Abstract P153: Association of Habitual Milk Intake with Cognitive Decline From Mid-life to Late-life: The Atherosclerosis Risk in Communities (ARIC) Neurocognitive Study (NCS)

Abstract

Background: Animal models and large population studies have related habitual milk intake to impaired cognitive performance via oxidative stress, but results are inconclusive. We report on the association of milk intake at midlife with 20-year cognitive decline in ARIC-NCS cohort. Methods: Average daily milk intake was assessed in 15,766 ARIC study participants with a standardized FFQ administered at cohort inception in 1987-1989. Reported milk intake was categorized as almost never, <1glass/day, 1 glass/day, >1 glass/day. Domain-specific and global cognitive decline was assessed on three occasions from 1990 through 2013, testing Delayed Word Recall (memory), Digit Symbol Substitution (executive function), and Word Fluency (language). A mixed effects model adjusted for age, race, sex, education, smoking, physical activity and ApoE4 was used to evaluate the association of milk intake with change in cognitive performance over 20 years. The MICE method was used to account for attrition. Results: Among 13,821 participants in the analytic set, the distribution of milk intake across the above four categories was 18.7%, 35.8%, 27.3%, and 18.1%, respectively. In race-combined fully adjusted models greater milk intake was associated with greater cognitive decline (Figure) as indicated by a significant trend (p=0.018). The average difference in the 20-year rate of change in global z between those who reported almost never drinking milk and those who consumed >1glass/day was -0.053 [95%CI: -0.092, -0.014], approximately equivalent to 1 year of additional cognitive aging. The trend met nominal statistical significance in Whites, but not African Americans. Conclusions: Results of this study add to reports indicating that greater milk intake may be associated with greater rate of cognitive decline. Replication of these results and insights into interactions with the intestinal microbiota and other putative modifying factors of the reported association is warranted considering the widely prevalent intake of milk in US adults.