Abstract
Background
Premenstrual syndrome (PMS) can negatively affect the quality of life of women of childbearing age, and symptom severity may increase in relation to visceral adiposity. The visceral adiposity index (VAI) is a novel indicator of adipose tissue dysfunction and may be useful in assessing risks associated with PMS and obesity. This study aimed to determine the relationships between PMS, VAI, and eating attitudes in women.
Methods
This cross-sectional study was conducted with 252 female volunteers aged 18–40 years who applied to a Ministry of Health Healthy Nutrition and Active Life Unit in Istanbul, Türkiye. Data were collected using a sociodemographic information form, the Premenstrual Syndrome Scale (PMSS), the Eating Attitudes Test (EAT-26), and a retrospective 24-hour dietary recall. Anthropometric measurements (weight, height, waist circumference) were taken, body composition was determined via bioelectrical impedance analysis (BIA), and VAI was calculated. Mann-Whitney U, Chi-square, and Logistic Regression tests were used for statistical analyses.
Results
PMS was detected in 57.9% of participants. The PMS group showed a significantly higher body fat percentage (38.1% vs. 37.1%, p = 0.039) compared to non-PMS group. Although VAI levels did not differ significantly between the groups, regression analysis revealed that high BMI, rather than PMS status, was the primary independent risk factor for eating disorders.
Conclusion
These findings specifically suggest that body fat percentage plays a more critical role than VAI in PMS symptomatology. While no significant differences in energy or nutrient intake were identified between groups in this study, the results highlight a significant interaction between PMS symptom severity and the risk of eating disorders. This suggests that weight management strategies encompassing both nutritional interventions and body composition targets should be prioritized for individuals with PMS.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12905-026-04412-3.
Keywords: Body mass index, Eating attitudes, Premenstrual syndrome, Visceral adiposity, Women’s Health
Introduction
Premenstrual syndrome (PMS) is a common health problem characterized by a cluster of physical, emotional, and behavioral symptoms that typically occur in the luteal phase of the menstrual cycle and resolve with the onset of menstruation [1]. Although its global prevalence is estimated to be approximately 47.8% [2], this rate varies widely between 20% and 90% depending on diagnostic criteria and cultural factors [3–6]. According to a meta-analysis, the overall prevalence of PMS in Türkiye has been reported as 52.2% [7]. Beyond reducing individual quality of life, PMS is considered a significant public health issue due to its impact on work productivity and social functioning. Although its etiology has not been fully elucidated, current evidence suggests that cyclic fluctuations in progesterone and estrogen levels play a substantial role, particularly through their effects on neurotransmitters such as serotonin and gamma-aminobutyric acid [3, 8].
The most common symptoms of PMS include breast tenderness, bloating, headaches, mood swings, depression, anxiety, anger, and irritability [9]. Additionally, increased appetite, intense cravings for sweet or salty foods, and binge-eating episodes are frequently reported [10]. Recent studies have demonstrated a strong association between PMS symptom severity and emotional eating as well as the risk of disordered eating attitudes, further supporting the potential link between these conditions [11–13]. Literature indicates that reduced serotonin levels during the luteal phase of the menstrual cycle trigger carbohydrate intake to induce temporary relief; however, this cycle may predispose individuals to weight gain, depressive symptoms, and disordered eating behaviors over time [11]. In young adult populations—particularly university students—the combination of body image concerns and PMS symptoms emerges as a significant factor that may increase the risk of developing eating disorders [12]. Furthermore, dietary habits, eating behaviors, body composition, and overall adiposity are recognized as essential determinants in the etiology of PMS [14]. The accumulation of visceral adipose tissue is thought to induce chronic low-grade inflammation, thereby disrupting endocrine homeostasis and potentially exacerbating the severity of PMS symptoms [14–17].
In contrast to the inherent limitations of the conventional body mass index (BMI), emerging anthropometric indices such as the visceral adiposity index (VAI) offer a more sensitive representation of visceral fat distribution and have demonstrated strong associations with metabolic risk profiles and inflammatory processes [18]. Nevertheless, current evidence addressing the interrelationship between PMS, disordered eating risk, and visceral adiposity—particularly through the simultaneous assessment of anthropometric and biochemical parameters—remains limited. Accordingly, the present study seeks to elucidate the influence of PMS on disordered eating attitudes and visceral adiposity in women of reproductive age.
Materials and methods
Study design and participants
This descriptive, cross-sectional study was conducted between June 2024 and April 2025. The study population consisted of 252 volunteer women aged 18–49 years who applied for nutritional counseling at the Healthy Nutrition and Active Life Unit affiliated with the Republic of Türkiye Ministry of Health, Istanbul Sariyer District Health Directorate. The sample size was calculated using G*Power (v3.1.9.7) based on an independent samples t-test design. Assuming a medium effect size (d = 0.5) as suggested by Cohen [19], a type I error rate of 0.05, and a power of 95%, the minimum required sample size was determined to be 210. To compensate for potential data loss, the target sample size was increased by 20%, resulting in a total of 252 participants. Individuals were excluded from the study if they were under 18 or over 49 years of age, postmenopausal, pregnant, or lactating. Furthermore, individuals with a physician-confirmed diagnosis of polycystic ovary syndrome (PCOS) or irregular menstruation, as well as those currently using hormonal contraceptives or psychiatric medications, were not included. Additionally, to ensure that biochemical parameters reflected baseline nutritional status without the confounding effects of supplementation, individuals reporting regular use of vitamin or mineral supplements within the last six months were also excluded from the study. The STROBE flow diagram of the study is presented in Fig. 1. All procedures involving human participants were conducted in accordance with the ethical standards outlined in the 1964 Declaration of Helsinki and its subsequent amendments. Ethical approval was obtained from the Istanbul Bilgi University Human Research Ethics Committee (Date: November 28, 2023, Approval No: 2023-20160-146), and written institutional permission was secured from the relevant health directorate. Written informed consent was obtained from all individual participants included in the study.
Fig. 1.
STROBE flow diagram of the study
Data collection
Data were collected via face-to-face interviews using a questionnaire developed by the researcher based on a review of the relevant literature. During these interviews, participants’ sociodemographic characteristics, and nutritional habits were queried and recorded. Anthropometric measurements were taken directly by the researcher. Additionally, biochemical parameters from the preceding three months were retrieved from laboratory results registered in the medical database.
Sociodemographic information form
The sociodemographic information form was structured to systematically assess key demographic and health-related characteristics of the participants, including age, educational level, socioeconomic status, occupational factors, history of chronic diseases, regular medication use, and smoking and alcohol consumption habits.
Premenstrual syndrome scale (PMSS)
The PMSS, developed by Gençdoğan [20], was employed to evaluate the severity of premenstrual symptoms. This 44-item scale utilizes a 5-point Likert-type scoring system and assesses symptoms experienced during the week preceding menstruation. Total scores range from 44 to 220, with higher scores indicating greater symptom severity. Consistent with the original validation study [20] and recent literature [21], a cut-off score of 110, corresponding to 50% of the maximum possible score, was used to classify the presence of PMS. Participants scoring > 110 were classified as having PMS. The scale comprises nine sub-dimensions: depressive affect, anxiety, fatigue, irritability, depressive thoughts, pain, appetite changes, sleep changes, and bloating [20]. While the Cronbach’s alpha coefficient was reported as 0.75 in the original study, it was calculated as 0.94 in the present study.
