- Split View
-
Views
-
Cite
Cite
Raquel de Deus Mendonça, Aline Cristine Souza Lopes, Adriano Marçal Pimenta, Alfredo Gea, Miguel Angel Martinez-Gonzalez, Maira Bes-Rastrollo, Ultra-Processed Food Consumption and the Incidence of Hypertension in a Mediterranean Cohort: The Seguimiento Universidad de Navarra Project, American Journal of Hypertension, Volume 30, Issue 4, 1 April 2017, Pages 358–366, https://doi.org/10.1093/ajh/hpw137
- Share Icon Share
Abstract
Some available evidence suggests that high consumption of ultra-processed foods (UPFs) is associated with a higher risk of obesity. Collectively, this association and the nutritional characteristics of UPFs suggest that UPFs might also be associated with hypertension.
We prospectively evaluated the relationship between UPF consumption and the risk of hypertension in a prospective Spanish cohort, the Seguimiento Universidad de Navarra project. We included 14,790 Spanish adult university graduates who were initially free of hypertension at baseline who were followed for a mean of 9.1 years (SD, 3.9 years; total person-years: 134,784). UPF (industrial formulations of chemical compounds which, beyond substances of common culinary use such as salt, sugar, oils, and fats, include substances also derived from foods but not used in culinary preparations) consumption was assessed using a validated semi-quantitative 136-item food-frequency questionnaire. Cox proportional hazards models were used to estimate adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) for hypertension incidence.
During follow-up, 1,702 incident cases of hypertension were identified. Participants in the highest tertile of UPF consumption had a higher risk of developing hypertension (adjusted HR, 1.21; 95% CI, 1.06, 1.37; P for trend = 0.004) than those in the lowest tertile after adjusting for potential confounders.
In this large prospective cohort of Spanish middle-aged adult university graduates, a positive association between UPF consumption and hypertension risk was observed. Additional longitudinal studies are needed to confirm our results.
Hypertension accounts for approximately 10.4 million deaths, 208.1 million disability-adjusted life-years, and 7% of the disease burden worldwide.1,2 Moreover, hypertension is a risk factor for cardiovascular disease and responsible for at least 45% and 51% of deaths due to heart disease and stroke, respectively.3 The prevalence of hypertension was approximately 22% worldwide in 2014 and 25% in the European region.1
Modifiable risk factors for hypertension include an unhealthy diet (consumption of foods containing excess salt and saturated fat and insufficient fruit and vegetables intake), harmful alcohol use, lack of physical activity, and excess weight.4 In several countries, ultra-processed foods (UPFs) are common sources of salt. According to Monteiro et al. UPFs are defined as drink and food products which, beyond substances of common culinary use such as salt, sugar, oils, and fats, include substances also derived from foods but not used in culinary preparation and are ready to eat, drink, or heat.5–7 They have high amounts of salt, total fat, saturated fat, and trans fat, free sugar, and high energy density, and low fiber and micronutrients content.8–11
Consumption of UPFs has been associated with higher risks of overweight/obesity,12 metabolic syndrome in adolescents,13 and increased total cholesterol and low-density lipoprotein cholesterol levels in children.14 Collectively, the nutritional characteristics of UPFs and the association between UPF and overweight/obesity likely support an association between UPF consumption and hypertension. Large prospective studies involving an assessment of UPF consumption in the development of hypertension are lacking; however, foods that can be classified as UPFs have been assessed in previous studies.15–18 For example, sugar-sweetened beverage consumption increases the risk of hypertension according to meta-analyses of prospective cohort studies.15,16 Furthermore, processed red meat consumption was associated with hypertension in a prospective cohort study in French women17 and in the Coronary Artery Risk Development in Young Adults (CARDIA) study.18
Therefore, the aim of this study was to evaluate the potential association between UPF consumption and hypertension risk in a Spanish cohort during a long follow-up period.
METHODS
Study population
The Seguimiento Universidad de Navarra (SUN) project is a dynamic and multipurpose prospective cohort study that has been conducted in Spain among university graduates since December 1999 to assess the associations between diet and the occurrence of several diseases and chronic conditions, including hypertension. The recruitment of participants is permanently open, and participants are followed-up biennially using questionnaires distributed by post or electronic mail. Details of the design, recruitment strategy, and methods of the SUN project have been published elsewhere.19,20
As of March 2013, the dataset of the SUN project included 21,678 participants who had answered the baseline questionnaire. In the present study, we excluded participants with prevalent hypertension (self-reported medical diagnosis of hypertension, antihypertensive medication use, or self-reported systolic blood pressure ≥140 mm Hg and/or diastolic blood pressure ≥90 mm Hg)21 at baseline (n = 2,378), who reported total energy intake values outside of predefined limits (low: <3,347 kJ/d or <800 kcal/d in men and <2,092 kJ/d or <500 kcal/d in women; high: >16,736 kJ/d or >4,000 kcal/d in men and >14,644 kJ/d or >3,500 kcal/d in women; n = 1,826),22 who reported a previously diagnosed chronic disease at baseline (diabetes, cancer, or cardiovascular disease; n = 1,310), or who were lost to follow-up (n = 1,374); 90.7% of the participants were retained in the study. Finally, 14,790 participants were included in the final analyses (Figure 1).
Ethics
This study was conducted according to the Declaration of Helsinki guidelines, and all procedures involving human subjects were approved by the Institutional Review Board of the University of Navarra. Voluntary completion of the baseline self-administrated questionnaire was considered to imply informed consent.
Ascertainment of hypertension
In both the baseline and follow-up questionnaires, participants were asked whether they had received a medical diagnosis of hypertension. Additionally, the baseline questionnaire inquired about the most recent systolic and diastolic blood pressures. The follow-up questionnaire also requested the date of hypertension diagnosis.
A participant was considered to have hypertension at baseline if he/she reported a medical diagnosis of hypertension, was taking an antihypertensive medication, or reported systolic and/or diastolic blood pressures ≥140 mm Hg and/or ≥90 mm Hg, respectively.23 New cases of hypertension were defined as individuals reporting a physician-based diagnosis of hypertension in the follow-up questionnaire who did not have hypertension at baseline.
