The increasing consumption of industrially processed convenience foods, soft drinks, and fast foods has been associated with a rise in non-communicable diseases, overweight and obesity.
This building block explores the concept of ultra-processed food: how it has been defined, and differing views as to whether it is a useful way of thinking about food and its relation to health and wider sustainability concerns.
Last update: 2 July, 2019
Ultra-processed food (UPF): Sometimes used loosely to refer to snacks and fast foods, UPF generally refers to one of the four categories of the NOVA food classification (see below). NOVA describes UPFs as ‘industrial formulations’ of food products, typically mass-produced, that contain few ‘natural’ ingredients. Advocates of NOVA point out that UPFs consist of many additives and food-derived ingredients such as whey, protein isolates and invert sugar, which are produced and combined through processes that are uncommon in domestic kitchens. They understand these foods to be designed so as to be so appealing that they displace the consumption of healthier, less processed foods, thereby generating high profits for their manufacturers. Foods in the UPF category include biscuits, mass-produced buns and breads, sweetened cereals, margarines and spreads, packaged snacks, ice cream, flavoured yogurts, soft drinks, powdered meals, ready-made meals, and instant sauces and stocks. Proponents of the concept have argued that the consumption of UPF is the primary driver of the global ‘pandemic’ of overweight and obesity, while contributing to non-communicable diseases such as metabolic syndrome and certain cancers. It has been argued that the production and consumption of UPFs undermine social and environmental sustainability while perpetuating unequal power dynamics in the food system. Opponents of the concept have contested these claims. They argue that the concept is imprecise and groups together foods with different nutritional characteristics.
NOVA classification: is a system of food classification created by a team of nutrition and health researchers at the University of São Paulo led by Carlos Monteiro. NOVA categorises food products by the ‘extent’ and ‘purpose’ of food processing on the grounds that today this is the main determinant of a food’s nutritional and environmental characteristics. NOVA has four categories: minimally processed food; processed culinary ingredients; processed foods; and ultra-processed foods. Introduced as a framework to measure the impacts of processed foods on human health, NOVA has also been promoted as an alternative to traditional government-approved dietary guidance such as the US MyPlate, the UK Eatwell plate or the Chinese Food Pagoda. While criticised, NOVA is increasingly used as a framework in nutrition science, especially in nutritional epidemiology. The dietary guidelines of Brazil (2014), Uruguay (2016), Ecuador (2018) and Peru (2019) and some reports from the PAHO-WHO have drawn upon the NOVA classification while the French (2019) dietary recommendations advise to reduce the consumption of ‘ultra-transformed’ foods.
Food processing: Defined by stakeholders in many different ways, food processing broadly refers to modifications made to raw food ingredients after they have been gathered or harvested and before they are consumed or prepared for consumption in a kitchen. Examples of food processing include the pickling of vegetables, smoking or mincing of meat, pasteurisation of milk, milling of wheat, and the hydrogenation of oils. By this definition, most foods are processed in some way (e.g. to improve their taste, extend their shelf-life, make them edible or increase their nutritional value), but there are growing concerns over the health impacts of industrially processed foods.
Minimally processed food: is a category of foods in the NOVA classification that have not been processed (e.g. fresh fruits and vegetables) or that have been modified only minimally by processes such as drying, roasting, boiling, freezing, and the removal of inedible parts. Advocates of NOVA understand minimally processed foods to be processed to extend their lifetime, enable their storage, make them easier to prepare, or to increase the number of ways they can be consumed. Examples include fresh meat and vegetables, whole wheat grains, wholemeal flour, ground coffee and pasteurised milk.
Processed culinary ingredients: is a category of food ingredients in the NOVA classification that result from the further treatment of unprocessed and minimally processed foods by processes such as pressing, refining, grinding, milling, and spray drying. Examples include salt, spices, cane sugar, flour, honey and olive oil. Processed culinary ingredients are added to other foods in the preparation of dishes. Advocates of NOVA understand their purpose to be enabling the preparation of more varied dishes from minimally processed foods and improving the taste and appearance of such dishes. Processed culinary ingredients are usually not consumed on their own.
Processed food: In public understanding, the concept of processed food is often used loosely to refer to mass-produced ready-to-eat foods such as instant flavoured noodles and soda drinks. The concept, however, also refers to one of the four categories of the NOVA classification, which classifies many of these foods as ultra-processed foods. Within NOVA’s processed food category are minimally processed foods to which one or more processed culinary ingredient has been added, and which have been further modified by processes such as smoking, salting and canning. Advocates of NOVA understand processed foods to be produced primarily to increase the durability of minimally processed foods and to enhance the taste and appearance of such foods. Examples of processed foods include freshly made breads, pickled vegetables, salted nuts, smoked meats and canned fish.
Globally, diets increasingly include more mass-produced products such as convenience foods, soft drinks, and fast foods (see our building block on the nutrition transition)1 . These foods are now often referred to as ‘ultra-processed foods’ (UPFs).
