chocolate, pies, jams, ice cream and other sweets), eggs, soups and broths, fruits (banana, lemon, apple, guava, mango, grape, watermelon and peach), milk, cheese and yogurt and polenta.
Factors effect on farmers’ dietary pattern
Associated with the greater adherence to the “traditional local” pattern was the male sex, the age group from 30 to 39 years old, the race/white color, having 4 to 8 years of study, transport using their own vehicle, owning the land, working with 5 or more crops and more than 40 h per week, having direct contact with pesticides, consumption of alcoholic beverages and eating out frequently (Table 5). Lower adherence to this first pattern was associated with being separated, divorced or widowed and working only with permanent crops. Regarding the greater adherence to the “traditional Brazilian” pattern, men were associated with transport using their own vehicle, having direct contact with pesticides, consuming alcoholic beverages and being a current or former smoker. In addition, it was associated with greater adherence to the “industrialized” pattern to be male, aged up to 29 years old, single and non-white, socioeconomic class A or B, over 8 years old of schooling, have a place to practice physical activity in the residence surroundings, transport using their own vehicle, working only with temporary crops, having direct contact with pesticides, drink alcohol, be a current or former smoker, have sedentary leisure, buy food in 3 places or more, spend R$ 200 or more (per capita/month) on food, eating often outside, and eating away from a table.
From the found associations, multiple analyses were performed for each dietary pattern (Tables 6, 7 and 8). Thus, the variables "age group, race/color, number of worked crops and physical activity extra-field were associated with the “local traditional” pattern (Table 6). Individuals aged 50 years and older were 56% less likely to adhere to this dietary pattern than those up to 29 years old (OR 0.44, 95% CI 0.25–0.74, p = 0.003). Farmers of non-white race/color were 58% less likely to be more adherent to this dietary pattern (OR 0.42, 95% CI 0.23–0.74, p = 0.003) and those who practiced extra-physical activity, 47% less likely to be more adherent to the “local traditional” pattern (OR 0.46, 95% CI 0.25–0.86, p = 0.014). However, workers who cultivated 5 or more crops were 1.59 times more likely to adhere to this pattern (OR 1.59, 95% CI 1.15–2.19, p = 0.005).
Regarding the “traditional Brazilian” pattern, farmers working in non-conventional agriculture were 54% less likely to adhere more to this dietary pattern (OR 0.46, 95% CI 0.25–0.86, p = 0.014) (Table 7).
For the third pattern, it was identified that the higher the age group, the lower the chances of farmers adhering to the “industrialized” pattern (Table 8). Individuals aged 30 to 39 years were 44% less likely (OR 0.56, 95% CI 0.36–0.87, p = 0.009), those aged 40 to 49 years were 67% less likely (OR 0.33, 95% CI 0.21–0.53, p < 0.001) and those aged 50 and over, 82% less likely (OR 0.18, 95% CI 0.10–0.30, p < 0.001) to adhere to this pattern than those aged 30 and under. Still, individuals of socioeconomic class D or E were 52% less likely to adhere to this pattern (OR 0.48, 95% CI 0.24–0.96, p = 0.037), whereas those who worked with temporary and permanent tillage had 1.57 times more likely to adhere to the “industrialized” pattern (OR 1.57, 95% CI 1.21–2.41, p = 0.002) than those working only on temporary tillage. Still, farmers who spent R$ 200 or more per capita to buy food were more than twice as likely to adhere to this food pattern (OR 2.22, 95% CI 1.32–3.73, p = 0.003). Likewise, those who had the habit of frequently eating out were 1.62 times more likely to adhere to this pattern (OR 1.62, 95% CI 1.11–2.36, p = 0.012) and those without the habit of eating at a table, 1.56 times more likely to adhere more to the “industrialized” pattern (OR 1.56, 95% CI 1.05–2.31, p = 0.028).
Three dietary patterns were identified in the sample of farmers studied: “local traditional”, “traditional Brazilian” and “industrialized”, which shows the eating habits of this group of workers. Such patterns represent the eating habits of this group, with the “local traditional” pattern being typical of the region’s farmer, associated with the white race/color, not within the extreme age group, working with many crops and lack of exercise extra-field. In addition, it was evident that the “traditional Brazilian” pattern, with beans, rice and vegetables, remains in the eating habits of this population, and this pattern is associated with the conventional type of production. An “industrialized” pattern was also identified, which demonstrates that the contemporary rural farmer also opts for a diet with processed products. This pattern was associated with a lower age range; the highest socioeconomic class; productivity with temporary and permanent crops; the largest number of locations and the highest monthly expenditure on food purchases; eating away from home and eating away from a table, typical of a more urbanized rural setting. A low consumption of fruits was also found for this population.