Adequate validation of the self-reported hypertension diagnosis was observed in a specific study within a subsample of the cohort.24 Hypertension was confirmed using conventional measurements of blood pressure for 82.3% (95% confidence interval [CI], 72.8–92.8) of participants. Furthermore, we validated each component of metabolic syndrome (including high blood pressure) and found adequate intraclass correlation coefficients for high systolic (0.47 [95% CI, 0.36–0.57]) and diastolic (0.46 [95% CI, 0.34–0.56]) blood pressures using direct assessments.25
Exposure assessment: UPF consumption
Dietary intake was assessed at baseline using a self-administered 136-item semi-quantitative food-frequency questionnaire (FFQ) for the previous year, which was previously validated in Spain and recently reevaluated.26,27 The FFQ included a typical Spanish portion size for each item, and consumption frequencies were categorized using 9 cate gories (ranging from never/almost never to >6 servings/day). Daily food consumption was estimated by multiplying the portion size by the consumption frequency for each food item.
The UPFs were defined according to the NOVA classification,5–7 which classifies foods into 4 groups based on the extent and purpose of industrial processing: (i) unprocessed or minimally processed foods (fruits and vegetables, grains [cereals] in general, nuts and seeds, fresh and pasteurized milk, and natural yogurt with no added sugar or artificial sweeteners); (ii) processed culinary ingredients (salt, sugar, honey, vegetable oils, butter, lard, and vinegar); (iii) processed foods (canned or bottled vegetables and legumes, fruits in syrup, canned fish, unpackaged cheeses, freshly made bread, and salted or sugared nuts and seeds); and (iv) UPF and drink products (carbonated drinks, processed meat, biscuits [cookies], candy [confectionery], ‘instant’ packaged soups and noodles, sweet or savory packaged snacks, and sugared milk and fruit drinks).
The frequency of UPF consumption per person was estimated using the sum of the UPF items in the FFQ (total of 33 items). Total UPF consumption was adjusted for total energy intake using the residual method.22 The sample was divided into tertiles according to total consumption (servings/d).
Assessment of covariates
At baseline, participants completed a questionnaire about a wide array of characteristics, including sociodemographic information (sex and age), medical history and medication use, anthropometric measurements (weight and height), lifestyle (smoking status, physical activity during leisure time, hours of television watched per day, and hours of sleeping and siesta per day), and consumption of a special diet (vegetarian diet, hypocaloric diet, diet to control lactose intolerance, and diet to prevent food allergy).
Self-reports of weight and height were validated in a specific study within a subsample.28 Body mass index was calculated as the self-reported weight (kg) divided by height squared (m2). Physical activity was evaluated using a validated 17-item questionnaire.29
Adherence to the Mediterranean dietary pattern was evaluated using a well-known score.30 Intakes of total energy, macronutrients, sodium, potassium, caffeine, fiber, and alcohol were calculated as the frequency multiplied by the nutrient composition of specified portion sizes using the FFQ. The nutrient databank was updated by a trained team of dieticians using the food composition tables for Spain. The consumption of dairy, fruits, vegetables, fast foods (sausage, hamburgers, and pizza), fried foods, processed meat (sausages, hamburgers, and ham), meat, and sugar-sweetened-beverages was also assessed using the FFQ.26,27
Statistical analyses
Differences in the baseline characteristics of participants according to the UPF consumption tertiles were evaluated using chi-square tests for trends (categorical variables) or linear regression models (continuous variables). The baseline characteristics were expressed as percentages for categorical variables and mean (SD) for continuous variables.
The follow-up time was defined as the interval between the date of recruitment and death date when the last follow-up questionnaire was returned or the date of hypertension diagnosis (for incident cases) whichever came first.
We used Cox regression models with age as the underlying time variable. Multivariable Cox regression models were fitted to estimate hazards ratios (HRs) and 95% CIs for hypertension risk during the follow-up for UPF consumption tertiles. We used the lowest tertile as the reference category. Tests of linear trends were performed by determining the median UPF consumption for each category, which was considered a continuous variable in the respective Cox regression model.
The Cox regression models were adjusted for several potential confounders defined a priori. As recommended to perform multivariate analyses, we identified potential confounders based on a previous causal knowledge of the existing literature, in spite of recurring to statistical criteria.31,32 We stratified all models by year of entry into the cohort. The models were without any adjustment (crude), adjusted for age and sex (model 1), additionally adjusted for physical activity, hours of television watching, baseline body mass index, smoking status, use of analgesics, following a special diet at baseline, family history of hypertension, hypercholesterolemia, and alcohol consumption (model 2), and additionally adjusted for total energy intake, olive oil intake, and fruit and vegetable consumption (model 3).
We evaluated the interaction between UPF consumption tertiles and the most relevant variables (sex, age, and body mass index) using a likelihood ratio test (2 degrees of freedom) that compared the fully adjusted Cox regression model and the same model with interaction product terms.
Sensitivity analyses were conducted by rerunning the multivariable-adjusted Cox regression models with the following changes: (i) additional adjustment for weight gain during follow-up (at least 2 years before developing hypertension); (ii) additional adjustment for energy-adjusted sodium intake; (iii) exclusion of participants with energy intake <5th percentile and >95th percentile; (iv) inclusion of participants who reported systolic blood pressure ≥140 mm Hg and/or diastolic blood pressure ≥90 mm Hg at baseline; (v) exclusion of participants with early incident hypertension (reported within the first 2 years of follow-up); (vi) exclusion of participants with >3 kg weight gain in the 5 years before entering the cohort; and (vii) exclusion of participants with incident chronic diseases.
All analyses were performed with the statistical software package STATA/SE version 12.1 (Stata Corp, College Station, TX), and the statistical significance was set at 5% (based on 2-tailed tests).
RESULTS
During a mean follow-up period of 9.1 years (SD, 3.9 years; person-years, 134,784), 1,702 incident cases of hypertension were identified. Compared with participants in the lowest tertile of UPF consumption at baseline, those in the highest UPF consumption tertile were significantly more likely to be men, younger, have gained weight in recent years, current smokers, watch more television, less physically active, have the highest consumption of total fat, saturated fatty acids, sodium, caffeine, and alcohol, have the lowest consumption of protein, total fiber, potassium, and olive oil, and were less likely to have a family history of hypertension and history of hypercholesterolemia (Table 1). On average, participants in the highest UPF consumption tertile also consumed more fast food, processed meats, fried foods, and sugar-sweetened beverages. In contrast, they had the lowest intakes of fruits, vegetables, and dairy products, were less likely to follow special diets, and had the lowest adherence to the Mediterranean diet (Table 1).
