Degree of food processing – a health indicator?

Date         17.09.2025

Summary

The debate about so-called ultra-processed foods (UPF) has become increasingly relevant in recent years, both in science and in public perception. The reason is that observational studies have identified a connection between a high consumption of UPF and an increase in the incidence of diseases such as obesity, type 2 diabetes, cardiovascular diseases and total mortality figures. But observational studies do not allow reliable inferences to be drawn about causal effect mechanisms and, so far, there are still very few corresponding intervention studies. Moreover, the complex effects of food processing are only partially captured by systems that classify foods according to their degree of processing. According to current research, the widespread NOVA classification system does not provide for a differentiated evaluation of the influence of processing on the health impact of food. Given the current shortage of data, the conclusions drawn are abbreviated and sometimes premature. In order to derive robust recommendations for policy-makers, healthcare and society on the influence of processing on the health-related value of foodstuffs, in-depth scientific studies and further intensive research are necessary.

A closer look

An in-depth examination of the topic reveals two main problem areas: One is the inconsistent definition of the term 'processing' and the other the limited data currently available on which specific properties of processed foods are responsible for the potentially undesirable effects on human health.

The terms 'processed' and, in particular, 'ultra-processed' are not defined identically in food legislation or the scientific literature. At present, science uses different systems to classify foods according to the degree of processing. A detailed description of the classification systems published so far can be found in the 15th DGE Nutrition Report [1]. In general, these systems use the term 'processing' to refer to the influence of the recipe and the technical processes used, but they are evaluated quite differently in detail. This complicates scientific and public discourse [2].

In the studies currently available, classification according to the degree of processing is mainly based on the NOVA classification [1]. This system divides foodstuffs into four groups on the basis of certain ingredients, the number of ingredients and the purpose of food processing [3, 4]. Group IV are the so-called 'ultra-processed foods'. There is, however, considerable criticism of the use of the NOVA system [5-10], a frequent point of criticism being that the classification does not systematically distinguish between the type and quantity of all ingredients used, that is, the recipe, and the processes utilised (e.g. drying, heating, extraction). In the case of the processes, it should be noted that in some cases only the processing goal (e.g., preservation) is used for classification and not the specific process. This is problematic because for preservation, for example, there are fundamentally different processes which are based on very different chemical, physical and microbiological principles. Furthermore, the same basic processes may be carried out under different processing conditions, e.g., different temperatures. Although this affects food quite differently, it leads to the same classification in the NOVA system. In addition, the NOVA classification focuses on industrial processing. This means that the same processing steps are evaluated differently depending on whether they are carried out in catering and households or in industry. However, other classification systems also fail to reflect in any detail the complexity of food processing and its effects on food ingredients and the food matrix. To explain the complexity of food processing, key aspects are outlined briefly below.

Effects of processing on foodstuffs

In principle, process-related changes can have both negative and positive effects on the nutritional quality of the food produced. For example, valuable ingredients such as vitamins can be degraded by heating and harmful substances such as acrylamide can be produced. The effects can also be positive, however, such as the inactivation of harmful microorganisms and enzymes or the reduction of antinutrients, such as phytates, oxalates, lectins, saponins and tannins [11, 12]. A low level of antinutrients usually has a positive effect on the bioavailability of minerals and trace elements, for example [13]. Especially when it comes to food safety, food processing is therefore an indispensable building block in the value chain.

