(P19-030-23) Associations of Dairy Foods and Cardiometabolic Markers Differ by Food Type, Fat Content and Dietary Pattern: The Boston Puerto Rican Health Study

Objectives: To evaluate associations between dairy intake and cardiometabolic markers (CM), among adults with healthy, vs less healthy, dietary patterns.

Methods: Baseline (n=1414) and visit 2 (n=1210 at ~2.2 y) data on diet and CM were available for adults aged 46-78 y. Diet was assessed with a validated FFQ. Dairy groups (total, milk, yogurt, cheese, total with fat (DWF), total non-fat (DNF) were created from disaggregated foods as servings/2000 kcal/d (s/d). CM included waist circumference (WC), HDL and LDL cholesterol, TG, insulin, C-reactive protein (CRP), blood pressure (BP), HbA1c, HOMA-IR, and homocysteine (Hcy) at visit 2. Generalized linear regression was used to test associations between dairy (mean baseline + visit 2) and CM markers (visit 2), adjusted for covariates. Interactions were tested by Mediterranean Diet Score, excluding the dairy component (MDS, high/low at the median=4.5, range 0-8).

Results: Baseline mean age was 59.3±7.7y, 70% female, 80% with metabolic syndrome. Total dairy intake was 2.3±1.1, DWF 2.1±1.0 and DNF 0.2±0.5 s/d. Total dairy, DWF, and milk were associated with lower Hcy at visit 2 (β -0.37, SE 0.12, P=0.001; β -0.39, SE 0.13, P=0.003; β -0.42, SE 0.13, P=0.002; respectively); DWF with lower WC (β -1.41, SE 0.44, P=0.001); cheese with lower TG (β -18.5, SE 9.2, P=0.04); and yogurt with lower A1c (β -0.46, SE 0.17, P=0.008). Significant interactions with MDS were observed for total dairy and WC (P=0.009); DNF and WC (PP=0.08) and HbA1c (P=0.09); DWF and WC (P=0.06); milk and WC (P=0.04); yogurt and LDL (P=0.06); cheese and WC (P=0.07) and Hcy (P=0.03). Among adults with MDS below the median, inverse associations between Hcy and total dairy, milk, DWF remained. Further, cheese was positively related to WC (β 3.01, SE 1.18, P=0.01) and inversely related with Hcy (β -1.3, SE 0.63, P=0.04). Among adults with MDS above the median, inverse associations between Hcy and total dairy, DWF, milk remained. Further, inverse associations between WC and total dairy (β -0.72, SE 0.35, P=0.04) and DNF (β -2.3, SE 0.73, P=0.001) were shown; cheese was inversely related to TG (β -25.9, SE 11.6, P=0.03).

Conclusions: Associations between dairy intake and CM markers appear to differ by dairy food, fat content and dietary quality.

Funding Sources: NIH