Abstract

The Nova system was proposed as a means of categorising food products according to the degree and methods of processing. The term ‘ultra-processed’ was used to indicate foods with the most extensive processing. When applied to research cohorts, foods in this category have shown widely heterogeneous associations with diabetes, cardiovascular disease and other chronic conditions, with some foods associated with increased risk and others associated with significantly reduced risk, raising important considerations for public policy. The purpose of this review is to summarise the findings of studies analysing the associations of ultra-processed foods with the risk of cardiometabolic diseases. Out of 309 records found on PubMed, 14 publications that differentiated between ultra-processed food groups were included in the review. Positive associations between ultra-processed food consumption and adverse health outcomes have been found for processed meats, sweetened beverages (both sugar-sweetened and artificially sweetened), and fats, spreads and sauces. Other ultra-processed foods have been associated with lower risk of these conditions, mainly breads, biscuits and breakfast cereals, and whole grains. Randomised clinical trials show that replacing animal products with plant-based foods, even when processed, leads to health benefits. These studies suggest that the Nova system may be improved by differentiating between animal and plant foods.

Introduction

The Nova classification was proposed to differentiate food products into four categories according to the degree and methods of processing. Consumption of some ultra-processed foods (ie, those with the most extensive processing) has been associated with increased incidence of diabetes, cardiovascular disease and other chronic conditions,1 while consumption of others is associated with a decreased risk.2 The purpose of this review is to summarise the studies that analyse the associations of different ultra-processed foods with the risk of cardiometabolic diseases.

Methods

A search of the PubMed database was conducted using the search terms ‘ultra-processed food subgroups’, without limit regarding language or date of publication. This search resulted in 309 records. A more refined search was also conducted, using additional search terms ‘ultra-processed foods and diabetes’; ‘processed foods and diabetes’; ‘ultra-processed foods and vascular disease’; ‘processed foods and vascular disease’; ‘ultra-processed foods and obesity’; ‘processed foods and obesity’; ‘ultra-processed foods and bodyweight’; and ‘processed foods and bodyweight’. After excluding duplicate records and papers that did not differentiate between ultra-processed foods (UPF) subgroups, 14 papers met the criteria and were included in the review (see figure 1). These included 12 prospective studies and two randomised clinical trials. The summary of all papers included in this review is shown in online supplemental table 1.

Figure 1

Search flow chart.

Ultra-processed foods and diabetes

Seven prospective studies reported associations between consumption of specific ultra-processed food categories and the risk of diabetes. Across these studies, the consumption of ultra-processed foods, as a group, was associated with an increased risk of type 2 diabetes. However, major differences between the various food groups were apparent.

A 2023 analysis of the data from three large US cohorts showed that while the consumption of animal-based ultra-processed foods and artificially sweetened and sugar-sweetened beverages was associated with an increased risk of type 2 diabetes, consumption of ultra-processed breads and cereals and packaged sweet and savoury snacks was associated with a reduced risk.2

These findings were confirmed by a 2023 analysis of a large cohort study from seven European countries in the European Prospective Investigation into Cancer and Nutrition (EPIC) study. The consumption of ultra-processed foods overall was associated with an increased risk of multimorbidity of cancer and cardiometabolic diseases, including diabetes. Looking at specific food groups, animal-based products and artificially sweetened and sugar-sweetened beverages were associated with an increased risk, while ultra-processed breads and cereals were associated with a lower risk.3

Similarly, a prospective analysis of 311 892 individuals in the EPIC cohort found that each 10% g/day higher intake of savoury snacks (HR 2.77; 95% CI 1.09 to 7.05), animal-based products (HR 2.25; 95% CI 1.96 to 2.57), ready-to-eat/heat mixed dishes (HR 1.16; 95% CI 1.01 to 1.35) and artificially sweetened or sugar-sweetened beverages (HR 1.25; 95% CI 1.22 to 1.28) was associated with a higher incidence of type 2 diabetes. In contrast, breads, biscuits and breakfast cereals (HR 0.65; 95% CI 0.57 to 0.73), sweets and desserts (HR 0.89; 95% CI 0.84 to 0.95) and plant-based alternatives (HR 0.46; 95% CI 0.26 to 0.82) were associated with a lower incidence of diabetes.4

