Anthropometric, Nutritional, and Lifestyle Factors Involved in Predicting Food Addiction: An Agnostic Machine...

Area of Science:

• Behavioral Science
• Nutritional Science
• Psychiatry

Background:

• Food addiction (FA) is an emerging psychiatric condition with similarities to other addictions.
• Early identification of FA is crucial to prevent the development of more severe disorders.

Purpose of the Study:

• To determine if anthropometric measures, eating habits, eating disorder (ED) symptoms, and lifestyle features can predict FA symptoms.
• To develop and validate prediction models for FA in a university student population.

Main Methods:

• A cross-sectional study involving 702 university students.
• Data collection included the Yale Food Addiction Scale 2.0 (YFAS 2.0), Eating Attitudes Test (EAT-26), Food Frequency Questionnaire (FFQ), anthropometric measurements, and lifestyle questionnaires.
• Multivariate logistic regression and machine learning approaches were used for analysis.

Main Results:

• 6.4% of participants showed symptoms compatible with FA; 8.1% were at risk for ED.
• Individuals with FA had higher BMI, waist circumference, and body fat, and a higher risk of ED.
• Non-dairy beverage consumption, vitamin D deficiency, and waist circumference significantly predicted FA symptoms (R²Nagelkerke = 0.349).

Conclusions:

• Prediction models accurately identified FA in university students.
• An individualized approach, considering eating behavior, body composition, and nutritional deficits, improved FA identification.
• The study highlights the complex interplay of factors contributing to food addiction.