Stress, Social Support, and Health of Urban Latinas

Latinas are one of the largest and fastest growing minority groups in the United States (U.S.) and are at risk for poor health outcomes due to stress. However, little is known about the buffering effects of social support on stress and health in Latinas. In this cross-sectional study, we applied a structural equation modeling (SEM) approach to examine stress, social support and health outcomes in urban Latinas who participated in a community survey. We conducted iterative development of conceptual and analytic models through exploratory factor analysis (EFA), confirmatory factor analysis (CFA), and SEM. Four hypotheses were tested with SEM: 1) Higher levels of stress are associated with worse health outcomes; 2) Lower levels of social support are associated with worse health outcomes; 3) Being older or having lower levels of education is associated with worse health outcomes; and 4) Social support moderates the relationship between stress and health outcomes. The study sample comprised 2,035 Latinas who were predominantly immigrants from the Dominican Republic, had a mean age of 50; the majority reported low levels of chronic stress, lower than national average psychological distress scores, and rated their health as good or better. Levels of online social support were low – about 20%. The EFA rendered a three-factor solution with strong factor loadings: chronic stress, online social support, and health. The CFA models had adequate model fit indices. The hypotheses were not supported in the SEM. This is the first study to apply EFA and SEM approaches to develop and evaluate measurements of stress, social support, and health in a large sample of Latinas. Consequently, this research provides an important foundation for further examination of these constructs. This is an essential prerequisite to development of culturally appropriate social support interventions including those that are online for the rapidly growing U.S. Latina population.