ABSTRACT
OBJECTIVE
To describe the Nova score for the consumption of ultra-processed foods (UPF) and evaluate its potential in reflecting the dietary share of UPF in Brazil.
METHODS
This study was conducted in São Paulo with a convenience sample of 300 adults. Using a tablet, participants answered a 3-minute electronic self-report questionnaire on the consumption of 23 subgroups of UPF commonly consumed in Brazil, regarding the day prior the survey. Each participant score corresponded to the number of subgroups reported. The dietary share of UPF on the day prior to the survey, expressed as a percentage of total energy intake, was calculated based on data collected on a 30-minute complete 24-hour dietary recall administered by trained nutritionists. The association between the score and the dietary share of UPF was evaluated using linear regression models. The Pabak index was used to assess the agreement in participants’ classification according to the fifths of Nova score and the fifths of dietary share of UPF.
RESULTS
The average dietary share of UPF increased linearly and significantly with the increase of the Nova score for the consumption of ultra-processed foods. We found a substantial agreement in participants’ classification according to the fifths of the distribution of scores and the fifths of the dietary share of UPF (Pabak index = 0.67). Age was inversely associated with a relatively high frequency of UPF consumption (upper fifth of the distribution) for both score and dietary share of UPF.
CONCLUSION
The Nova score for the consumption of ultra-processed foods, obtained in a quick and practical manner, shows a good potential in reflecting the dietary share of UPF in Brazil.
Keywords: Food Consumption; Ultra-processed Food; Diet Surveys, methods; Surveys and Questionnaires; Validation Study
INTRODUCTION
According to the Nova food classification system1, ultra-processed foods (UPF) are industrial formulations of food-derived substances that contain little or no whole food, often including flavorings, colorings, emulsifiers, and other cosmetic additives to provide them palatability or even hyperpalatability. Most ingredients and processes used to manufacture these foods are exclusively used by the food industry2.
Nutritional surveys conducted with probabilistic samples from the population of several countries show that UPF intake, measured by the percentage of total energy intake related to these foods, is strongly and inversely related to the nutritional quality of the diet3. Systematic reviews of well-conducted and large cohort studies show that the percentage of total energy intake from UPF is directly associated with the risk of chronic non-communicable diseases such as obesity, diabetes, hypertension, dyslipidemias, cardiovascular and cerebrovascular diseases, cancer in general, breast cancer, and depression, as well as with premature deaths from any cause4.
Studies addressing the association between UPF consumption and nutritional quality of the diet or risk of chronic diseases measured UPF dietary contribution using data-collection tools that require experienced interviewers and time and disposition from interviewees, as 24-hour dietary recalls; or tools that require participants with high education level, time, and disposition, as in food records or food frequency questionnaires11. Given the complexity of these data-collection instruments, the intake of ultra-processed foods in many populations is still unknown and, even more, difficult to be monitored, thus hampering the formulation and evaluation of public policies aimed at reducing UPF consumption.
To monitor the consumption of UPF by the Brazilian adult population, and as part of the Surveillance System for Risk and Protective Factors for Chronic Diseases by Telephone Survey (VIGITEL), authors of this article developed a simplified instrument addressing questions on the previous-day dietary intake of a list of 13 subgroups of ultra-processed foods (answered with “yes” or “no”). Part of the VIGITEL’s annual questionnaire since 201812, the instrument enables the calculation of a score of UPF consumption ranging from zero to thirteen13 – equivalent to the number of subgroups consumed in the previous day by the interviewees. A study conducted with a convenience sample of 150 participants showed a good agreement between the score and the dietary share of UPF14.
As part of the development of the NutriNet Brasil cohort study, we built a second simplified instrument to evaluate UPF consumption that waives interviewers’ participation, conducted in mobile phones, tablets, or computers15 – the Nova screener for the consumption of ultra-processed foods. This instrument provides the Nova score for the consumption of ultra-processed foods, whose description and ability to reflect the dietary share of UPF in Brazil are presented below.