Determining food patterns in this population is an unprecedented finding, because of the method used to determine dietary patterns, highly recommended for approaching the actual behavior of the population studied [40, 41], and for extrapolated the combination of nutrients and antinutritional factors involved in the human diet . In addition, can meet the scarcity of studies that investigate the food consumption of farmers or rural populations and their associated factors [15,16,17,18,19,20,21,22,23,24,25,26,27,28,29], especially regarding their labor characteristics .
As in the present study, Andrade et al. (2018) , in a survey of 34,003 individuals from the 2008/2009 Household Budget Survey (POF), showed three dietary patterns in the Brazilian population, with traditional meals, “typical Brazilian breakfast/tea” and the last one with “ultra-processed foods”. Nogueira et al. (2019) , in a study with metallurgists from Fortaleza/CE, Brazil, also identified three eating patterns: the first common northeastern, the second popular general and the third western. These patterns shows similarities with respect to the consumption of a traditional local standard, a traditional Brazilian and an industrialized one.
Other studies on the determination of dietary patterns in working classes have also shown consistency with the patterns found in farmers, with appropriate interpretations in their working environments, such as “vegetables, fruits, cereals and tubers”, “sweets and snacks” and “traditional and protein” in banking . Likewise, the “traditional”, “fruit and vegetables”, “pastry” and “diet/light” patterns in civil servants  and the “healthy”, “western” and “traditional” in teachers .
Typical “breakfast” or “snack” foods stood out in the “traditional local” pattern, due to the continued habit of some families producing their own foods such as breads, cookies, pies and home-made cakes . This first pattern elucidates the eating habits of the region’s typical white-colored farmer, colonist (Pomeranian peasant immigrant), who still produces many of their own foods  and doesn’t use their time in physical activities beyond those already arising from work in the field, in particular by working with many crops for olericulture .
In addition to the snack pattern, the “traditional Brazilian pattern” was also identified. This patter represents the typical Brazilian diet, reinforced by the high frequency of consumption of rice, coffee and beans in this population. This indicates that, as in other population studies in rural areas, typical foods remain in the eating habits of these individuals [16, 21, 23]. In Brazil, three quarters of the population regularly consume beans (71.9, 95% CI 71.2–72.6%), and the state of Espírito Santo is the place with the highest consumption of this food (86.5, 95% CI 84.4–88.6), which is even higher in the rural area (76.3, 95% CI 74.8–77.9) (6).
Unconventional production was less likely to be more adherent to the “traditional Brazilian” pattern. Conventional farmers represented the vast majority of farmers surveyed (90% of the sample, n = 666), which reflects the type of crop in the municipality . Possibly, farmers who work with organic and agroecological farmers have different habits, determined by different work logistics, income and access to food .
It is also worth noting that fruit consumption was reduced among farmers (48.9% consume some kind of fruit). In Brazil, between 1987/1988 and 2008/2009, there was some stability and low levels of fruit and vegetable acquisition, including a small reduction in the consumption of vegetables , with lower consumption in the rural area of these two food groups [17, 19, 28].
Fruit and vegetable consumption was expected to be higher in rural areas, given the possibility of access to land and the cultivation of these products. However, the food produced by these farmers can be understood as commodities and intended for sale and income, and not perceived as self-consumption products . It should be noted that, on average, subsistence production accounts for 58% of the caloric consumption of rural households, and thus 42% of the calories consumed from purchased food .
In accordance with this fact, the third dietary pattern detected was the “industrialized” pattern, which shows that, although farmers have a still typically local and Brazilian dietary habit, factors associated with globalization are affecting their eating habits. This, which, among other consequences, makes the contemporary rural farmer also opt for a diet with processed and ultraprocessed products [9, 21, 23, 29, 49].
Regarding the factors associated with the “industrialized” pattern, it was possible to identify an epidemiological gradient regarding age. It is worth mentioning that the younger the individual, the greater the chances of adherence to this pattern. This may show that changes in dietary patterns are occurring mainly in the younger population, as seen in rural Pelotas/RS/Brazil  and in Brazil as a whole .