. | Tertiles of total ultra-processed food consumption (energy-adjusted intake) . | . | ||
---|---|---|---|---|
. | 1 (n = 4,930) . | 2 (n = 4,930) . | 3 (n = 4,930) . | P for trenda . |
Total ultra-processed food consumption (servings/d) | 2.1 (0.9) | 3.1 (1.0) | 5.0 (1.7) | <0.001 |
Men | 33.0 | 36.0 | 40.0 | <0.001 |
Age (years) | 40.0 (11.3) | 35.9 (10.5) | 32.9 (9.1) | <0.001 |
Baseline BMI (kg/m2) | 23.2 (3.2) | 23.1 (3.3) | 23.1 (3.3) | 0.31 |
Weight gain >3 kg over the past 5 years | 28.1 | 30.1 | 32.0 | <0.001 |
Family history of hypertension | 43.1 | 39.1 | 37.9 | <0.001 |
Smoking status | <0.001 | |||
Current smoker | 19.6 | 21.6 | 25.9 | |
Former smoker | 32.7 | 25.2 | 21.5 | |
Never smoked | 45.0 | 50.8 | 49.9 | |
Physical activity (METS-h/wk) | 23.5 (24.3) | 20.8 (21.1) | 20.9 (22.7) | <0.001 |
Television watching (h/d) | 1.5 (1.2) | 1.6 (1.2) | 1.7 (1.2) | <0.001 |
Sleeping/siesta (h/d) | 0.3 (0.7) | 0.3 (0.8) | 0.3 (0.9) | 0.05 |
Analgesic use | 3.0 | 2.3 | 3.2 | 0.72 |
Hypercholesterolemia | 16.2 | 13.6 | 11.7 | <0.001 |
Total energy intake (kcal/d) | 2,425 (579) | 2,215 (608) | 2,401 (621) | 0.05 |
Macronutrients (% energy) | ||||
Carbohydrate intake | 44.5 (7.7) | 42.9 (7.0) | 43.1 (7.1) | <0.001 |
Protein intake | 18.6 (3.3) | 18.6 (3.3) | 17.4 (3.0) | <0.001 |
Fat intake | 35.6 (7.0) | 37.1 (6.1) | 38.0 (6.1) | <0.001 |
SFA | 11.6 (3.2) | 12.8 (3.0) | 13.4 (3.1) | <0.001 |
MUFA | 15.8 (4.1) | 15.8 (3.5) | 15.9 (3.4) | 0.41 |
PUFA | 5.0 (1.5) | 5.2 (1.5) | 5.5 (1.6) | <0.001 |
Total dietary fiber intake (g/d) | 33.3 (13.8) | 25.7 (10.2) | 23.9 (9.8) | <0.001 |
Sodium intakeb (g/d) | 2.8 (1.3) | 3.3 (1.7) | 3.9 (2.7) | <0.001 |
Potassium intakeb (g/d) | 5.3 (1.4) | 4.7 (1.0) | 4.2 (1.0) | <0.001 |
Caffeine intake (mg/d) | 41.6 (37.8) | 40.2 (35.9) | 49.1 (43.9) | <0.001 |
Alcohol consumption (g/d) | 4.3 (7.7) | 4.0 (5.9) | 4.9 (7.8) | <0.001 |
Olive oil intake (g/d) | 23.3 (17.2) | 17.0 (13.2) | 15.6 (12.6) | <0.001 |
Low-fat dairy consumption (g/d) | 253 (276) | 216 (232) | 203 (225) | <0.001 |
High-fat dairy consumption (g/d) | 212 (222) | 194 (189) | 197 (179) | <0.001 |
Fruit consumption (g/d) | 449 (368) | 309 (249) | 257 (209) | <0.001 |
Vegetable consumption (g/d) | 638 (393) | 493 (277) | 432 (284) | <0.001 |
Sugar-sweetened-beverage consumption (ml/d) | 24.6 (40.7) | 45.6 (58.4) | 128 (188) | <0.001 |
Fast food consumptionc (g/d) | 16.2 (16.2) | 22.0 (18.1) | 29.5 (25.1) | <0.001 |
Processed meat consumptiond (g/d) | 36.0 (24.3) | 42.6 (28.0) | 52.9 (36.5) | <0.001 |
Meat consumption (g/d) | 78.5 (47.0) | 77.6 (45.2) | 76.6 (45.6) | <0.001 |
Fried-food consumption (servings/wk) | 3.3 (4.2) | 3.5 (3.7) | 4.2 (4.7) | <0.001 |
Mediterranean dietary pattern (0–9 points)e | 5.0 (1.6) | 4.1 (1.7) | 3.6 (1.7) | <0.001 |
Special diet at baseline | 7.9 | 6.9 | 5.5 | <0.001 |
. | Tertiles of total ultra-processed food consumption (energy-adjusted intake) . | . | ||
---|---|---|---|---|
. | 1 (n = 4,930) . | 2 (n = 4,930) . | 3 (n = 4,930) . | P for trenda . |
Total ultra-processed food consumption (servings/d) | 2.1 (0.9) | 3.1 (1.0) | 5.0 (1.7) | <0.001 |
Men | 33.0 | 36.0 | 40.0 | <0.001 |
Age (years) | 40.0 (11.3) | 35.9 (10.5) | 32.9 (9.1) | <0.001 |
Baseline BMI (kg/m2) | 23.2 (3.2) | 23.1 (3.3) | 23.1 (3.3) | 0.31 |
Weight gain >3 kg over the past 5 years | 28.1 | 30.1 | 32.0 | <0.001 |
Family history of hypertension | 43.1 | 39.1 | 37.9 | <0.001 |
Smoking status | <0.001 | |||
Current smoker | 19.6 | 21.6 | 25.9 | |
Former smoker | 32.7 | 25.2 | 21.5 | |
Never smoked | 45.0 | 50.8 | 49.9 | |
Physical activity (METS-h/wk) | 23.5 (24.3) | 20.8 (21.1) | 20.9 (22.7) | <0.001 |
Television watching (h/d) | 1.5 (1.2) | 1.6 (1.2) | 1.7 (1.2) | <0.001 |
Sleeping/siesta (h/d) | 0.3 (0.7) | 0.3 (0.8) | 0.3 (0.9) | 0.05 |
Analgesic use | 3.0 | 2.3 | 3.2 | 0.72 |
Hypercholesterolemia | 16.2 | 13.6 | 11.7 | <0.001 |
Total energy intake (kcal/d) | 2,425 (579) | 2,215 (608) | 2,401 (621) | 0.05 |
Macronutrients (% energy) | ||||
Carbohydrate intake | 44.5 (7.7) | 42.9 (7.0) | 43.1 (7.1) | <0.001 |
Protein intake | 18.6 (3.3) | 18.6 (3.3) | 17.4 (3.0) | <0.001 |