Eating attitudes test-26 (EAT-26)
The EAT-26 was utilized to assess the risk and symptoms of eating disorders. Developed initially as EAT-40 by Garner and Garfinkel [22] and revised by Garner et al. [23], the Turkish validity and reliability of the scale were established by Ergüney-Okumuş and Sertel-Berk [24]. This 6-point Likert-type scale consists of three sub-dimensions: Dieting, Bulimia/Food Preoccupation, and Oral Control. Total scores range from 0 to 78, with a clinical cut-off point of 20. Participants scoring 20 or higher were classified as “High Risk for Eating Disorders” while those scoring below 20 were considered to have “Low Risk for Eating Disorders” [24]. The Cronbach’s alpha coefficient, reported as 0.84 in the Turkish adaptation, was calculated as 0.72 in the present study.
Dietary assessment
Dietary intake was assessed using the retrospective 24-hour dietary recall method, and energy and nutrient intake levels were analyzed using the BeBIS (Nutrition Information System) software, version 7.2. The inclusion of individuals with and without PMS who had similar energy and macronutrient intakes enhanced the comparability between groups (data are presented as supplementary material).
Anthropometric measurements
Body weight, fat mass, lean body mass, body fat percentage, and total body water were analyzed using a calibrated TANITA BC-418 MA Bioelectrical Impedance Analysis (BIA) device, following standard measurement protocols [25]. Height was measured using a calibrated stadiometer with the participant standing barefoot, upright, and with the head positioned in the Frankfurt horizontal plane [26]. The BMI was calculated as weight in kilograms divided by the square of height in meters (kg/m²). Participants were categorized according to the World Health Organization (WHO) classification: underweight (< 18.5 kg/m²), normal weight (18.5–24.9 kg/m²), overweight (25.0–29.9 kg/m²), and obese (≥ 30 kg/m²) [27]. Waist, hip, and neck circumferences were measured using a non-stretchable tape measure in accordance with standard guideline [28]. Additionally, Waist-to-Hip Ratio (WHR) and Waist-to-Height Ratio (WHtR) were calculated and evaluated.
Biochemical parameters
Biochemical data were obtained retrospectively from the institutional medical database (Hospital Information Management System) of the Healthy Nutrition and Active Life Unit within the Istanbul Sariyer District Health Directorate, a primary healthcare facility affiliated with the Republic of Türkiye Ministry of Health. As this unit primarily serves individuals applying for routine health screenings and weight management, the parameters reflect a generally healthy population without active inflammatory pathologies. The analysis encompassed lipid profiles (total cholesterol, high density lipoprotein-cholesterol (HDL-C), low density lipoprotein-cholesterol (LDL-C), triglycerides (TG)), inflammation markers (C-reactive protein (CRP), ferritin), minerals (iron, magnesium), and vitamins (vitamin B12, folic acid, vitamin D). To ensure that CRP levels reflected baseline metabolic status rather than acute clinical conditions, individuals with physician-noted acute illnesses or laboratory values indicating active infection at the time of screening were excluded. All recorded values were based on 12-hour fasting blood samples ordered by a physician and analyzed within the three months preceding the study. Since these biochemical data reflect real-world clinical screening rather than a controlled trial setting, standardization of the menstrual cycle phase was not feasible. However, the selected parameters (e.g., vitamin D, vitamin B12, ferritin, lipid profiles) serve as relatively stable markers of long-term nutritional and metabolic status and are significantly less susceptible to acute daily hormonal fluctuations compared to sex steroids.
Visceral adiposity index (VAI)
The VAI is a composite marker that reflects visceral fat accumulation and dysfunction, integrating both anthropometric (BMI, waist circumference) and metabolic (TG, HDL-C) parameters. For female participants, VAI scores were determined using the sex-specific equation:
 |
Where WC is expressed in cm, and TG and HDL-C levels are in mmol/L [29].
Statistical analysis
Statistical analyses were performed using the SPSS software (Statistical Package for the Social Sciences, version 30.0; IBM Corp., Armonk, NY, USA). Appropriate descriptive and inferential methods were applied. The normality of the data distribution was assessed using the Shapiro–Wilk test. Continuous variables were presented as mean ± standard deviation or median [25th–75th percentile], depending on the data’s normality. Categorical variables were expressed as frequencies and percentages. Between-group comparisons were conducted using the Independent Samples t-test for normally distributed variables and the Mann–Whitney U test for non-normally distributed variables. The Pearson Chi-square test was used to analyze associations between categorical variables, and the Kruskal–Wallis H test was applied for comparisons across more than two independent groups. To identify factors associated with the risk of eating disorders, a multivariate binomial logistic regression analysis was performed, including PMS status, age, BMI, and chronic disease as independent variables. Model fit was evaluated using Cox and Snell R², Nagelkerke R², and model chi-square statistics. VAI was excluded from the logistic regression model to avoid multicollinearity, as BMI and waist circumference are already integral components of the VAI formula. A p-value of < 0.05 was considered statistically significant.
Results
Among the 252 women included in the study, 57.9% were identified as having PMS, while 42.1% did not meet the PMS criteria. Examination of lifestyle habits revealed a significant association between smoking and PMS status, with a higher proportion of smokers in the PMS group (28.8%) compared with the non-PMS group (16.0%) (p = 0.018). When specific nutritional habits were examined, patterns such as alcohol consumption was also associated with higher PMSS scores (p = 0.022). Although the prevalence of eating disorder risk did not significantly differ between PMS and non-PMS groups (p = 0.392), participants at high risk for eating disorders had substantially higher PMSS scores (p < 0.001) (Table 1).
Table 1.
