Exploring the association between hedonic hunger and obesogenic eating behaviors in females with overweight/obesity using machine learning methods

Hedonic hunger, the drive to eat for pleasure, may be associated with some obesogenic eating behaviors. Therefore, this study investigated the association between hedonic hunger and these behaviors in females with overweight/ obesity. In this cross-sectional study, 405 healthy female volunteers with overweight/obesity were recruited. The Power of Food Scale was used to evaluate hedonic hunger. The External Eating subscale of the Dutch Eating Behavior Questionnaire, the Salzburg Emotional Eating Scale, the Food Craving Inventory, and the Salzburg Stress Eating Scale were used to assess eating behaviors. Crude and adjusted linear regression models, along with two machine learning methods (conditional tree and random forest), were used for data analysis. The mean age and BMI of participants were 35.93 ± 9.07 years and 30.71 ± 3.65 kg/m2, respectively. Both crude and adjusted linear regression models demonstrated a significant positive association between hedonic hunger and the total score of various studied obesogenic eating behaviors. According to the conditional tree analysis, hedonic hunger was the main predictor in classifying individuals based on the scores of all examined eating behaviors. Additionally, random forest results indicated that hedonic hunger was the most important predictor of these behaviors. In conclusion, this study provides evidence for a significant positive association between hedonic hunger and obesogenic eating behaviors. Future research is recommended to investigate the roles of possible mediators in this association.