This study aimed to fill the gap in research regarding the longitudinal studies on the association between urban neighborhood characteristics and mental health of older populations. Individual level register-based data sets from Finland (10 largest cities), Sweden (Stockholm), and Italy (Turin) including satellite based land cover data were used. We followed individuals for antidepressant purchases for 5 years in 2001-2015, and used hierarchical negative binomial models to assess whether there was an association between social and physical area characteristics and antidepressant use and to what extent was this association attributable to individual sociodemographic characteristics of the residents and whether the findings were consistent across countries. We found weak and inconsistent evidence of high levels of area characteristics related to dense physical urban structure being predictive of increased antidepressant use in ages 50+. However, generally the extent to which antidepressant use was clustered by areas in the studied contexts was minimal.
Presented in Session 26. Flash Session: Neighborhood Processes in Health