Comparison of sociodemographic characteristics and nutritional habits according to the presence of PMS
| Variables | PMS Group (n = 146) | Non-PMS Group (n = 106) | p-Value1 | PMSS Scores | p-Value2 |
|---|---|---|---|---|---|
| Age (years) | 34.47 ± 8.99 | 35.75 ± 8.14 | 0.246* | NA | NA |
| Menstruation duration (d) | 6.36 ± 1.84 | 6.12 ± 1.60 | 0.358** | NA | NA |
| Marital status | 0.068+ | 0.017 ** | |||
| Married | 87 (59.6) | 75 (70.8) | 114.0 [92.8-137.5] | ||
| Single | 59 (40.4) | 31 (29.2) | 126.5 [100.0-148.5] | ||
| Educational status | 0.145+ | 0.177*** | |||
| Primary education | 25 (17.1) | 27 (25.5) | 107.5 [92.3-135.8] | ||
| High school | 51 (34.9) | 27 (25.5) | 122.5 [98.0-144.8] | ||
| University | 70 (47.9) | 52 (49.0) | 119.0 [96.8–142.0] | ||
| Employment status | 0.090+ | 0.036 ** | |||
| Unemployed | 56 (38.4) | 52 (49.1) | 113.0 [88.0-138.5] | ||
| Employed | 90 (61.6) | 54 (50.9) | 122.0 [99.3–144.0] | ||
| Income status | 0.258+ | 0.413*** | |||
| Less than expenses | 39 (26.7) | 19 (17.9) | 122.5 [101.0-142.5] | ||
| Equal to expenses | 80 (54.8) | 66 (62.3) | 118.0 [96.0-142.0] | ||
| Higher than expenses | 27 (18.5) | 21 (19.8) | 118.5 [86.3-141.8] | ||
| Presence of chronic disease | 0.261+ | 0.258** | |||
| No | 92 (63.0) | 74 (69.8) | 122.0 [98.0-144.0] | ||
| Yes | 54 (37.0) | 32 (30.2) | 116.5 [95.0-140.5] | ||
| Smoking status | 0.018 + | 0.010 ** | |||
| Smoker | 42 (28.8) | 17 (16.0) | 129.0 [104.0-148.0] | ||
| Non-smoker | 104 (71.2) | 89 (84.0) | 114.0 [94.0-141.0] | ||
| Alcohol consumption | 0.052+ | 0.022 ** | |||
| Yes | 30 (20.5) | 12 (11.3) | 132.0 [103.0-159.0] | ||
| No | 116 (79.5) | 94 (88.7) | 115.5 [95.0-140.0] | ||
| Daily snack consumption | 0.649+ | 0.766*** | |||
| None | 22 (15.1) | 17 (16.0) | 122.0 [93.0-145.0] | ||
| 1 | 48 (32.9) | 35 (33.0) | 119.0 [98.0-137.0] | ||
| 2 | 53 (36.3) | 43 (40.6) | 116.5 [92.3–142.0] | ||
| ≥ 3 | 23 (15.8) | 11 (10.4) | 122.5 [98.8–151.0] | ||
| Number of main meals | 0.572+ | 0.512** | |||
| ≤ 2 | 81 (55.5) | 55 (51.9) | 120.0 [97.0–144.0] | ||
| 3 | 65 (44.5) | 51(48.1) | 118.0 [95.0–140.8] | ||
| Skipping main meals | 0.214+ | 0.212*** | |||
| Yes | 78 (53.4) | 46 (43.4) | 123.0 [97.3-144.8] | ||
| No | 26 (17.8) | 27 (25.5) | 107.0 [83.0-140.5] | ||
| Sometimes | 42 (28.8) | 33 (31.1) | 117.0 [95.0-140.0] | ||
| EAT-26 Classification | 0.392+ | < 0.001 ** | |||
| High Risk for ED | 40 (27.4) | 24 (22.6) | 125.0 [100.8-158.8] | ||
| Low Risk for ED | 106 (72.6) | 82 (77.4) | 94.0 [0.0-126.0] | ||
Data are presented as Mean ± Standard Deviation, Median [25th − 75th percentile], and number (percentage)
Abbreviations: BMI Body Mass Index, EAT-26 Eating Attitudes Test-26, ED Eating Disorders, NA Not Applicable, PMS Premenstrual Syndrome, PMSS Premenstrual Syndrome Scale
Statistical Analysis: *Independent Samples t-test, **Mann-Whitney U test, ***Kruskal-Wallis H, +Pearson Chi-square test was used
p1: Within-group comparison p-value (PMS Group vs. Non-PMS Group)
p2: Comparison of PMSS Scores by Variables
p < 0.05 was considered statistically significant and indicated in bold
Comparison of anthropometric measurements revealed that the median body fat percentage was significantly higher in the PMS group (38.1%) compared with the non-PMS group (37.1%) (p = 0.039); however, no significant difference was observed in VAI levels between the groups (p > 0.05). When each PMS subgroup was compared by eating disorder risk, the non-PMS group but at high risk for eating disorders had significantly higher waist circumferences (p = 0.014), waist-to-hip ratios (p = 0.006), and waist-to-height ratios (p = 0.015) than those at low risk. In the PMS group, neck circumference was significantly higher in participants at high risk for eating disorders compared to those at low risk (p = 0.049) (Table 2).
Table 2.
Comparison of anthropometric measurements and biochemical parameters of participants according to the presence of PMS status and eating attitude disorder risk
| Variables | n | PMS Group (n = 146) | p 1 | Non-PMS Group (n = 106) | p 1 | p 2 | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| Low Risk for ED (n = 106) | High Risk for ED (n = 40) | Total | Low Risk for ED (n = 82) | High Risk for ED (n = 24) | Total | |||||
| Anthropometric Measurements | ||||||||||
| Body weight (kg)* | 252 | 75.9 [68.9–85.2] | 79.9 [72.2–90.3] | 76.7 [69.5–87.2] | 0.089 | 75.2 [65.6–85.1] | 79.0 [69.6–87.1] | 76.0 [66.3–85.9] | 0.359 | 0.179 |
| BMI (kg/m2)* | 252 | 28.8 [25.6–34.9] | 30.7 [27.8–36.2] | 29.6 [26.3–35.0] | 0.078 | 29.0 [25.6–32.1] | 30.1 [27.0-33.6] | 29.3 [26.1–32.5] | 0.230 | 0.287 |
| Fat (%)* | 252 | 37.8 [32.1–43.0] | 39.0 [37.0-42.9] | 38.1 [33.3–43.0] | 0.117 | 37.1 [31.2–39.6] | 37.2 [34.5–40.9] | 37.1 [31.9–40.1] | 0.284 | 0.039 |
| Fat mass (kg)* | 252 | 27.8 [22.5–36.4] | 30.1 [26.9–38.7] | 29.2 [23.7–37.1] | 0.063 | 28.2 [21.4–34.5] | 29.3 [23.6–35.1] | 28.4 [21.4–35.0] | 0.485 | 0.177 |
| Fat free mass (kg)* | 252 | 47.4 [43.3–51.5] | 48.6 [45.7–51.2] | 47.8 [44.7–51.5] | 0.244 | 46.4 [42.8–50.3] | 48.2 [44.0-52.3] | 47.0 [43.0–51.0] | 0.251 | 0.167 |
| Total body water (%)* | 252 | 45.3 [41.5–48.8] | 44.7 [41.8–46.2] | 45.2 [41.6–48.2] | 0.269 | 45.7 [43.7–49.9] | 45.9 [43.2–48.0] | 45.8 [43.7–49.5] | 0.561 | 0.062 |
| Waist circumference (cm)** | 252 | 91.0 ± 15.2 | 95.5 ± 16.2 | 92.21 ± 15.51 | 0.114 | 87.8 ± 14.0 | 95.8 ± 13.3 | 89.61 ± 14.23 | 0.014 | 0.229 |
| Neck Circumference (cm)** | 252 | 34.3 ± 3.1 | 35.4 ± 3.0 | 34.0 [32.0–37.0] | 0.049 | 33.9 ± 2.6 | 34.8 ± 2.4 | 34.0 [32.1–36.0] | 0.115 | 0.348 |
| Waist/Hip Ratio* | 252 | 0.8 [0.7–0.9] | 0.8 [0.8–0.9] | 0.82 [0.8–0.9] | 0.224 | 0.8 [0.7–0.8] | 0.8 [0.8–0.9] | 0.81 [0.8–0.9] | 0.006 | 0.285 |
| Waist/Height Ratio* | 252 | 0.6 [0.5–0.6] | 0.6 [0.5–0.7] | 0.57 ± 0.10 | 0.129 | 0.6 [0.5–0.6] | 0.6 [0.5–0.7] | 0.55 ± 0.09 | 0.015 | 0.202 |