METHODS
Sample
This study was conducted in the city of São Paulo with a convenience sample of 300 adults aged 18 years or older, users of two health centers of the Universidade de São Paulo (USP) and employees and students at USP.
Data Collection
Two nutritionists trained by one of the authors of this article (CSC) collected the study data between September and November 2019. All participants were informed about the study purpose and invited to participate. After agreeing to participate by signing the consent form, the participants informed their gender, age, and education level. Then, using a tablet and without the nutritionist aid, participants answered to the Nova screener for the consumption of UPF, checking all items within a list that had been consumed the day before (checkbox format). The average time spent to complete the answers was three minutes. After completion, the nutritionist conducted a 24-hour dietary recall (24-hR), where participants informed all foods and the amount they had consumed the day before. The dietary recall took on average 30 minutes.
Nova screener for the consumption of ultra-processed foods
As in the instrument employed by the VIGITEL system, the Nova screener for the consumption of ultra-processed foods was developed to include UPF subgroups with greater participation in the daily energy intake, estimated by the national food consumption survey conducted by the 2008–2009 Household Budget Survey(POF) of the Brazilian Institute of Geography and Statistics (IBGE)16. After unfolding some of the 13 subgroups of the original instrument, the Nova screener presents a list of 23 subgroups of UPF. The questions addressing the intake of each of these subgroups are presented on three categories: beverages (six subgroups); products that replace or accompany meals (ten subgroups); and products often consumed as snacks (seven subgroups), as shown in Figure 1. Questions were uploaded into the tablet with the Epicollect5 Data Collectiona software, which stores participants’ answers as a database.
Figure 1. Nova screener for the consumption of ultra-processed foods on the Epicollect5 Data Collection® platform.
24-hour Dietary Recall (24-hR)
The 24-hR was applied by the nutritionists, using the five-stage multiple-pass method17. First, participants report, quick and uninterruptedly, all foods and beverages consumed. Then, the nutritionist asks for other foods or beverages that the interviewee might have forgotten to report, based on a list. The participant is then queried about the type, time, and place of each meal, followed by the provision of details such as preparation mode, origin, quantities, homemade measures and sizes, as well as other foods addition (e.g., sugar). To finalize, the interviewer lists the entire report to the interviewee, reviewing and stimulating the report of foods possibly forgotten and/or omitted.
This study was approved by the Research Ethics Committee of the School of Public Health of the Universidade de São Paulo (CAAE: 83221317.5.0000.5421; opinion no.: 2.517.894) and all participants signed the informed consent form.
Data Analysis
The Nova score of each participant was calculated from the sum of UPF subgroups reported among the 23 listed, thus ranging from 0 to 23. To estimate the dietary share of UPF in the previous-day food consumption, each item reported in the 24-hR was initially classified into ultra-processed or non-ultra-processed, according to the Nova classification1,2. Then, the consumed quantity of each item, reported in homemade measures, was transformed into grams and converted into calories using the Composition Table of Foods Consumed in Brazil18. Finally, we calculated the total calories consumed, the calories from UPF, and the percentage of total calories from UPF.
To assess the association between the Nova score and the dietary share of UPF, we examined the variation in the average percentage of calories from UPF according to the score variation, expressed continuously and also at intervals corresponding to fifths of its distribution. In both cases, linear regression models were used to test the linear trend. Then, the degree of agreement in participants’ classification according to the fifths of the distribution of the percentage of calories from UPF and the fifths of the distribution of the Nova score was evaluated by calculating the prevalence-adjusted and bias-adjusted kappa (Pabak) index19. Values greater than 0.80 indicate an almost perfect agreement; between 0.61 and 0.80, a substantial agreement; between 0.41 and 0.60, moderate; between 0.21 and 0.40, fair; and equal to or less than 0.20, slight20. We also compared the variation in the prevalence of relatively high consumption of UPF according to age group, determined by two alternative criteria: consumption equivalent to that observed in the upper fifth of the distribution of Nova score; and consumption equivalent to that observed in the upper fifth of the distribution of the percentage of calories from UPF.
