Importance: Previous research has found that vegetarian diets are associated with lower bone mineral density and higher risk of fractures, but these studies did not differentiate the quality of the plant-based foods.
Objective: To examine the association between the quality of plant-based diets (not necessarily vegan but also omnivorous) and hip fracture risk among postmenopausal women in the Nurses' Health Study.
Design, setting, and participants: This cohort study analyzed data from 70 285 postmenopausal women who participated in the US Nurses' Health Study from 1984 through 2014. Data were analyzed from January 1 to July 31, 2023.
Main outcomes and measures: Hip fractures were self-reported on biennial questionnaires. Diet was assessed every 4 years using a validated semiquantitative food frequency questionnaire. Plant-based diet quality was assessed using 2 previously established indices: the healthful Plant-Based Diet Index (hPDI), for which healthy plant foods (whole grains, fruits, vegetables, nuts, legumes, vegetable oils, and tea or coffee) received positive scores, whereas less healthy plant foods (fruit juices, sweetened beverages, refined grains, potatoes, and sweets or desserts) and animal foods received reversed scores; and the unhealthful Plant-Based Diet Index (uPDI), for which positive scores were given to less healthy plant foods and reversed scores to healthy plant and animal foods. Quintile scores of 18 food groups were summed, with a theoretical range for both indices of 18 to 90 (highest adherence). Cox proportional hazards regression with time-varying covariates was used to compute hazard ratios (HRs) and 95% CIs for hip fracture.
Results: In total, 70 285 participants (mean [SD] age, 54.92 [4.48] years; 100% White women) were included, and 2038 cases of hip fracture were ascertained during the study and for up to 30 years of follow-up. Neither the hPDI (HR for highest vs lowest quintile, 0.97 [95% CI, 0.83-1.14]) nor the uPDI (HR for highest vs lowest quintile, 1.02 [95% CI, 0.87-1.20]) for long-term diet adherence was associated with hip fracture risk. However, when examining recent intake for the highest vs lowest quintiles, the hPDI was associated with 21% lower risk of hip fracture (HR, 0.79 [95% CI, 0.68-0.92]; P = .02 for trend), and the uPDI was associated with 28% higher risk (1.28 [95% CI, 1.09-1.51]; P = .008 for trend).
Conclusions and relevance: Findings of this cohort study indicated that long-term adherence to healthful or unhealthful plant-based diets as assessed by hPDI and uPDI scores was not associated with hip fracture risk. Future research should clarify whether the associations observed with recent dietary intake are due to short-term effects of these dietary patterns, reverse causality, or both.