Fat intake | 35.6 (7.0) | 37.1 (6.1) | 38.0 (6.1) | <0.001 |
SFA | 11.6 (3.2) | 12.8 (3.0) | 13.4 (3.1) | <0.001 |
MUFA | 15.8 (4.1) | 15.8 (3.5) | 15.9 (3.4) | 0.41 |
PUFA | 5.0 (1.5) | 5.2 (1.5) | 5.5 (1.6) | <0.001 |
Total dietary fiber intake (g/d) | 33.3 (13.8) | 25.7 (10.2) | 23.9 (9.8) | <0.001 |
Sodium intakeb (g/d) | 2.8 (1.3) | 3.3 (1.7) | 3.9 (2.7) | <0.001 |
Potassium intakeb (g/d) | 5.3 (1.4) | 4.7 (1.0) | 4.2 (1.0) | <0.001 |
Caffeine intake (mg/d) | 41.6 (37.8) | 40.2 (35.9) | 49.1 (43.9) | <0.001 |
Alcohol consumption (g/d) | 4.3 (7.7) | 4.0 (5.9) | 4.9 (7.8) | <0.001 |
Olive oil intake (g/d) | 23.3 (17.2) | 17.0 (13.2) | 15.6 (12.6) | <0.001 |
Low-fat dairy consumption (g/d) | 253 (276) | 216 (232) | 203 (225) | <0.001 |
High-fat dairy consumption (g/d) | 212 (222) | 194 (189) | 197 (179) | <0.001 |
Fruit consumption (g/d) | 449 (368) | 309 (249) | 257 (209) | <0.001 |
Vegetable consumption (g/d) | 638 (393) | 493 (277) | 432 (284) | <0.001 |
Sugar-sweetened-beverage consumption (ml/d) | 24.6 (40.7) | 45.6 (58.4) | 128 (188) | <0.001 |
Fast food consumptionc (g/d) | 16.2 (16.2) | 22.0 (18.1) | 29.5 (25.1) | <0.001 |
Processed meat consumptiond (g/d) | 36.0 (24.3) | 42.6 (28.0) | 52.9 (36.5) | <0.001 |
Meat consumption (g/d) | 78.5 (47.0) | 77.6 (45.2) | 76.6 (45.6) | <0.001 |
Fried-food consumption (servings/wk) | 3.3 (4.2) | 3.5 (3.7) | 4.2 (4.7) | <0.001 |
Mediterranean dietary pattern (0–9 points)e | 5.0 (1.6) | 4.1 (1.7) | 3.6 (1.7) | <0.001 |
Special diet at baseline | 7.9 | 6.9 | 5.5 | <0.001 |
Values are expressed as means and standard deviations or %. Abbreviations: BMI, body mass index; MET, metabolic equivalents of task; MUFA, monounsaturated fatty acid; PUFA, polyunsaturated fatty acid; SFA, saturated fatty acid.
aChi-square test for trend (categorical variables) and linear regression models (continuous variables) across tertiles of ultra-processed food consumption.
bEnergy adjusted.
cSum of hamburgers, sausages, and pizza.
dSum of sausages, hamburgers, and ham.
eHigher scores indicate greater adherence.
. | Tertiles of total ultra-processed food consumption (energy-adjusted intake) . | . | ||
---|---|---|---|---|
. | 1 (n = 4,930) . | 2 (n = 4,930) . | 3 (n = 4,930) . | P for trenda . |
Total ultra-processed food consumption (servings/d) | 2.1 (0.9) | 3.1 (1.0) | 5.0 (1.7) | <0.001 |
Men | 33.0 | 36.0 | 40.0 | <0.001 |
Age (years) | 40.0 (11.3) | 35.9 (10.5) | 32.9 (9.1) | <0.001 |
Baseline BMI (kg/m2) | 23.2 (3.2) | 23.1 (3.3) | 23.1 (3.3) | 0.31 |
Weight gain >3 kg over the past 5 years | 28.1 | 30.1 | 32.0 | <0.001 |
Family history of hypertension | 43.1 | 39.1 | 37.9 | <0.001 |
Smoking status | <0.001 | |||
Current smoker | 19.6 | 21.6 | 25.9 | |
Former smoker | 32.7 | 25.2 | 21.5 | |
Never smoked | 45.0 | 50.8 | 49.9 | |
Physical activity (METS-h/wk) | 23.5 (24.3) | 20.8 (21.1) | 20.9 (22.7) | <0.001 |
Television watching (h/d) | 1.5 (1.2) | 1.6 (1.2) | 1.7 (1.2) | <0.001 |
Sleeping/siesta (h/d) | 0.3 (0.7) | 0.3 (0.8) | 0.3 (0.9) | 0.05 |
Analgesic use | 3.0 | 2.3 | 3.2 | 0.72 |
Hypercholesterolemia | 16.2 | 13.6 | 11.7 | <0.001 |
Total energy intake (kcal/d) | 2,425 (579) | 2,215 (608) | 2,401 (621) | 0.05 |
Macronutrients (% energy) | ||||
Carbohydrate intake | 44.5 (7.7) | 42.9 (7.0) | 43.1 (7.1) | <0.001 |
Protein intake | 18.6 (3.3) | 18.6 (3.3) | 17.4 (3.0) | <0.001 |
Fat intake | 35.6 (7.0) | 37.1 (6.1) | 38.0 (6.1) | <0.001 |
SFA | 11.6 (3.2) | 12.8 (3.0) | 13.4 (3.1) | <0.001 |
MUFA | 15.8 (4.1) | 15.8 (3.5) | 15.9 (3.4) | 0.41 |
PUFA | 5.0 (1.5) | 5.2 (1.5) | 5.5 (1.6) | <0.001 |
Total dietary fiber intake (g/d) | 33.3 (13.8) | 25.7 (10.2) | 23.9 (9.8) | <0.001 |
Sodium intakeb (g/d) | 2.8 (1.3) | 3.3 (1.7) | 3.9 (2.7) | <0.001 |
Potassium intakeb (g/d) | 5.3 (1.4) | 4.7 (1.0) | 4.2 (1.0) | <0.001 |
Caffeine intake (mg/d) | 41.6 (37.8) | 40.2 (35.9) | 49.1 (43.9) | <0.001 |
Alcohol consumption (g/d) | 4.3 (7.7) | 4.0 (5.9) | 4.9 (7.8) | <0.001 |
Olive oil intake (g/d) | 23.3 (17.2) | 17.0 (13.2) | 15.6 (12.6) | <0.001 |
Low-fat dairy consumption (g/d) | 253 (276) | 216 (232) | 203 (225) | <0.001 |
