(P26-30-26) Substitution Models of Non-Ultra-Processed Food Alternatives for Ultra-Processed Foods: Association With Cardiovascular Disease in Three Prospective US Adult Cohorts

Objectives: To assess associations between replacing ultra-processed food (UPF) groups with non-UPF alternatives and cardiovascular disease (CVD), coronary heart disease (CHD), and stroke risk in 3 prospective US adult cohorts.

Methods: Repeated food frequency questionnaires assessed diet in the Nurses’ Health Study (NHS; 1984–2016; n = 75,735), NHSII (1991–2017; n = 90,813), and Health Professionals Follow-Up Study (1986–2016; n = 40,409). Cohort-specific multivariable Cox models estimated CVD (16,800 cases), CHD (10,401 cases), and stroke (6,758 cases) risk from isocaloric substitutions of 10 points of the cumulatively averaged percentage contribution to total energy intake from 10 Nova-defined UPF groups with the same percentage from 10 non-UPF groups. One-serving substitutions were modeled for non-caloric beverages. For each substitution pair, UPF and non-UPF groups were jointly modeled; log-hazard ratios (HRs) and 95% CIs were computed as β-coefficient differences, and standard errors derived from variances/covariance; cohort estimates were pooled with fixed-effect models.

Results: Each 10-percentage-point energy intake replacement of ultra-processed (UP) sugary drinks with non-UP fruit juice was associated with lower CVD (HR: 0.90; 95% CI: 0.85, 0.96), CHD (0.92; 0.86, 0.99), and stroke (0.87; 0.79, 0.97) risk. Replacing UP meats with non-UP fish/seafood/poultry/meat/eggs was associated with lower CVD (0.78; 0.72, 0.85), CHD (0.76; 0.69, 0.84), and stroke (0.81; 0.70, 0.92) risk. Each serving replacement of UP diet soda with tea/coffee/water was associated with lower CVD (0.96; 0.94, 0.98), CHD (0.97; 0.95, 0.99), and stroke (0.94; 0.92, 0.97) risk. In contrast, replacing UP cold cereal with non-UP cooked cereal (e.g., oats, wheat germ) was associated with higher CVD (1.11; 1.04, 1.19) and CHD (1.12; 1.03, 1.22) risk, and replacing overall UP bread/cold cereal (1.09; 1.02, 1.17) or UP refined-grain bread (1.11; 1.01, 1.22) with non-UP cooked cereal was associated with higher stroke risk. Other substitutions were not significant.

Conclusions: Replacing UP sugary drinks, meats, and diet soda with less-processed alternatives may lower CVD risk in US adults. Future research should determine how preparation/formulation of non-UP cooked cereal and nutrient profile/processing features of UP bread/cold cereal influence CVD risk.

Funding Sources: NIH