Different typologies of neighborhood resources are associated with cognitive function

Abstract

AbstractBackgroundNeighborhood resources (NRes) such as libraries, grocery stores, and child/family centers may promote cognitive health via opportunities for cognitive stimulation and other health benefits. Resources are distributed differently across neighborhoods, but many studies examine effects of individual NRes on health outcomes. This study uses typologies of NRes and examines whether different types of NRes patterns contribute to Black‐White disparities in cognitive function.MethodThis cross‐sectional study of 462 non‐Hispanic Black and White older adults (Mage = 63.6; 53% Black) was drawn from the Michigan Cognitive Aging Project, a regionally‐representative, prospective longitudinal study of adults transitioning to late life in the Detroit metropolitan area. Participants’ current addresses were geocoded and linked to neighborhood data from the National Neighborhood Data Archive for the year 2017. First, a latent class analysis (LCA) using Mplus identified typological classes of NRes using fifteen indicators across all 2,813 census tracts in Michigan. Next, path analysis estimated associations among race, NRes class, and global cognition, operationalized as a composite of five factor scores (episodic memory, executive function, processing speed, language, visuospatial ability). Models adjusted for age, sex/gender, education, and population size.ResultLCA indicated three classes of NRes: low‐resourced (48%), well‐resourced (42%), and highly‐resourced (10%) neighborhoods. Black participants had lower cognitive health and were more likely to live in low‐resourced or highly‐resourced neighborhoods than White participants. White participants were more likely to live in well‐resourced neighborhoods than Black participants. Surprisingly, living in well‐resourced neighborhoods was associated with worse cognition, compared to living in low‐resourced neighborhoods, but this association was only found among Black participants (βBlack = ‐.181, βWhite = .008). Living in highly‐resourced neighborhoods was not associated with cognition, but only 4% of the sample lived in these neighborhoods.ConclusionParadoxical findings suggest additional work is needed to better disentangle the role of NRes in cognitive disparities. Findings may be driven by factors such as racial residential differences (i.e., home rental and homeownership) and/or access to resources within a neighborhood, as well as neighborhood social climate. This study adds to—and highlights the need for more comprehensive models in—research on neighborhoods and ADRD inequalities.