The UPF concept was coined in 2009 by the Brazilian nutrition and health researcher Carlos Monteiro to refer to a specific group of industrially processed foods that are often energy dense and high in fat, salt, sugar and additives, while lacking dietary fibre and micronutrients 2 . Monteiro and colleagues argue that since the 1980s, food systems have been increasingly dominated by these foods (see Figure 1)3 ,4 . They understand the increased consumption of UPFs to be a leading cause of the global ‘pandemic ’ of overweight, obesity and related non-communicable diseases , while their production and consumption is also thought to undermine sustainability and general human well-being2 ,3 .
The UPF concept is gaining traction in some academic and policy quarters and is increasingly used by NGOs and the media. The concept is used in different ways and the research on UPF is sometimes misinterpreted. Some question the overall usefulness of the concept and others suggest that the concept can be useful for health education or possibly for other purposes when further research and clarification are provided.
2. What is ultra-processed food?
Sometimes used loosely by commentators to denote food products such as fast foods and ready-made meals, UPF is also a category of the NOVA food classification developed by Monteiro and colleagues4 ,11 .
NOVA classifies food into four categories (see Figure 2) that are distinguished by the extent and purpose of food processing11 :
Healthy dietary patterns, advocates of NOVA argue, consist of meals that are based on minimally processed foods and small quantities of processed culinary ingredients . The use of processed food is to be limited and UPFs avoided12 .
As to the purpose of food processing, it is argued that minimally processed foods are processed to extend their lifetime, enable their storage, make them easier to prepare, or increase the variety of their uses4 ,11 ,12 . Processed culinary ingredients are made to aid the preparation of a variety of dishes from minimally processed foods and to improve the taste and appearance of these dishes4,11,12. Processed foods, produced by adding processed culinary ingredients to minimally processed foods, are produced to increase the durability of minimally processed foods and to enhance their taste and appearance4 ,11 ,12 . UPFs, in contrast, are thought to be made primarily to generate high profits for their manufacturers11 ,12 . These foods are understood to be generally cheap to make, attractive to buy, convenient to consume, and thereby likely to displace the consumption of other, less processed, foods.
Regarding the extent of processing: the argument here is that while minimally processed foods, processed culinary ingredients, and processed foods are processed in varying degrees, they are all produced by applying at most a few processes – for example drying, salting, fermenting, canning or removing inedible parts – to raw food ingredients. These processes, moreover, are relatively uncomplicated; they are not based on advanced and large-scale technological processes and many of them are common in domestic kitchens.
UPFs, instead, are thought to be produced by applying sequences of processes to often already heavily processed food ingredients4 ,11 ,13 . UPF-ingredients such as whey, gluten, invert sugar, and reconstituted meat are mostly derived by heavily processing commodity crops (for example corn or wheat) and animals12 . UPFs often contain much fat, salt, and sugar but also additives such as flavour enhancers and emulsifiers as well as added micronutrients, all of which are not found in the processed culinary ingredients-category and are produced using advanced chemical techniques. Some of the processes that are used to produce UPFs, such as the hydrolysation of proteins or the manufacture of aspartame, cannot be undertaken in domestic kitchens.
3. UPFs, nutrition and health
3.1 How do UPFs impact upon health?
Proponents of the NOVA classification argue that UPFs impact upon health in many ways3 ,14 ,15 . They sometimes criticise what has been termed ‘nutritionism ’16 ,17 , which refers to the idea that the healthfulness (or otherwise) of a food is determined by its constituent nutrients. This approach is seen to be prevalent in nutrition science, and to shape mainstream dietary recommendations and food industry policies.
Proponents of the UPF concept point out that UPFs generally have a poor nutrient profile as defined by the nutritionist approach (they are often high in sugar, salt, fat, or additive contents while lacking dietary fibre and micronutrients). But they also argue that a focus solely on these aspects is inadequate as it ignores the health impacts of how foods are designed, produced, distributed, advertised, and consumed. The NOVA framework, accordingly, is not a nutrient profiling model that is based on maximum and minimum intake levels of nutrients, but a food classification that considers the function of a food in the wider context of meals and diets.
- UPFs are often ‘hyperpalatable’. Unlike food from the other NOVA categories, they have been designed to taste and look attractive; compare water or tea, for example, with a cola drink.
- Because they are often high in sugar, salt, colourants and flavourings, nutritionally poor UPFs may appeal to consumers more than healthier meals and dishes.
- UPFs often come in large portion sizes.
- They are generally convenient (requiring little preparation) and readily available. People may choose ready-to-eat or ready-to-heat UPFs over foods that are healthier but that require more preparation.
- They tend to be heavily marketed – sometimes using health claims that mask their unhealthy qualities (e.g. a soda beverage with added vitamins can still be rich in sugar).
