What are ultra-processed foods (UPF)?
Within the food and drink portfolio of the world's largest food corporations, the majority of the items produced can be described as unhealthy, possibly amounting to a staggering 70-90% of the total [Evans 2021; Nestle 2021; Bandy et al 2023]. Such unhealthy foods are often described as ultra-processed foods (UPF). However, there is still debate on how UPF should be understood and defined. Essentially three important characteristics can be identified: 1) the degree of processing, 2) the inferior quality of the ingredients, 3) and the intention to displace other foods.
Degree of processing
There have been deliberate attempts by multinational corporations [e.g., WBCSD 2021] and their associated researchers to create confusion about the food category of ultra-processed foods, for instance by conflating it with 'processed foods' in general or by claiming that the category is too vague to be workable [Lawrence 2022]. The problem, however, should not be restricted to the simple degree of processing (in many cases, processing may be beneficial), so that a more comprehensive description is needed.
Inferior quality of the ingredients
The inferior quality of their ingredients is a hallmark of UPF. They are 'mostly constructed from a narrow range of cheap, extracted, refined and fractionated ingredients [..] while lacking in whole, intact and fresh, ingredients' [Scrinis & Monteiro 2022]. As such, they are 'formulations of often chemically manipulated cheap ingredients such as modified starches, sugars, oils, fats, and protein isolates, with little if any whole food added, made palatable and attractive by using combinations of flavours, colours, emulsifiers, thickeners, and other additives' [Monteiro 2024].
The intention to displace other foods
In addition to those technological criteria, ultra-processed foods are typically generated by transnational corporations 'to create branded, convenient (durable, ready-to-consume), attractive (hyper-palatable) and highly profitable (low-cost ingredients) food products often designed to displace all other food groups' [Monteiro et al. 2018, 2019]. They 'bypass the body's satiety signals and elicit quasi-addictive behaviours' [Scrinis & Monteiro 2022]. At the social level, they tend to replace the rich experiences of shared traditional meals with a plethora of often individualized ready-made products, leading to both mindless eating and further atomization of society [Anastasiou et al. 2022].
Aggressive marketing, greenwashing, and nutri-washing, are used to 'drive up demand and create new food cultures, construct global supply chains to obtain cheap ingredients, and use extensive packaging that encourages mass production, long-distance transportation, and waste related to their consumption' [Seferidi et al. 2020]. As such, they reinforce the position of large corporations, to the detriment of smallholder production [Baker et al. 2020].
What is the current prevalence of UPF in the diet?
With respect to volume of sales, the annual per capita amount of UPFs has been estimated at about 100 kg (Western Europe) to 120 kg (North America and Australasia); additionally, ultra-processed drinks accounted for 180 kg and 120 kg, respectively [Vandevijvere et al. 2019]. In France, not only 2/3 of conventional industrial foods are to be categorized as ultra-processed, but also half of the products that are labelled as organic [Davidou et al. 2021]. Typically, they are based on refined oils, extracts, starches, and/or sugars.
The share of ultra-processed foods in the diets of adults in high-income countries has been estimated at 15-60% of the total caloric intake [Anastasiou et al. 2022; Bertoni Maluf et al. 2022; Touvier et al. 2023], and at 13-21% for Brazil [Nilson et al. 2022]. In the US, >70% of packaged foods are ultra-processed [Baldridge et al. 2019; et al. 2023], i.e., 'industrial formulations made entirely or mostly from substances extracted from foods (oils, fats, sugar, starch & proteins)’ [Baldridge et al. 2019]. In the US, UK, and Australia, ultra-processed foods represent 70-80% of the diet in the top fifth of consumers of these products [Scrinis & Monteiro 2022].
Children in the Anglosphere now even obtain 55-65% of their caloric intake from ultra-processed foods, such as crisps, biscuits, juices and sodas, while the intake level in low- and middle-income countries is already at 18-35% [Khandpur et al. 2020]. Total energy consumed from ultra-processed foods by US youths (2-19 years) increased from 61% to 67% between 1999 and 2018 [Wang et al. 2021]. In France, intake is estimated at 46% for children, which is higher than the 35% for adults [Fardet et al. 2021].
What are the perspectives?
The situation is expected to worsen globally due to aggressive marketing and the capturing and co-opting of policies, science, and civil society [Mialon & da Silva Gomes 2019; Moodie et al. 2021]. To make matters worse, they are disproportionally consumed by lower socioeconomic groups that already suffer from disbalanced nutrition [Baker et al. 2020]. When children of such groups are exposed to ultra-processed foods at young age, metabolic disturbances may follow [Ribeiro Gomes et al. 2024]. However, higher socio-economic groups are also vulnerable, for instance by their sensitivity to environmental ('plant-based' foods) and health messages (via Nutrient Profiling Systems, such as NutriScore).
The push for a 'plant-based' transition contributes to UPF production
The market for 'plant-based' mock animal source foods [see elsewhere] has been embraced by large food multinationals, given that such foods are usually highly processed concoctions [Bohrer 2019], yet can be greenwashed based on narratives relating to 'planetary health' [see elsewhere]. In a French cohort study, vegetarians and vegans (especially those who started such diets more recently) had a higher intake of ultra-processed products than meat eaters, driven by a higher consumption of mock foods [Gehring et al. 2021]. Moreover, such foods, when fortified, can reach certain micronutrient targets where whole-food plant-based diets are failing, as for vitamins B12 and D, calcium, and iron [Bunge et al. 2024]. However, despite being more micro-nutrient adequate, they do contribute to an increasing share of UPF the vegetarian diets.