The analyses were performed using the Stata® 16.1 software, and the Pabak index was calculated using the WINPEPI (PEPI-for-Windows) 11.65 softwareb.
RESULTS
Among the 300 volunteers interviewed, most were female (71.3%), aged between 25 and 59 years (72.3%), who completed secondary education, or completed or are attending tertiary education (80.7%), as shown in Table 1.
Table 1. Distribution according to sociodemographic variables of adult users of health centers and employees and students at the Universidade de São Paulo (n = 300). São Paulo, Brazil, 2019.
| Variables | n | % |
|---|---|---|
| Gender | ||
| Male | 86 | 28,7 |
| Female | 214 | 71,3 |
| Age (years) | ||
| 18–24 | 34 | 11,4 |
| 25–34 | 79 | 26,3 |
| 35–44 | 71 | 23,7 |
| 45–59 | 67 | 22,3 |
| 60+ | 49 | 16,3 |
| Education level | ||
| Some secondary education | 58 | 19,3 |
| Secondary education | 114 | 38,0 |
| Some tertiary education or tertiary education | 128 | 42,7 |
Table 2 describes the consumption frequency of each subgroup included in the Nova screener for the consumption of UPF on the day prior to the interview. About one in every three participants reported having consumed margarine (38.0%), loaf, hot dogs, or hamburger bread (33.0%), and regular or diet soda (30.7%). Between 15% and 20% reported having consumed biscuits with or without filling (19.7%), packaged snacks, shoestring potatoes or crackers (16.3%), and chocolate bar or bonbon (15.0%). Less than 15% of the interviewees reported consuming food from the other subgroups on the day prior to the interview.
Table 2. Consumption frequency (%) of foods included in the Nova screener for the consumption of ultra-processed foods on the day prior the interview. Adult users of health centers and employees and students at the Universidade de São Paulo (n = 300). São Paulo, Brazil, 2019.
| Foods | % |
|---|---|
| Margarine | 38.0 |
| Loaf, hot dog, or hamburger bread | 33.0 |
| Regular or diet soda | 30.7 |
| Biscuits with or without filling | 19.7 |
| Packaged snacks, shoestring potatoes or crackers | 16.3 |
| Chocolate bar or bonbon | 15.0 |
| Ham, salami or mortadella | 14.7 |
| Sausage, hamburger or nuggets | 13.3 |
| Fruit- or chocolate-flavored yogurt | 12.7 |
| Canned or bottled fruit juice (Del Valle-type) | 12.7 |
| Powdered drink mix (Tang-type) | 12.0 |
| Mayonnaise, ketchup or mustard | 11.7 |
| Ice cream or popsicle | 10.3 |
| Chocolate drink (Nescau-type) | 8.3 |
| French fries, either frozen or from restaurant chains such as McDonald’s | 5.3 |
| Instant noodles (Miojo-type) or packaged soup | 5.3 |
| Tea-based beverage (ice tea-type) | 4.0 |
| Pizza, either frozen or from restaurant chains, such as Pizza Hut or Domino’s | 3.7 |
| Frozen lasagna or other frozen ready-made meals | 3.3 |
| Ready-made salad sauce | 3.0 |
| Packaged cake | 2.7 |
| Cereal bar | 2.7 |
| Breakfast cereal (Sucrilhos-type) | 1.7 |
Table 3 describes the distribution of the Nova score for the consumption of UPF, which is equivalent to the number of subgroups consumed on the day before the interview. Scores ranged from 0 to 9, but 1 (19.7%), 2 (20.3%), 3 (19.3%), and 4 (14.0%) were the most common; 9.0% of participants reached null scores and 17.7% equal to or higher than 5. As shown in Table 3, the average percentage of dietary share of UPF, calculated based on the 24-hour dietary recall, increases linearly and significantly with the increase in the UPF consumption score.