In the present study, farmers from lower economic classes were found to be less likely to adhere to the “industrialized” pattern, possibly because the average value of ultra-processed foods in Brazil is higher than that of other foods , in contrast to what is observed in developed countries .
Farmers who work with temporary and permanent crops were also associated with higher adherence to this food pattern, likely due to the income proxy that this variable may present, since larger areas of land are needed for the planting of some permanent crops, such as coffee, one of the most important economic activities of the municipality .
Other habits of a more urbanized rural pattern could also be identified as associated with greater adherence to the third dietary pattern. The purchase of food in a variety of places, such as local shops and supermarkets, may be responsible for diversifying the purchase and consumption of products that deviate from the pattern of fresh foods produced on many farms [23, 47, 48]. In addition, higher monthly expenses on food purchases may also lead to more access to food away from home, since with the 10% increase in household income, for example, there is already a 3% increase in the share of food consumed away from home .
Finally, not eating at a table increased the chances of greater adherence to the industrialized pattern. It is important to remember that the family institution influences the eating habits of its members, since the practice of eating together has a significant role in learning healthy eating practices. Therefore, meals taken away from a table and away from the family unit may promote the highest consumption of ultra-processed foods .
Possible limitations of this study include its cross-sectional nature, which requires greater caution in interpreting the results, due to the possibility of reverse causality. In addition, the factorial analysis employed in the derivation of dietary patterns involves some subjectivity in its decision-making. However, such limitations are mitigated by detailing all deliberations . It is noteworthy that the dietary patterns detected in this study were comparable with those of other studies, which validates the results obtained externally [27, 30, 34, 41,42,43].
Despite the study being conducted in only one Brazilian municipality, the results and discussion may indicate changes in dietary patterns in rural areas of the country as a global trend. The supporting to implement and strengthen public health surveillance systems in food consumption and education for an adequate and healthy diet could be impact a large portion of the economically active population, representing by agricultural employs. Thus, measures that increase cultivation, consumption and access to traditional foods and as well as encouraging the maintenance of culinary skills, preserving the local food heritage, are considered of utmost importance and should be encouraged.
Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
- 95% CI:
Confidence interval of 95%
Bartlett’s test of sphericity
Principal Component Analysis
2008/2009 Household Budget Survey
Traitler H, Dubois M, Heikes K, Pétiard V, Zilberman D. Megatrends in Food and Agriculture: Technology, Water Use and Nutrition. 1st ed. New Jersey: Wiley Online Library; 2018. p. 336. https://doi.org/10.1002/9781119391173.
Lowder SK, Skoet J, Raney T. The number, size, and distribution of farms, smallholder farms, and family farms worldwide. World Dev. 2016;87:16–29. https://doi.org/10.1016/j.worlddev.2015.10.041.
Hawkes C, Ruel MT. From agriculture to nutrition: pathways, synergies and outcomes. Agric Rural Dev Notes. 2008;40:1–4.
Fanzo J. The role of farming and rural development as central to our diets. Physiol Behav. 2018;193(B):291–7. https://doi.org/10.1016/j.physbeh.2018.05.014.
U.S. Bureau of Labor Statistics. Fatal occupational injuries in 2016 (charts). 2017. https://www.bls.gov/iif/oshwc/cfoi/cfch0015.pdf. Accessed 29 Dec 2018.
Brasil. Ministério da Saúde. Portaria n° 1.823, de 23 de agosto de 2012. Institui a Política Nacional de Saúde do Trabalhador e da Trabalhadora [Ordinance No. 1,823, of August 23, 2012. Establishes the National Policy for worker health]. Brasília: Diário Oficial da União; 2012. http://bvsms.saude.gov.br/bvs/saudelegis/gm/2012/prt1823_23_08_2012.html. Accessed 04 Jun 2018.
Pickett W, King N, Lawson J, Dosman JA, Trask C, Brison RJ, Hagel L, Janssen I. Saskatchewan farm injury cohort study team. Farmers, mechanized work, and links to obesity. Prev Med. 2015;70:59–63. https://doi.org/10.1016/j.ypmed.2014.11.012.
Torske MO, Bjørngaard JH, Hilt B, Glasscock D, Krokstad S. Farmers' mental health: a longitudinal sibling comparison – the HUNT study. Norway Scand J Work Environ Health. 2016;42(6):547–56. https://doi.org/10.5271/sjweh.3595.