- As a result of their dependence on UPFs, people may lose the knowledge, skills and interest to prepare healthy dishes themselves, leading to a situation where they become increasingly reliant on UPFs.
In current research on UPF, most of these reasons are not directly investigated but are supported by reference to other research.
3.2 Research on the health impacts of UPF consumption
To date, most research on UPFs consists of epidemiological studies that use the NOVA framework to study the associations between particular dietary patterns and overall health outcomes. Many of these studies have been conducted by Monteiro and his colleagues at the University of São Paulo, but the NOVA framework is increasingly being adopted by other research teams.
- UPFs account for a major share of total food energy (see Figure 2). This share has been increasing since the 1980s.
- Foods in the UPF category have, on average, higher fat, salt and sugar levels and fewer micronutrients compared to foods on average found in the other NOVA categories.
- Increases in the national consumption of UPF have been associated with a higher prevalence of obesity and overweight. For most countries studied, this association remains significant after adjusting for potential confounding factors such as physical exercise and smoking.
Increased intake of UPFs is associated with a range of medical conditions. Most studies focus on overweight and obesity; some studies find significant associations with, for example, overall and breast cancer18 , overall mortality19 , hypertension20 , metabolic syndrome21 , and cardiovascular diseases22 . For most studies, these associations remain significant after adjusting for common confounding factors such as lack of physical exercise or smoking. Many of these studies are longitudinal cohort studies and suggest that these associations remain significant over time.
Although most UPF research is epidemiological, a 2019 clinical trial by Hall et al.39 compared energy intake and weight gain among 20 participants who alternately followed a UPF diet and a minimally-processed diet with meals that were matched for energy density, macronutrients , sugar, sodium, and fibre. Participants were instructed to consume as much or as little as they desired. This study found that participants on a UPF diet consumed more calories relative to a minimally-processed diet, and gained weight, while subjects on the unprocessed diet lost weight. It did not, however, establish the mechanisms through which this causal relation appeared.
3.3 Critiques and disagreements
To date, only a few peer-reviewed commentaries on the UPF concept and NOVA research have been published40 ,41 ,42 ,43 ,44 . Advocates of NOVA have responded to these criticisms and have questioned their authors’ relationship with ‘Big Food ’14 ,45 . Critics of the UPF concept can also be found in the media and on social media.
Criticisms can be broadly clustered into the following themes:
3.3.1 What defines the UPF category?
A major part of the criticism of the NOVA framework deals with how the UPF concept is defined, and with the nutritional characteristics of foods that belong to this category.
First, critics indicate that the UPF concept lumps together foods with different nutrient profiles while foods with similar nutrient profiles fall both in and outside the UPF category40 ,42 . For example, home-made brownies (made from minimally processed foods and processed culinary ingredients) and factory-made brownies (a UPF) are put in different categories. Mass-produced wholemeal bread, Quorn, fortified unsweetened soymilk, packaged crisps, and gummy candies, however, all get classed as UPFs. Many of these foods, critics argue, do not have a poor nutritional profile and can contribute to healthy diets40 ,41 . In any case, Gibney et al. note, it is difficult to see how they are ‘hyperpalatable’ and stimulate overconsumption41 .
Second, Gibney argues that the definition of the UPF concept has changed since it was first coined and is open to multiple interpretations42 . Older definitions emphasise salt and additive levels and the accessibility, convenience, and palatability of UPFs whereas newer definitions tend to be longer and highlight factors such as micronutrient fortification and ingredients that are not usually used in domestic food preparation (e.g. soy protein isolates). In addition, Gibney argues, different articles about the NOVA framework give different examples of foods from each category, leading to confusion and foods being classified in a subjective fashion42 .
Monteiro et al. respond that these and similar criticisms fail to consider the key characteristic of the NOVA framework: it classifies foods by the extent and purpose of food processing, not by their nutrient profiles14 . By this very definition, the UPF category is bound to contain foods with different nutrient compositions. Furthermore, they argue, the definition of UPF has not significantly changed since 2012 when the category of processed foods was added to the NOVA framework in order to better distinguish between different types of industrially processed foods. Very few NOVA studies had been published before this change in the model. In a 2019 article Monteiro and others point out that, in line with earlier definitions, UPFs can be readily identified by checking whether their ingredient labels mention one or more of the following ultra-processed ingredients12 :
- a “food substance that [is] never or rarely used in kitchens”12 (e.g. hydrolysed protein, soy protein isolate, gluten, whey, fructose, dextrose, or hydrogenated oil); or
- a “[food] additive whose function is to make the final product palatable or more appealing”12 (e.g. flavour enhancers, colours, artificial sweeteners, gelling agents).
Slight differences in examples, moreover, originate from fine-tuning the classification to different contexts (e.g. ‘sausages’ may commonly be ultra-processed in one country but elsewhere be mainly consumed in a form that is merely processed).