High-fat dairy consumption (g/d) | 212 (222) | 194 (189) | 197 (179) | <0.001 |
Fruit consumption (g/d) | 449 (368) | 309 (249) | 257 (209) | <0.001 |
Vegetable consumption (g/d) | 638 (393) | 493 (277) | 432 (284) | <0.001 |
Sugar-sweetened-beverage consumption (ml/d) | 24.6 (40.7) | 45.6 (58.4) | 128 (188) | <0.001 |
Fast food consumptionc (g/d) | 16.2 (16.2) | 22.0 (18.1) | 29.5 (25.1) | <0.001 |
Processed meat consumptiond (g/d) | 36.0 (24.3) | 42.6 (28.0) | 52.9 (36.5) | <0.001 |
Meat consumption (g/d) | 78.5 (47.0) | 77.6 (45.2) | 76.6 (45.6) | <0.001 |
Fried-food consumption (servings/wk) | 3.3 (4.2) | 3.5 (3.7) | 4.2 (4.7) | <0.001 |
Mediterranean dietary pattern (0–9 points)e | 5.0 (1.6) | 4.1 (1.7) | 3.6 (1.7) | <0.001 |
Special diet at baseline | 7.9 | 6.9 | 5.5 | <0.001 |
. | Tertiles of total ultra-processed food consumption (energy-adjusted intake) . | . | ||
---|---|---|---|---|
. | 1 (n = 4,930) . | 2 (n = 4,930) . | 3 (n = 4,930) . | P for trenda . |
Total ultra-processed food consumption (servings/d) | 2.1 (0.9) | 3.1 (1.0) | 5.0 (1.7) | <0.001 |
Men | 33.0 | 36.0 | 40.0 | <0.001 |
Age (years) | 40.0 (11.3) | 35.9 (10.5) | 32.9 (9.1) | <0.001 |
Baseline BMI (kg/m2) | 23.2 (3.2) | 23.1 (3.3) | 23.1 (3.3) | 0.31 |
Weight gain >3 kg over the past 5 years | 28.1 | 30.1 | 32.0 | <0.001 |
Family history of hypertension | 43.1 | 39.1 | 37.9 | <0.001 |
Smoking status | <0.001 | |||
Current smoker | 19.6 | 21.6 | 25.9 | |
Former smoker | 32.7 | 25.2 | 21.5 | |
Never smoked | 45.0 | 50.8 | 49.9 | |
Physical activity (METS-h/wk) | 23.5 (24.3) | 20.8 (21.1) | 20.9 (22.7) | <0.001 |
Television watching (h/d) | 1.5 (1.2) | 1.6 (1.2) | 1.7 (1.2) | <0.001 |
Sleeping/siesta (h/d) | 0.3 (0.7) | 0.3 (0.8) | 0.3 (0.9) | 0.05 |
Analgesic use | 3.0 | 2.3 | 3.2 | 0.72 |
Hypercholesterolemia | 16.2 | 13.6 | 11.7 | <0.001 |
Total energy intake (kcal/d) | 2,425 (579) | 2,215 (608) | 2,401 (621) | 0.05 |
Macronutrients (% energy) | ||||
Carbohydrate intake | 44.5 (7.7) | 42.9 (7.0) | 43.1 (7.1) | <0.001 |
Protein intake | 18.6 (3.3) | 18.6 (3.3) | 17.4 (3.0) | <0.001 |
Fat intake | 35.6 (7.0) | 37.1 (6.1) | 38.0 (6.1) | <0.001 |
SFA | 11.6 (3.2) | 12.8 (3.0) | 13.4 (3.1) | <0.001 |
MUFA | 15.8 (4.1) | 15.8 (3.5) | 15.9 (3.4) | 0.41 |
PUFA | 5.0 (1.5) | 5.2 (1.5) | 5.5 (1.6) | <0.001 |
Total dietary fiber intake (g/d) | 33.3 (13.8) | 25.7 (10.2) | 23.9 (9.8) | <0.001 |
Sodium intakeb (g/d) | 2.8 (1.3) | 3.3 (1.7) | 3.9 (2.7) | <0.001 |
Potassium intakeb (g/d) | 5.3 (1.4) | 4.7 (1.0) | 4.2 (1.0) | <0.001 |
Caffeine intake (mg/d) | 41.6 (37.8) | 40.2 (35.9) | 49.1 (43.9) | <0.001 |
Alcohol consumption (g/d) | 4.3 (7.7) | 4.0 (5.9) | 4.9 (7.8) | <0.001 |
Olive oil intake (g/d) | 23.3 (17.2) | 17.0 (13.2) | 15.6 (12.6) | <0.001 |
Low-fat dairy consumption (g/d) | 253 (276) | 216 (232) | 203 (225) | <0.001 |
High-fat dairy consumption (g/d) | 212 (222) | 194 (189) | 197 (179) | <0.001 |
Fruit consumption (g/d) | 449 (368) | 309 (249) | 257 (209) | <0.001 |
Vegetable consumption (g/d) | 638 (393) | 493 (277) | 432 (284) | <0.001 |
Sugar-sweetened-beverage consumption (ml/d) | 24.6 (40.7) | 45.6 (58.4) | 128 (188) | <0.001 |
Fast food consumptionc (g/d) | 16.2 (16.2) | 22.0 (18.1) | 29.5 (25.1) | <0.001 |
Processed meat consumptiond (g/d) | 36.0 (24.3) | 42.6 (28.0) | 52.9 (36.5) | <0.001 |
Meat consumption (g/d) | 78.5 (47.0) | 77.6 (45.2) | 76.6 (45.6) | <0.001 |
Fried-food consumption (servings/wk) | 3.3 (4.2) | 3.5 (3.7) | 4.2 (4.7) | <0.001 |
Mediterranean dietary pattern (0–9 points)e | 5.0 (1.6) | 4.1 (1.7) | 3.6 (1.7) | <0.001 |
Special diet at baseline | 7.9 | 6.9 | 5.5 | <0.001 |
Values are expressed as means and standard deviations or %. Abbreviations: BMI, body mass index; MET, metabolic equivalents of task; MUFA, monounsaturated fatty acid; PUFA, polyunsaturated fatty acid; SFA, saturated fatty acid.
aChi-square test for trend (categorical variables) and linear regression models (continuous variables) across tertiles of ultra-processed food consumption.
bEnergy adjusted.
cSum of hamburgers, sausages, and pizza.
dSum of sausages, hamburgers, and ham.
eHigher scores indicate greater adherence.
