Abstract:This study aimed to develop a k-means clustering algorithm to identify distinct food intake patterns through cluster analysis.This was a secondary analysis of the Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-Be (nuMoM2b), including nulliparous individuals with singleton pregnancies. Dietary intake data from the 3 months preceding pregnancy were collected using a validated questionnaire. The primary outcome was a composite measure including preterm birth, stillbirth, preeclampsia, eclampsia, gestational diabetes, and small for gestational age. Clusters were formed using a k-means clustering algorithm with Euclidean distance, based on 335 dietary variables. The association between dietary clusters and adverse pregnancy outcomes (APOs) was assessed. Relative risks with 95% confidence intervals (95% CIs) were calculated using modified Poisson regression, adjusting for predefined confounders. A random forest model was also employed to identify features predictive of cluster allocation.The analysis included 7,599 participants, distributed across three clusters: Cluster 1 (n = 4,243, 55.8%), Cluster 2 (n = 2,768, 36.4%), and Cluster 3 (n = 588, 7.7%). Cluster 2, which serves as the referent cluster, is characterized by a higher intake of vitamin E as α-tocopherol, vitamin A retinol activity equivalents, vegetables, and fruits, aligning most closely with a healthy diet pattern. Compared with Cluster 2, Cluster 1, characterized by a lower intake of the same nutrients, did not show a significant association with increased odds of APOs (22.7 vs. 25.4%; adjusted relative risk [aRR], 1.07 [95% CI: 0.98–1.18]). In contrast, Cluster 3, characterized by higher intake of trans fats, dietary polyunsaturated fatty acids 20:4, red meat, and sugary beverages, was significantly associated with APOs compared with Cluster 2 (31.0 vs. 22.7%; aRR, 1.19 [95% CI: 1.01–1.39]).A dietary pattern characterized by a high intake of trans fats, polyunsaturated fatty acids, red meat, and sugary beverages is significantly associated with an increased risk of APOs.