Figure 3.

Modified Clinical Dementia Rating Scale (Sum of Boxes) at Baseline (pre-Intervention), 6-months (post-Intervention), and 1-year follow-up across Exercise factor (AE alone and AE+DASH), non-exercise (DASH and HE), DASH factor (DASH alone and AE+DASH) and non-DASH (AE and HE) groups, controlling for age, education, gender, race, Framingham Stroke Risk Profile, CVD medication burden, chronic anti-inflammatory use, baseline MoCA score, and baseline mCDR-SB scores. Participants in the Exercise conditions (AE alone and AE+DASH) exhibited better performance on the mCDR-SB scores (P = .027) compared to participants in the non-Exercise conditions (DASH alone and HE). A non-significant trend was observed for greater improvements in CDR-SB scores among participants in the DASH factor groups (DASH and AE+DASH) compared to non-DASH groups (AE and HE) (P = .054). Values for post-intervention and one-year post-intervention (18 months post-randomization) were derived using least squares means from our repeated measures, mixed model analyses with adjustment for all covariates. Baseline values were derived using least squares means from simple linear regression using the same covariates as our primary mixed model.