Degree of food processing as a health indicator?

As of August 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 for this is that observational studies have found a link between a high consumption of 'ultra-processed foods' and an increased incidence of diseases such as obesity, type 2 diabetes or cardiovascular diseases as well as overall mortality. However, observational studies do not allow reliable conclusions to be drawn about causal mechanisms of action, and corresponding intervention studies are still largely lacking. In addition, the complex effects of food processing are only partially captured by systems that classify foods according to their degree of processing. According to current studies, the widespread classification according to the NOVA system is not suitable for a differentiated assessment of the influence of processing on health effects of food. Given the current lack of data, conclusions drawn are shortened and partly premature. In order to derive well-founded recommendations for politics, health care and society regarding the influence of processing on the health-related value of food, in-depth scientific examination and intensive further research are required.

A closer look
An in-depth examination of the topic reveals two main areas of concern. These are the inconsistent definition of the term 'processing' and the currently limited data 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 uniformly defined in food legislation or in scientific literature. Currently, there are different classification systems in science for classifying 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 their detailed assessments vary greatly. This complicates the scientific and public discourse [2].

In the studies currently available, the classification according to the degree of processing is mainly based on the NOVA system [1]. This system divides food into four groups based on certain ingredients, the number of ingredients, and the purpose of food processing [3, 4]. Group IV are the so-called 'ultra-processed foods'. However, there is substantial criticism of the use of the NOVA system [5-10]. A frequent point of criticism is that the classification does not systematically distinguish between the type and quantity of all ingredients used, i.e. the recipe, and the application of processes (e.g. drying, heating, extraction). In the case of the processes, it is important to note that in some cases only the treatment goal (e.g. preservation) and not the specific process is used for classification. This is problematic because, e.g. for preservation, there are fundamentally different processes. Preservation processes can be based on very different chemical, physical or microbiological principles. Even if the same basic processes are used, they can be carried out under different process conditions, e.g. different temperatures. Although this has very different effects on food, it leads to the same classification in the NOVA system. Another point is that 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 gastronomy and household or industry. However, other classification systems also fail to reflect the complexity of food processing and its effects on food ingredients and the food matrix in great detail. To explain the complexity of food processing, key aspects are briefly outlined below.

Effects of processing on food
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, or harmful substances such as acrylamide can be created. However, positive effects can also be achieved, such as the inactivation of harmful microorganisms and enzymes or the reduction of anti-nutritional substances such as phytates, oxalates, lectins, saponins and tannins [11, 12]. A low content of anti-nutritional substances generally has a positive effect on the bioavailability of minerals and trace elements, for example [13]. Food processing is therefore an indispensable building block in the value chain, especially when it comes to food safety.

In addition to 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 given when the original structure of the food is no longer visible after processing. Examples include the use of processes to produce purees or juices from fruit or vegetables, or to produce protein isolates from animal or plant sources. New structures can also be formed, e.g. in products with a meat-like structure made from soybean flour by extrusion process. Such processes are also used in the sustainable utilization of agricultural raw materials. An example is the processing of residues from food production, such as press cakes from juice or vegetable oil production, to obtain nutrients still contained in them for use in food. 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 lead to the body having access to larger quantities of valuable nutrients in food [14]. This is usually beneficial for micronutrients, such as vitamins, minerals and antioxidants [15]. If, for example, mono- and disaccharides or fats become more available as a result of the matrix effect, this has an impact on blood sugar development [16-18], for example, and/or energy intake [19]. If malnutrition prevails, this effect is rather positive. However, in countries such as Germany, where there is no shortage of food and a high obesity prevalence, this effect can increase the risk of developing diet-related diseases. Nevertheless, a general correlation between parameters such as the glycaemic index or the glycaemic load and the degree of processing according to the NOVA classification has not yet been reliably proven [20]. A change in matrix can also mean that a food product requires less chewing and is therefore consumed more quickly and in larger quantities, for example when consuming fruit or vegetable juice. This can lead to a delayed onset of satiety and thus to increased energy intake [1, 21, 22]. 

Assessing health effects of 'ultra-processed foods'
Assessing the effects of consuming certain foods or food groups on the health of people of all ages is a complex and multifaceted field of research. Research on the impact of 'ultra-processed foods' is also a young field of research in which there are hardly any intervention studies that could be used to establish causal relationships. However, the existing observational studies largely show a link between a high consumption of 'ultra-processed foods' (according to the NOVA system) and diseases such as obesity, type 2 diabetes or cardiovascular diseases [23-26]. Studies also show undesirable effects on risk factors for metabolic diseases in children and adolescents [27-29], although the study situation is much better in adults. Many foods that belong to the 'ultra-processed foods' according to the NOVA classification have an unfavourable nutrient composition even according to traditional assessment [30, 31], which is based on the recipe. However, studies show that the negative effects of 'ultra-processed foods' cannot be explained solely by the unfavourable nutrient composition [32, 33]. This shows that there are effects of food processing that are not based solely on the nutrient composition. But there are also studies that show that not all foods classified as 'ultra-processed' by the NOVA system 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 found for other food groups that are also classified as 'ultra-processed' by NOVA, such as certain breakfast cereals and plant-based alternatives to animal products. For example, the Scientific Advisory Council on Agricultural Policy, Nutrition and Consumer Health Protection (WBAE) concludes in its report on plant-based alternative products that the assessment of the nutritional quality of these foods should be assessed on the ingredients and the general study situation rather than on the classifications such as NOVA [8].

Overall, it is clear that not all foods classified as 'ultra-processed' according to the NOVA system necessarily have poor nutritional quality. As described above, this is due to the fact that the NOVA classification does not capture the complex effects of food processing in a sufficiently differentiated way. Therefore, with the NOVA classification, it is impossible to draw conclusions about which specific aspects of food processing are responsible for the observed health effects. Often, a differentiated view of the health effects of the degree of food processing is also not possible due to limitations in the level of detail of the nutritional data from existing epidemiological studies. Prospective cohort studies that include detailed dietary surveys are therefore needed to assess long-term health effects. The question of the fundamental causes and biological mechanisms of the observed relationships remains largely unanswered at present. In order to be able to answer this question and derive causal relationships, targeted randomised controlled intervention studies would be required. In addition, more knowledge must be generated about the effects of different processes on food ingredients in order to correlate them with health effects.

Conclusion
Food processing plays a crucial role in our diet today. When done correctly, its advantages, such as improved nutrient supply, greater food safety and higher sustainability, are undeniable. A blanket devaluation of processed foods, including so-called 'ultra-processed foods', therefore fails to do justice to the multifaceted nature of the issue. Classification systems such as NOVA do not sufficiently differentiate between the complex effects of food processing. This makes it impossible to identify which specific aspects of food processing are responsible for the observed health effects. 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. For this purpose, in cohort studies it is necessary to collect as detailed nutritional data as possible, and the data situation must be complemented by intervention studies.