Purpose and Vision:
Preliminary research suggests that adults with lower engagement in lifestyle health behaviors, including physical activity, cognitive activity, healthy diet, and social activity are at a greater risk for cognitive decline. However, even modest engagement these behaviors could provide a meaningful, clinically significant benefit related to cognition, particularly among those with low engagement at baseline. These preliminary observational results need to be examined in a larger, more diverse study design.
Studying the influence of lifestyle health behaviors on cognitive outcomes in observational studies has significant limitations such as insufficient control for confounding and measurement bias. The benefit of pooling Randomized Control Trials is that change in behavior is examined in a context where the impact of lifestyle behavior interventions can be accurately measured with less concern for these biases, as well as leveraging sample size and trial measurement overlap to assess impact across populations. Results from this research have the potential to move both the science and clinical guidelines for lifestyle health behaviors and cognitive outcomes forward and provide realistic guidance to create a positive impact
Study Design
What is Harmonization in the context of InLife? The process of aligning and integrating data from multiple sources into a unified, participant-level data set, allowing researchers to answer questions with greater precision and power
InLife is built around a core strength: harmonizing participant-level data across multiple U.S.-based lifestyle intervention trials. By pooling individual-level data, we will be able to move beyond post-hoc meta-analyses of lifestyle interventions and cognitive decline and instead gain the power to test de novo hypotheses with precision. Our harmonization framework builds on techniques rooted in item response theory (IRT) and qualitative assessments of measures of interest. To ensure transparency and scientific rigor, domain teams have been convened to guide harmonization across key areas, demographics, cognitive outcomes, health, and lifestyle health behaviors. The result is a scalable, generalizable framework for data harmonization that can enhance statistical power, support subgroup analyses, and directly respond to NIA’s priority of getting the most out of existing data. For InLife, harmonization is not just a technical step, it is the driving force behind the ability to better inform precise recommendations for behavior change, while moving forward the science on reducing risk of Alzheimer’s disease, dementia, and cognitive impairment.
Cognitive Outcomes
Our study investigates how changes in everyday behaviors may shape cognitive abilities as people age. Our primary outcomes include episodic memory, defined as the ability to learn and recall verbal information, and executive function, which is a measure of planning, inhibition, and cognitive flexibility. Our secondary outcomes include perceptual speed, measured through timed tasks that assess ability for rapid information processing, and global cognition, which is a composite score derived from multiple cognitive functions. When available, incidence of mild cognitive impairment (MCI) and all-cause dementia will also be measured. These outcomes will be harmonized across trials using an item response theory (IRT) framework, which aligns scores based on overlapping test items and adjusts for differential item functioning (DIF), while retaining trial-specific items to enhance precision.
Behavior Change
With regard to modifiable lifestyle behaviors such as physical activity, cognitive activity, healthy diet, social activity, there remains a critical need to further understand which of these lifestyle behaviors is most impactful on improving cognitive outcomes. Behavior change is as the difference between baseline and follow-up measures within measured lifestyle behaviors. Physical activity is measured as minutes per day of moderate-to-vigorous activity. This is derived from accelerometer data when available, and secondarily through self-report. Cognitive activity is measured as time spent in cognitively stimulating tasks such as reading, puzzles, or training programs. Healthy diet is assessed using the MIND diet score, calculated using servings per week of brain-healthy and unhealthy food groups. Social activity is measured as time spent on social interactions such as structured group activities. Among harmonized measures, we will test associations with behavior change cognitive outcomes across trials.
Exploratory Analyses
Subgroup Differences– We will examine whether the impact of lifestyle behavior changes on cognitive outcomes varies across participant characteristics commonly collected in clinical trials. These include demographic factors such as age, sex, educational attainment, health status/history, and geographic and neighborhood
Combined Behavior– We will assess the potential for multiple behavior changes to have greater effects than individual changes alone.
Funding
InLife is supported by the National Institutes of Aging (NIA) 1RF1AG089550-01 and the performance site is the University of Illinois Chicago. The study is a planned five year project that began on 9/15/2024 and will continue through 2029. IRB approval was obtained from Office of the Protection of Research Subjects at the University of Illinois Chicago.