Enhancing Neurocognitive Health via Activity, Nutrition and Cognitive Exercise (ENHANCE): A Randomized Controlled Trial

Abstract

Background: While multidomain interventions show promise for promoting healthy aging, their impact on brain structure remains unclear. This randomized controlled trial (ENHANCE) assessed the efficacy of a 12-month group-based multidomain intervention on brain structure and function in community-dwelling older adults, with particular attention to urban-rural disparities.

Methods: The ENHANCE trial delivered twice-weekly group-based multidomain sessions (physical exercise, cognitive training and nutrition education) in urban and rural communities for 12 months, while the control group received quarterly telephone education. A total of 88 participants completed the trial (attendance rates > 60% across all sites), with 76 completing longitudinal MRI assessments (intervention: n = 35; control: n = 41). The intervention group (n = 44; 75.0% female) was significantly older than the control group (n = 44; 70.5% female) (75.0 ± 6.6 vs. 72.3 ± 5.0 years, p = 0.035) and had lower BMI (23.4 vs. 25.2 kg/m², p = 0.016) at baseline. The primary outcomes were brain structures (voxel-based changes in brain grey matter volume [GMV]), while secondary outcomes were functional outcomes (physical performance, nutritional status, cognitive function, psychosocial assessments and cardiometabolic biomarkers).

Results: Using two-stage tensor-based morphometry analysis, the intervention group demonstrated significantly less GMV reduction in regions over the left inferior temporal lobe compared with controls over 12 months (p < 0.05). Using generalized estimating equation (GEE) model, the intervention group showed enhanced physical performance at 6 months (5-time chair rise test: -1.19 s; 95% CI, -2.24 to -0.13; p = 0.028) and improved cognitive function by 12 months (Montreal Cognitive Assessment: +1.32 points; 95% CI, 0.10-2.54; p = 0.034). Cardiometabolic improvements included increased HDL-C (+6.65 mg/dL, p < 0.001) and decreased triglycerides (-16.07 mg/dL, p = 0.025) at 12 months in GEE models. In subgroup analyses, rural participants showed preserved GMV in additional regions including the cerebellum (Crus I and II) and occipital cortex with greater cognitive improvements (MoCA: +3.06 points; 95% CI, 0.84-5.27; p = 0.007), while urban participants showed greater GMV reduction in the left temporal-occipital fusiform cortex but achieved superior physical performance gains (5-time chair rise test: -1.85 s; 95% CI, -3.07 to -0.64; p = 0.003).

Conclusions: This study demonstrates that multidomain interventions can induce neuroplasticity in older adults, with differential effects on brain structure and function between urban and rural participants, emphasizing the need for tailored approaches that consider sociocultural factors to optimize healthy aging across diverse populations.

Keywords: brain structure; multidomain intervention; neuroplasticity; older adults; urban–rural disparities.

FIGURE 1

Neuroimage processing: two‐stage tensor‐based morphometry. A longitudinal registration of the baseline and follow‐up T1‐weighted images for each participant was performed (upper part). This process employing an inverse‐consistent non‐linear registration approach calculated the mid‐point average images and the corresponding voxel‐wise Jacobians determinants. These determinants were used to calculate the rate of brain changes over time, where warmer colours (positive values) denoted volume expansion and cooler colours (negative values) represented volume reduction. Subsequent standard voxel‐based morphometry (VBM) was applied to these average images (lower part), segmenting them into different tissue types and utilizing this segmentation to create group‐specific templates via a fast diffeomorphic registration algorithm. The deformation fields were applied to the individual Jacobian rate maps, which were then transformed into standard Montreal Neurological Institute (MNI) space and smoothed for voxel‐wise statistical analysis of brain expansion or contraction rates.

FIGURE 2

Study recruitment algorithm.

FIGURE 3

Differences of neuroanatomic changes between intervention and control groups. Grey matter volume (GMV) change rate differences between the intervention and control groups after 12 months, adjusted for study site, age, years of education and body mass index (BMI), were discovered in regions over the left inferior temporal lobe (left inferior temporal gyrus) (top row). When stratified by site, the Yilan subcohort analysis showed that, in addition to the left inferior temporal gyrus, the intervention group exhibited less GMV reduction or even expansion in regions such as the cerebellum (Crus I and II) and the occipital cortex (left occipital pole and bilateral lateral occipital cortex) after similar adjustments (middle row). In contrast, the Taipei subcohort demonstrated greater GMV reduction in the left temporal‐occipital fusiform cortex in the intervention group (bottom row). Hot colours represent areas with significantly increased GMV change, while cold colours indicate decreased GMV change in the intervention group compared with the control group.