Kapelios CJ, Kyriazis I, Ioannidis I, Dimosthenopoulos C, Hatziagelaki E, Liatis S, PERSEAS Study Group. Diet, life-style and cardiovascular morbidity in the rural, free living population of Elafonisos island. BMC Public Health. 2017;17(1):147. https://doi.org/10.1186/s12889-017-4053-x.
Martins-Silva T, De Mola CL, Vaz JS, Tovo-Rodrigues L. Obesidade geral e abdominal em adultos residentes em zona rural no Sul do Brasil. Rev Saúde Pública. 2018;52:7s. https://doi.org/10.11606/S1518-8787.2018052000264.
Park S, Kim S-K, Kim J-Y, Lee K, Choi JR, Chang S-J, Chung CH, Park KS, Oh SS, Koh SB. Exposure to pesticides and the prevalence of diabetes in a rural population in Korea. Neurotoxicology. 2019;70:12–8. https://doi.org/10.1016/j.neuro.2018.10.007.
Monteiro CA, Levy RB, Claro RM, Castro IRR, Cannon G. A new classification of foods based on the extent and purpose of their processing. Cad Saúde Pública. 2010;26(11):2039–49. https://doi.org/10.1590/S0102-311X2010001100005.
Claro RM, Levy RB, Bandoni DH. Influence of income on food expenditures away from home among Brazilian families, 2002-2003. Cad Saude Publica. 2009;25(11):2489–96. https://doi.org/10.1590/S0102-311X2009001100018.
Martins APB, Levy RB, Claro RM, Moubarac JC, Monteiro CA. Increased contribution of ultra-processed food products in the Brazilian diet (1987-2009). Rev Saúde Pública. 2013;47(4):1–10. https://doi.org/10.1590/S0034-8910.2013047004968.
Araújo MC, Bezerra IN, Barbosa FS, Junger WL, Yokoo EM, Pereira RA, Sichieri R. Macronutrient consumption and inadequate micronutrient intake in adults. Rev Saúde Pública. 2013;47(Suppl 1):177s–89s. https://doi.org/10.1590/S0034-89102013000700004.
Bezerra IN, Souza AM, Pereira RA, Sichieri R. Consumption of foods away from home in Brazil. Rev Saude Publica. 2013;47(Suppl 1):200s–11s. https://doi.org/10.1590/S0034-89102013000700006.
Jaime PC, Stopa SR, Oliveira TP, Vieira ML, Szwarcwald CL, Malta DC. Prevalence and sociodemographic distribution of healthy eating markers, National Health Survey, Brazil 2013. Epidemiol Serv Saúde. 2015;24(2):267–76. https://doi.org/10.5123/S1679-49742015000200009.
Santos JC, Ristow MR. Suicides: social fact and developmentalism on the basis of attacks against life. Emancipação. 2010;10(2):563–76. https://doi.org/10.5212/Emancipacao.v.10i2.563576.
Dean WR, Sharkey JR. Rural and urban differences in the associations between characteristics of the community food environment and fruit and vegetables intake. J Nutr Educ Behav. 2011;43(6):426–33. https://doi.org/10.1016/j.jneb.2010.07.001.
Carvalho EO, Rocha EF. Consumption to feed of resident adult population in rural area of the city of Ibatiba (ES, Brazil). Cien Saude Colet. 2011;16(1):179–85. https://doi.org/10.1590/S1413-81232011000100021.
Reinaldo EDF, Silva MRF, Nardoto GB, Garavello MEPE. Changes in eating habits in rural communities of the semi-arid, region of northeastern of Brazil. Interciencia. 2015;40(5):330–6.
Claro RM, Santos MAS, Oliveira TP, Pereira CA, Szwarcwald CL, Malta DC. Unhealthy food consumption related to chronic non-communicable diseases in Brazil: National Health Survey, 2013. Epidemiol Serv Saúde. 2015;24(2):257–65. https://doi.org/10.5123/S1679-4974201500020000.
Rodrigues LP, Carvalho RC, Maciel A, Otanasio PN, Garavello ME, Nardoto GB. Food insecurity in urban and rural areas in Central Brazil: transition from locally produced foods to processed items. Ecol Food Nutr. 2016;55(4):365–77. https://doi.org/10.1080/03670244.2016.1188090.
Claro RM, Maia EG, Costa BVL, Diniz DP. Food prices in Brazil: prefer cooking to ultra-processed foods. Cad Saúde Pública. 2016;32(8):e00104715. https://doi.org/10.1590/0102-311X00104715.
