Abstract
BACKGROUND: "Precision nutrition" aims to recognize variation in response to dietary patterns to inform tailored advice based on behavioral, social, environmental, genetic, and metabolic factors. OBJECTIVES: We sought to identify characteristics of pregnant individuals that modify the associations between a high fat, sugar, and sodium diet and poor perinatal outcomes. METHODS: We used data from 8054 participants in the Nulliparous Pregnancy Outcomes Study: monitoring mothers-to-be (8 United States medical centers, 2010‒2013), a prospective cohort study. Usual periconceptional dietary intake was assessed at 6‒13 wk of gestation using a food frequency questionnaire. The exposure was a high fat, sugar, and sodium dietary pattern compared with all other diet patterns. The outcome was a composite of 1 or more perinatal outcomes: preeclampsia, gestational diabetes, preterm birth, or small-for-gestational-age birth. We used the doubly robust learner, which enables the use of machine learning to identify maternal characteristics that modify the effect of the dietary pattern on the composite outcome. RESULTS: Approximately 29% had a dietary pattern that was high in fat, sugar, and sodium. One quarter had any adverse pregnancy outcome. The confounder-adjusted association between a high fat, sugar, and sodium dietary pattern and risk of the adverse composite pregnancy outcome was stronger among certain subgroups of the cohort than others, including individuals with a higher BMI, lower socioeconomic status, and non-Hispanic Black race/ethnicity. For instance, compared with other diet patterns, intake of a diet high in fat, sugar, and sodium was associated with 5.9 excess cases per 100 pregnancies {adjusted risk difference 0.059 [95% confidence interval (CI): 0.012, 0.11]} among individuals living in a high-poverty neighborhood, but 2.3 excess cases per 100 pregnancies (0.023; 95% CI: -0.011, 0.057) among those residing in a low-poverty neighborhood. CONCLUSIONS: This work may provide clues that contribute to a deeper understanding of the heterogeneity in dietary responses in pregnancy.