(P32-10-26) A Food-Based Global Diet Quality Score Is Associated With Lower Risk of Cardiovascular Disease: Multinational Prospective Cohort Analyses in the United States and Europe

Objectives: To examine the association of the food-based Global Diet Quality Score (GDQS) with cardiovascular disease (CVD) risk and compare its performance with nutrient-centric diet quality indices (DQIs) in US and European populations.

Methods: We analyzed prospective data from US cohorts (Nurses’ Health Study [NHS], NHS II, Health Professionals Follow-up Study) and the European Prospective Investigation into Cancer and Nutrition. US adults (36-53 y) were followed for 24-28 y, with questionnaires updating covariates every 2-4 y. European adults (mean (SD) age = 51 y) completed baseline questionnaires and were followed for a median of 12.8 y (interquartile range 2.8 y). The GDQS, Alternative Healthy Eating Index-2010, Adapted Mediterranean Diet Score, Dietary Approaches to Stop Hypertension (DASH), and Planetary Health Diet Index were derived from cohort-specific dietary data. Incident CVD, coronary heart disease (CHD), and stroke were ascertained. Multivariable Cox models estimated associations per 1-SD higher DQIs and by quintiles; US cohorts were pooled using fixed-effects models. Restricted cubic splines assessed dose-response associations. Scores were further standardized and modeled jointly to compare coefficients across DQIs (Wald tests). Harrell’s C-statistic evaluated predictive performance for CVD outcomes.

Results: We included 517,333 participants from seven countries and documented 31,920 incident CVD events (21,023 CHD, 11,788 strokes). In the US, hazard ratios (HRs) per 1‑SD higher GDQS were 0.96 (95% CI 0.95, 0.98) for CVD, 0.96 (0.94, 0.98) for CHD, and 0.96 (0.94, 0.98) for stroke. In Europe, corresponding HRs were 0.94 (0.92, 0.97), 0.95 (0.92, 0.98), and 0.94 (0.90, 0.99). Higher GDQS quintiles were associated with lower CVD risk compared to lower quintiles, with significant dose-response associations. All DQIs yielded similar associations, with no significant differences between standardized DQIs (p-Wald >0.05). GDQS models showed C-statistics for CVD prediction comparable to those of other DQIs.

Conclusions: In adults from seven countries, higher GDQS was associated with lower CVD risk in a dose-dependent manner and showed discrimination performance comparable to DQIs that require local adaptation and specialized nutrient data analysis, supporting its use for CVD risk assessment in diverse settings.

Funding Sources: NIH