**Background:** Late eating has been associated with adverse metabolic health outcomes, possibly through poor diet quality and higher energy intake. However, no prior studies examined whether meal timing is associated with the degree of food processing, which is an independent predictor of health outcomes. Ultra-processed foods (UPFs) are increasingly recognized as a major health threat, linked to obesity, cardiovascular disease, cancer, and mortality.
**Methods:** Data were from the Italian Nutrition & HEalth Survey (INHES), a telephone-based survey conducted from November 2010 to November 2013. After exclusions (age <20 years, missing diet data, extreme energy intakes, missing meal timing), 8,688 participants (4,053 men, 4,635 women; mean age 56.9±14.6 years) were analyzed. Dietary data were collected via a single 24-hour dietary recall. Foods were categorized using the NOVA classification into: (1) minimally processed foods, (2) culinary ingredients, (3) processed foods, and (4) ultra-processed foods. Consumption was calculated as the proportion (%) of each NOVA group on total weight of food eaten (g/d). Meal timing was assessed using population median times: breakfast after 7am, lunch after 1pm, dinner after 8pm. Participants scoring ≥2 points (out of 3) were classified as late eaters. Multivariable-adjusted linear regression models were used, with two models: Model 1 adjusted for age, sex, and energy intake; Model 2 additionally adjusted for education, geographical area, place of residence, sport activity, occupation, marital status, smoking, BMI, cardiovascular disease, cancer, hypertension, diabetes, and hyperlipidaemia. Missing data were handled using multiple imputation.
**Key Results:** The average weight contributions to the diet were: 73.7% (±12.0) unprocessed/minimally processed foods, 2.6% (±1.2) culinary ingredients, 15.9% (±10.7) processed foods, and 7.8% (±7.0) UPFs. Late eaters were younger, more likely to live in Southern Italy and urban environments, had higher educational levels, and were more likely to be non-manual workers compared to early eaters. In fully adjusted models (Model 2), late eaters consumed significantly less minimally processed food (β = −0.10; 95% CI −0.14 to −0.06) and more UPFs (β = 0.13; 95% CI 0.09 to 0.18) compared to early eaters. Late eating was also associated with higher consumption of processed culinary ingredients (β = 0.05; 95% CI 0.01 to 0.10), lower adherence to the Mediterranean Diet (β = −0.07; 95% CI −0.12 to −0.03), and a higher (worse) FSAm-NPS dietary index (β = 0.10; 95% CI 0.05 to 0.14). The absolute difference in UPF consumption was approximately 1% higher among late eaters. Subgroup analyses confirmed these associations across age groups and both sexes, though relationships with unprocessed/minimally processed food were stronger in younger participants. Meal-specific analyses showed that late breakfast was associated with higher UPF intake at breakfast, and late dinner was associated with higher processed food and UPF intake at dinner.
**Clinical Implications:** This study provides novel evidence that late eating is associated with higher consumption of UPFs and lower intake of minimally processed foods, independent of overall diet quality. Given that even small increments in UPF consumption (approximately 1% absolute increase) have been linked to higher mortality risk in prior cohort studies, these findings suggest that increased UPF intake may be a mechanism linking mistimed meals to adverse cardiometabolic outcomes. Anticipating meal timing may represent a complementary strategy for reducing UPF consumption. However, the cross-sectional design limits causal inference, and the findings may not generalize beyond the Italian population. Future studies should examine whether UPF consumption mediates the association between late eating and adverse health outcomes.