Abstract P044: The Role of Self-Perceived Attributes of the Neighborhood Environment on the Association between Depressive Symptoms and Physical Activity in Older Latino Adults

Background: Physical inactivity is associated with the development of cardiovascular risk factors (e.g., hypertension, hypercholesterolemia, obesity, and diabetes) and an increased risk for cardiac-related morbidity and mortality. The impact of depressive symptomatology on self-perceptions of the neighborhood environment may explain the low levels of physical activity (PA) among older Latino adults with depression. The aims of the current study are to examine whether individuals with depression are less likely to view their neighborhood environment as conducive for walking behavior, and to examine the role of self-perceptions of the neighborhood environment in the association between depression and PA. Methods: We analyzed baseline in-person interview and pedometer data collected from Latinos ≥60 years participating in an exercise intervention (Caminemos!) at 27 senior centers. Depression was assessed using the Geriatric Depression scale and the Neighborhood Environment Walkability Scale (NEWS) was use to capture self-perceived attributes of the neighborhood environment associated with transport and recreational walking. The NEWS includes multiple subscales capturing the presence and accessibility of sidewalks, neighborhood aesthetics, pedestrian/traffic safety, and crime safety. PA was objectively measured using pedometer data. Bivariate analyses were conducted to examine the association between depression status and measures of PA, overall neighborhood walkability and associated subscales. We followed Baron and Kenny’s four steps to conduct a mediation analysis to examine whether self-perceived attributes of the neighborhood environment served as a mediator between depressive symptoms and PA. Results: Baseline characteristics for this sample (n=572) include: mean age of 73 years; 77% female; 83% with incomes below $20,000; and 59% with less than an 8th grade education. Approximately 27.6% screened positive for depression. When using continuous pedometer data as the outcome of interest, depressed individuals were less likely to engage in walking behavior when compared to their non-depressed counterparts (p=0.004), independent of basic demographic variables. T-tests revealed that depression status was only associated with the subscale of neighborhood walkability capturing criminal activity. Older Latino adults with depressive symptoms were more likely to perceive their neighborhood as a high-crime area (p=0.01). Results from the Sobel test did not find self-perceptions of criminal activity in the neighborhood to mediate the relationship between depressive symptoms and PA (p-value=0.31). Conclusion: While depressive symptoms are linked to self-perceptions of neighborhood criminal activity in older Latino adults, perceptions of neighborhood criminal activity does not act as a mediator in the relationship between depression and PA.