In multivariable models, participants in the highest UPF consumption tertile had a greater risk of developing hypertension than those in the lowest tertile (multivariable-adjusted HR, 1.21 [95% CI, 1.06–1.37]). The estimates showed a statistically significant linear trend (P = 0.004) after adjusting for potential confounders (Table 2). When we excluded total energy intake in our model because of the potential existence of an over adjustment the results did not substantially change (adjusted HR tertile 3 vs. tertile 1: 1.21, 95% CI: 1.07–1.37; P for trend = 0.003).
. | Tertiles of total ultra-processed food consumption (energy-adjusted intake) . | . | ||
---|---|---|---|---|
. | 1 . | 2 . | 3 . | P for trend . |
Crude | 1 (ref) | 1.08 (0.97–1.21) | 1.43 (1.27–1.61) | <0.001 |
Age and sex adjusteda | 1 (ref) | 1.01 (0.90–1.13) | 1.27 (1.12–1.43) | <0.001 |
Multivariable adjustedb | 1 (ref) | 1.00 (0.89–1.12) | 1.23 (1.09–1.38) | 0.001 |
Multivariable adjustedc | 1 (ref) | 0.99 (0.88–1.12) | 1.21 (1.06–1.37) | 0.004 |
. | Tertiles of total ultra-processed food consumption (energy-adjusted intake) . | . | ||
---|---|---|---|---|
. | 1 . | 2 . | 3 . | P for trend . |
Crude | 1 (ref) | 1.08 (0.97–1.21) | 1.43 (1.27–1.61) | <0.001 |
Age and sex adjusteda | 1 (ref) | 1.01 (0.90–1.13) | 1.27 (1.12–1.43) | <0.001 |
Multivariable adjustedb | 1 (ref) | 1.00 (0.89–1.12) | 1.23 (1.09–1.38) | 0.001 |
Multivariable adjustedc | 1 (ref) | 0.99 (0.88–1.12) | 1.21 (1.06–1.37) | 0.004 |
Data are shown as Cox proportional hazard ratios and 95% confidence intervals.
aAdjusted for age and sex.
bAdjusted for sex, age, physical activity, hours of TV watching, baseline body mass index, smoking status, use of analgesics, following a special diet at baseline, family history of hypertension, hypercholesterolemia, and alcohol consumption.
cAdditionally adjusted for total energy intake, olive oil intake, and consumption of fruits and vegetables.
. | Tertiles of total ultra-processed food consumption (energy-adjusted intake) . | . | ||
---|---|---|---|---|
. | 1 . | 2 . | 3 . | P for trend . |
Crude | 1 (ref) | 1.08 (0.97–1.21) | 1.43 (1.27–1.61) | <0.001 |
Age and sex adjusteda | 1 (ref) | 1.01 (0.90–1.13) | 1.27 (1.12–1.43) | <0.001 |
Multivariable adjustedb | 1 (ref) | 1.00 (0.89–1.12) | 1.23 (1.09–1.38) | 0.001 |
Multivariable adjustedc | 1 (ref) | 0.99 (0.88–1.12) | 1.21 (1.06–1.37) | 0.004 |
. | Tertiles of total ultra-processed food consumption (energy-adjusted intake) . | . | ||
---|---|---|---|---|
. | 1 . | 2 . | 3 . | P for trend . |
Crude | 1 (ref) | 1.08 (0.97–1.21) | 1.43 (1.27–1.61) | <0.001 |
Age and sex adjusteda | 1 (ref) | 1.01 (0.90–1.13) | 1.27 (1.12–1.43) | <0.001 |
Multivariable adjustedb | 1 (ref) | 1.00 (0.89–1.12) | 1.23 (1.09–1.38) | 0.001 |
Multivariable adjustedc | 1 (ref) | 0.99 (0.88–1.12) | 1.21 (1.06–1.37) | 0.004 |
Data are shown as Cox proportional hazard ratios and 95% confidence intervals.
aAdjusted for age and sex.
bAdjusted for sex, age, physical activity, hours of TV watching, baseline body mass index, smoking status, use of analgesics, following a special diet at baseline, family history of hypertension, hypercholesterolemia, and alcohol consumption.
cAdditionally adjusted for total energy intake, olive oil intake, and consumption of fruits and vegetables.
No significant interactions between UPF consumption and sex (P interaction = 0.65), age (P interaction = 0.32), and body mass index (P interaction = 0.59) were observed. The results from the sensitivity analyses did not substantially change in any of the scenarios (Table 3).
. | Tertiles of total ultra-processed food consumption (energy-adjusted intake) . | . | |||
---|---|---|---|---|---|
. | Cases/person-years, n . | 1 . | 2 . | 3 . | P for trend . |
Overalla | 1,702/134,784 | 1 (ref) | 0.99 (0.88–1.12) | 1.21 (1.06–1.37) | 0.004 |
Additionally adjusted for weight gain during the follow-up | 1,702/134,784 | 1 (ref) | 0.99 (0.88–1.12) | 1.20 (1.07–1.37) | 0.01 |
Additionally adjusted for energy-adjusted sodium intake | 1,702/134,784 | 1 (ref) | 1.00 (0.88–1.13) | 1.22 (1.07–1.38) | 0.003 |
Energy limits between the 5th and 95th percentiles | 1,640/134,872 | 1 (ref) | 1.02 (0.90–1.15) | 1.22 (1.07–1.39) | 0.01 |
Including those who reported SBP ≥140 mm Hg and/or DBP ≥90 mm Hg at baseline | 1,869/137,406 | 1 (ref) | 1.04 (0.93–1.17) | 1.23 (1.09–1.39) | 0.002 |
Excluding early incident cases of hypertension (until 2 years of follow-up) | 1,266/133,721 | 1 (ref) | 1.03 (0.90–1.19) | 1.21 (1.04–1.40) | 0.01 |
Excluding participants with >3 kg weight gain over the past 5 years | 1,023/94,525 | 1 (ref) | 0.94 (0.80–1.10) | 1.22 (1.04–1.45) | 0.02 |
Excluding participants with incident chronic diseases | 1,443/125,118 | 1 (ref) | 1.04 (0.91–1.18) | 1.25 (1.09–1.44) | 0.002 |
. | Tertiles of total ultra-processed food consumption (energy-adjusted intake) . | . | |||
---|---|---|---|---|---|
. | Cases/person-years, n . | 1 . | 2 . | 3 . | P for trend . |
Overalla | 1,702/134,784 | 1 (ref) | 0.99 (0.88–1.12) | 1.21 (1.06–1.37) | 0.004 |
Additionally adjusted for weight gain during the follow-up | 1,702/134,784 | 1 (ref) | 0.99 (0.88–1.12) | 1.20 (1.07–1.37) | 0.01 |
Additionally adjusted for energy-adjusted sodium intake | 1,702/134,784 | 1 (ref) | 1.00 (0.88–1.13) | 1.22 (1.07–1.38) | 0.003 |
Energy limits between the 5th and 95th percentiles | 1,640/134,872 | 1 (ref) | 1.02 (0.90–1.15) | 1.22 (1.07–1.39) | 0.01 |
Including those who reported SBP ≥140 mm Hg and/or DBP ≥90 mm Hg at baseline | 1,869/137,406 | 1 (ref) | 1.04 (0.93–1.17) | 1.23 (1.09–1.39) | 0.002 |
Excluding early incident cases of hypertension (until 2 years of follow-up) | 1,266/133,721 | 1 (ref) | 1.03 (0.90–1.19) | 1.21 (1.04–1.40) | 0.01 |
Excluding participants with >3 kg weight gain over the past 5 years | 1,023/94,525 | 1 (ref) | 0.94 (0.80–1.10) | 1.22 (1.04–1.45) | 0.02 |
Excluding participants with incident chronic diseases | 1,443/125,118 | 1 (ref) | 1.04 (0.91–1.18) | 1.25 (1.09–1.44) | 0.002 |
Abbreviations: DBP, diastolic blood pressure; SBP, systolic blood pressure.
