The feeding control study aims to answer one of the biggest questions in nutritional science: Is UPF harmful because of the way it is made or the ingredients it contains?
Recent articles published in journals modern clinical trials We described a protocol for a randomized controlled trial (RCT) designed to investigate how the unique characteristics of ultra-processed foods (UPFs) influence cardiometabolic risk.
UPF is becoming increasingly prevalent in diets around the world. Meta-analytic evidence suggests that high UPF intake is associated with increased cancer, cardiovascular disease (CVD), diabetes, and mortality. Despite increasing research on UPF, the characteristics underlying its negative health effects remain unclear. Some have suggested that food processing and industrial processes are responsible for the harmful effects of UPF.
The poor nutritional profile of UPF may explain its negative effects on cardiometabolic risk. The 2025 Scientific Recommendations noted that there is no consensus on whether all UPF constitutes a cardiovascular health concern. They also emphasized the need to evaluate individual aspects of UPF, such as industrial processing and nutrient content, on a case-by-case basis. Therefore, identifying the factors underlying the negative effects of UPF is important to inform public health policy.
UPF trial design and participant criteria
In this study, researchers presented a protocol for a 2 × 2 factorial RCT to investigate how UPF modulates cardiometabolic risk. The first factor is the degree of industrial processing of the UPF according to the NOVA definition, and the second is the nutritional content. The main objective of this study is to analyze the individual effects of industrial processing of UPF on nutrient content on cardiometabolic risk in healthy adults. The authors hypothesize that the high saturated fat, added sugar, and sodium content of some UPFs increases cardiometabolic risk factors, but the degree of industrial food processing itself does not.
Participants will be recruited through social media advertisements and the research center’s e-newsletter. Eligible participants must be between 18 and 75 years old, have a body mass index (BMI) between 18 kg/m2 and 35 kg/m2, stable weight, low-density lipoprotein cholesterol (LDL-C) <5 mmol/L, blood pressure (BP) <150/90 mmHg, and glycated hemoglobin (HbA1c) <6.5%.
People with diabetes, CVD, hypertension, or cancer and those taking blood pressure-lowering, lipid-lowering, or diabetes medications are excluded. Also excluded are those who are pregnant or breastfeeding, consume more than 7 drinks of alcohol per week, have associated food allergies or aversions, or follow a diet that may limit compliance. Participants will be randomized to consume a low UPF diet low in saturated fat, added sugars, and sodium (SFSS), a low UPF diet high in SFSS, a high UPF diet low in SFSS, or a high UPF diet high in SFSS for 6 weeks.
Controlled nutritional diet and measurement of its results
A cyclical 7-day daily menu standardized to 2,500 kcal of protein (15% kcal), carbohydrate (50% kcal), and fat (35% kcal) was developed for each meal. Each item on the menu was classified into one of the four groups of the NOVA classification. Group 1 – minimally processed or unprocessed foods. Group 2 – Processed cooking ingredients. Group 3 – Processed foods. Group 4 – UPF.
High UPF meals primarily include commercially available foods classified as NOVA Group 4, such as flavored yogurt, oatmeal raisin cookies, waffles, prepackaged spaghetti with sauce, and ready-to-eat coleslaw. Low-UPF diets primarily include custom-made recipes and foods classified in NOVA groups 1-3, such as plain yogurt with maple syrup and frozen fruit, spaghetti with homemade sauce, homemade oatmeal raisin cookies, and pancakes.
Participants will visit the study cafeteria on two to three weekdays to consume supervised meals and collect meals/snacks for the rest of the day and the next day. They will abstain from alcohol, probiotics, natural health products, artificially sweetened beverages, and fiber supplements during the study period. Dietary adherence is assessed daily using a self-report checklist. Blood lipids, glucose, and insulin will be measured at baseline and after the intervention.
Primary outcomes included changes from baseline in LDL-C, daytime ambulatory systolic blood pressure (dtSBP), and Homeostasis Assessment Model of Insulin Resistance (HOMA-IR) at the end of the intervention. Secondary outcomes included changes from baseline in triglycerides, ApoB-100, high-density lipoprotein cholesterol (HDL-C), 24-hour ambulatory systolic blood pressure, 24-hour ambulatory diastolic blood pressure, daytime ambulatory diastolic blood pressure, insulin, and glucose.
In addition, questionnaires will be administered to investigate perceived healthiness of the meal, palatability of the meal, feelings of hunger and satiety, dimensions of eating pleasure, and physical activity. Dual-energy X-ray absorptiometry scans will be performed to assess body composition and adiposity at the beginning and end of the intervention. A mixed model will be used for data analysis, following an intention-to-treat approach.
Implications for UPF nutrition policy
In summary, the proposed RCT will help elucidate which characteristics of UPF are responsible for its negative health effects, advance current knowledge, and help inform policy makers to better incorporate UPF into future guidelines. RCTs have several strengths. Factorial designs, sample sizes, and controlled feeding conditions allow us to disentangle the individual effects of industrially processed UPF on cardiometabolic risk. However, because meals are provided under isocaloric conditions to maintain body weight, this study does not directly test whether UPF promotes weight gain under excess caloric intake or ad libitum conditions.
Reference magazines:
- Rochette M, Charest A, Gigroux I, Couture P, Provencher V, Lamarche B (2026). How ultra-processed foods modulate cardiometabolic risk: A 2 × 2 factorial randomized controlled trial protocol in healthy adults. Modern Clinical Trials, 165, 108323. Doi: 10.1016/j.cct.2026.108323, https://www.sciencedirect.com/science/article/pii/S1551714426001096
