Skip to main content
. Author manuscript; available in PMC: 2021 Sep 29.
Published in final edited form as: J Alzheimers Dis. 2021;83(2):683–692. doi: 10.3233/JAD-210107

Table 7.

Association of MIND diet score with global cognition proximate to death and cognitive decline

Model N β-estimate (SE) P-value
A 391 0.121 (0.042) 0.005
B 569 0.098 (0.037) 0.008
C 363 0.120 (0.044) 0.007
D 569 0.108 (0.037) 0.004
E 374 0.114 (0.050) 0.023
F 569 0.012 (0.005) 0.022

Each model (A-E) is a separate linear regression model adjusted for age at death, sex, APOE4, education, total energy intake, and late-life cognitive activities and global AD pathology (not shown). Model F is a linear mixed-effect model adjusted for age at death, sex, APOE4, education, total energy intake, and late-life cognitive activities (not shown), and their interactions with time (years before death).

Estimates are reported for the interaction of predictor with time. β-estimate for MIND diet score is for 1 standard deviation (SD=1.42) increase.

Model A excludes individuals with mild cognitive impairment or other conditions contributing to cognitive impairment at the baseline.

Model B MIND diet score is based on dietary information collected at the baseline.

Model C restrict analysis on participants whose MIND diet scores were relatively constant (e.g., consecutive change less than 20%) during the study period.

Model D is further adjusted for physical activity

Model E includes cases diagnosed with postmortem diagnosis of Alzheimer’s disease based on NIA-Reagan consensus recommendations.

Model F evaluates the rates of cognitive decline. β-estimate indicates rate of change in cognitive decline per year.