To verify if the intake of ultra-processed foods is associated with higher BMI and waist circumference (WC) among participants of the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil) cohort.

Cross-sectional analysis of the ELSA-Brasil baseline (2008–2010). Dietary information obtained through an FFQ was classified according to characteristics of food processing (NOVA) and used to estimate the percentage energy contribution from ultra-processed foods (i.e. industrial formulations, elaborated from food processing, synthetic constituents and food additives) to individuals’ total energy intake. BMI and WC and their respective cut-off points served as response variables. Associations were estimated through linear and multinomial logistic regression models, after adjusting for confounders and total energy intake.

Six Brazilian capital cities, 2008–2010.

Active and retired civil servants, aged 35–64 years, from universities and research organizations (n 8977).

Ultra-processed foods accounted for 22·7 % of total energy intake. After adjustments, individuals in the fourth quartile of percentage energy contribution from ultra-processed foods presented (β; 95 % CI) a higher BMI (0·80; CI 0·53, 1·07 kg/m2) and WC (1·71; 1·02, 2·40 cm), and higher chances (OR; 95 % CI) of being overweight (1·31; 1·13, 1·51), obese (1·41; 1·18, 1·69) and having significantly increased WC (1·41; 1·20, 1·66), compared with those in the first quartile. All associations suggest a dose–response gradient.

Results indicate the existence of associations between greater energy contribution from ultra-processed foods and higher BMI and WC, which are independent of total energy intake. These findings corroborate public policies designed to reduce the intake of this type of food.