Tripathy JP, Thakur JS, Jeet G, Chawla S, Jain S, Prasad R. Urban rural differences in diet, physical activity and obesity in India: are we witnessing the great Indian equalisation? Results from a cross-sectional STEPS survey. BMC Public Health. 2016;16(1):816. https://doi.org/10.1186/s12889-016-3489-8.
Machado PP, Claro RM, Martins APB, Costa JC, Levy RB. Is food store type associated with the consumption of ultra-processed food and drink products in Brazil? Public Health Nutr. 2018;21(1):201–9. https://doi.org/10.1017/S1368980017001410.
Andrade GC, Louzada MLC, Azeredo CM, Ricardo CZ, Martins APB, Levy RB. Out-of-home food consumers in Brazil: what do they eat? Nutrients. 2018;10:2018. https://doi.org/10.3390/nu10020218.
Canella DS, Louzada MLC, Claro RM, Costa JC, Bandoni DH, Levy RB, Martins APB. Consumo de hortaliças e sua relação com os alimentos ultraprocessados no Brasil [Vegetable consumption and its relationship with ultra-processed foods in Brazil]. Rev Saude Publica. 2018;52:50. https://doi.org/10.11606/S1518-8787.2018052000111.
Fernandes MP, Bielemann RM, Fassa AG. Factors associated with the quality of the diet of residents of a rural area in Southern Brazil. Rev Saude Publica. 2018;52(Suppl 1):6s. https://doi.org/10.11606/s1518-8787.201805200026.
Cattafesta M, Zandonade E, Bissoli NS, Salaroli LB. Eating patterns of bank workers and their association with socioeconomic, behavioral and labor factors Cien. Saude Colet. 2019;24:3909–22. https://doi.org/10.1590/1413-812320182410.31342017.
Núcleo de Estudos e Pesquisas em Alimentação (NEPA)/Universidade Estadual de Campinas (UNICAMP). Tabela Brasileira de Composição de Alimentos (TACO) [Brazilian Table of Food Composition (TACO)]. 4th ed. Campinas: NEPA/UNICAMP; 2011. p. 161.
Willett W. Issues in Analysis and Presentation of Dietary Data. In: Nutritional Epidemiology, vol. 40. 3rd ed. New York: Oxford University Press; 2013. p. 305–32.
Selem SS, Castro MA, César CL, Marchioni DM, Fisberg RM. Associations between dietary patterns and self- reported hypertension among Brazilian adults: a cross-sectional population-based study. J Acad Nutr Diet. 2014;114:1216–22. https://doi.org/10.1016/j.jand.2014.01.007.
Nogueira VC, Arruda SPM, Sampaio HAC, Rodrigues BC, Silva EB, Farias BO, Sabóia KM. Socioeconomic, demographic and lifestyle factors associated with eating patterns of shift workers. Cien Saude Colet. 2019;24(3):761–9. https://doi.org/10.1590/1413-81232018243.03362017.
Olinto MTA. Padrões alimentares: análise de componentes principais. In: Kac G, Sichieri R, Gigante DP, editors. Epidemiologia nutricional [Nutritional Epidemiology]. 1st ed. Rio de Janeiro: Atheneu; 2007. p. 213–25.
Associação Brasileira de Empresas de Pesquisa (ABEP). Critério de Classificação Econômica Brasil [Brazil Economic Classification Criteria]. 1st ed. São Paulo: ABEP; 2014. p. 6.
Instituto Brasileiro de Geografia e Estatística (IBGE). Censo agropecuário 2006 [Agricultural Census 2006]. Rio de Janeiro: IBGE; 2006. https://biblioteca.ibge.gov.br/visualizacao/periodicos/51/agro_2006.pdf. Accessed 29 Dec 2018.
Brasil. Ministério do Trabalho e Emprego. Portaria MTE n.° 86, de 03 de março de 2005. NR 31 - Segurança e Saúde no Trabalho na Agricultura, Pecuária, Silvicultura, Exploração Florestal e Aquicultura [MTE Ordinance No. 86, of March 03, 2005. NR 31 - occupational safety and health in agriculture, livestock, forestry, forestry and aquaculture]. Diário Oficial da União: Brasília; 2005. http://trabalho.gov.br/images/Documentos/SST/NR/NR31.pdf. Accessed 04 Jun 2018.
