To assess ultra-processed food (UPF) consumption and its socio-demographic, psychosocial and behavioural correlates in a general population of Italian children, adolescents and adults.

Cross-sectional telephone-based surveySetting:Italy, 2010–2013.

In total, 9078 participants (5–97 years) from the Italian Nutrition & Health Survey. Dietary intakes were collected by a 1-d 24-h dietary recall. UPF was defined by the NOVA classification and expressed as percentage of total energies.

Average energy intake from UPF (95 % CI) was 17�3 % (17�1 %, 17�6 %) among adults and 25�9 % (24�8 %, 27�0 %) in children/adolescents. Top sources of UPF were processed meats (32�5 %) and bread substitutes (16�7 %). Among adults, age (β = −3�10; 95 % CI (−4�40, −1�80) for >65 years v. 20–40 years; βs are dimensionless) and residing in Southern Italy (β = −0�73; 95 % CI (−1�32, −0�14) v. Northern) inversely associated with UPF. Screen view during meals was directly linked to UPF, as well as poor self-rated health (β = 5�32; 95 % CI (2�66, 7�99)), adverse life events (β = 2�33; 95 % CI (1�48, 3�18)) and low sleep quality (β = 2�34; 95 % CI (1�45, 3�23)). Boys consumed two-point percent more UPF of the total energy than girls (β = 2�01; 95 % CI (0�20, 3�82)). For all ages, a Mediterranean diet was inversely associated with UPF (β = −4�86; 95 % CI (−5�53, −4�20) for good v. poor adherence in adults and (β = −5�08; 95 % CI (−8�38, −1�77) for kids).

UPF contributes a modest proportion of energy to the diets of Italian adults while being one-quarter of the total energies in children/adolescents. UPF was associated with several psychosocial factors and eating behaviours. Increased adherence to Mediterranean diet would possibly result in lower UPF consumption.