Table 3. Dietary share of ultra-processed foods calculated by the 24-hour food recall according to the Nova score. Adult users of health centers and employees and students at the Universidade de São Paulo (n = 300). São Paulo, Brazil, 2019.
| Nova score for the consumption of ultra-processed foods | Sample n (%) | Dietary share of ultra-processed foods (% of total energy) Average (95%CI) |
|---|---|---|
| 0 | 27 (9.0) | 9.4 (2.3–16.6) |
| 1 | 59 (19.7) | 23.8 (19.0–28.6) |
| 2 | 61 (20.3) | 31.6 (26.9–36.4) |
| 3 | 58 (19.3) | 31.1 (26.2–35.9) |
| 4 | 42 (14.0) | 35.6 (29.8–41.3) |
| 5 | 23 (7.7) | 37.5 (29.8–45.2) |
| 6 | 17 (5.7) | 54.1 (45.2–63.1) |
| 7 | 8 (2.7) | 47.0 (34.0–60.1) |
| 8 | 1 (0.3) | 27.5 (-9.4–64.5) |
| 9 | 4 (1.3) | 35.6 (17.1–54.0)a |
| 0–1 | 86 (28.7) | 19.3 (15.2–23.4) |
| 2 | 61 (20.3) | 31.6 (26.8–36.5) |
| 3 | 58 (19.3) | 31.1 (26.1–36.1) |
| 4 | 42 (14.0) | 35.6 (29.7–41.4) |
| 5 or + | 53 (17.7) | 43.9 (38.7–49.1)a |
95%CI: 95% confidence interval
a P-value for linear trend < 0.001.
Table 4 shows that participants’ distribution, considering their classification based on the fifths of the dietary share of UPF (calculated by the 24-hR) and of the Nova score (0–1, 2, 3, 4 and ≥ 5) indicates substantial agreement between the two criteria (Pabak index of 0.67).
Table 4. Distribution (%) according to the fifths of the dietary share of ultra-processed foods and (approximate) fifths of the Nova score for the consumption of ultra-processed foods. Adult users of health centers and employees and students at the Universidade de São Paulo (n = 300). São Paulo, Brazil, 2019.
| Fifths of the Nova score for the consumption of ultra-processed foods | Fifths of the dietary share of ultra-processed foods (% of total calories) | 0–1 | 2 | 3 | 4 | 5 or + | Total |
|---|---|---|---|---|---|---|---|
| Q1 (≤ 11.0) | 13.0 | 2.0 | 2.3 | 2.0 | 0.7 | 20.0 | |
| Q2 (11.1–20.4) | 6.0 | 5.0 | 4.0 | 2.7 | 2.3 | 20.0 | |
| Q3 (20.5–34.8) | 4.0 | 4.7 | 6.0 | 2.3 | 3.0 | 20.0 | |
| Q4 (34.9–49.5) | 3.3 | 5.0 | 3.7 | 3.0 | 5.0 | 20.0 | |
| Q5 (≥ 49.6) | 2.3 | 3.7 | 3.3 | 4.0 | 6.7 | 20.0 | |
| Total | 28.6 | 20.4 | 19.3 | 14.0 | 17.7 | 100.0 | |
Note: Pabak index (prevalence-adjusted bias-adjusted Kappa) = 0.67.
Figure 2 shows the variation in the prevalence of relatively high consumption of UPF according to age groups, defined based on the consumption observed, alternatively, in the upper fifth (approximate) of the distribution of the Nova score (≥ 5) and of the distribution of UPF participation in the total caloric intake (≥ 49.6% of the total calories). Using these two criteria, we verified that the prevalence of relatively high consumption of UPF linearly decreases with increasing age (p = 0.038 and p = 0.001, respectively).
Figure 2. Variation in the prevalence (%) of high consumption of ultra-processed foods according to age group based on two criteria. Adult users of health centers and employees and students at the Universidade de São Paulo (n = 300). São Paulo, Brazil, 2019.
a P-value for linear trend = 0.038.
b P-value for linear trend = 0.001.