| VAI* | 252 | 3.0 [2.0-4.8] | 3.5 [2.5–4.7] | 3.09 [2.1–4.7] | 0.281 | 2.8 [2.0-4.8] | 3.7 [2.8–5.4] | 3.06 [2.0-4.9] | 0.057 | 0.864 |
| Biochemical Parameters | ||||||||||
| TC (mg/dL)* | 252 | 190.5 [166.5-222.8] | 195.5 [178.0-219.2] | 193.0 [171.0-221.8] | 0.479 | 190.0 [170.0-219.8] | 183.0 [163.5-202.2] | 189.0 [167.2-214.5] | 0.207 | 0.347 |
| HDL-C (mg/dL)* | 252 | 55.0 [45.0-63.8] | 55.0 [44.8–65.0] | 55.0 [45.0–64.0] | 0.673 | 53.4 [47.0-60.8] | 50.0 [45.0–57.0] | 53.0 [47.0-59.8] | 0.082 | 0.476 |
| LDL-C (mg/dL)* | 241 | 114.0 [95.2-141.5] | 120.0 [101.0-137.0] | 116.0 [96.5-140.5] | 0.592 | 113.0 [95.0-137.5] | 102.0 [92.5–131.0] | 110.5 [95.0-135.5] | 0.166 | 0.404 |
| TG (mg/dL)* | 252 | 86.0 [65.0-127.8] | 101.0 [75.8–129.0] | 90.5 [65.1-127.8] | 0.222 | 82.0 [65.2-118.8] | 94.0 [71.5–138.0] | 87.0 [66.2–131.0] | 0.151 | 0.678 |
| Ferritin(ng/ml)* | 215 | 15.9 [9.4–26.7] | 11.6 [8.7–17.8] | 14.0 [9.2–25.2] | 0.072 | 12.3 [8.1–21.6] | 16.5 [9.4–19.8] | 14.0 [8.2–21.5] | 0.539 | 0.387 |
| CRP (mg/L)* | 157 | 2.9 [1.1–6.8] | 2.5 [1.4-5.0] | 2.8 [1.3–5.7] | 1.000 | 2.0 [1.0-7.2] | 4.2 [2.0-6.3] | 2.3 [1.1–6.5] | 0.420 | 0.755 |
| Serum Iron (µg/L)* | 167 | 64.0 [45.2–96.8] | 54.0 [43.0-81.5] | 60.0 [45.0–92.0] | 0.530 | 68.0 [47.0–88.0] | 71.0 [52.0–97.0] | 68.0 [47.2–91.0] | 0.603 | 0.576 |
| Magnesium (mg/dL)* | 171 | 2.0 [1.9-2.0] | 1.9 [1.8-2.0] | 1.9 [1.9- 2.0] | 0.313 | 2.0 [1.9–2.1] | 2.0 [1.8-2.0] | 2.0 [1.9–2.1] | 0.533 | 0.446 |
| Vitamine B12 (pg/ml)* | 161 | 224.0 [166.5-340.5] | 182.5 [149.0-272.2] | 214.0 [160.0-313.0] | 0.139 | 202.0 [151.0-249.0] | 217.0 [184.5–318.0] | 205.0 [153.2-271.5] | 0.235 | 0.298 |
| Serum Folate (ng/ml)* | 114 | 6.7 [5.4–8.9] | 7.0 [6.2–9.3] | 6.8 [5.7–9.1] | 0.346 | 7.4 [5.5–9.8] | 7.2 [4.4–12.3] | 7.4 [5.2–11.5] | 0.938 | 0.574 |
| Vitamine D (ng/ml)* | 79 | 17.1 [12.1–24.6] | 20.1 [14.0-23.4] | 17.5 [12.4–24.0] | 0.720 | 19.4 [11.4–23.4] | 22.9 [19.3–29.2] | 20.6 [15.1–23.5] | 0.129 | 0.415 |
Data are presented as Mean ± Standard Deviation, Median [25th − 75th percentile]
Abbreviations: n Number of participants, BMI Body Mass Index, CRP C-Reactive Protein, ED Eating Disorders, HDL-C High Density Lipoprotein Cholesterol, LDL-C Low Density Lipoprotein-Cholesterol, PMS Premenstrual Syndrome, TC Total Cholesterol, TG Triglycerides, VAI Visceral Adiposity Index
Statistical Analysis: *Mann-Whitney U test, and **Independent Samples t-test were used
p1: Within-group comparison p-value (Low vs. High Eating Disorder Risk)
p2: Between-group comparison p-value (PMS Group vs. Non-PMS Group)
p < 0.05 was considered statistically significant and indicated in bold
PMSS total scores were significantly higher in the high-risk eating disorder group (median = 146.5) compared with the low-risk group (median = 136.5) (p = 0.010). Analysis of PMSS subdimensions revealed that depressive affect (p = 0.038), anxiety (p = 0.012), depressive thoughts (p = 0.027), and appetite changes (p = 0.016) were significantly higher among participants at high eating disorder risk. No significant differences were observed in the other subdimensions (p > 0.05) (Table 3).
Table 3.
Comparison of PMSS scores according to participants’ PMS status and eating attitude disorder risk
| Variables | PMS Group (n = 146) | p 1 | Non-PMS Group (n = 106) | p 1 | ||||
|---|---|---|---|---|---|---|---|---|
| Low Risk for ED (n = 106) | High Risk for ED (n = 40) | Total | Low Risk for ED (n = 82) | High Risk for ED (n = 24) | Total | |||
| PMSS Total | 136.5 [123.0-149.0] | 146.5 [131.0-167.0] | 139.0 [124.0-155.8] | 0.010 | 91.0 [75.0-100.0] | 96.5 [79.0-102.0] | 91.5 [75.2–100.0] | 0.543 |
| Depressive affect | 23.0 [19.0–28.0] | 25.5 [21.0–32.0] | 23.0 [20.0–28.0] | 0.038 | 11.0 [7.0–16.0] | 9.5 [7.0–12.0] | 11.0 [7.0–15.0] | 0.311 |
| Anxiety | 15.0 [11.0–18.0] | 17.5 [13.0–27.0] | 15.0 [12.0–20.0] | 0.012 | 9.0 [7.0–10.0] | 9.0 [7.0–10.0] | 9.0 [7.0–10.0] | 0.848 |
| Fatigue | 22.0 [19.0–24.0] | 24.0 [17.0–28.0] | 23.0 [19.0–25.0] | 0.206 | 14.0 [10.0–17.0] | 14.0 [11.0–19.0] | 14.0 [10.0-17.8] | 0.607 |
| Irritation | 17.0 [14.0–21.0] | 19.0 [14.0–22.0] | 17.5 [14.0–21.0] | 0.182 | 10.0 [6.0–13.0] | 11.0 [7.0–13.0] | 10.0 [6.2–13.0] | 0.347 |
| Depressive thoughts | 18.0 [13.0–22.0] | 21.5 [14.0–26.0] | 19.0 [14.0–23.0] | 0.027 | 9.0 [7.0–13.0] | 7.0 [7.0–9.0] | 9.0 [7.0–13.0] | 0.168 |
| Pain | 10.0 [7.0–12.0] | 10.0 [7.0–12.0] | 10.0 [7.0–12.0] | 0.972 | 6.0 [5.0–8.0] | 7.0 [7.0–9.0] | 7.0 [5.0–9.0] | 0.032 |
| Appetite changes | 13.0 [11.0–15.0] | 15.0 [11.0–15.0] | 13.0 [11.0–15.0] | 0.016 | 10.0 [8.0–13.0] | 11.5 [10.0–13.0] | 11.0 [8.0–13.0] | 0.448 |
| Sleep changes | 9.0 [6.0–11.0] | 10.0 [7.0–12.0] | 9.0 [7.0–11.0] | 0.124 | 5.0 [3.0–7.0] | 6.5 [3.0–7.0] | 5.0 [3.0–7.0] | 0.587 |
| Bloating | 13.0 [10.0–15.0] | 15.0 [9.0–15.0] | 14.0 [10.0–15.0] | 0.342 | 11.0 [6.0–15.0] | 8.0 [7.0–13.0] | 11.0 [7.0–15.0] | 0.836 |
Data are presented as Median [25th − 75th percentile]
Abbreviations: PMS Premenstrual Syndrome, PMSS Premenstrual Syndrome Scale, ED Eating Disorders, n Number of participants
Statistical Analysis: The Mann-Whitney U test was used
p1: Within-group comparison p-value (Low vs. High Eating Disorder Risk)
p < 0.05 was considered statistically significant and indicated in bold
In the correlation analysis, a positive and weak statistically significant relationship was found between the EAT-26 score and the VAI score (r = 0.128, p = 0.042) (Table 4). The multivariate binomial logistic regression model examining predictors of eating disorder risk was statistically significant (χ² = 9.93, p < 0.05). After adjusting for age and chronic disease, PMS status was not a significant predictor of eating disorder risk (p = 0.625). BMI emerged as the only significant independent predictor in the model, indicating that each 1-unit increase in BMI increased the likelihood of being at risk for an eating disorder by 1.05 times (aOR = 1.049, 95% CI: 1.002–1.098; p = 0.041) (Table 5).
