Household income and food addiction symptoms predict ultra-processed foods intake in Brazilian university students

Implications for policy

Bibliographic Data

ID24226554
AuthorsGuilherme Macedo Soares Coutinho, Jasmin Bruna Stariolo, Thayane Carvalho Lemos, Camila Monteiro Fabricio Gama (0000-0003-4435-3177), Arthur Viana Machado (0000-0001-8824-1615), Daniela Silva Canella (0000-0001-9672-4983), Mirtes G Pereira (0000-0003-4682-223X), Leticia De Oliveira (0000-0002-8450-2264), Isabel Antunes David (0000-0003-3894-8043)
Year2026
Volume39
Issue1
Publication date2026-07-01
Open AccessYes
TypeARTICLE
VenuePsicologia Reflexão e Crítica (JOURNAL)
Journal identifiersISSN: 0102-7972 • E-ISSN: 1678-7153
PublisherSpringer Science and Business Media LLC (PUBLISHER)
DOI10.1186/s41155-026-00405-1
PMID42384318
OpenAlexW7166848167
LanguageEN
References cited71

In Brazil, ultra-processed food consumption has increased disproportionately among lower-income groups, and food insecurity has been linked to food addiction. Previous studies have examined income and food addiction separately as predictors of ultra-processed food intake. This study evaluated the joint contribution of household income and food addiction to ultra-processed food intake in Brazilian university students. METHODS: A cross-sectional online study was conducted with 368 Brazilian university students aged 18-30 years. Ultra-processed food intake was assessed using a score based on the Nova classification, reflecting the number of ultra-processed food subgroups consumed on the previous day. Food addiction was measured with the modified Yale Food Addiction Scale 2.0 and operationalized as diagnostic severity (none, mild, moderate, severe) and symptom presence (zero vs. one or more symptoms). Household income was categorized into four classes (A-D/E). Associations were examined using negative binomial regression, with multivariable models adjusted for body mass index and additional models including gender, race, and perceived stress related to the COVID-19 pandemic. RESULTS: Lower household income was consistently associated with higher intake of ultra-processed foods. Compared with the highest income class, participants in income classes C and D/E showed expected intakes 39% and 42% higher, respectively. When food addiction was analyzed using diagnostic criteria, only the mild diagnosis group differed from the non-food addiction group, showing lower expected intake after adjustment for body mass index. In contrast, symptom-based analyses revealed that participants with one or more food addiction symptoms had an approximately 18% higher expected ultra-processed food intake than asymptomatic individuals. Lower income and symptom presence were particularly associated with consumption of ultra-processed foods that replace or accompany meals, as well as ultra-processed beverages. CONCLUSIONS: Household income and food addiction symptoms jointly influence ultra-processed food intake among Brazilian university students. Symptom-based measures appear more sensitive than diagnostic categories for identifying population-level vulnerability. These findings highlight the need for public policies that address both socioeconomic inequalities and addictive-like eating behaviors to reduce excessive ultra-processed food consumption.

Brazilian university students · Food addiction · Food insecurity · Household income · NOVA classification · Public health policy · socioeconomic disparities · ultra-processed foods · university students · Yale Food Addiction Scale · Consumer Attitudes and Food Labeling · Food Security and Health in Diverse Populations · Nutrition, Health and Food Behavior

Tools

Open DOIOpen Access