Impact of change in lifestyle health behaviors on cognition: Harmonizing and leveraging participant-level data from randomized controlled trials

Lifestyle health behaviors, such as physical activity, cognitive activity, healthy diet, and social activity, are low- risk, non-pharmaceutical strategies that may help protect against cognitive decline and risk of Alzheimer's disease and related dementias (AD/ADRD). However, causal evidence across observational and trial research

is inconsistent. There remains a need to further understand the impact of lifestyle health behaviors on cognitive outcomes, including what type(s) of behavior and magnitude of behavior change have greatest impact, and if impact differs across diverse at-risk subgroups. The purpose of our proposed study, InLife (Individualized

Lifestyle Change For Every Aging Brain), is to harmonize existing participant-level data from randomized controlled trials (RCTs) to examine how changes in lifestyle health behavior impact cognitive outcomes across diverse adults. We will leverage previously collected de-identified data from 12 U.S. trials representing 9,570

older adults for de novo analyses. In trials that agreed to contribute data, one or more lifestyle health behaviors (physical activity, cognitive activity, healthy diet, and/or social activity) were targeted and cognitive outcomes were assessed, including change in cognitive function (primary outcome: episodic memory and executive

function, secondary outcome: perceptual speed and global cognition) and (in some) incidence of mild cognitive impairment or dementia (exploratory outcome). Aim 1: Quantify impact of changes in lifestyle health behaviors on cognitive outcomes. Following our innovative harmonization methods based on an item response theory

framework, we will harmonize data across trials to test hypotheses that (a) the association between amount of change in behaviors and cognitive outcomes will be strongest for those with lower initial levels of the targeted behavior and (b) there will be nonlinear effects of increases in behaviors on cognitive outcomes such that

benefits will diminish as behavior change increases (i.e., diminishing returns). Aim 2: Examine determinants of health that moderate (enhance or inhibit) impact of lifestyle health behavior change on cognitive outcomes. We will evaluate whether associations between lifestyle health behavior and cognitive outcomes in Aim 1 differ

between subgroups. Informed by National Institute on Aging and World Health Organization frameworks, we focus on subgroups based on determinants of health that are structural/social (age, sex, education, racial background, ethnic background), biological (e.g., cardiometabolic disease), and environmental (e.g., rural vs.

urban, U.S. region, area deprivation index [API]). Aim 3 (exploratory): Determine the impact of lifestyle health behavior combinations on cognitive outcomes. We will evaluate whether two or more behaviors together have more impact than one alone and quantify which combinations of behaviors are most impactful. The proposed

InLife study will result in the first harmonized dataset of U.S. trials testing lifestyle interventions for cognitive outcomes. This research will be influential in shaping clinical guidelines focused on realistic lifestyle health behaviors for cognitive outcomes across diverse aging populations.