Table 4.
Correlation between variables
| PMSS | EAT-26 | VAI | ||
|---|---|---|---|---|
| PMSS | r | 1 | 0.086 | 0.078 |
| p | - | 0.175 | 0.217 | |
| EAT-26 | r | 0.086 | 1 | 0.128 |
| p | 0.175 | - | 0.042 | |
| VAI | r | 0.078 | 0.128 | 1 |
| p | 0.217 | 0.042 | - | |
Abbreviations: PMSS Premenstrual Syndrome Scale, EAT-26 Eating Attitudes Test-26, VAI Visceral Adiposity Index
Spearman correlation test
p < 0.05 was considered statistically significant and indicated in bold
Table 5.
Multivariate binomial logistic regression analysis identifying risk factors for eating disorders
| Variables | β | S.E. | Wald | p | aOR | 95% C.I. |
|---|---|---|---|---|---|---|
| PMS Presence (Ref: Absent) | 0.149 | 0.305 | 0.239 | 0.625 | 1.161 | 0.639–2.109 |
| Age (years) | -0.019 | 0.019 | 1.022 | 0.312 | 0.981 | 0.946–1.018 |
| BMI (kg/m²) | 0.048 | 0.023 | 4.186 | 0.041 | 1.049 | 1.002–1.098 |
| Chronic Disease (Ref: Absent) | 0.350 | 0.317 | 1.219 | 0.270 | 1.419 | 0.762–2.643 |
| Constant | -2.094 | 0.831 | 6.345 | 0.012 | 0.123 |
Dependent Variable: Risk of Eating Disorder. Independent Variables: Presence of PMS, Age, BMI, Chronic Disease
Abbreviations: β Regression Coefficient, S.E. Standard Error, aOR Adjusted Odds Ratio, 95% C.I. Confidence Interval, BMI Body Mass Index, PMS Premenstrual Syndrome
Model Fit Statistics: Cox & Snell R2 = 0.026, Nagelkerke R2 = 0.038, Model X2 = 9.93, p < 0.05
p < 0.05 was considered statistically significant and indicated in bold
Discussion
In this study, we investigated the impact of PMS on the risk of eating disorders, the VAI, and biochemical parameters in women of reproductive age. PMS was identified in 57.9% of the participants. Although this rate is slightly higher than the global prevalence of 47.8% reported in the literature [3, 30, 31], it is consistent with findings from other study conducted in Türkiye [7]. Variations in PMS prevalence across studies are thought to reflect differences in diagnostic criteria, cultural factors, and regional variations in lifestyle habits.
Body composition and obesity have become increasingly important factors in the etiology of PMS. In the present investigation, individuals identified with PMS exhibited a significantly higher body fat percentage compared to the non-PMS group, supporting the hypothesis that increased adiposity contributes to the syndrome’s pathophysiology. Adipose tissue functions not merely as a passive energy reservoir but as a metabolically active endocrine organ that modulates estrogen metabolism and secretes pro-inflammatory cytokines [32]. Previous research has demonstrated that obesity induced chronic low-grade inflammation and subsequent disruptions in neurotransmitter balance can act as a potential trigger for PMS symptoms [33, 34]. Notably, our finding that body fat percentage—rather than BMI—differed significantly suggests that BMI may be an insufficient proxy for capturing the specific endocrine and metabolic activities of adipose tissue associated with PMS. This discrepancy highlights the clinical importance of identifying individuals with “normal-weight obesity”, a condition in which individuals maintain a ‘normal’ BMI but have high body fat percentages, thereby remaining at risk due to adipose-derived hormonal and inflammatory pathways [35]. Consequently, rather than relying solely on static weight measurements, comprehensive body composition assessment provides more granular, clinically relevant information for identifying at-risk populations and tailoring preventive strategies.
One of the primary objectives of our study was to investigate the relationship between PMS and the VAI. This composite indicator differs from traditional anthropometric measurements by incorporating both anatomical (waist circumference and BMI) and physiological (TG and HDL-C) parameters. The literature suggests that visceral adipose tissue may contribute to the etiology of PMS through chronic low-grade inflammation driven by the secretion of pro-inflammatory cytokines [4]. However, despite the higher overall body fat percentage observed in the PMS group, no significant difference in VAI levels was found between the groups. This lack of significant difference can be explained by the “Metabolically Healthy Obese” phenotype often seen in young women. In this age group, excess fat is preferentially stored subcutaneously rather than viscerally. Since VAI is specifically sensitive to visceral adipose dysfunction and TG levels, it may not fully capture the subcutaneous adiposity load that drives PMS-related inflammation in this younger, non-diabetic cohort [36]. Because the majority of participants were young and metabolically healthy, triglyceride and HDL-C levels — key components of the VAI formula — likely remained within normal ranges, as confirmed by our biochemical analyses showing no significant differences in lipid profiles. Additionally, waist circumference may be influenced by abdominal bloating and fluid retention frequently observed during the luteal phase, potentially masking actual differences. Regardless of PMS status, our study revealed a significant positive correlation between EAT-26 scores and VAI levels. This relationship may indicate specific dietary habits often associated with disordered eating patterns. Previous research has shown that individuals with high EAT-26 scores frequently follow restrictive diets followed by binge eating episodes characterized by the consumption of high-energy-density foods rich in refined sugars and saturated fats [37–39]. Such nutritional profiles have been reported to promote visceral lipogenesis and metabolic dysfunction preferentially [40]. Therefore, VAI can serve as a sensitive indicator for detecting early metabolic risks, particularly those stemming from eating disorders, and offer a perspective beyond general obesity indicators.
The current analysis revealed no statistically significant differences between the PMS and non-PMS groups in biochemical parameters. The literature, however, provides strong evidence that deficiencies in micronutrients such as magnesium, calcium, vitamin D, and B-group vitamins may influence PMS etiology by altering neurotransmitter synthesis and exacerbating symptom severity [41]. For example, the regulatory effects of magnesium on serotonin receptors and the role of vitamin D in calcium homeostasis underscore the importance of these micronutrients in managing PMS [42]. Several potential explanations exist for the absence of significant biochemical differences in our findings. First, the biochemical data were obtained retrospectively from laboratory records ordered by physicians within the last three months, rather than through standardized blood sampling for research purposes. This limited our ability to obtain uniform biochemical measurements for the entire sample and may have resulted in missing data, thereby reducing statistical power. Second, the timing of blood collection is a critical factor. Hormonal and biochemical fluctuations related to PMS are particularly pronounced during the luteal phase of the menstrual cycle. Because the retrospective records did not indicate the cycle day on which samples were collected, measurements taken during the follicular phase may have masked potential luteal-phase reductions in micronutrient levels. Third, homeostatic mechanisms likely played a role. The body tightly regulates serum concentrations to maintain physiological stability, meaning that blood levels may remain within normal ranges even when intracellular stores are depleted. The young and generally healthy nature of our sample likely contributed to the effectiveness of these compensatory mechanisms. In conclusion, clarifying the relationship between PMS and micronutrient status will require prospective studies in which biochemical samples are collected during the luteal phase, and intracellular levels are also assessed.