It is not clear, however, whether the purpose of food processing is necessarily different for foods from different NOVA categories. For instance, whether a manufacturer’s primary intention for producing an UPF is to generate high profits is difficult to measure. It is also not clear whether a producer of a food from another NOVA category would be any less motivated by profit or the desire to appeal to consumers. A further methodological challenge for research on UPF is that common nutritional epidemiological methods such as 24-hour recalls and food frequency questionnaires do not generally measure whether a food was made at home or in a factory and thus cannot reveal whether a health outcome is associated with UPFs, home-made foods, or both44 .
3.3.2 Association does not equal causation
Commentaries on UPFs often emphasise that association does not equal causation (see for example this NHS explainer on UPFs and cancer and this NHS explainer and this Science Media Centre commentary on UPFs and overall mortality)46 . For instance, when a study finds that (A) an increase in UPF intake is associated with (B) an increase in the prevalence of non-communicable diseases or medical conditions such as obesity, the relationship between A and B does not have to be causal but could be coincidental, result from a reverse causation, or be influenced by other factors. A and B may for example share a common cause, or B can be caused by another factor that for some reason correlates with A.
Other factors that influence a certain association are called confounding factors. In relation to UPF intake and health, potential confounding factors include smoking, physical inactivity, or income, and other dietary factors (people with a high UPF intake could turn out to be eating more fatty foods such as (artisanally produced) meats and cheeses). Different factors may confound UPF’s associations with specific health outcomes differently. Some argue that therefore a greater number of studies per individual health outcome is needed with attention to more confounding factors44 ,46 .
In clinical trials such as the study by Hall et al., the influence of confounding factors is generally ruled out by randomly assigning participants to an intervention group and a control group47 . Due to the use of non-experimental methods such as dietary surveys, this is impossible in nutritional epidemiological research and confounding is therefore more likely. Most NOVA and other epidemiological studies measure and statistically control for the influence of potential confounding factors. An association is more likely to be causal when it remains significant after adjustment for many confounding factors. The number of potential confounding factors, however, is unlimited and it is impossible to adjust for all of them.
Hall et al. argue that their randomised clinical trial is the first to study and find a causal link between UPF intake, energy intake and body weight change. Their finding is significant because Jones40 has suggested that UPF-consumers may be consuming excess dietary energy for reasons unrelated to UPF-intakes (see Figure 3). If this is the case, then the challenge is to reduce energy intakes and this could be achieved while still consuming UPFs if people ate UPFs that (for example) replaced sugar with artificial sweeteners (the relationship between artificial sweeteners and changes in body weight, is complex, contested and not discussed further here48 ,49 ). The findings of Hall et al., however, suggest that excess calorie intake is not separate from, but stimulated by, UPF consumption – the relationship is causal. This supports the view that UPFs induce overconsumption (see above). That said, Hall et al.’s study does not reveal the mechanisms underlying the causality and leaves open the question of whether reformulation could eliminate UPFs’ contribution to weight gain. In a commentary, Ludwig et al.50 also urge for caution about extrapolating Hall et al.’s findings, in particular because they are based upon short-term data on food intake and weight gain.
The question about reformulation relates to the more general point made by critics that even when UPF as a category is associated with poor health outcomes, particular UPFs such as mass-produced wholemeal bread, mycoprotein-based products (Quorn), or reformulated products (e.g. no added sugar granola with added whey powder) may contribute to improved public health. Monteiro and others question this reasoning on the ground that these foods can still contribute to unhealthy dietary patterns while some of their ingredients might cause health impacts that have not yet been explored. Scrinis and Monteiro17 also point out that reformulation often includes the substitution of ingredients such as sugar for ultra-processed ingredients such as artificial sweeteners whose impacts on health are often little known. Besides, they note, people may still overconsume reformulated UPFs.
3.3.3 Underlying disagreements
Underlying debate on UPF and poor health outcomes is the question of how this association relates to other well-known associations, for example between a varied diet and health or between dietary sugar and bodyweight. Opponents may think that the UPF concept is too general and of limited use because it provides little additional detail in understanding the relations between food and health. Proponents, however, may find UPF relevant as an overarching concept that groups a wide range of foods and their associations with different health outcomes. To a certain extent, this difference may be rooted in a preference for ‘lumping’ or ‘splitting’: for UPF proponents, lumping foods into four broad categories sheds light on a general trend in the food system, whereas for opponents, the presence of inevitable exceptions invalidates the framework. The simplicity of NOVA is sometimes also understood to make the framework especially helpful for health education and the model has been used by several governments as the basis for national dietary guidelines.