Nutrient Profiling Systems further aggravate the situation
Nutrient Profiling Systems (NPS) tend to underestimate the risks associated with ultra-processed foods, while exaggerating the effects of animal source foods [Ortenzi et al. 2022]. It comes as no surprise that such nutrient-centric dietary models and Front-of-Pack labels (e.g., NutriScore or Health Stars) are usually embraced by food corporations [e.g., Nestlé 2020]. On the Spanish market, for instance, 1/4 and 1/2 products within the favourable NutriScore categories A and B, respectively, are ultra-processed foods [Romero Ferreiro et al. 2021]. Such models require small sacrifices but meanwhile associate the harms of the foods they produce with high energy density and elevated concentrations of specific nutrients (sodium, sugar, saturated fat, etc.), which can be conveniently addressed through what is the core expertise of such corporations: ultra-processing. Products are then reformulated to somewhat reduce the levels of some of these nutrients (e.g., with synthetic sweeteners, salt replacers, texturizers, flavouring agents), or by adding ingredients with a healthy aureole (e.g., fibre or vitamins). This says little about true improvement of the healthiness of a food.
Why reformulation is not the solution
The adverse consequences of ultra-processed foods are generally independent of dietary quality or patterns, which undermines the idea that reformulation would mitigate against negative health outcomes [Dicken & Batterham 2022]. In some cases, such interventions may even be potentially harmful, for instance when adding artificial sweeteners [Dalenberg et al. 2020; Debras et al. 2022; Suez et al. 2022], when using emulsifiers in fat-lowering formulations [Sandall et al. 2020; Viennois et al. 2020], or when generally restructuring the food matrix through ultra-processing [Fardet 2016].
What is the impact of UPFs on health?
Taken together, UPFs may have to be considered as a primary modifiable target for early prevention of non-communicable diseases [Touvier et al. 2023], including colorectal cancer [Hang et al. 2022]. An umbrella review of epidemiological meta-analyses has shown an increased risk of all-cause mortality, combined overweight and obesity, and type-2 diabetes, with 'moderate' certainty of evidence according to GRADE, and a large variety of other negative health outcomes at lower certainty levels [Lane et al. 2024; Monteiro 2024].
What is known from intervention trials?
It is not fully understood yet how UPFs negatively affect metabolic health in a clear and causal manner. However, it is known that their consumption generally leads to higher energy intake levels and weight gain compared to an unprocessed diet [Hall et al. 2019]. Animal studies indicate that 'hedonic overdrive', due to hyperpalatable foods overriding homeostatic intake regulation, leads to weight gain [Gao et al. 2024]. Such results from intervention trials are to be combined with increasing concerns about the constituents and biochemical make-up of UPFs [e.g., Fardet 2016].
What are the likely mechanisms?
The harmful effects of Western foods may be less a matter of poor nutritional value than of the impact of ultra-processing as such [Bonaccio et al. 2022; Touvier et al. 2023; Monteiro 2024]. This occurs largely through food matrix degradation, due to dietary reconstitution, extrusion, intense heating, etc. [Fardet & Rock 2022], effects on the gut microbiome [Zinöcker & Lindseth 2018; Smith et al. 2022], and addictive eating [Schulte et al. 2015; Gearhardt & Difeliceantonio 2022]. This may ultimately drive alterations of the homeostatic body weight 'set point', for instance via changes in brain dopamine, which would explain the development of overweight and obesity [Darcey et al. 2023]. Consumption of ultra-processed foods can also result in inflammatory responses ("fast food fever") [Myles, 2014; Hoffman 2022], reduced water intake [Galastri Baraldi et al. 2021], overeating due to soft textures [Forde et al. 2020] and altered glycemic responses [Fardet 2016], and harmful effects due to a higher intake of additives (colourants, artificial sweeteners, flavours, emulsifiers, ...), fillers, products from chemical hydrogenation, and various neo-formed substances [Ogata et al. 1999; Dalenberg et al. 2020; Sandall et al. 2020; Viennois et al. 2020; Debras et al. 2022; Neumann & Fasshauer 2022; Scrinis & Monteiro 2022; Suez et al. 2022; Vissers et al. 2022; Zangara et al. 2022; Martínez Steele et al. 2023; WHO 2023a, 2023b]. Harmful contaminants, e.g. endocrine-disrupting chemicals in the abundant packaging materials, could also play a role [cf. Diamanti-Kandarakis et al. 2009; Buckley et al. 2019]. For plant-based 'meat', for instance, a weakened gastrointestinal digestive function and lower digestibility has been found in mice, when compared to real meat [Xie et al. 2022].
What is the impact of UPF on the environment?
The impact of ultra-processed foods on various environmental dimensions is indeed very substantial [Scott 2018; Fardet & Rock 2020; Seferidi et al. 2020; Anastasiou et al. 2022]. In Brazil, impact of dietary change on change in greenhouse gas emissions, water footprint, and ecological footprint was ascribed to ultra-processed foods, while there was no change in these environmental indicators for unprocessed or minimally processed foods [da Silva et al. 2021].