aAdjusted for sex, age, physical activity, hours of TV watching, baseline body mass index, smoking status, use of analgesics, following a special diet at baseline, family history of hypertension, hypercholesterolemia, alcohol consumption, total energy intake, olive oil intake, and consumption of fruits and vegetables.
. | Tertiles of total ultra-processed food consumption (energy-adjusted intake) . | . | |||
---|---|---|---|---|---|
. | Cases/person-years, n . | 1 . | 2 . | 3 . | P for trend . |
Overalla | 1,702/134,784 | 1 (ref) | 0.99 (0.88–1.12) | 1.21 (1.06–1.37) | 0.004 |
Additionally adjusted for weight gain during the follow-up | 1,702/134,784 | 1 (ref) | 0.99 (0.88–1.12) | 1.20 (1.07–1.37) | 0.01 |
Additionally adjusted for energy-adjusted sodium intake | 1,702/134,784 | 1 (ref) | 1.00 (0.88–1.13) | 1.22 (1.07–1.38) | 0.003 |
Energy limits between the 5th and 95th percentiles | 1,640/134,872 | 1 (ref) | 1.02 (0.90–1.15) | 1.22 (1.07–1.39) | 0.01 |
Including those who reported SBP ≥140 mm Hg and/or DBP ≥90 mm Hg at baseline | 1,869/137,406 | 1 (ref) | 1.04 (0.93–1.17) | 1.23 (1.09–1.39) | 0.002 |
Excluding early incident cases of hypertension (until 2 years of follow-up) | 1,266/133,721 | 1 (ref) | 1.03 (0.90–1.19) | 1.21 (1.04–1.40) | 0.01 |
Excluding participants with >3 kg weight gain over the past 5 years | 1,023/94,525 | 1 (ref) | 0.94 (0.80–1.10) | 1.22 (1.04–1.45) | 0.02 |
Excluding participants with incident chronic diseases | 1,443/125,118 | 1 (ref) | 1.04 (0.91–1.18) | 1.25 (1.09–1.44) | 0.002 |
. | Tertiles of total ultra-processed food consumption (energy-adjusted intake) . | . | |||
---|---|---|---|---|---|
. | Cases/person-years, n . | 1 . | 2 . | 3 . | P for trend . |
Overalla | 1,702/134,784 | 1 (ref) | 0.99 (0.88–1.12) | 1.21 (1.06–1.37) | 0.004 |
Additionally adjusted for weight gain during the follow-up | 1,702/134,784 | 1 (ref) | 0.99 (0.88–1.12) | 1.20 (1.07–1.37) | 0.01 |
Additionally adjusted for energy-adjusted sodium intake | 1,702/134,784 | 1 (ref) | 1.00 (0.88–1.13) | 1.22 (1.07–1.38) | 0.003 |
Energy limits between the 5th and 95th percentiles | 1,640/134,872 | 1 (ref) | 1.02 (0.90–1.15) | 1.22 (1.07–1.39) | 0.01 |
Including those who reported SBP ≥140 mm Hg and/or DBP ≥90 mm Hg at baseline | 1,869/137,406 | 1 (ref) | 1.04 (0.93–1.17) | 1.23 (1.09–1.39) | 0.002 |
Excluding early incident cases of hypertension (until 2 years of follow-up) | 1,266/133,721 | 1 (ref) | 1.03 (0.90–1.19) | 1.21 (1.04–1.40) | 0.01 |
Excluding participants with >3 kg weight gain over the past 5 years | 1,023/94,525 | 1 (ref) | 0.94 (0.80–1.10) | 1.22 (1.04–1.45) | 0.02 |
Excluding participants with incident chronic diseases | 1,443/125,118 | 1 (ref) | 1.04 (0.91–1.18) | 1.25 (1.09–1.44) | 0.002 |
Abbreviations: DBP, diastolic blood pressure; SBP, systolic blood pressure.
aAdjusted for sex, age, physical activity, hours of TV watching, baseline body mass index, smoking status, use of analgesics, following a special diet at baseline, family history of hypertension, hypercholesterolemia, alcohol consumption, total energy intake, olive oil intake, and consumption of fruits and vegetables.
DISCUSSION
In this prospective study, higher baseline consumption of UPFs was associated with a higher risk of incident hypertension, even after adjustment for potential confounding factors. To our knowledge, this is the first epidemiologic study that has evaluated the association between UPFs, a new category of foods based on their degree of processing, and the risk of hypertension.
