**Background:** Obesity is a global pandemic and a major risk factor for chronic non-communicable diseases (NCDs). Health risk behaviors such as inadequate eating, physical inactivity, and sedentary behavior often cluster together, amplifying health risks. University students are particularly vulnerable due to lifestyle changes during the transition to higher education. This study aimed to estimate the prevalence of co-occurring obesogenic behaviors and identify associated sociodemographic and health-related factors among Brazilian university students.
**Methods:** This cross-sectional study was conducted at Universidade Federal de Ouro Preto (UFOP), Minas Gerais, Brazil, using baseline data from the PADu longitudinal study. Participants were 351 university students (from 355 enrolled) entering undergraduate courses in the first and second semesters of 2019, aged ≥18 years. Data were collected via self-administered questionnaires between April and September 2019. The outcome was the co-occurrence of three obesogenic behaviors: inadequate eating practices (measured using the validated Gabe and Jaime scale, with scores ≤31 classified as inadequate), leisure-time physical inactivity (assessed via VIGITEL survey question), and sedentary behavior (screen time >6 hours/day based on adapted VIGITEL questions). Explanatory variables included sociodemographic characteristics (sex, age, skin color, sexual orientation, marital status, income, housing, area of knowledge, work status) and health conditions (self-rated health, BMI, medication use, anxiety/depression/stress symptoms via DASS-21). Statistical analyses included frequency distributions, Venn diagrams for visualizing co-occurrence, chi-square tests (P<0.20 for inclusion), and multivariate logistic regression with backward selection (P<0.05 retained). Three models compared no behavior vs. 1, 2, or 3 behaviors, all adjusted for sex.
**Key Results:** The sample was 57.6% female, 65.8% aged ≤20 years (range 18–31), 51.1% white, 95.4% single, 79.5% heterosexual, 66.4% living without family, and 89.2% not employed. Inadequate eating practices were the most prevalent isolated risk behavior (80.6%; 95% CI: 76.5%–84.8%), followed by sedentary behavior (49.2%; 95% CI: 44.0–54.5%) and leisure-time physical inactivity (37.3%; 95% CI: 32.2–42.4%). The most common combination was inadequate eating practices plus sedentary behavior (23.6%), followed by all three behaviors (17.9%), and inadequate eating plus physical inactivity (15.7%). Only 9.7% had no risk behaviors. In multivariate analysis, age ≤20 years was significantly associated with having 1 behavior (OR 3.68; 95% CI: 1.58–8.59), 2 behaviors (OR 2.77; 95% CI: 1.23–6.26), and 3 behaviors (OR 3.34; 95% CI: 1.23–9.05) compared to older students. Non-white skin color was associated with 1 behavior (OR 3.09; 95% CI: 1.23–7.74), 2 behaviors (OR 4.61; 95% CI: 1.88–11.31), and 3 behaviors (OR 3.31; 95% CI: 1.15–9.58). Poor self-rated health was associated with 2 behaviors (OR 2.70; 95% CI: 1.09–6.71). Depressive symptoms were strongly associated with 3 behaviors (OR 6.15; 95% CI: 2.10–18.05).
**Clinical Implications:** These findings demonstrate that obesogenic behaviors are highly prevalent and cluster among university students, particularly younger students, those with non-white skin color, those reporting poor health, and those with depressive symptoms. The strong association between depressive symptoms and having all three risk behaviors (OR 6.15) underscores the need for integrated mental health and lifestyle interventions in university settings. The high prevalence of inadequate eating practices (80.6%) suggests that university food environments and time constraints are critical targets. Limitations include the single-university sample, cross-sectional design preventing causal inference, and reliance on self-reported data. Nonetheless, results support developing comprehensive health promotion strategies addressing multiple risk behaviors simultaneously, with attention to racial/ethnic disparities and mental health, to reduce long-term NCD risk in this population.