Women With PCOS Who Eat Ultra-Processed Foods at Night Face Worse GI Anxiety

For women living with polycystic ovary syndrome, the digestive system is rarely a bystander. PCOS is already linked to bloating, irregular bowel habits, and a heightened risk of irritable bowel syndrome, but a new cross-sectional study published in June 2026 in Food Science and Nutrition suggests the problem runs deeper than the gut itself. Researchers found that among 205 women with confirmed PCOS, night eating symptoms emerged as the single strongest independent predictor of GI-specific anxiety, a measure of fear and cognitive distress around gastrointestinal sensations.

Night Eating Syndrome Predicts GI Anxiety in PCOS Women

The study recruited women aged 18 to 50 with physician-confirmed PCOS and asked them to complete three validated questionnaires: the Screening Questionnaire for Highly Processed Food Consumption (sQ-HPF), the Night Eating Questionnaire (NEQ), and the Visceral Sensitivity Index (VSI). The VSI is not a simple pain scale; it captures cognitive-affective dimensions, specifically the fear of GI symptoms and the way that fear shapes behavior and daily life. About 21% of participants met the criteria for Night Eating Syndrome (NES), defined as a score of 25 or higher on the NEQ, while 34.6% reported high ultra-processed food (UPF) consumption. In adjusted regression models, night eating score was the sole significant independent predictor of GI-specific anxiety, with a standardized beta of 0.430 and an R-squared of 0.222.

Ultra-Processed Foods Add Fuel to an Already Disrupted System

The connection between ultra-processed foods and gut health is well-established in the broader literature. A review in PMC found that UPFs drive gut dysbiosis and increase intestinal permeability, setting conditions for functional GI disorders to take hold. In this PCOS cohort, women who scored high on UPF consumption also had higher NEQ scores, greater obesity rates, and more frequent consumption of sugary foods and restaurant meals. The correlation between NEQ and sQ-HPF was moderate but meaningful (r = 0.245, p < 0.001). However, when researchers ran their full adjusted regression model, UPF consumption's direct influence on night eating scores lost statistical significance (p = 0.111), suggesting that other factors, particularly GI-specific anxiety and BMI, carry more explanatory weight. This does not clear UPFs of suspicion. The proposed mechanisms are plausible: UPFs may alter appetite-regulating hormones like GLP-1, peptide YY, and ghrelin, and shift neurotransmitter activity involving serotonin and norepinephrine. In a hormonal environment already disrupted by hyperandrogenism, those shifts could amplify emotional hunger and erode normal satiety signaling.

How the Study Was Designed and What It Measured

Between September 2024 and January 2025, the research team enrolled 205 women across a clinical setting. To qualify, participants had to meet Rotterdam criteria for PCOS and be free of oral contraceptive use, pregnancy, and major comorbidities. More than half (53.7%) were overweight or obese. The sample skewed educated: 97.6% held a university degree or higher. Common PCOS symptoms in the group included fatigue (92.2%), hirsutism (80.5%), and menstrual irregularities (80.5%). Nearly 70% reported skipping main meals, while 54.1% snacked regularly. The three instruments all showed acceptable internal consistency, with the VSI scoring particularly high (Cronbach's alpha = 0.95). Researchers ran multiple linear regression models adjusted for BMI, smoking, alcohol use, marital status, and income. In the model predicting night eating scores, GI-specific anxiety (beta = 0.337), smoking (beta = 0.286), and BMI (beta = 0.201) all emerged as independent predictors, together explaining about 39% of the variance.

Cross-Sectional Design Limits What the Data Can Prove

This study should be read as early-stage, hypothesis-generating work. Its cross-sectional design means it cannot establish whether night eating causes GI anxiety, whether GI anxiety drives nocturnal snacking, or whether both are downstream of a common third factor like circadian disruption or chronic stress. Observational studies and randomized trials offer complementary but distinct kinds of evidence, and cross-sectional data cannot answer temporal questions. Self-reported dietary data and self-measured anthropometrics introduce additional uncertainty, particularly social desirability bias around food intake. The study also did not stratify participants by PCOS phenotype (A through D), which vary substantially in metabolic severity and hormonal profile. Physical activity levels and total daily energy intake were not captured. Generalizability is further constrained by the high education level of the sample and the single clinical recruitment site. The regression models' R-squared values (0.390 for night eating, 0.222 for GI anxiety) indicate that a large share of the variance remains unexplained, consistent with what inconsistencies across nutritional epidemiology studies have long signaled about unmeasured confounders in this field.

What Clinicians and Researchers Might Do Next

Functional gastrointestinal disorders affect more than 40% of people worldwide, according to the Rome Foundation Global Study, and women with PCOS appear to be disproportionately vulnerable. The authors argue that nutritional care for PCOS should expand beyond caloric restriction and weight management to explicitly target diet quality and eating timing. Screening for night eating patterns and GI-specific anxiety could become part of standard PCOS assessment, particularly given how strongly these two measures track together. Future research would benefit from longitudinal designs, objective dietary measures like 24-hour recalls or food diaries, and phenotype stratification to determine whether these associations hold across the full spectrum of PCOS presentations. Intervention studies testing whether reducing UPF intake or shifting eating windows toward earlier in the day can reduce GI anxiety in this population have yet to be done, but the case for running them is now a little more concrete.

Evidence rating

INVESTIGATIONAL — 71/100

Caveats:

• Cross-sectional design — associations shown here cannot establish cause and effect or determine which came first.
• All dietary, anthropometric, and symptom data were self-reported, introducing potential recall and social desirability bias.
• The sample (n=205, 97.6% university-educated, single clinical site) limits how broadly these findings apply.
• Regression models explained only 22–39% of the variance, meaning many contributing factors remain unmeasured.
• Funding source was not disclosed by the authors.
• Gut microbiome mechanisms are proposed but not directly measured in this study.

Cross-sectional study of 205 women with physician-confirmed PCOS (single clinical site, September 2024–January 2025). Findings reflect associations only — the design cannot establish causality or temporal direction between ultra-processed food intake, night eating, and gastrointestinal-specific anxiety. Gut microbiome mechanisms are proposed theoretically but were not directly assessed. Not yet replicated in an independent population.

References

1. Emine Merve Ekici, Sena Dilşad Akçakaya, Vefa Şakar, Özge Mengi Çelik (2026). Associations of Ultra‐Processed Food Consumption and Night Eating With GI Symptom‐Specific Anxiety in Reproductive‐Age Women With Polycystic Ovary Syndrome. Food Science & Nutrition. https://doi.org/10.1002/fsn3.72007
2. Debora Rondinella, Pauline Celine Raoul, Eleonora Valeriani, et al. (2025). The Detrimental Impact of Ultra-Processed Foods on the Human Gut Microbiome and Gut Barrier. Nutrients. https://doi.org/10.3390/nu17050859
3. Mikkel Zöllner Ankarfeldt (2014). Comment on “ Limitations of Observational Evidence: Implications for Evidence-Based Dietary Recommendations ”. Advances in Nutrition. https://doi.org/10.3945/an.114.005777
4. Mariona Pinart, Katharina Nimptsch, Sofia K. Forslund, et al. (2021). Identification and Characterization of Human Observational Studies in Nutritional Epidemiology on Gut Microbiomics for Joint Data Analysis. Nutrients. https://doi.org/10.3390/nu13093292
5. Worldwide Prevalence and Burden of Functional Gastrointestinal Disorders, Results of Rome Foundation Global Study – PubMed. NIH. https://pubmed.ncbi.nlm.nih.gov/32283129/