An analysis of 31 172 Americans in the Atherosclerosis Risk in Communities Study found that the highest quartile consumption of sugar-sweetened and artificially sweetened beverages, ultra-processed meats and sugary snacks was associated with a 29%, 21% and 16% higher risk of diabetes, respectively, compared with the lowest quartile. In contrast, baked goods and ice cream were associated with a lower risk of diabetes (by 12% each).5

A prospective study of 7438 adults in the Korean Genome and Epidemiology Study Ansan-Ansung cohort determined that a higher intake of ham/sausage, ice cream, instant noodles and carbonated beverages was associated with a 40%, 8%, 7% and 2% increased risk of diabetes, whereas a higher intake of candy/chocolate was associated with a decreased risk of diabetes by 22%.6

A study of 15 105 adults in the Brazilian Longitudinal Study of Adult Health found that among ultra-processed food subgroups, only the consumption of processed meats and sweetened beverages was associated with increased diabetes risk.7

Finally, a study of 7774 British civil servants from the Whitehall II cohort found that out of all ultra-processed food subgroups, only the intake of soft drinks was significantly associated with an increased risk of diabetes (HR 1.02).8

Ultra-processed foods and sex differences

No sex differences were observed in the Nurses’ Health Study, Nurses’ Health Study II, Health Professionals Follow-Up Study or in the EPIC cohort.2 4 A prospective study conducted with 896 participants from the 1978/79 Ribeiro Preto cohort in Brazil found no association between ultra-processed food consumption and metabolic syndrome, but consumption of ultra-processed foods was associated with increased risk of abdominal obesity and low high-density lipoprotein (HDL)-cholesterol in women but not men.9

Ultra-processed foods and cardiovascular disease

A study of 3042 men and women from Greece found that among ultra-processed food subgroups, only artificially sweetened and sugar-sweetened beverages (OR 1.06; 95% CI 1.02 to 1.13) and spreads and sauces (OR 1.06; 95% CI 1.02 to 1.13) had a positive association with cardiometabolic multimorbidity. For breads and cereals, an inverse association was observed (OR 0.95; 95% CI 0.92 to 0.98) and there was no significant association found in the remaining categories.10

An analysis of 8293 participants in the Atherosclerosis Risk in Communities Study found that participants in the highest quartile of consumption of sugar‐sweetened beverages, red and processed meat, and margarine had a 16%, 10% and 6% higher risk of incident hypertension, respectively, when compared with the lowest quartiles of consumption. However, those in the highest quartile of intake of cold breakfast cereal and dairy had a 10% and 11% lower risk of incident hypertension compared with the lowest quartiles.11

A study of 64 934 women in the Mexican Teachers’ Cohort found that comparing extreme categories showed that higher total and solid ultra-processed food consumptions were not associated with incident hypertension. However, liquid ultra-processed foods (which included sugar-sweetened beverages and distilled liquor) and processed meats were associated with increased hypertension (incidence rate ratio 1.32; 95% CI 1.10 to 1.65; and 1.17; 95% CI 1.01 to 1.36, respectively). Added fats were inversely associated with hypertension (incidence rate ratio 0.82; 95 % CI 0.73 to 0.94). Dairy products, sugary products, cereals, fast food, snacks and alcoholic beverages were not associated with hypertension.12

