Greater protein intake at breakfast or as snacks and less at dinner is associated with cardiometabolic health in adults

BACKGROUND & AIMS Protein intake is inversely associated with waist circumference and positively associated with HDL-cholesterol concentrations. However, the relationship between protein intake during specific eating occasions and cardiometabolic health is not well documented. This cross-sectional study measured protein intake at meals and combined snacking occasions and evaluated associations between protein intake at meals or snacking occasions and markers of cardiometabolic health in adults. METHODS Deciles of individual usual intake (IUI) for protein at meals and combined snacking occasions were calculated using NHANES 2013-2016 data (n = 10,112; ≥19 y). Associations between protein intake at meals or snacks and markers of cardiometabolic health were determined using regression analysis. Covariates included age, age2, gender, ethnicity, physical activity level, poverty income ratio, protein IUI at other eating occasions, and total energy IUI at the eating occasion being analyzed (model 1). P < 0.0042 was considered statistically significant. Registered as ISRCTN11120152. RESULTS Protein intake ranged (10th to 90th percentiles) 5.9 ± 0.1 to 22.6 ± 0.3 g/d at breakfast, 14.0 ± 0.1 to 34.6 ± 0.4 g/d at lunch, 24.3 ± 0.3 to 46.8 ± 0.2 g/d at dinner, and 4.9 ± 0.1 to 16.5 ± 0.2 g/d at combined snacking occasions. Protein intake at breakfast was inversely associated with diastolic (-0.39 ± 0.10, P = 0.0003) and systolic (-0.40 ± 0.13 mmHg per decile, P = 0.0038) blood pressure. Protein intake at breakfast was positively related to HDL-cholesterol (0.75 ± 0.16 mg/dL per decile, P = 0.0001). Protein intake at dinner was positively associated with insulin concentrations (0.77 ± 0.23 uU/mL per decile, P = 0.0025) and the homeostatic model assessment of insulin resistance (HOMA-IR, 0.32 ± 0.09 per decile, P = 0.0017). Protein intake from snacks was inversely associated with diastolic blood pressure (-0.41 ± 0.09 mmHg per decile, P < 0.0001) and CVD risk score (-0.0018 ± 0.0004 per decile, P = 0.0001). Protein intakes at meals and snacks were not associated with BMI, waist circumference, glucose, total cholesterol, LDL-cholesterol, or triglycerides. CONCLUSIONS In adults, protein consumption at breakfast is inversely associated with systolic and diastolic blood pressure and positively associated with HDL-cholesterol, while protein consumption at dinner is positively associated with HOMA-IR and insulin concentrations.