Sociodemographic, Economic, and Health Factors Associated with Ultra-Processed Food Intake Among Older Adults in Chile.

Background/

Objectives

Consumption of ultra-processed foods (UPF) has been linked to poorer diet quality and adverse health outcomes. Although Chile ranks among the highest consumers of UPFs in Latin America, studies using primary dietary data, especially among older adults, are scarce. This study aimed to describe the food intake of Chilean older adults according to the degree of food processing, and to explore the association between UPF intake and sociodemographic, economic and health factors.

Methods

A cross-sectional study of 434 non-institutionalized older adults (≥60 years) living in the Metropolitan Region of Chile was conducted. Dietary intake was assessed using interviewer-administered 24h recall, with a second assessment 8-15 days later in a random subsample (n = 60). Foods were classified according to the NOVA system into minimally processed foods (MPFs), culinary ingredients, processed foods (PF), or UPF. Usual energy intake was estimated using the MSM. Sociodemographic (sex, age, area), economic (income, education, health system), and health-related variables (chronic conditions, sedentary lifestyle, tobacco use) were collected through home-visit questionnaires. Anthropometric and functional measurements were taken by trained nutritionists. The association between UPF intake and studied variables was evaluated using multivariate fractional probit regression, with mean marginal effects presented.

Results

Most of the participants were women (86.2%), aged 70-79 years (47.9%), and residents of urban areas (76.3%). Most of their calories came from MPF (45.7%), followed by PF (25.5%) and UPF (16.6%). Higher UPF intake was associated with living in an urban area (+3.8%; 95% CI 1.2-6.3%), higher education (+3.5%; 95% CI 1.1-6.0%), and being affiliated with the private health system (+9.1%; 95% CI 4.1-14.0%).

Conclusions

In this community-based sample of Chilean older adults, UPF intake was associated with socioeconomic factors but not health status.