Ultra-processed foods and all-cause mortality

An analysis of 74 563 women and 39 501 men from the Nurses’ Health Study and Health Professionals Follow-Up Study found that meat/poultry/seafood-based ready-to-eat products consistently showed strong associations with mortality outcomes (HRs ranged from 1.06 to 1.43). Sugar-sweetened and artificially sweetened beverages (HR 1.09; 95% CI 1.07 to 1.12), dairy-based desserts (HR 1.07; 95% CI 1.04 to 1.10) and ultra-processed breakfast foods (HR 1.04; 95% CI 1.02 to 1.07) were also associated with higher all-cause mortality. In contrast, consumption of sweet snacks and desserts was associated with a lower cancer mortality (HR 0.94; 95% CI 0.89 to 0.98).13

Ultra-processed foods in randomised clinical trials

A secondary analysis of a randomised clinical trial in 244 overweight adults in the USA showed that replacing animal products, regardless of their level of processing, with plant-based foods may be an effective weight-loss strategy, even when processed plant-based foods are included.14 Similarly, a vegan diet that included unprocessed, processed and ultra-processed plant foods was associated with a significant weight loss and a reduction in severe hot flashes in postmenopausal women.15

Summary of the findings

The summary of the findings is shown in table 1. The ultra-processed food groups most consistently positively associated with diabetes and other adverse health outcomes (ie, their higher consumption was associated with an increased risk) were processed meats (such as ham, sausage, ready-to-eat meat, ultra-processed meat),2–5 7 11 13 sweetened beverages (both sugar-sweetened and artificially sweetened)2–5 7 8 10 11 13 and fats, spreads and sauces (eg, margarine, added fats/condiments).10 11

Table 1

•

Summary of the findings. UPF food groups associated with an increased risk and a decreased risk, and food groups with no association

In contrast, the ultra-processed food groups most consistently negatively associated with diabetes and other adverse health outcomes (ie, their higher consumption was associated with a lower risk) were breads, biscuits and breakfast cereals, and2 4 10 11 whole grains16 17 and plant-based meat alternatives,4 although one study showed no association.3

Mixed results have been observed for sweets and desserts, depending on which foods were included in this category. Specifically, consumption of ice cream, sugary snacks and dairy-based desserts was associated with an increased risk of diabetes and other health outcomes,5 6 13 while consumption of packaged sweet snacks, candy and/or chocolate was negatively associated with the incidence of diabetes and other health outcomes.2 4 6

Discussion

The published studies indicate that within the Nova ultra-processed food category, foods have highly variable associations with disease incidence. Processed meats, artificially sweetened and sugar-sweetened beverages, and fats, spreads and sauces are the main drivers of the association of ultra-processed food consumption and diabetes and other adverse health outcomes.

Processed meat consumption has been shown to increase the risk of type 2 diabetes,18 cardiovascular disease19 and all-cause mortality20 through several mechanisms. Processed meats are rich in saturated fat, sodium and nitrates/nitrites, which promote hypertension, dyslipidaemia, endothelial dysfunction and systemic inflammation—all key drivers of cardiometabolic disease.19 Processed meat is also typically high in haem-iron, which enhances lipid peroxidation and the endogenous formation of N-nitroso compounds, contributing to oxidative stress and insulin resistance.21 22 Additionally, certain components such as L-carnitine and choline are metabolised by the gut microbiome to form trimethylamine-N-oxide, a metabolite associated with atherosclerosis and major adverse cardiovascular events and death.23 Furthermore, high levels of advanced glycation end products formed during processing and high-heat cooking further impair insulin signalling and promote metabolic dysfunction.24 Together, these mechanisms create a proinflammatory, pro-oxidative environment that increases the risk of diabetes, cardiovascular disease and other chronic conditions.

