Effect of the food processing degree on cardiometabolic health outcomes: A prospective approach in childhood

Abstract

Background & aims: Ultra-processed foods (UPF) consumption has increased greatly over the last decades but its impact on health remains unclear. This study aimed to evaluate the prospective effect of different degrees of food processing on children's cardiometabolic profile.

Methods: Children from the Portuguese population-based birth cohort Generation XXI evaluated at 7 and 10 years of age (y), with dietary information at 7 y and anthropometric measurements at 10 y, were included in the present study (n = 3034). Dietary data were collected by 3-day food diaries and all reported food items were classified according to the degree of processing using the NOVA classification system. The daily consumption of the NOVA classification groups was calculated (in grams, as a proportion of total grams and total energy intake). The cardiometabolic outcomes included were serum markers [fasting glucose, insulin, blood lipids], anthropometric measurements [body mass index (BMI) and waist circumference (WC)], body composition [fat-free mass (FFM) and fat mass (FM)] and blood pressure (BP). Age- and sex-specific sample z-scores were obtained for all outcomes. Principal Component Analysis was used to identify a potential cluster of cardiovascular risk factors. Generalized linear models were fitted to evaluate the association between the consumption according to the processing degree (as absolute gram intake) and the individual cardiometabolic risk factors and cluster, adjusted for child's sex, maternal age and education and remaining NOVA groups. A mediation role of children's BMI in the previous analysis were tested.

Results: At 7 y, the daily median consumption of un/minimally processed, processed and UPF was 1210 g, 113 g and 433 g (68%, 6% and 25% of the total grams, as well as 51%, 15% and 31% of the total energy intake, respectively). After adjustment, an increase of 100 g in the consumption of un/minimally processed at 7 y was associated with a lower BMI (βˆ = -0.028; 95%CI: -0.043; -0.014), WC (βˆ = -0.020; 95%CI: -0.032; -0.008), FM (βˆ = -0.023; 95%CI: -0.035; -0.011), insulin (βˆ = -0.022; 95%CI: -0.036; -0.008) and BP (systolic BP: βˆ = -0.014; 95%CI: -0.025;-0.004; diastolic BP: βˆ = -0.013; 95%CI: -0.020;-0.005), at 10 y. BMI was a mediator between the consumption of un/minimally processed foods and BP. Positive associations were found between consumption of processed foods and the cluster 'higher blood pressure', while an inverse association was described between un/minimally processed and the cluster 'higher blood pressure'. No significant associations were found between the UPF consumption and any cardiometabolic outcomes.

Conclusions: High consumption of unprocessed or minimally processed foods has a favourable effect on later children's cardiometabolic health, namely lower body weight and body fat, lower waist circumference, blood pressure insulin serum levels.

Keywords: Cardiometabolic health; Children; Cohort studies; Obesity; Ultra-processed foods.