Perceptions of Neighborhood Safety and Depressive Symptoms Among Older Minority Urban Subsidized Housing Residents: The Mediating Effect of Sense of Community Belonging

Background:   Depression not only impairs older adults’ physical, cognitive and social functioning; its presence adversely affects the course and complicates the treatment of other chronic diseases.  Furthermore, depression is one of the strongest predictors of suicide in the older population.  Older minority subsidized housing residents represent a population that is particularly vulnerable to depression.  While estimates of the prevalence of depression in the U.S. older population range from 8% to 15%; older subsidized housing residents report depression rates ranging from 25% to 33%. Although research shows that neighborhood characteristics influence mental health, it has not been explored in this target population.  Using social disorganization and social capital theories, this study’s aim was to explore if perceptions of neighborhood safety are associated with depressive symptoms; and, whether a sense of community belonging has a mediating effect on this potential relationship. 

Methods:  The data are from interviews, conducted in English or Spanish, with 216 of 300 tenants age 55-plus of a subsidized housing development in a minority neighborhood of a Northeastern city (a 72% response rate). Approximately 50% of the participants identified as Black and 45% identified as Latino. To test for a mediation effect, Baron and Kenney’s (1986) three-step hierarchical regression approach was used.   In the first equation, the criterion is regressed on the independent variable; in the second equation, the mediator is regressed on the independent variable; and, in the third equation, the criterion is regressed on the independent variable and the mediator simultaneously. A mediation is inferred if the independent variable has a significant effect on the criterion (equation 1) and mediator (equation 2); and, if equation 3 reveals that firstly, the mediator has a significant effect on the criterion and secondly, the effect of the independent variable on the criterion is either no longer or less significant than found for this relationship in equation 1.

Results: Among participants, 80% identified feeling very safe during the day while 63% expressed feeling very safe at night; 60% possessed a strong sense of community belonging and 26% had clinically relevant depressive symptoms.    After controlling for number of chronic conditions and functional disability (covariates), neighborhood safety perceptions were associated with depressive symptoms (Equation 1, β = -.29, p < .001) and sense of community belonging (Equation 2, β = .19, p < .01).  In Equation 3, both perceptions of neighborhood safety (β = -.23, p < .001) and community belonging (β = -.26, p < .001) are significant predictors of depressive symptoms; however the relationship between sense of neighborhood safety and depression is weaker in Equation 3 than in Equation 1 (a decline in β).   Thus, our core hypothesis of the buffering effect of community belonging on the relationship between neighborhood safety and depression is supported.

Implications:  The findings suggest that having a sense of connection to one’s surrounding and environment can be a protective factor and underscore the importance of continued exploration of the role of social capital in relation to feelings of safety in later life for urban older adults.