Linking to the concept of the nutrition transition, the general trend advocates of the UPF concept speak of is an undesirable change in diets that has been driven by the food industry. Through the design, production and marketing of UPFs, traditional diets are shifting towards those that are associated with poor health outcomes such as overweight, obesity and diabetes. The NOVA classification thus blends in with a political ideology that understands companies to be reaping large profits at the expense of people’s health, thereby evading a moral responsibility. Critics of the UPF concept, however, may see this political framing as driving the development of the UPF concept and the NOVA framework in the first place. As such, they understand the research on UPF to be pervaded, and therefore invalidated, by an ideological prejudice against profit making and big businesses. Monteiro et al., in response, are careful to note that their problem is not with the food industry per se, but only with that part that enables and encourages the consumption of UPFs.
4. UPFs and social and environmental sustainability
Besides impacting upon health, it is argued that UPFs undermine social and environmental sustainability2 ,11 . For example, the 2014 Brazilian dietary guidelines51 , which used the NOVA classification system, drew attention to the links between UPFs and concerns such as the loss of traditional food cultures and smallholder farmers’ livelihoods, the environmental impacts of monocultural farming, and the power of ‘Big Food’ in determining what sits on our plates. The dominance of UPFs in today’s food system can be seen as a proxy indicator for many of the social, ethical and environmental harms that this system causes (see Figure 4).
At present, since the main advocates of the UPF concept have been health professionals, there is as yet limited research into these non-health dimensions of UPFs. Some of the main arguments and questions raised on the environmental impacts of UPFs are, however, summarised here.
4.1 Environmental impacts
The 2014 Brazilian dietary guidelines suggest that UPFs are interlinked with unsustainable forms of intensive agriculture , whereas minimally processed foods are often produced on small farms in environmentally benign ways. Currently, more research is needed to explore the connections between foods from the different NOVA categories and the agricultural systems on which they are based. That said, the production of crops sold as minimally processed foods does not necessarily generate fewer environmental impacts than the production of crops that end up being made into UPFs. Compare, for example, intensive almond production leading to the sale of raw, unprocessed nuts and intensive wheat production leading to the production of biscuits. The relative environmental impacts of large scale vs. small scale farming are also the subject of much disagreement (see our building block on the land sparing - land sharing continuum).
When it comes to food preparation, several life-cycle assessment (LCA) studies comparing GHG emissions from (ultra-processed) ready-made meals and home-made meals with similar ingredients, show that on balance the ready-made foods that were studied generate about 30-50% higher emissions than the home-made meals52 ,53 . This mainly results from higher energy use during the processing of these UPFs, their dependence on the cold chain , and higher levels of food waste in the stages before consumption. While these figures will differ depending on the UPF (and alternative home cooked product) in question, it is unlikely that people exactly substitute home-made foods with their UPF equivalents and vice versa53 ,54 . Accordingly, the relative environmental impacts of diets based on UPFs or on minimally processed foods are difficult to determine.
The Brazilian dietary guidelines also state that UPFs cause higher transport-related emissions and pollute the environment through their plastic packaging, whereas minimally processed foods are associated with the seasonal consumption of locally produced foods. Many fruits and vegetables, however, are also imported and exported and have long supply chains that depend upon refrigeration and require a great deal of primary and secondary packaging. In the case of seasonal surpluses, some processed and ultra-processed foods could potentially help to limit food waste because they have long shelf-lives and do not need to be refrigerated54 . In many developed countries, minimally processed foods are generally packaged. It is unclear whether overall reliance on packaging is inherently lower in diets that are less dependent on UPFs.
Finally, NOVA’s engagement with the debate on the environmental impacts of meat and dairy consumption is currently minimal. The 2014 Brazilian dietary guidelines, for example, do not mention meat or dairy in their concluding ‘ten steps to healthy diets’51 .
5. UPF, health and sustainability
In summary, the UPF concept is gaining traction in some academic and policy quarters, and it is also increasingly used by NGOs and the media. Overall, the concept and the accompanying NOVA classification offer a different way of thinking about food that focuses less on nutrients and nutrient profiling, and more on the processing of foods and ingredients; on overall dietary patterns and the contexts in which we eat food; on the concentration of power in the hands of corporations; on the environmental impacts of food systems; and on food’s centrality to people’s social and cultural lives. Several governments – from Latin American countries in particular – have found utility in adopting NOVA and the UPF concept for health education in their national dietary guidelines.
The UPF concept, however, has also been criticised as imprecise and for encouraging oversimplified views about the food industry’s influence on dietary health, while adding little that is new to the idea that the consumption of certain processed foods (e.g. processed meats or soft drinks) may link to certain medical conditions, or to existing dietary advice to reduce the consumption of foods high in fat, sugar and salt. To date, most of the focus – by advocates and critics alike – has been on the link between UPFs and health outcomes. More work is needed to establish what, if any, relationship UPFs have with the wider social and environmental dimensions of food systems (Figure 5).
