Trajectories of Cognitive Decline Among Older Adults in the United States: The Influence of Neighborhood Physical Disadvantage and Education

Background and Purpose: It is estimated that one in four adults aged 65 and older in the United States experience cognitive decline before death. Cognitive decline can influence the quality of life and independence among older adults. Cognitive decline typically occurs over a prolonged period of 15-25 years with varying rates of decline among individuals. While mounting studies documented the significant role of neighborhood environments on cognition, previous studies have overlooked the longitudinal impact of living in a physically disadvantaged neighborhood on trajectories of cognitive function. Using 15 years of longitudinal data, this study aims to identify the trajectories of cognitive function and examine the associations between the trajectories and neighborhood factors.

Methods: We used data from the Health and Retirement Study (HRS) from 2006 to 2020. HRS is a nationally representative data of U.S. adults aged 50 years and older. Respondents (n=7460) were included in the sample if they answered neighborhood-related questions in 2006. Neighborhood physical disadvantage is measured by four-item questions asking respondents observations on their neighborhood environment. We used group-based trajectory models to identify latent groups for trajectories of cognitive function over time. We used Bayesian Information Criteria (BIC), average posterior probability (Ave>0.7), and the odds of correction classification (OCC>5) to decide the best-fitted model. We applied multinomial logistic regression to examine how neighborhood physical disadvantage may be associated with cognitive function trajectories. We also tested the moderating effect of education in the associations.

Results: We identified five trajectory groups based on their cognitive function changes over time. Group one (“high start, stable”) had the highest baseline cognitive function but remained stable with cognitive function over time. Group two (“medium-high start, stable”) and group three (“medium-high start, slight decrease”) had lower baselines and experienced a slow cognitive decline over time. Group four (“medium start, sharp decrease”) had the fastest cognitive decline, while group five (“low start, slight decrease”) had the lowest baseline and experienced moderate cognitive decline. Our multinomial logistic regression results showed that neighborhood physical disadvantage was associated with a higher odds ratio of being a member of Group two (RRR=1.14, p<0.001), three (RRR=1.29, p<0.001), four (RRR=1.36, p<0.001), and five (RRR=1.26, p<0.01). These groups reported a lower baseline and faster cognitive decline compared to Group one. Moreover, a significant interaction term between neighborhood physical disadvantage and education was found. Respondents who received lower than a high school education and lived in a more physically disadvantaged neighborhood were more likely to be in Groups two (RRR=2.10, p<0.01), three (RRR=2.75, p<0.001), four (RRR=2.58, p<0.01), and five (RRR=2.81, p<0.05), which exhibited a lower baseline and faster decline compared to Group one.

Conclusion and Implications: Our findings highlighted the important role of neighborhood physical disadvantage and education in older adults' cognitive function trajectories. Future policy and practice should provide extra support to older adults with lower levels of education and living in physically disadvantaged neighborhoods.