Apart from the direct effect on ingredients, processes often have a so-called matrix effect, too. This refers to the change in the food structure. A strong matrix effect is manifest when the original structure of the food is no longer visible after processing. Examples include the production of purees or juices from fruit or vegetables, or the extraction of protein isolates from animal or plant sources. New structures can also be formed, for example in products with a meat-like structure made from soybean meal by the extrusion process. Similar processes are also used in the sustainable utilisation of agricultural raw materials, such as in processing residues from food production to obtain nutrients still contained in them for use in food. Examples include press cakes made from juice and plant oil production. A matrix effect can have positive or negative effects on the nutritional value of a food. For example, the opening of plant cells through processing can mean the body gains access to larger quantities of valuable nutrients in food [14]. In the case of micronutrients, such as vitamins, minerals and antioxidants, this is usually beneficial [15]. If more mono- and disaccharides or fats become available because of the matrix effect, this has an impact on blood sugar development, for example [16-18] and/or energy intake [19]. Where there is malnutrition, this effect can be seen as positive. However, in countries like Germany, where there is no shortage of food and a high prevalence of obesity this effect can lead to an increase in the risk of developing diet-related diseases. Nevertheless, a general correlation between parameters such as the glycemic index or the glycemic load and the degree of processing according to the NOVA classification has not yet been reliably proven [20]. A change in the matrix can also mean that a foodstuff requires less chewing and is thus consumed faster and in larger quantities, such as fruit and vegetable juice. This can lead to a delayed onset of satiety and thus to increased energy intake [1, 21, 22].

Assessing the health impact of ‘ultra-processed foods’

Assessing the impact of consuming certain foodstuffs or food groups on the health of people of all ages is a complex and mutlifacted research field. Research on the impact of ultra-processed foods is also a young field of research in which there are hardly any intervention studies to date that could help to establish causal relationships. The existing observational studies, however, largely indicate a link between high consumption of ultra-processed foods (according to the NOVA system) and diseases such as obesity type 2 diabetes and cardiovascular diseases [23-26]. In children and adolescents, too, studies reveal undesirable effects on risk factors for metabolic diseases [27-29], although there are significantly more studies available on adults. The composition of many foodstuffs, which are classified as ultra-processed foods by NOVA, is unfavourable also by traditional assessment methods [30, 31], which are based on the recipe. However, studies on the subject show that the negative effects of ultra-processed foods cannot be explained solely by the unfavourable composition of their nutrients [32, 33]. This clearly indicates that food processing has effects that are not solely based on their nutrient composition. However, there are also studies that reveal that not all of the foodstuffs classified as ultra-processed according to the NOVA system do necessarily increase the risk of disease [34]. These studies describe links between the consumption of certain food groups, such as soft drinks or animal foods, and an increased risk of disease. But no such disease risks were identified in other food groups that are also classified as ultra-processed by NOVA, such as certain breakfast cereals and plant-based alternatives to animal products. In its report on plant-based alternative products, for example, the Scientific Advisory Council on Agricultural Policy, Nutrition and Consumer Health Protection (WBAE) concludes that the assessment of the nutritional quality of these foods should be based on the ingredients and the general state of research rather than on classifications like NOVA [8].

All in all, it becomes clear that not all the foodstuffs classified as ultra-processed by NOVA are necessarily of poor nutritional quality. As described above, this is due to the fact that the NOVA classification does not differentiate sufficiently between the complex effects of food processing. Using the NOVA classification it is, therefore, impossible to draw conclusions about which specific aspects of food processing are responsible for the health impacts observed. Often, it is also not possible to take a differentiated view of the health effects of the degree of food processing due to limitations in the level of detail of nutritional data from existing epidemiological studies. For this reason, prospective cohort studies to assess the long-term health effects that contain detailed dietary surveys are required. The questions of the fundamental causes for and biological mechanisms involved in the links observed remain largely unanswered at present. In order to address this question and derive causal relationships, targeted, randomised, controlled intervention studies would be required. Moreover, more knowledge should be generated on the effects of various processes on food ingredients so that they can be correlated with health effects.

Conclusion

Food processing plays a crucial role in our diets today. Utilised correctly, the advantages, such as improved nutrient supply, greater food safety and higher sustainability, are indisputable. A blanket devaluation of processed foods, including so-called ultra-processed foods, consequently does not do justice to the multifaceted nature of the issue. Classification systems like NOVA fail to differentiate sufficiently between the complex effects of food processing. It is therefore impossible to identify which specific aspects of food processing are responsible for the health impacts observed. For a scientifically sound assessment, it is essential to know the effects of processes and recipes on the nutritional quality of processed foods and to evaluate them in a differentiated manner. To do so, the most detailed nutrition survey possible is required in cohort studies, and the data should be supplemented by intervention studies.