(P19-004-23) Association Between the Dietary Inflammatory Index and Premenstrual Syndrome Among College-Aged Women in the UMass Vitamin D Study (2006–2014)

Objectives: Prior studies have shown an association between inflammatory markers and premenstrual symptoms, but few studies have examined the role of pro-inflammatory diets on PMS. We examined whether the DII is associated with PMS symptom severity and prevalence of PMS among women aged 18-31. We hypothesized that higher Dietary Inflammatory Index (DII) scores (i.e., more pro-inflammatory diet) will be correlated with higher physical, affective, and total PMS symptom severity scores, and higher prevalence of premenstrual syndrome (PMS).

Methods: This was a cross-sectional analysis of the UMass Vitamin D Study (n=412). We calculated DII using dietary data collected with the Harvard Food Frequency Questionnaire. DII was distributed normally and was used as a continuous variable in the models. Menstrual symptom occurrence and severity were self-reported using a modified version of the Calendar of Premenstrual Experiences. Using Mortola-based criteria, 89 women were classified as cases and 78 as controls. We used logistic regression to assess the association of DII and prevalent PMS. Linear regression was used to assess the association between DII and physical, affective, and total PMS symptom severity scores among 412 participants. Likelihood-ratio test was used to find the most parsimonious model, and analyses were adjusted for age at menarche.

Results: No significant associations were found in the adjusted models between DII and any of the PMS outcomes. For each unit increase in the DII, there was a 0.8-unit decrease in affective scores (95% CI -0.36, 0.18; P=0.52), 0.33-unit decrease in physical score (95% CI -0.66, 0.006; P=0.054), and 0.40-unit decrease in total menstrual symptom severity score (95% CI -0.97, 0.16; P=0.16). We did not find any significant associations between DII and PMS prevalence (OR=1.02; 95% CI 0.87, 1.20, P=0.78). Based on the r-squared of the different regression models, DII only explained 4-7% of the variability in PMS outcomes.

Conclusions: We did not find any significant associations between the DII and severity of menstrual symptom or PMS prevalence. Due to the small age range of participants, these results cannot be generalized to all women of reproductive age. Future research should examine these associations among women above the age of 30, when PMS is most prevalent.

Funding Sources: N/A