- 1Popkin, B. M. Global nutrition dynamics: the world is shifting rapidly toward a diet linked with noncommunicable diseases–. Am J Clin Nutr 84, 289–298 (2006). doi:10.1093/ajcn/84.1.289
- 2 a b c Monteiro, C. A. Nutrition and health. The issue is not food, nor nutrients, so much as processing. Public Health Nutrition 12, 729–731 (2009). doi:10.1017/S1368980009005291
- 3 a b c d Monteiro, C. The big issue is ultra-processing. World Nutrition 1, 33 (2010).
- 4 a b c d e f g Monteiro, C. A. et al. NOVA. The star shines bright. World Nutrition 7, 28–38 (2016).
- 5 a b Vandevijvere, S., De Ridder, K., Fiolet, T., Bel, S. & Tafforeau, J. Consumption of ultra-processed food products and diet quality among children, adolescents and adults in Belgium. Eur J Nutr (2018). doi:10.1007/s00394-018-1870-3
- 6 a b Louzada, M. L. da C. et al. Consumption of ultra-processed foods and obesity in Brazilian adolescents and adults. Preventive Medicine 81, 9–15 (2015). doi:10.1016/j.ypmed.2015.07.018
- 7Moubarac, J.C. et al. Processed and Ultra-processed Food Products: Consumption Trends in Canada from 1938 to 2011. Canadian Journal of Dietetic Practice and Research 75, 15–21 (2014). doi:10.3148/75.1.2014.15
- 8Adams, J. & White, M. Characterisation of UK diets according to degree of food processing and associations with socio-demographics and obesity: cross-sectional analysis of UK National Diet and Nutrition Survey (2008-12). Int J Behav Nutr Phys Act 12, 160 (2015). doi:10.1186/s12966-015-0317-y
- 9 a b Baraldi, L. G., Steele, E. M., Canella, D. S. & Monteiro, C. A. Consumption of ultra-processed foods and associated sociodemographic factors in the USA between 2007 and 2012: evidence from a nationally representative cross-sectional study. BMJ Open 8, e020574 (2018). doi:10.1136/bmjopen-2017-020574
- 10Parra, D. C. et al. Association between ultra-processed food consumption and the nutrient profile of the Colombian diet in 2005. Salud Pública de México 61, 147–154 (2019). doi:10.21149/9038
- 11 a b c d e f g h Monteiro, C. A. et al. The UN Decade of Nutrition, the NOVA food classification and the trouble with ultra-processing. Public Health Nutr 21, 5–17 (2018). doi:10.1017/S1368980017000234
- 12 a b c d e f g h Monteiro, C. A. et al. Ultra-processed foods: what they are and how to identify them. Public Health Nutrition 1–6 (2019). doi:10.1017/S1368980018003762
- 13 a b Monteiro, C. A., Levy, R. B., Claro, R. M., Castro, I. R. R. de & Cannon, G. A new classification of foods based on the extent and purpose of their processing. Cadernos de Saúde Pública 26, 2039–2049 (2010). doi:10.1590/S0102-311X2010001100005
- 14 a b c Monteiro, C. A. et al. Ultra-processing. An odd ‘appraisal’. Public Health Nutrition 21, 497–501 (2018). doi:10.1017/S1368980017003287
- 15Monteiro, C. A. et al. Dietary guidelines to nourish humanity and the planet in the twenty-first century. A blueprint from Brazil. Public Health Nutrition 18, 2311–2322 (2015). doi:10.1017/S1368980015002165
- 16Scrinis, G. Nutritionism: The Science and Politics of Dietary Advice. (Columbia University Press, 2013).
- 17 a b Scrinis, G. & Monteiro, C. A. Ultra-processed foods and the limits of product reformulation . Public Health Nutrition 21, 247–252 (2018). doi:10.1017/S1368980017001392
- 18 a b Fiolet, T. et al. Consumption of ultra-processed foods and cancer risk: results from NutriNet-Santé prospective cohort. BMJ k322 (2018). doi:10.1136/bmj.k322
- 19 a b Schnabel, L. et al. Association Between Ultraprocessed Food Consumption and Risk of Mortality Among Middle-aged Adults in France. JAMA Intern Med 179, 490–498 (2019). doi:10.1001/jamainternmed.2018.7289
- 20 a b Mendonça, R. de D. et al. Ultra-Processed Food Consumption and the Incidence of Hypertension in a Mediterranean Cohort: The Seguimiento Universidad de Navarra Project. Am J Hypertens 30, 358–366 (2017). doi:10.1093/ajh/hpw137
- 21 a b Tavares, L. F., Fonseca, S. C., Rosa, M. L. G. & Yokoo, E. M. Relationship between ultra-processed foods and metabolic syndrome in adolescents from a Brazilian Family Doctor Program. Public Health Nutrition 15, 82–87 (2012). doi:10.1017/S1368980011001571
- 22 a b Rauber, F., Campagnolo, P. D. B., Hoffman, D. J. & Vitolo, M. R. Consumption of ultra-processed food products and its effects on children’s lipid profiles: A longitudinal study. Nutrition, Metabolism and Cardiovascular Diseases 25, 116–122 (2015). doi:10.1016/j.numecd.2014.08.001
- 23Monteiro, C. A. et al. Household availability of ultra-processed foods and obesity in nineteen European countries. Public Health Nutrition 21, 18–26 (2018). doi:10.1017/S1368980017001379