DISCUSSION
Our results indicate that the Nova score for the consumption of ultra-processed foods, obtained with a 3-minute electronic self-report questionnaire, was directly and linearly associated with the percentage of total energy intake from UPF, obtained with a 24-hour dietary recall applied by a trained nutritionist in about 30 minutes. We also found a substantial agreement in participants’ classification according to the fifths of the distribution of scores and the fifths of the dietary share of UPF and an inverse association between age and the relatively high frequency of UPF consumption (upper fifth of the distribution) for both score and dietary share of UPF.
Despite being conducted with a convenience sample, the distribution of participants’ consumption frequency of UPF subgroups found in our study was quite similar to that estimated by the VIGITEL system for the adult Brazilian population of the capitals of the 27 units of the Federation13. In both scenarios, the three subgroups more frequently consumed on the day prior to the interview were margarine (38.0% in our study and 42.6% in VIGITEL’s), loaf bread and similar (33.0% and 32.8%), and soft drinks (30.7% and 27.7%). The score distribution in the convenience sample was also similar to that observed in the VIGITEL sample, with the upper fifth of the score distribution containing scores ≥ 5 in both cases13.
Added to the findings reported by a similar study on the version of the score used by the VIGITEL system14, our results indicate the feasibility in monitoring the participation of UPF in the dietary intake in an effective, quick, and practical manner. In Brazil and in several countries worldwide, such participation has been associated with the sharp deterioration in diet quality3 and the increased risk of obesity, diabetes, cardiovascular diseases, and several other chronic non-communicable diseases of great epidemiological relevance4. As part of the Innovative Methods and Metrics for Agriculture and Nutrition Actions program (IMMANA – based at the London School of Hygiene and Tropical Medicinec), the Nova screener for the consumption of ultra-processed foods is being adapted for use in India, Senegal, and Ecuador, which will enable other countries to study the performance of the Nova score.
The main limitation of this study is the impossibility of extrapolating its results to Brazilian populations with low education level, since four fifths of the participants had completed at least the secondary education. We also did not evaluate the scores of men and women and in specific age groups, representing another limitation. Our sample size (n = 300) was adequate to identify even weak correlations between two methods classifications21, but it did not allow analyses stratification according to sociodemographic strata. Scores performance, according to gender, age, and education level, will be soon evaluated based on data collected from a subsample by quotas of the NutriNet Brasil cohort (n = 900), which monitors more than 90,000 people from all regions of the countryd.
CONCLUSION
The Nova score for the consumption of ultra-processed foods, obtained in a quick and practical manner using an electronic self-report questionnaire, shows a good potential in reflecting the dietary share of this food group in Brazil.
Funding Statement
Funding: NutriNet-Brasil Cohort research project – Diet and Chronic Non-Communicable Diseases – CNPq (process no. 408365/2017-0) and Associação Samaritano/Umane; Research fellowship - FAPESP (process no. 2019/06852-5).
Epicollect5 [software]. Centre for Genomic Pathogen Surveillance; 2019 [cited 7 feb 2021]. Available at: https://five.epicollect.net/
Abramson JH. WinPepiSetup.exe. Version 11.65. London, UK: Brixton Health; 2016 [cited 07 feb 2021]. Available at: http://www.brixtonhealth.com/pepi4windows.html
Innovative Methods and Metrics for Agriculture and Nutrition Actions (IMMANA) [Internet]. Agriculture and Nutrition Actions (ANH) Academy; 2020. [cited feb. 7, 2021]. Available at: https://www.anh-academy.org/immana
Universidade de São Paulo, Faculdade de Saúde Pública, Núcleo de Pesquisas Epidemiológicas em Nutrição e Saúde. NutriNet Brasil. São Paulo: Nupens; c2018-2021 [cited 7 feb 2021]. Available at:https://nutrinetbrasil.fsp.usp.br
Funding: NutriNet-Brasil Cohort research project – Diet and Chronic Non-Communicable Diseases – CNPq (process no. 408365/2017-0) and Associação Samaritano/Umane; Research fellowship - FAPESP (process no. 2019/06852-5).