The multifaceted relationship between nutrition and PMS becomes even more complex when the prevalence of eating disorders is taken into account [12]. The present study found that, as eating disorder scores increased, the severity of PMS symptoms also rose proportionally. This finding is consistent with the study by Yi et al. [43], which reported a positive correlation between eating disorder severity and PMS. Similarly, an Iranian study found that women with a high risk of eating disorders experienced more severe PMS symptoms, even though the statistical significance was borderline [13]. The underlying mechanism of this association may relate to hormonal fluctuations throughout the menstrual cycle that disrupt appetite regulation. In particular, decreased serotonin levels during the luteal phase may trigger carbohydrate cravings, making it more challenging to regulate eating behavior and potentially predisposing individuals to disordered eating patterns [11]. However, our multivariate logistic regression results suggest that PMS does not solely drive this relationship. After adjusting for age and chronic diseases, PMS was not identified as an independent predictor of eating disorder risk. Although a symptomatic correlation exists between these two conditions, our findings weaken the hypothesis that PMS directly causes eating disorders. While our results do not establish a definitive causal relationship, they highlight the need for future large-scale studies to elucidate further the underlying mechanisms involved in this association.
Current results show that while PMS symptom severity was associated with disordered eating behaviors, the primary determinant of clinical eating disorder risk was linked more closely to a higher BMI rather than the presence of PMS itself. This suggests that while hormonal fluctuations during the premenstrual phase may trigger acute changes in eating patterns, they do not necessarily culminate in a clinical eating disorder unless other risk factors, such as high adiposity, are present. These findings are supported by Pearce et al. [19] who emphasized that increased BMI is one of the strongest predictors of the development of pathological eating attitudes. Therefore, in the clinical management of PMS, it is essential to distinguish between transient premenstrual appetite changes and a more stable, weight-related eating disorder risk.
Regarding anthropometric measurements, long-term research has shown that women with a BMI above 27.5 kg/m2 are significantly more likely to develop severe PMS compared to those with a BMI below 20 kg/m2 [44]. However, the relationship between BMI and PMS risk may not be strictly linear. Some studies have identified a negative or U-shaped association, particularly in young populations, in which both high adiposity and underweight are associated with increased symptom severity [45, 46]. While obesity triggers PMS through chronic inflammation and hormonal dysregulation, exceptionally low body weight may act as a different physiological stressor. In underweight individuals, inadequate nutritional intake and low energy availability can disrupt the hypothalamic-pituitary-ovarian axis, leading to hormonal fluctuations that exacerbate premenstrual symptoms [47]. Therefore, both extremes of the BMI spectrum compromise endocrine stability, albeit through different biological mechanisms.
In our subgroup analyses, we found that among the non-PMS group, those at risk for eating disorders had significantly higher waist circumference, waist-to-hip ratio, and waist-to-height ratio. However, this relationship disappeared within the PMS group, where only neck circumference remained significantly associated with eating disorder risk. Neck circumference is a stable anthropometric measure that is minimally affected by abdominal bloating and cycle-related fluid retention, yet it correlates strongly with visceral adiposity [48]. The lack of significance in waist-related variables within the PMS group may be attributed to abdominal edema caused by fluctuations in aldosterone and progesterone during the luteal phase, which could mask actual differences in abdominal fat accumulation between groups. Therefore, the finding that neck circumference—unaffected by edema—remained significant in the PMS group suggests that eating disorder risk in these women is still linked to adiposity. However, this relationship becomes difficult to detect using standard waist measurements.
In this research, participants were grouped according to the presence of PMS, and the impact of eating disorder risk on PMS symptom severity was examined. The analyses revealed that among women individuals identified with PMS, those at risk for eating disorders had significantly higher total PMSS scores, as well as higher scores in the subdimensions of depressive affect, anxiety, depressive thoughts, and appetite changes compared with those without such risk. In contrast, among the non-PMS group, eating disorder risk did not produce a significant difference in total PMSS scores or in psychological subdimensions. This pattern suggests that a tendency toward disordered eating may act as an “exacerbating factor” that intensifies an already existing PMS profile. The relationship between PMS and eating disorders is frequently interpreted through the serotonergic dysregulation hypothesis [11]. In individuals at risk for eating disorders, biological vulnerability may be compounded by cognitive factors such as fear of weight gain and restrictive eating, which can elevate stress levels and exacerbate anxiety and depressive symptoms [49]. Our findings indicate that anxiety and depressive affect scores were particularly high among individuals with PMS who are at risk for eating disorders, further supporting this psycho-biological burden hypothesis. Additionally, the significantly higher appetite change scores in the risk group suggest that these women may experience the physiological increase in appetite during the luteal phase more chaotically (i.e., cycles of binge eating or excessive restriction). Supporting this, Mighani et al. [13] reported that premenstrual appetite increases in women with disordered eating behaviors may intensify symptom perception when combined with emotional eating tendencies.
Interestingly, among the non-PMS group, eating disorder risk was not associated with heightened psychological symptoms; instead, it was linked only to higher pain scores. This may indicate that women at risk for eating disorders, but without PMS, may experience lowered pain thresholds or increased somatization due to inadequate or irregular eating patterns. Overall, however, the findings suggest that eating disorder risk specifically intensifies PMS-related symptomatology, whereas it does not generate a comparable psychological profile in the non-PMS group. Therefore, in clinical practice, we recommend that women presenting with PMS, particularly those reporting pronounced depressive or anxious symptoms, should also be assessed for eating disorder risk.
Limitations
The findings of this study should be interpreted considering several limitations. First, the cross-sectional design allows for the identification of associations and their direction but does not permit the establishment of definitive causal relationships between variables. Second, the data collection relied on self-reported scales. Although this approach carries the potential risk of recall bias, a standard limitation in nutrition and psychological research, we attempted to minimize this limitation by employing standardized instruments with established validity and reliability. Third, the biochemical parameters were obtained retrospectively from hospital records covering the previous three months. While this method provides valuable real-world data from a large sample, it limits our ability to standardize blood collection according to menstrual cycle phases (follicular/luteal). However, the biochemical markers assessed in this study—such as vitamin B12, vitamin D, ferritin, and lipid profile are relatively stable indicators reflecting long-term nutritional and metabolic status rather than acute hormonal fluctuations, which may have mitigated the impact of the lack of phase-specific sampling. Furthermore, while BIA measurements were conducted under strict fasting and resting protocols to ensure accuracy, the potential for cyclical fluid retention to influence body composition results cannot be entirely ruled out. This represents a common methodological challenge in cross-sectional PMS research, and results should be interpreted within the context of these physiological variations. Finally, although the single-center nature of the study may limit the generalizability of the findings, the achieved sample size was consistent with the calculated power analysis, thereby strengthening the statistical validity of the results.
