Exploring the Determinants of Stress: An Examination Using Spatial and Multilevel Analysis in Philadelphia

Tse-Chuan Yang, Pennsylvania State University
Stephen Matthews, Pennsylvania State University

Traditionally, studies of the predictors of stress are focused on the individual characteristics. In our study we also include ecological factors, particularly social and built environment characteristics of neighborhoods. We combine multiple georeferenced data sets with the Philadelphia Health Management Corporation’s Community Health database, a telephone survey of 4,095 respondents embedded in several hierarchical contexts; our results focus on 158 neighborhood areas. Exploratory spatial data analysis suggests a moderate spatial autocorrelation and uncovers a spatial clustering of self-rated stress in Philadelphia County. At the individual level, few social factors appear to matter, though self-rated neighborhood trust and food insecurity have independent effects on stress. At the neighborhood level, the presence of a hazardous waste site and traffic levels are determinants of self-rated stress even after controlling for other individual characteristics. The latter two factors are relevant to public health policy as they are potentially modifiable.