REFERENCES
- 1.1. Monteiro CA, Cannon G, Moubarac JC, Levy RB, Louzada MLC, Jaime PC. The UN Decade of Nutrition, the NOVA food classification and the trouble with ultra-processing. Public Health Nutr. 2018;21(1):5-17. 10.1017/S1368980017000234 [DOI] [PMC free article] [PubMed]; Monteiro CA, Cannon G, Moubarac JC, Levy RB, Louzada MLC, Jaime PC. The UN Decade of Nutrition, the NOVA food classification and the trouble with ultra-processing. Public Health Nutr. 2018;21(1):5–17. doi: 10.1017/S1368980017000234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.2. Monteiro CA, Cannon G, Levy RB, Moubarac JC, Louzada ML, Rauber F, et al. Ultra-processed foods: what they are and how to identify them. Public Health Nutr. 2019;22(5):936-41. 10.1017/S1368980018003762 [DOI] [PMC free article] [PubMed]; Monteiro CA, Cannon G, Levy RB, Moubarac JC, Louzada ML, Rauber F, et al. Ultra-processed foods: what they are and how to identify them. Public Health Nutr. 2019;22(5):936–941. doi: 10.1017/S1368980018003762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.3. Monteiro CA, Cannon G, Lawrence M, Louzada MLC, Machado PP. Ultra-processed foods, diet quality, and health using the NOVA classification system. Rome: FAO; 2019.; Monteiro CA, Cannon G, Lawrence M, Louzada MLC, Machado PP. Ultra-processed foods, diet quality, and health using the NOVA classification system. Rome: FAO; 2019. [Google Scholar]
- 4.4. Askari M, Heshmati J, Shahinfar H, Tripathi N, Daneshzad E. Ultra-processed food and the risk of overweight and obesity: a systematic review and meta-analysis of observational studies. Int J Obes (Lond). 2020;44(10):2080-91. 10.1038/s41366-020-00650-z [DOI] [PubMed]; Askari M, Heshmati J, Shahinfar H, Tripathi N, Daneshzad E. Ultra-processed food and the risk of overweight and obesity: a systematic review and meta-analysis of observational studies. Int J Obes (Lond) 2020;44(10):2080–2091. doi: 10.1038/s41366-020-00650-z. [DOI] [PubMed] [Google Scholar]
- 5.5. Chen X, Zhang Z, Yang H, Qiu P, Wang H, Wang F, et al. Consumption of ultra-processed foods and health outcomes: a systematic review of epidemiological studies. Nutr J. 2020;19(1):86. 10.1186/s12937-020-00604-1 [DOI] [PMC free article] [PubMed]; Chen X, Zhang Z, Yang H, Qiu P, Wang H, Wang F, et al. Consumption of ultra-processed foods and health outcomes: a systematic review of epidemiological studies. 86Nutr J. 2020;19(1) doi: 10.1186/s12937-020-00604-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.6. Leonie E, Machado P, Zinöcker M, Baker P, Lawrence M. Ultra-Processed foods and health outcomes: a narrative review. Nutrients. 2020;12(7):1955. 10.3390/nu12071955 [DOI] [PMC free article] [PubMed]; Leonie E, Machado P, Zinöcker M, Baker P, Lawrence M. Ultra-Processed foods and health outcomes: a narrative review. Nutrients. 2020;12(7):1955. doi: 10.3390/nu12071955. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.7. Meneguelli TS, Hinkelmann JV, Hermsdorff HHM, Zulet MA, Martínez JA, Bressan J. Food consumption by degree of processing and cardiometabolic risk: a systematic review. Int J Food Sci Nutr. 2020;71(6):678-92. 10.1080/09637486.2020.1725961 [DOI] [PubMed]; Meneguelli TS, Hinkelmann JV, Hermsdorff HHM, Zulet MA, Martínez JA, Bressan J. Food consumption by degree of processing and cardiometabolic risk: a systematic review. Int J Food Sci Nutr. 2020;71(6):678–692. doi: 10.1080/09637486.2020.1725961. [DOI] [PubMed] [Google Scholar]
