**Background:** Emotional eating—overeating in response to negative emotions—is more common in women than men. During the luteal phase of the menstrual cycle, many women experience premenstrual syndrome (PMS) or the more severe premenstrual dysphoric disorder (PMDD), which includes symptoms such as mood swings, depression, anxiety, and changes in appetite. Previous research has linked PMS/PMDD to altered eating behaviors, but the role of perceived stress as a mediating factor was not well understood. This study aimed to investigate how PMS/PMDD and negatively perceived stress relate to emotional eating, and whether negative perceived stress mediates the PMS–emotional eating relationship differently between PMDD and non-PMDD groups.
**Methods:** A web-based survey was conducted among 409 women aged 20–39 yrs with a BMI of 18.5–29.9 kg/m², recruited via quota sampling by age and marital status in Seoul and Gyeonggi-do, South Korea. Participants completed the Shortened Premenstrual Assessment Form (SPAF; 10 items, 3 subfactors: affective, pain, water retention), the Negative Perceived Stress Scale (NPSS; 5 items), and the Emotional Eater Questionnaire (EEQ; 10 items, 2 items deleted after confirmatory factor analysis). Participants were divided into a PMDD group (SPAF score ≥27; n = 296, 72.4%) and a non-PMDD group (n = 113, 27.6%). Independent t-tests compared group means, and mediation analyses using the PROCESS macro (5,000 bootstrap estimates) tested whether negative perceived stress mediated the effect of PMS on emotional eating.
**Key Results:** The PMDD group had significantly higher mean scores for emotional eating (10.58 ± 5.40 vs. 8.19 ± 4.59, P < 0.001), PMS (39.24 ± 7.87 vs. 19.03 ± 4.75, P < 0.001), and negative perceived stress (10.64 ± 4.21 vs. 8.70 ± 3.83, P < 0.001) compared to the non-PMDD group. No significant difference in BMI was found between groups. In the PMDD group, when all 10 PMS items were the independent variable, PMS significantly predicted negative perceived stress (β = 0.15; 95% CI, 0.09–0.21; P < 0.001), negative perceived stress significantly predicted emotional eating (β = 0.33; 95% CI, 0.18–0.47; P < 0.001), and the indirect effect of PMS on emotional eating via negative perceived stress was significant (β = 0.05; Boot SE = 0.02; 95% CI, 0.02–0.09), indicating partial mediation. When only the 4 affective factors of PMS were the independent variable, the direct effect of PMS on emotional eating was not significant (β = 0.12; P = 0.12), while the indirect effect was significant (β = 0.14; Boot SE = 0.04; 95% CI, 0.07–0.23), indicating full mediation. In the non-PMDD group, negative perceived stress significantly predicted emotional eating (β = 0.33; 95% CI, 0.11–0.55; P < 0.01), but PMS did not significantly predict negative perceived stress or emotional eating, and no significant indirect effect was found.
**Clinical Implications:** This study demonstrates that in women with PMDD, PMS—particularly its affective (emotional) symptoms—influences emotional eating primarily through the mechanism of negatively perceived stress. The finding that affective factors of PMS fully mediated by negative perceived stress suggests that interventions targeting stress perception and coping strategies may be more effective for reducing emotional eating than focusing on PMS symptoms alone. The absence of this mediation pathway in the non-PMDD group indicates that only women who experience clinically significant premenstrual symptoms are at risk for this stress–eating cascade. Given that women experience menstruation for approximately 40 years, managing both PMS symptoms and the cognitive appraisal of PMS-related stress may be important for promoting healthy eating behaviors and overall well-being. Limitations include the cross-sectional design (which precludes causal inference), the use of self-report measures, and the need for further research on covariates and specific subfactors of eating behavior.