LAUSR

BACKGROUND&AIMS It is controversial to preferentially choose low-fat milk or full-fat items. This study aimed to investigate the association of total and cause-specific mortality with 2 g/100 g or ≤ 1 g/100 g low-fat milk consumption compared with whole milk in general population. METHODS Overall, 29,283 adults aged ≥20 years from US National Health and Nutrition Examination Survey 1999-2014 were recruited with a median follow-up of 8.3 years. The types of milk consumption at baseline (e.g., whole-fat, 2 g/100 g low-fat, and ≤1 g/100 g low-fat) were reported during in-house interviews. Hazard ratios for the associations between milk types and mortality were assessed with the weighted Cox proportional regression. RESULTS During 241,572 person-years of follow-up, 4170 deaths occurred including 730 heart disease-related deaths and 846 cancer deaths. Consumption of milk contained lower fat exhibited an inverse association with total and cardiovascular mortality after multivariable adjustment. Compared with participants consuming whole-fat milk, those consuming 2 g/100 g or ≤1 g/100 g low-fat milk had a 14%-22% decrease in total mortality (p trend ≤0.001). Individuals consuming 2 g/100 g and ≤1 g/100 g low-fat milk had hazard ratios (95%CI) of 0.73 (0.55-0.97) and 0.67 (0.49-0.91) for heart-related mortality (p trend = 0.009). No significant difference was noted between whole-fat and lower-fat milk for mortality due to cancer, Alzheimer's disease, or diabetes mellitus. A similar trend was noted in the stratification and sensitivity analyses. CONCLUSION Compared with whole milk, low-fat or skim milk intake was associated with reduced total and heart-related mortality. Low-fat milk may be more conducive than whole milk for promoting cardiovascular health in general adults.