Conclusion
This study is significant as it reveals the multidimensional relationship between PMS, eating behavior, body composition, and biochemical parameters in women of reproductive age. Current results indicate that individuals identified with PMS had a significantly higher body fat percentage than the non-PMS group, underscoring the role of increased adiposity and lifestyle factors in the pathophysiology of the syndrome. A key finding of this research concerns the complex nature of the relationship between eating disorder risk and PMS. Although our findings showed that PMS symptom severity increased in parallel with higher eating disorder scores, the multivariate regression analysis revealed that the primary determinant of eating disorder risk was not the presence of PMS itself, but rather an elevated BMI. Furthermore, even among healthy individuals without PMS, eating disorder risk was associated with abdominal obesity indicators such as waist circumference and waist-to-hip ratio, confirming the strong link between disordered eating tendencies and central adiposity. In conclusion, PMS, eating disorders, and obesity are complex conditions that may mutually reinforce one another and share overlapping physiological mechanisms. Given that BMI, rather than PMS status, emerged as the primary driver of eating disorder risk, clinicians should prioritize weight management strategies over solely symptom-based treatments. While lifestyle interventions targeting body fat reduction are essential, precise assessment is equally critical. Therefore, neck circumference emerges as a practical, edema-independent alternative to waist circumference for assessing adiposity risk in individuals experiencing severe PMS bloating. Future research employing prospective designs with biochemical monitoring across different phases of the menstrual cycle will help clarify the underlying mechanisms of these interconnected relationships.
Supplementary Information
Acknowledgements
The authors would like to thank the volunteer women participants and the Sarıyer District Health Directorate for their support of the study.
Authors’ contributions
E.O. and H.S.A. designed the study. E.O. contributed to sample collection. E.O., NK and HSA conducted the research, analyzed and interpreted the data. E.O., N.K., and H.S.A. wrote the draft. E.O., N.K., and H.S.A. had primary responsibility for the final content, and all authors carefully reviewed the manuscript and approved the final version submitted for publication.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Data availability
The data presented in this study are available on request from the corresponding author due to privacy or ethical restrictions.
Declarations
Ethics approval and consent to participate
This study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving research study participants were approved by the Human Research Ethics Committee of Istanbul Bilgi University (Date: November 28, 2023, Approvel No: 2023-20160-146). The current study was in compliance with the Declaration of Helsinki.
Consent for publication
The written consent was taken from the participants.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Modzelewski S, Oracz A, Żukow X, Iłendo K, Śledzikowka Z, Waszkiewicz N. Premenstrual syndrome: new insights into etiology and review of treatment methods. Front Psychiatry. 2024;15:1363875. 10.3389/fpsyt.2024.1363875. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Direkvand-Moghadam A, Sayehmiri K, Delpisheh A, Kaikhavandi S. Epidemiology of premenstrual syndrome (PMS): a systematic review and meta-analysis study. J Clin Diagn Res. 2014;8(2):106–9. 10.7860/JCDR/2014/8024.4021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gudipally PR, Sharma GK, Premenstrual syndrome. StatPearls. 2023. https://pubmed.ncbi.nlm.nih.gov/32809533. Accessed 31 Jan 2026. [PubMed]
- 4.Takeda T. Premenstrual disorders: premenstrual syndrome and premenstrual dysphoric disorder. J Obstet Gynaecol Res. 2023;49(2):510–8. 10.1111/jog.15484. [DOI] [PubMed] [Google Scholar]
- 5.Gao M, Zhang H, Gao Z, Cheng X, Sun Y, Qiao M, et al. Global and regional prevalence and burden for premenstrual syndrome and premenstrual dysphoric disorder: a systematic review and meta-analysis. Med (Baltim). 2022;101(1):e28528. 10.1097/MD.0000000000028528. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Dutta A, Sharma A. Prevalence of premenstrual syndrome and premenstrual dysphoric disorder in India: a systematic review and meta-analysis. Health Promot Perspect. 2021;11(2):161–70. 10.34172/hpp.2021.20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Erbil N, Yücesoy H. Premenstrual syndrome prevalence in Turkey: a systematic review and meta-analysis. Psychol Health Med. 2023;28(5):1347–57. 10.1080/13548506.2021.2013509. [DOI] [PubMed] [Google Scholar]
- 8.Dubol M, Epperson CN, Lanzenberger R, Sundström-Poromaa I, Comasco E. Neuroimaging premenstrual dysphoric disorder: a systematic and critical review. Front Neuroendocrinol. 2020;57:100838. 10.1016/j.yfrne.2020.100838. [DOI] [PubMed] [Google Scholar]
- 9.Green LJ, O’Brien PMS, Panay N, Craig M. Management of premenstrual syndrome. BJOG. 2017;124(3):e73–105. 10.1111/1471-0528.14260. [DOI] [PubMed] [Google Scholar]
- 10.Tiranini L, Nappi RE. Recent advances in understanding/management of premenstrual dysphoric disorder/premenstrual syndrome. Fac Rev. 2022;11:11. 10.12703/r/11-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Oboza P, Ogarek N, Wójtowicz M, Rhaiem TB, Olszanecka-Glinianowicz M, Kocełak P. Relationships between premenstrual syndrome (PMS) and diet composition, dietary patterns and eating behaviors. Nutrients. 2024;16(12):1911. 10.3390/nu16121911. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yoshinari Y, Morino S, Shinohara Y, Chen CY, Onishi M, Akase Y, et al. Association between premenstrual syndrome and eating disturbance in college students: a cross-sectional study. BMC Womens Health. 2024;24:330. 10.1186/s12905-024-03158-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Mighani S, Shivyari FT, Razzaghi A, Amerzadeh M, Javadi M. Relationship between dietary intake, eating attitudes, and premenstrual syndrome severity among Iranian women: insights from a cross-sectional study. J Eat Disord. 2025;13:131. 10.1186/s40337-025-01326-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Rad M, Sabzevary MT, Dehnavi ZM. Factors associated with premenstrual syndrome in female high school students. J Educ Health Promot. 2018;7:64. 10.4103/jehp.jehp_126_17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Xu H, Li PH, Barrow TM, Colicino E, Li C, Song R, et al. Obesity as an effect modifier of the association between menstrual abnormalities and hypertension in young adult women: Results from Project ELEFANT. PLoS ONE. 2018;13(11):e0207929. 10.1371/journal.pone.0207929. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Dang N, Khalil D, Sun J, Naveed A, Soumare F, Hamidovic A. Waist circumference and its association with premenstrual food craving: The PHASE Longitudinal Study. Front Psychiatry. 2022;13:784316. 10.3389/fpsyt.2022.784316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sharifan P, Jafarzadeh Esfehani A, Zamiri A, Ekhteraee Toosi MS, Najar Sedgh Doust F, Taghizadeh N, et al. Factors associated with the severity of premenstrual symptoms in women with central obesity: a cross-sectional study. J Health Popul Nutr. 2023;42:9. 10.1186/s41043-022-00343-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Pearce E, Jolly K, Jones LL, Matthewman G, Zanganeh M, Daley A. Exercise for premenstrual syndrome: a systematic review and meta-analysis of randomised controlled trials. BJGP Open. 2020;4(3). 10.3399/bjgpopen20X101032. [DOI] [PMC free article] [PubMed]
- 19.Cohen J. Set correlation and contingency tables. Appl Psychol Meas. 1988;12(4):425–34. 10.1177/014662168801200410. [Google Scholar]
- 20.Gençdoğan B. Premenstruel sendrom için yeni bir ölçek. Turk Psikiyatri Derg. 2006;8(2):81–7. [Google Scholar]
- 21.Kucukkelepce DS, Unver H, Nacar G, Tashan ST. The effects of acupressure and yoga for coping with premenstrual syndromes on premenstrual symptoms and quality of life. Complement Ther Clin Pract. 2021;42:101282. 10.1016/j.ctcp.2020.101282. [DOI] [PubMed] [Google Scholar]
- 22.Garner DM, Garfinkel PE. The Eating Attitudes Test: an index of the symptoms of anorexia nervosa. Psychol Med. 1979;9:273–9. 10.1017/s0033291700030762. [DOI] [PubMed] [Google Scholar]
- 23.Garner DM, Olmsted MP, Bohr Y, Garfinkel PE. The eating attitudes test: psychometric features and clinical correlates. Psychol Med. 1982;12(4):871–8. 10.1017/s0033291700049163. [DOI] [PubMed] [Google Scholar]
- 24.Ergüney-Okumuş FE, Sertel-Berk HÖ. Yeme tutum testi kısa formunun (YTT-26) üniversite örnekleminde Türkçeye uyarlanması ve psikometrik özelliklerinin değerlendirilmesi. Psikoloji Çalışmaları. 2020;40(2). 10.26650/SP2019-0039.