Our results support those of most previous cross-sectional prospective studies17,18,33 and meta-analyses.15,16 Consuming processed meat was associated with the incidence of elevated blood pressure in black and white men and women in the CARDIA study.14 Furthermore, sweetened beverage consumption (≥1 serving/d) was directly associated with hypertension in 2 meta-analyses.11,12
We believe that the observed associations between UPF consumption and hypertension risk could be explained by a higher total intake of salt, saturated fat, and sugar and an inadequate intake of fiber and micronutrients.8–11 Participants who consumed a greater amount of UPFs in our cohort had higher intakes of salt, saturated fatty acids, and sugar-sweetened beverages, lower intakes of fiber, potassium, vegetables, and fruits, and a lower adherence to the Mediterranean dietary pattern. A meta-analysis of prospective studies and randomized control trials provided evidence of the relationship between high adherence to the Mediterranean dietary pattern and lower risk of cardiovascular disease and metabolic syndrome. The Mediterranean diet is based on consumption of fresh foods such as vegetable, fruit, legumes, and olive oil.34,35 Therefore, the Mediterranean diet could act as a confounder in the association between UPF consumption and hypertension. When we adjusted for Mediterranean diet score instead of adjusting for alcohol, olive oil, fruits, and vegetables the results did not substantially change (adjusted HR tertile 3 vs. tertile 1: 1.22; 95% CI: 1.07–1.38; P for trend = 0.002). According to the International Study of Macro-/Micronutrients and Blood Pressure (INTERMAP), individuals with a low cardiovascular risk had higher intakes of fiber, vitamins and minerals, vegetable protein, and nutrient-dense foods (fruits, vegetables, grains, fish) and lower intakes of animal protein, sodium, saturated fatty acids, and high-calorie foods (processed meats and sugar-sweetened beverages).33 Our results are similar.
Excessive dietary sodium intake is associated with an increased risk of hypertension and cardiovascular-related deaths.36,37 Salt, the main source of sodium, is present in processed foods and ready-to-eat meals in many countries.4 Most UPFs have a high sodium content, especially condiments (complete seasonings and sauces), broths, soup powders, and processed meats.38,39 In a meta-analysis of 34 trials, a longer-term reduction in salt intake of 4.4 g/d resulted in an important reduction in blood pressure (systolic blood pressure: −4.18 mm Hg [95% CI, −5.18, −3.18]; diastolic blood pressure: −2.06 mm Hg [95% CI, −2.67, −1.45]), and that a greater reduction in salt intake provided a greater decrease in systolic blood pressure.40
Similar results have been observed for saturated fat intake. Reduced saturated fat intake was associated with a reduced cardiovascular disease risk in a recent Cochrane review of 15 randomized controlled trials that used a variety of interventions to reduce saturated fat intake.41 Furthermore, replacing saturated fat with unsaturated plant fats reduces the risk of coronary heart disease.42 Another hypothesis to explain the association between UPF and hypertension risk could be weight gain related to UPF consumption during follow-up. In a previous study in the same cohort, a positive association between higher UPF consumption and risk of overweight/obesity was shown.43 However, we conducted additional analyses adjusting for weight gain during follow-up, which showed a small change in the results, suggesting that additional mechanisms might explain the relationship between UPF and hypertension.
In contrast, diets rich in fruits and vegetables are associated with reduced blood pressure, potentially owing to their effect on body weight control and fiber, vitamin, and mineral intake. In an analysis of 3 cohorts in the United States (Nurses’ Health Study, Nurses’ Health Study II, and Health Professionals Follow-up Study), long-term fruit consumption was associated with a reduced risk of developing hypertension in adults.44 In this context, a high consumption of UPFs could mean that calories from healthy fruits and vegetables are replaced by UPF calories, which is also a plausible explanation of our findings. In the INTERMAP study, an inverse association between raw and cooked vegetable consumption and blood pressure, with a stronger association for raw vegetables, was found in 4,680 men and women aged 40–59 years from Japan and the People’s Republic of China, United Kingdom, and United States.45
Although we focused on the association between hypertension risk and consumption of a specific food group (UPFs), it is important to consider that the overall dietary pattern and lifestyle have more effect than individual food intake. We believe that individuals with excessive UPFs consumption have other characteristics that comprise an unhealthy lifestyle, such as less physical activity, more hours of watching television, and tobacco use. In contrast, individuals who consume a smaller amount of UPFs tend to adopt healthier lifestyles, such as following the Mediterranean diet, which can be beneficial for controlling and preventing hypertension.46 These aspects underscore the importance of cultural food practices, compared with a pattern of consuming industrialized products, which are artificially produced and typically unhealthy.
To the best of our knowledge, this is the first longitudinal evaluation of the relationship between UPF consumption and hypertension. Other strengths of this study include the prospective design, relatively large sample size, long follow-up period, and use of validated methods.24,26,28
Some limitations are also present. The FFQ used to evaluate UPF consumption was not specifically designed for this purpose, which may underestimate or overestimate UPF consumption. However, the FFQ was previously validated and used to collect information about the main foods consumed by the study population.26 Furthermore, although a self-reported diagnosis of hypertension was used, this method was previously validated in our cohort.24 Although we adjusted for potential confounders related to lifestyle and diet, potential residual confounding (i.e., factors associated with a healthy lifestyle and the Mediterranean dietary pattern) may still exist. Furthermore, we conducted multiple sensitivity analyses to account for additional mechanisms which might explain the relationship between UPF and hypertension, potential uncertainties in our assumptions regarding the induction period and also for possible sources of bias including measurement errors. A healthier profile was observed in the SUN project participants, who were college graduates and may have more health awareness, which potentially implies an even greater consumption of UPFs in the general population. Therefore, caution is required when extrapolating the results to the general population. Furthermore, the participants had high educational levels and were likely capable of providing higher quality self-reported data.
According to our study, excessive UPF consumption was associated with an increased risk of developing hypertension in a well-defined cohort of university graduates. Methods to encourage the consumption of plant foods, adoption of a healthy lifestyle, and reestablish the food culture, such as the Mediterranean diet, are required. However, these results should be interpreted with caution, and other confirmatory studies are necessary.
DISCLOSURE
The authors declared no conflicts of interest.
ACKNOWLEDGMENTS
We thank the participants of the SUN project for their continued cooperation and participation. We thank all of the members of the SUN project for their administrative, technical, and material support. R.D.M. would also like to thank Capes Foundation, Ministry of Education of Brazil for the scholarship to continue his training at the University of Navarra, Spain. The SUN project received funding from the Spanish Government-Instituto de Salud Carlos II and European Regional Development Fund (Grants PI10/02993, PI13/00615, PI14/01668, PI14/01798, PI14/1764, G03/140); the Navarra Regional Government (45/2011, 122/2014); and the University of Navarra. R.D.M., A.C.S.L., and M.B.-R. analyzed the data and drafted the manuscript. M.B.-R. obtained the funding and designed the study. A.M.P. and A.G. contributed to the study design and data interpretation. M.A.M.-G. contributed to the study design and is the founder of the SUN Project. All authors critically revised the manuscript for important intellectual content and approved the final version to be submitted for publication.
REFERENCES
Author notes
Correspondence: Maira Bes-Rastrollo (mbes@unav.es).