Sugar-sweetened and artificially sweetened beverage consumption has been linked to an increased risk of type 2 diabetes, cardiovascular disease and all-cause mortality.25 For sugar-sweetened beverages, one possible explanation is the excess energy intake at high doses when sugars from sugar-sweetened beverages make more than 20% of total energy or more than 100 g/day.26 It has been demonstrated that this excess energy intake of liquid calories is not fully compensated for at subsequent meals.27 The associations of artificially sweetened beverage consumption with adverse health outcomes are still to be explored. The role of altering the gut microbiome and the possible link with insulin resistance has been discussed, but the evidence is still inconclusive.28 However, the similar magnitude of the association of artificially sweetened and sugar-sweetened beverages with adverse health outcomes suggests that their consumption may be just a proxy for an unhealthy dietary pattern.

The positive association between dietary fats, spreads and sauces with type 2 diabetes, cardiovascular disease and all-cause mortality can be explained through two main mechanisms. First, the relatively high energy density of these foods may lead to a positive energy balance and weight gain.29 And second, their high saturated fat content may promote hepatic fat accumulation,30 decrease insulin sensitivity31 and increase inflammation and cardiovascular and all-cause mortality.32

In contrast, breads, biscuits, breakfast cereals and whole grains have been negatively associated with diabetes, cardiovascular disease and all-cause mortality. This is in line with previous studies.33 For example, a 2024 cross-sectional analysis of the 2015–2018 National Health and Nutrition Examination Survey has demonstrated that an exclusion of whole grains from the Nova ultra-processed food category made no difference in cardiometabolic risk markers.34 Furthermore, in a meta-analysis comparing high versus low consumption, whole grains were associated with a 16% lower risk of all-cause mortality, a decreased risk of cardiovascular mortality by 18% and a 12% reduced risk of cancer mortality.16 Data from three large US prospective studies showed that after adjusting for lifestyle and dietary risk factors for diabetes, participants in the highest category for total whole grain consumption had a 29% lower risk of type 2 diabetes compared with those in the lowest category.17

According to the Dietary Guidelines for Americans, 2020–2025, only 2% of the US population eats three or more servings of whole grains a day. Ultra-processed breads and cereals may assist in increasing this percentage. It is noteworthy that milling of grains does not necessarily remove fibre, which is responsible for many of the beneficial effects of these foods. Dietary fibre promotes satiety and healthy weight control, attenuates postprandial glucose and insulin responses and improves insulin sensitivity. Whole grains are also rich in micronutrients, antioxidants, phytochemicals (eg, phenolic acids, lignans) and magnesium, each of which contributes to reduced oxidative stress and lower blood pressure.35 Furthermore, fermentation of fibre found in grains by the gut microbiota generates short-chain fatty acids, which improve glycaemic control and lipid metabolism, and reduce systemic inflammation and endothelial dysfunction.36

The association between plant-based meat alternatives and a lower risk of diabetes may be explained by several mechanisms. Compared with conventional meat products, which these alternatives usually replace, they typically contain substantially less saturated fat and haem iron, and more fibre, which may help increase insulin sensitivity and reduce oxidative stress and serum levels of trimethylamine-N-oxide.37 Moreover, replacing animal meat with plant-based alternatives reduces exposure to nitrites,38 advanced glycation end products24 39 and branched-chain amino acids,40 all of which have been shown to play a role in insulin resistance and risk of diabetes.37

Limitations

This review has two main limitations. First, our findings are predominantly based on observational studies, and therefore, potential residual confounding cannot be excluded. Second, we used the Nova classification, which is a commonly used classification of foods based on the level and method of processing, but which may need further refinement in its description of source foods and processing methods, mainly by differentiating between animal-based and plant-based foods.

Conclusion

Processed meats, artificially sweetened and sugar-sweetened beverages, and fats, spreads and sauces have been consistently shown to be the main driver of the association of ultra-processed food consumption and incidence of diabetes and other adverse health outcomes in observational studies, while consumption of breads, biscuits and breakfast cereals, and whole grains is typically associated with reduced risk. Randomised clinical trials show that replacing animal products with plant-based foods, even when processed, leads to health benefits. These studies suggest that the Nova system may benefit from differentiating between foods of animal versus plant origin.