- 24Monteiro, C. A., Levy, R. B., Claro, R. M., Castro, I. R. R. de & Cannon, G. Increasing consumption of ultra-processed foods and likely impact on human health: evidence from Brazil. Public Health Nutrition 14, 5–13 (2010). doi:10.1017/S1368980010003241
- 25Moubarac, J.-C. et al. Consumption of ultra-processed foods and likely impact on human health. Evidence from Canada. Public Health Nutrition 16, 2240–2248 (2013). doi:10.1017/S1368980012005009
- 26Monteiro, C. A., Moubarac, J.-C., Cannon, G., Ng, S. W. & Popkin, B. Ultra-processed products are becoming dominant in the global food system. Obes Rev 14, 21–28 (2013). doi:10.1111/obr.12107
- 27Juul, F. & Hemmingsson, E. Trends in consumption of ultra-processed foods and obesity in Sweden between 1960 and 2010. Public Health Nutrition 18, 3096–3107 (2015). doi:10.1017/S1368980015000506
- 28Simões, B. dos S. et al. Consumption of ultra-processed foods and socioeconomic position: a cross-sectional analysis of the Brazilian Longitudinal Study of Adult Health. Cad. Saúde Pública, Cad. saúde pública 34, e00019717 (2018). doi:10.1590/0102-311x00019717
- 29Steele, E. M. et al. Ultra-processed foods and added sugars in the US diet: evidence from a nationally representative cross-sectional study. BMJ Open 6, e009892 (2016). doi:10.1136/bmjopen-2015-009892
- 30Costa, C. S. et al. Ultra-processed food consumption and its effects on anthropometric and glucose profile: a longitudinal study during childhood. Nutrition, Metabolism and Cardiovascular Diseases (2018). doi:10.1016/j.numecd.2018.11.003
- 31Mendonça, R. de D. et al. Ultraprocessed food consumption and risk of overweight and obesity: the University of Navarra Follow-Up (SUN) cohort study . Am J Clin Nutr 104, 1433–1440 (2016). doi:10.3945/ajcn.116.135004
- 32Canella, D. S. et al. Ultra-Processed Food Products and Obesity in Brazilian Households (2008–2009). PLOS ONE 9, e92752 (2014). doi:10.1371/journal.pone.0092752
- 33Julia, C. et al. Contribution of ultra-processed foods in the diet of adults from the French NutriNet-Santé study. Public Health Nutrition 21, 27–37 (2018). doi:10.1017/S1368980017001367
- 34Marrón-Ponce, J. A., Sánchez-Pimienta, T. G., Louzada, M. L. da C. & Batis, C. Energy contribution of NOVA food groups and sociodemographic determinants of ultra-processed food consumption in the Mexican population. Public Health Nutrition 21, 87–93 (2018). doi:10.1017/S1368980017002129
- 35Louzada, M. L. da C. et al. The share of ultra-processed foods determines the overall nutritional quality of diets in Brazil. Public Health Nutrition 21, 94–102 (2018). doi:10.1017/S1368980017001434
- 36Batal, M. et al. Quantifying associations of the dietary share of ultra-processed foods with overall diet quality in First Nations peoples in the Canadian provinces of British Columbia, Alberta, Manitoba and Ontario. Public Health Nutrition 21, 103–113 (2018). doi:10.1017/S1368980017001677
- 37Cediel, G. et al. Ultra-processed foods and added sugars in the Chilean diet (2010). Public Health Nutrition 21, 125–133 (2018). doi:10.1017/S1368980017001161
- 38Nasreddine, L. et al. A minimally processed dietary pattern is associated with lower odds of metabolic syndrome among Lebanese adults. Public Health Nutrition 21, 160–171 (2018). doi:10.1017/S1368980017002130
- 39Hall, K. D. et al. Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain: An Inpatient Randomized Controlled Trial of Ad Libitum Food Intake. Cell Metabolism (2019). doi:10.1016/j.cmet.2019.05.008
- 40 a b c d Jones, J. M. Food processing: criteria for dietary guidance and public health? Proceedings of the Nutrition Society 1–15 (2018). doi:10.1017/S0029665118002513
- 41 a b c Gibney, M. J., Forde, C. G., Mullally, D. & Gibney, E. R. Ultra-processed foods in human health: a critical appraisal. Am J Clin Nutr 106, 717–724 (2017). doi:10.3945/ajcn.117.160440
- 42 a b c d Gibney, M. J. Ultra-Processed Foods: Definitions and Policy Issues. Curr Dev Nutr 3, (2019). doi:10.1093/cdn/nzy077
- 43Botelho, R., Araújo, W. & Pineli, L. Food formulation and not processing level: Conceptual divergences between public health and food science and technology sectors. Critical Reviews in Food Science and Nutrition 58, 639–650 (2018). doi:10.1080/10408398.2016.1209159
- 44 a b c Poti, J. M., Braga, B. & Qin, B. Ultra-processed Food Intake and Obesity: What Really Matters for Health—Processing or Nutrient Content? Curr Obes Rep 6, 420–431 (2017). doi:10.3945/ajcn.114.100925
- 45Mialon, M., Sêrodio, P. & Scagliusi, F. B. Criticism of the NOVA classification: who are the protagonists? 1 9, 176–240 (2018). doi:10.26596/wn.201893176-240
- 46 a b Monge, A. & Lajous, M. Ultra-processed foods and cancer. BMJ 360, k599 (2018). doi:10.1136/bmj.k599
- 47Didelez, V. Statistical causality. Consilience: Interdisciplinary Communications 2005/2006 115–120 (2007).
