Emotional Stability and Emotional Eating in Medical Students: Exploring Indirect Associations Involving Anxiety and Depressive Symptoms

Background/Objectives: Emotional stability is a personality trait associated with emotion regulation and adaptation to stress. Low emotional stability has been associated with anxiety, depressive symptoms, and emotional eating, but the mechanisms underlying these relationships among medical students remain unclear. This study examined the association between emotional stability and emotional eating and explored indirect statistical associations involving anxiety and depressive symptoms using exploratory parallel and serial path analyses. Methods: A cross-sectional observational study was conducted among 291 medical students. Emotional stability was assessed using the DECAS Personality Inventory, emotional eating using the DEBQ (Dutch Eating Behavior Questionnaire), anxiety symptoms using the GAD-7 (Generalized Anxiety Disorder-7), and depressive symptoms using the PHQ-9 (Patient Health Questionnaire-9). Spearman correlations and exploratory parallel and serial path analyses were performed using the PROCESS macro for SPSS Statistical Calculator v26 with 5000 bootstrap samples. Results: Emotional stability was negatively correlated with anxiety (ρ = −0.546; p < 0.001), depressive symptoms (ρ = −0.603; p < 0.001), and emotional eating (ρ = −0.383; p < 0.001). Anxiety (ρ = 0.380; p < 0.001) and depressive symptoms (ρ = 0.454; p < 0.001) were positively correlated with emotional eating. Only depressive symptoms showed a significant indirect statistical association linking emotional stability and emotional eating (B = −0.0139; 95% bootstrap CI: −0.0223 to −0.0068). In the serial path analysis, only the model specifying anxiety as the first intermediate variable showed a significant indirect effect (B = −0.0065; 95% bootstrap CI: −0.0109 to −0.0032). Conclusions: Depressive symptoms showed the strongest indirect association linking emotional stability and emotional eating in medical students. These findings should be interpreted as being consistent with the proposed statistical model and do not establish temporal or causal relationships.