Living Environments and Social Isolation Among Older Adults: A Longitudinal Analysis By Racial and Income Groups

Background and Purpose: The living environment provides settings for social interaction. How do physical and social environments affect social isolation among community-dwelling racially diverse older adults has not been extensively researched. The community environments might affect the social life of racial minority older adults differently from their non-Hispanic White counterparts. One potential explanatory factor of the racial disparity in social isolation is income. The current paper scrutinizes the extent to which physical and social environments influence the experience of social isolation among older adults by racial groups and household income levels.

Methods: This study used waves 5-9 data from the National Health and Aging Trend Study (NHATS) (N = 5807). Facility-dwelling participants were excluded from the analysis because their living environments might not be comparable to that of their community-dwelling counterparts. All participants were 65 and older. Interviewers visited respondents’ homes and evaluated two layers of physical environments: community disorder and in-home disorder. The social environment was measured by respondents' perceived community cohesion. Social isolation score was the sum of living alone, small network size, and rare social participation. Income was assessed with the self-reported household annual income from various sources. Data management, variable re-coding, and analysis were conducted using Stata 15 SE. The longitudinal association between social isolation and environmental variables was analyzed using multilevel modeling. The moderating effects of income levels were explored. Subgroup analyses were conducted for non-Hispanic Black, non-Hispanic White, and Hispanic participants. The complex sample design of NHATS was accounted for by applying sampling weight and strata variables using "svyset” command in Stata.

Results: More than 30% of participants were racial minorities. Physical environment variables were not significantly related to social isolation over time. For the non-Hispanic Black (B=-0.041, SE=0.018, p=0.023) and non-Hispanic White (B=-0.035, SE=0.007, p<.001) participants, a better social environment was related to reduced social isolation longitudinally. A higher household income was directly related to a lower level of social isolation in non-Hispanic Black participants (B=-0.002, SE=0.001, p=.024), yet associated with increased social isolation among Hispanic participants (B=0.004, SE=0.000, p<.001). Household income moderated the longitudinal influence of social environment on isolation among Hispanic participants (B=-0.002, SE=0.000, p<.001). Moreover, it also moderated the association between in-home environment and isolation in Hispanic (B=-0.014, SE=0.005, p=0.017), and non-Hispanic White subgroups (B=-0.001, SE=0.000, p=0.010). Among Hispanic participants with less than median household income, an increase in social environment score was related to increased social isolation. On the contrary, an advantaged social environment was negatively associated with social isolation among all participants with higher income levels.

Conclusion and Implications: Social environment longitudinally influences social isolation in later life. Household income both directly affected and moderated the longitudinal relationships between living environment and social isolation. Subgroup analysis results suggest the influence of income levels was especially salient among Hispanic participants. Intervention programs aim to address social isolation could adapt community cohesion approaches for minority older adults, especially those with lower household incomes.