Serena Harris
Research Mentor(s): Jonathan Reader
Mentor Department: Michigan Alzheimer’s Disease Research Center
Authors: Serena Harris, Jonathan Reader
Session: Session 5 (2:00pm – 2:50pm)
Presentation Type: Poster 107
Abstract
By 2050, approximately 152 million individuals in the United States will be at greater risk of age-related impairments like dementia. While previous research suggests several associations with dementia, including educational attainment and Area Deprivation, few studies have examined the association between school-type density (e.g., four-year universities) and cognitive impairment. Understanding the relationship between school density and dementia will provide insight into how urbanization impacts cognitive health. This study combines data from the National Neighborhood Data Archive (NaNDA), the Census Crosswalk, the Michigan Neurodegenerative Dataset (MINDSet), and the National Alzheimer’s Coordinating Center’s Uniform Dataset Version 3 (UDSv3). School densities of interest per 1,000 people were identified by NaNDA and grouped by census tract. Census tracts were converted into zip codes matched to individual participants in MiNDSet. County of residence and county-level density from the average density of zip codes within a county were derived from ZIP codes. Diagnostic status (cognitively unimpaired (CU), Impaired but not mild cognitive impairment (Impaired), mild cognitive impairment (MCI), or dementia), age, race, and sex were obtained from the UDSv3. Multinomial logistic regression was used to determine the likelihood of individuals having a specific diagnosis compared to being CU. Independent variables included educational attainment, school type density, county, age, sex, and race. Schools included K-12, community colleges, four-year universities, business, technology, flight, cosmetology, apprenticeship, fine arts, sports instructional, language, and tutoring services. Participants (N = 1,086) mostly resided in Michigan (99.66%), self-identified as White (65.02%) and were biologically female (62.2%). The mean age was 73.4 years (SD = 8.2) with an average of 16.3 years of education (SD = 2.5). Most (43.0%) individuals were CU, 30.9%, MCI, 22.6%, dementia, and 3.5% Impaired. There was a significant association for race, sex, age, and education (all ps < 0.05). Non-White individuals were more likely to have MCI (odds ratio (OR) = 4.27) but were less likely to have dementia (OR = 0.37) compared to White individuals. Females were less likely to have MCI (OR = 0.51) and dementia (OR = 0.41) compared to males. Older individuals were more likely to have MCI (OR = 1.07) and dementia (OR = 1.04). More years of education indicated less likelihood of MCI (OR = 0.83) or dementia (OR = 0.80). The density of four-year universities decreased MCI (OR = 0.90) and impairment (OR = 0.72) likelihood but did not significantly differentiate between those with dementia and CU (ns). The density of other educational institutions had no statistically significant association. Results suggest that a greater density of four-year universities per 1,000 individuals predicted a lower likelihood of cognitive impairment. This study contributes to existing literature on urbanization and dementia by analyzing a novel dataset composed of several smaller publically available datasets. However, several limitations should be considered including a disproportionately white, female sample. Analyses were limited to associations and future research should consider the longitudinal relationship between school densities and cognitive impairment.