- 8.8. Pagliai G, Dinu M, Madarena MP, Bonaccio M, Iacoviello L, Sofi F. Consumption of ultra-processed foods and health status: a systematic review and meta-analysis. Br J Nutr. 2020;125(3):308-18. 10.1017/S0007114520002688 [DOI] [PMC free article] [PubMed]; Pagliai G, Dinu M, Madarena MP, Bonaccio M, Iacoviello L, Sofi F. Consumption of ultra-processed foods and health status: a systematic review and meta-analysis. Br J Nutr. 2020;125(3):308–318. doi: 10.1017/S0007114520002688. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.9. Santos FS, Dias MS, Mintem GC, Oliveira IO, Gigante DP. Food processing and cardiometabolic risk factors: a systematic review. Rev Saude Publica 2020;54:70. 10.11606/s1518-8787.2020054001704 [DOI] [PMC free article] [PubMed]; Santos FS, Dias MS, Mintem GC, Oliveira IO, Gigante DP. Food processing and cardiometabolic risk factors: a systematic review. Rev Saude Publica. 2020;54:70. doi: 10.11606/s1518-8787.2020054001704. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.10. Lane MM, Davis JA, Beattie S, Gómez‐Donoso C, Loughman A, O’Neil A, et al. Ultraprocessed food and chronic noncommunicable diseases: a systematic review and meta‐analysis of 43 observational studies. Obes Rev. 2021;22(3)e13146. 10.1111/obr.13146 [DOI] [PubMed]; Lane MM, Davis JA, Beattie S, Gómez‐Donoso C, Loughman A, O’Neil A, et al. Ultraprocessed food and chronic noncommunicable diseases: a systematic review and meta‐analysis of 43 observational studies. Obes Rev. 2021;22(3):e13146. doi: 10.1111/obr.13146. [DOI] [PubMed] [Google Scholar]
- 11.11. Marchioni DML, Gorgulho BM, Steluti J. Consumo alimentar: guia para avaliação. Barueri, SP: Manole; 2019.; Marchioni DML, Gorgulho BM, Steluti J. Consumo alimentar: guia para avaliação. Barueri, SP: Manole; 2019. [Google Scholar]
- 12.12. Ministério da Saúde (BR), Secretaria de Vigilância em Saúde, Departamento de Análise em Saúde e Vigilância de Doenças Não Transmissíveis. Vigitel Brasil 2018: vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico: estimativas sobre frequência e distribuição sociodemográfica de fatores de risco e proteção para doenças crônicas nas capitais dos 26 estados brasileiros e no Distrito Federal em 2018. Brasília, DF; 2019. 132p.; Ministério da Saúde (BR) Secretaria de Vigilância em Saúde. Departamento de Análise em Saúde e Vigilância de Doenças Não Transmissíveis . Vigitel Brasil 2018: vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico: estimativas sobre frequência e distribuição sociodemográfica de fatores de risco e proteção para doenças crônicas nas capitais dos 26 estados brasileiros e no Distrito Federal em 2018. Brasília, DF: 2019. 132 [Google Scholar]
- 13.13. Costa CS, Sattamini IF, Martínez Steele E, Louzada MLC, Claro RM, Monteiro CA. Consumo de alimentos ultraprocessados e associação com fatores sociodemográficos na população adulta das 27 capitais brasileiras Rev Saude Publica. 2021;55. No prelo.; Costa CS, Sattamini IF, Martínez Steele E, Louzada MLC, Claro RM, Monteiro CA. Consumo de alimentos ultraprocessados e associação com fatores sociodemográficos na população adulta das 27 capitais brasileiras. 55Rev Saude Publica. 2021 No prelo. [Google Scholar]
