Dietary patterns derived using reduced rank regression in postmenopausal women and risk of mortality: A population-based study

Objectives: The menopause transition increases the risk of chronic conditions in women. Given the crucial role of diet in health, we identified dietary patterns that explain variations in factors related to major health concerns in postmenopausal women. Also, we explored their association with all-cause and cardiovascular mortality. Study design: This study was conducted on 1814 postmenopausal women from the population-based Rotterdam Study. Main outcome measures: Dietary patterns were identified using reduced rank regression. Response variables included bone mineral density, body composition parameters, lipid profile markers, insulin resistance, systolic blood pressure, cognitive function, depression, and sleep quality. The associations with risk of mortality were assessed using Cox proportional hazard models. Results: The first dietary pattern, characterized by higher intake of vegetables, whole grains, legumes, nuts, coffee, tea, alcoholic beverages, and cheese, explained 2.95 % of the variation in responses, accounted for 12.11 % of the variation in general cognitive function captured by G-factor, 5.62 % in systolic blood pressure, and 4.13 % in bone mineral density, and was correlated with less adiposity, lower blood pressure, lipid markers, and insulin resistance. The second dietary pattern, characterized by higher intakes of processed meat, unprocessed red meat, poultry, eggs, and coffee, and lower intakes of sweets and tea, explained 1.54 % of the variation in responses, accounted for 5.45 % of variation in fat mass percentage, 3.47 % in lean mass index, and 3.29 % in bone mineral density, and was correlated with higher adiposity, insulin resistance, and lipid markers. No associations with mortality risk were identified after adjusting for confounders such as demographics, socioeconomic status, lifestyle, disease history, and medication use. Conclusions: We identified dietary patterns explaining a range of variation in health factors related to postmenopausal health. While these dietary patterns explained a large variation in some of the individual factors, their combined explained variation across multiple risk factors simultaneously was limited and no significant association with mortality risk was observed. This study provides a foundation for future research aimed at identifying optimal dietary patterns, integrating diverse health aspects, to improve health in postmenopausal populations.