- 48Yang, Q. Gain weight by “going diet?” Artificial sweeteners and the neurobiology of sugar cravings. Yale J Biol Med 83, 101–108 (2010).
- 49Toews, I., Lohner, S., Küllenberg de Gaudry, D., Sommer, H. & Meerpohl, J. J. Association between intake of non-sugar sweeteners and health outcomes: systematic review and meta-analyses of randomised and non-randomised controlled trials and observational studies. BMJ 364, k4718 (2019). doi:10.1136/bmj.k4718
- 50Ludwig, D. S. et al. Ultra-Processed Food and Obesity: The Pitfalls of Extrapolation from Short Studies. Cell Metabolism (2019). doi:10.1016/j.cmet.2019.06.004
- 51 a b Ministry of Health of Brazil. Dietary Guidelines for the Brazilian Population. (Ministério da Saúde, 2014).
- 52Espinoza-Orias, N. & Azapagic, A. Understanding the impact on climate change of convenience food: Carbon footprint of sandwiches. Sustainable Production and Consumption 15, 1–15 (2018). doi:10.1016/j.spc.2017.12.002
- 53 a b Schmidt Rivera, X. C., Espinoza Orias, N. & Azapagic, A. Life cycle environmental impacts of convenience food: Comparison of ready and home-made meals. Journal of Cleaner Production 73, 294–309 (2014). doi:10.1016/j.jclepro.2014.01.008
- 54 a b Garnett, T. Wise moves. Exploring the relationship between food, road transport and CO2. Transport2000, London (2003).
- The commentary in which Carlos Monteiro coined the UPF concept: Nutrition and health. The issue is not food, nor nutrients, so much as processing
- A critical review of the UPF concept and a reply from Monteiro and others:
- Gibney et al. (2017): Ultra-processed foods in human health: a critical appraisal
- Monteiro et al. (2018): Ultra-processing. An odd ‘appraisal’
- A commentary by Monteiro et al.: Ultra-processed foods: what they are and how to identify them
- A randomised clinical trial by Hall et al.: Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain: An Inpatient Randomized Controlled Trial of Ad Libitum Food Intake
- A commentary by Ludwig et al. on the study by Hall et al.: Ultra-Processed Food and Obesity: The Pitfalls of Extrapolation from Short Studies
- A commentary by Monteiro et al. on the study by Hall et al.: Freshly Prepared Meals and Not Ultra-Processed Foods
- An English translation of the 2014 Brazilian dietary guidelines, which draw upon the work of Monteiro et al.
- The Open Food Facts-website runs a searchable database with many foods classified according to the NOVA classification.
Fraanje, W. & Garnett, T. (2019). What is ultra-processed food? And why do people disagree about its utility as a concept? (Foodsource: building blocks). Food Climate Research Network, University of Oxford.
- Walter Fraanje, Food Climate Research Network, University of Oxford
- Tara Garnett, Food Climate Research Network, University of Oxford
- Helen Breewood, Food Climate Research Network, University of Oxford
- Dr Phillip Baker, Faculty of Health, Deakin University
- Geoffrey Cannon and Professor Carlos Monteiro, Centre for Epidemiological Studies in Health and Nutrition, University of São Paulo
- Professor Mike Gibney, Institute of Food and Health, University College Dublin
- Professor Corinna Hawkes, Centre for Food Policy, City University of London
- Professor Mike Rayner, Nuffield Department of Population Health, University of Oxford
- Dr Gyorgy Scrinis, School of Agriculture and Food, University of Melbourne
Reviewing and advising do not constitute an endorsement. Final editorial decisions, including any remaining inaccuracies and errors, are the sole responsibility of the Food Climate Research Network.
- The Daniel and Nina Carasso Foundation
- The Oxford Martin School
- The Wellcome Trust, Our Planet Our Health (Livestock, Environment and People - LEAP), award number 205212/Z/16/Z