Sichieri R, Souza RA. Strategies for obesity prevention in children and adolescents. Cad Saúde Pública. 2008;24(Suppl 2):209–23. https://doi.org/10.1590/S0102-311X2008001400002.
Ocké MC. Evaluation of methodologies for assessing the overall diet: dietary quality scores and dietary pattern analysis. Proc Nutr Soc. 2013;72:191–9. https://doi.org/10.1017/S0029665113000013.
Massarani FA, Cunha DB, Muraro AP, Souza BS, Sichieri R, Yokoo EM. Familial aggregation and dietary patterns in the Brazilian population. Cad Saúde Pública. 2015;31(12):2535–45. https://doi.org/10.1590/0102-311X00090314.
Cardoso LO, Carvalho MS, Cruz OG, Melere C, Luft VC, Molina MC, Faria CP, Benseñor IM, Matos SM, Fonseca MJ, Griep RH, Chor D. Eating patterns in the Brazilian longitudinal study of adult health (ELSA-Brasil): an exploratory analysis. Cad Saúde Pública. 2016;32(5):e00066215. https://doi.org/10.1590/0102-311X00066215.
Esmaillzadeh A, Kimiagar M, Mehrabi Y, Azadbakht L, Hu FB, Willett WC. Dietary patterns, insulin resistance, and prevalence of the metabolic syndrome in women. Am J Clin Nutr. 2007;85:910–8. https://doi.org/10.1093/ajcn/85.3.910.
Schimidt A, Farias RCP. A comida e a sociabilidade na cultura pomerana [Food and sociability in Pomeranian culture]. Tessituras. 2015;3(2):195–218. https://doi.org/10.15210/tes.v3i2.5928.
Prefeitura Municipal de Santa Maria de Jetibá (PMSMJ). Dados gerais do município [General data of the municipality]. 2015. http://www.pmsmj.es.gov.br/portal/index.php/o-municipio/. Accessed 18 May 2017.
Oliveira KSC, Rozendo C. Food and nutritional security of the farmers association of organic producers of Ceara Mirim-RN. Agroalimentaria (Caracas). 2016;22(23):151–64.
Triches R, Schneider S. Feeding, Agrifood System, and Consumers: New Connections for Rural Development. Cuadernos de Desarrollo Rural. 2015;12(75):55–75. https://doi.org/10.11144/Javeriana.cdr12-75.asac.
Sibhatu KT, Qaim M. Rural food security, subsistence agriculture, and seasonality. Rural food security, subsistence agriculture, and seasonality. PLoS One. 2017;12(10):e0186406. https://doi.org/10.1371/journal.pone.0186406.
Lima RS, Ferreira Neto JÁ, Farias RCP. Dinâmicas alimentares na relação rural urbano: o caminho entre o tradicional e o moderno [Food dynamics in the urban rural relationship: the path between traditional and modern]. In Estudos socioculturais em alimentação e saúde: saberes em rede [Sociocultural studies in food and health: networked knowledge]. 1st ed. Prado SD, Amparo-Santos L, Silva LF, Arnaiz MG, Bosi MLM. EDUERJ, Sabor metrópole series: Rio de Janeiro; 2016;(5):59–79.
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. https://doi.org/10.1111/obr.12107.
Sichieri R, Casto JF, Moura AS. Factors associated with dietary patterns in the urban Brazilian population. Cad Saúde Pública. 2003;19(Suppl 1):s47–53. https://doi.org/10.1590/S0102-311X2003000700006.
To the Research and Innovation Support Foundation of Espírito Santo (FAPES) and the Research Program for SUS (PPSUS) for the financial support.
This research was funded by Research and Innovation Support Foundation of Espírito Santo (FAPES), grant number FAPES/CNPq/Decit - SCTIE-MS/SESA n° 05/2015 – PPSUS (Research Program for Brazilian Health Unic System).
Ethics approval and consent to participate
The study was approved by the Research Ethics Committee of the Health Sciences Center of the Federal University of Espírito Santo (Ufes), under number 1,856,331 (CAAE 52839116.3.0000.5060), and followed the precepts of the Declaration of Helsinki. All respondents signed the Informed Consent Form.
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Cattafesta, M., Petarli, G.B., da Luz, T.C. et al. Dietary patterns of Brazilian farmers and their relation with sociodemographic, labor, and lifestyle conditions. Nutr J 19, 23 (2020). https://doi.org/10.1186/s12937-020-00542-y