- 14.14. Sattamini IF. Instrumentos de avaliação da qualidade de dietas: desenvolvimento, adaptação e validação no Brasil [tese]. São Paulo: Faculdade de Saúde Pública da Universidade de São Paulo; 2019.; Sattamini IF. Instrumentos de avaliação da qualidade de dietas: desenvolvimento, adaptação e validação no Brasil. São Paulo: Faculdade de Saúde Pública da Universidade de São Paulo; 2019. tese. [Google Scholar]
- 15.15. Martinez Steele E, Rauber F, Costa CS, Leite MA, Gabe KT, Louzada ML, et al. Mudanças alimentares na coorte NutriNet Brasil durante a pandemia de covid-19. Rev Saude Publica. 2020;54:91. 10.11606/s1518-8787.2020054002950 [DOI] [PMC free article] [PubMed]; Martinez Steele E, Rauber F, Costa CS, Leite MA, Gabe KT, Louzada ML, et al. Mudanças alimentares na coorte NutriNet Brasil durante a pandemia de covid-19. Rev Saude Publica. 2020;54:91. doi: 10.11606/s1518-8787.2020054002950. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.16. Instituto Brasileiro de Geografia e Estatística. Pesquisa de Orçamentos Familiares: 2008-2009: análise do consumo alimentar pessoal no Brasil. Rio de Janeiro: IBGE; 2011 [cited 2021 Feb 7]. Available from: https://biblioteca.ibge.gov.br/visualizacao/livros/liv50063.pdf ; Instituto Brasileiro de Geografia e Estatística . Pesquisa de Orçamentos Familiares: 2008-2009: análise do consumo alimentar pessoal no Brasil. Rio de Janeiro: IBGE; 2011. [cited 2021 Feb 7]. https://biblioteca.ibge.gov.br/visualizacao/livros/liv50063.pdf [Google Scholar]
- 17.17. Moshfegh AJ, Rhodes DG, Baer DJ, Murayi T, Clemens JC, Rumpler WV, et al. The US Department of Agriculture Automated Multiple-Pass Method reduces bias in the collection of energy intakes. Am J Clin Nutr. 2008;88:324-32. 10.1093/ajcn/88.2.324 [DOI] [PubMed]; Moshfegh AJ, Rhodes DG, Baer DJ, Murayi T, Clemens JC, Rumpler WV, et al. The US Department of Agriculture Automated Multiple-Pass Method reduces bias in the collection of energy intakes. Am J Clin Nutr. 2008;88:324–332. doi: 10.1093/ajcn/88.2.324. [DOI] [PubMed] [Google Scholar]
- 18.18. Instituto Brasileiro de Geografia e Estatística. Pesquisa de Orçamentos Familiares 2008-2009: tabelas de composição nutricional dos alimentos consumidos no Brasil. Rio de Janeiro: IBGE; 2011 [cited 2021 Feb 7]. Available from: https://biblioteca.ibge.gov.br/visualizacao/livros/liv50002.pdf ; Instituto Brasileiro de Geografia e Estatística . Pesquisa de Orçamentos Familiares 2008-2009: tabelas de composição nutricional dos alimentos consumidos no Brasil. Rio de Janeiro: IBGE; 2011. [cited 2021 Feb 7]. https://biblioteca.ibge.gov.br/visualizacao/livros/liv50002.pdf [Google Scholar]
- 19.19. Byrt T, Bishop J, Carlin JB. Bias, prevalence and kappa. J Clin Epidemiol. 1993;46(5):423-9. 10.1016/0895-4356(93)90018-v [DOI] [PubMed]; Byrt T, Bishop J, Carlin JB. Bias, prevalence and kappa. J Clin Epidemiol. 1993;46(5):423–429. doi: 10.1016/0895-4356(93)90018-v. [DOI] [PubMed] [Google Scholar]
- 20.20. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159-74. [PubMed]; Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159–174. [PubMed] [Google Scholar]
- 21.21. Cade J, Thompson R, Burley V, Warm D. Development, validation and utilisation of food-frequency questionnaires: a review. Public Health Nutr. 2002;5(4):567-87. 10.1079/PHN2001318 [DOI] [PubMed]; Cade J, Thompson R, Burley V, Warm D. Development, validation and utilisation of food-frequency questionnaires: a review. Public Health Nutr. 2002;5(4):567–587. doi: 10.1079/PHN2001318. [DOI] [PubMed] [Google Scholar]