- 25.Kyle UG, Bosaeus I, De Lorenzo AD, Deurenberg P, Elia M, Gomez JM, et al. Bioelectrical impedance analysis—part I: review of principles and methods. Clin Nutr. 2004;23(5):1226–43. 10.1016/j.clnu.2004.06.004. [DOI] [PubMed] [Google Scholar]
- 26.World Health Organization. Physical status: the use and interpretation of anthropometry: report of a WHO Expert Committee. World Health Organ Tech Rep Ser. No. 854. Geneva: WHO; 1995. [PubMed] [Google Scholar]
- 27.World Health Organization. Obesity: preventing and managing the global epidemic. Report of a WHO consultation. World Health Organ Tech Rep Ser. No. 894. Geneva: WHO; 2000. [PubMed] [Google Scholar]
- 28.World Health Organization. WHO guidelines on physical activity and sedentary behaviour. 2020. https://iris.who.int/server/api/core/bitstreams/ca408ade-05c9-4b7c-8967-6ee5b5e0ccd8/content. Accessed 31 Jan 2026. [PubMed]
- 29.Amato MC, Verghi M, Galluzzo A, Giordano C. The oligomenorrhoic phenotypes of polycystic ovary syndrome are characterized by a high visceral adiposity index: a likely condition of cardiometabolic risk. Hum Reprod. 2011;26(6):1486–94. 10.1093/humrep/der088. [DOI] [PubMed] [Google Scholar]
- 30.Turan A, Güler Kaya İ, Çakır HB, Topaloğlu S. Prevalence and correlates of premenstrual syndrome and premenstrual dysphoric disorder among women aged 18–25 in Turkey. Int J Psychiatry Med. 2024;59(1):101–11. 10.1177/00912174231189936. [DOI] [PubMed] [Google Scholar]
- 31.Yesildemir O, Arslan N, Dubus EN, Guner H. Relationship between premenstrual syndrome and compliance with the Mediterranean diet among women of reproductive age. Med Sci. 2025;14(3):686–92. 10.5455/medscience.2025.03.072. [Google Scholar]
- 32.Kuryłowicz A. Estrogens in adipose tissue physiology and obesity-related dysfunction. Biomedicines. 2023;11(3):690. 10.3390/biomedicines11030690. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Deng H, Chen Y, Xing J, Zhang N, Xu L. Systematic low-grade chronic inflammation and intrinsic mechanisms in polycystic ovary syndrome. Front Immunol. 2024;15:1470283. 10.3389/fimmu.2024.1470283. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hestiantoro A, Kapnosa Hasani RD, Shadrina A, Situmorang H, Ilma N, Muharam R, et al. Body fat percentage is a better marker than body mass index for determining inflammation status in polycystic ovary syndrome. Int J Reprod Biomed. 2018;16(10):623–8. [PMC free article] [PubMed] [Google Scholar]
- 35.Franco LP, Morais CC, Cominetti C. Normal-weight obesity syndrome: diagnosis, prevalence, and clinical implications. Nutr Rev. 2016;74(9):558–70. [DOI] [PubMed] [Google Scholar]
- 36.Blüher M. Metabolically healthy obesity. Endocr Rev. 2020;41(3):bnaa004. 10.1210/endrev/bnaa004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Figueiredo N, Kose J, Srour B, Julia C, Kesse-Guyot E, Péneau S, et al. Ultra-processed food intake and eating disorders: Cross-sectional associations among French adults. J Behav Addict. 2022;11(2):588–99. 10.1556/2006.2022.00009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Wiss D. Clinical considerations of ultra-processed food addiction across weight classes: An eating disorder treatment and care perspective. Curr Addict Rep. 2022;9(4):255–67. 10.1007/s40429-022-00411-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Javadi Arjmand E, Bemanian M, Vold JH, Skogen JC, Sandal GM, Arnesen EK, et al. Emotional eating and changes in High-Sugar Food and drink consumption linked to psychological distress and worries: a Cohort Study from Norway. Nutrients. 2023;15(3):778. 10.3390/nu15030778. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Yu B, Sun Y, Wang Y, Wang B, Tan X, Lu Y, et al. Associations of artificially sweetened beverages, sugar-sweetened beverages, and pure fruit/vegetable juice with visceral adipose tissue mass. Diabetes Metab Syndr Clin Res Rev. 2023;17(10):102871. 10.1016/j.dsx.2023.102871. [DOI] [PubMed] [Google Scholar]
- 41.Siminiuc R, Ţurcanu D. Impact of nutritional diet therapy on premenstrual syndrome. Front Nutr. 2023;10:1079417. 10.3389/fnut.2023.1079417. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Robinson J, Ferreira A, Iacovou M, Kellow NJ. Effect of nutritional interventions on the psychological symptoms of premenstrual syndrome in women of reproductive age: a systematic review of randomized controlled trials. Nutr Rev. 2025;83(2):280–306. 10.1093/nutrit/nuae043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Yi SJ, Kim M, Park I. Investigating influencing factors on premenstrual syndrome (PMS) among female college students. BMC Womens Health. 2023;23:592. 10.1186/s12905-023-02752-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Bertone-Johnson ER, Hankinson SE, Willett WC, Johnson SR, Manson JE. Adiposity and the Development of Premenstrual Syndrome. J Womens Health (Larchmt). 2010;19(11):1955–62. 10.1089/jwh.2010.2128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Cheng SH, Shih CC, Yang YK, Chen KT, Chang YH, Yang YC. Factors associated with premenstrual syndrome—A survey of new female university students. Kaohsiung J Med Sci. 2013;29(2):100–5. 10.1016/j.kjms.2012.08.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Ju H, Jones M, Mishra GD. A U-shaped relationship between body mass index and dysmenorrhea: a longitudinal study. PLoS ONE. 2015;10(7):e0134187. 10.1371/journal.pone.0134187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Shufelt CL, Torbati T, Dutra E. Hypothalamic amenorrhea and the long-term health consequences. Semin Reprod Med. 2017;35(03):256–62. 10.1055/s-0037-1603581. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Scovronec A, Provencher A, Iceta S, Pelletier M, Leblanc V, Nadeau M, et al. Neck circumference is a better correlate of insulin resistance markers than other standard anthropometric indices in patients presenting severe obesity. Obes Res Clin Pract. 2022;16(4):307–13. 10.1016/j.orcp.2022.07.005. [DOI] [PubMed] [Google Scholar]
- 49.Finch JE, Xu Z, Baker JH. Understanding comorbidity between eating disorder and premenstrual symptoms using a network analysis approach. Appetite. 2023;181:106410. 10.1016/j.appet.2022.106410. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data presented in this study are available on request from the corresponding author due to privacy or ethical restrictions.