An Ecological Examination of Psychological Stress and Child Asthma: Family, Housing and Neighborhood Determinants

Objective. Asthma disproportionately affects non-white, urban, low- socioeconomic status populations, but trends and inequities are not well-explained by known risk factors. We hypothesized that disadvantaged populations experience housing and neighborhood stressors that produce psychological stress and impact child respiratory health through biologic and behavioral pathways. Methods. We examined relationships between material and social stressors and six child respiratory outcomes (measured as events/two weeks, except unplanned visits (six months) and controllability (time unspecified) and parent and child general health (GH) (time unspecified) in two cross-sectional studies using data from low-income, racially/ethnically heterogeneous Chicago families of children with respiratory problems. Adjusted binomial and negative binomial regression models produced risk differences (RDs), incidence rate differences (IRDs), and 95% confidence intervals (CIs). Housing Stressors (HS) was a continuous exposure representing number of stressors experienced in six months by 682 parents, weighted by parent-reported difficulty. Principal components analysis yielded two exposures representing 319 parents' perceptions of neighborhood collective efficacy (CE) and physical/social order (order); three-level exposures yielded low and mid vs. high (most favorable) contrasts. Results. HS was associated with nearly one extra day/two weeks of exercise intolerance [IRD=0.88 (95%CI: 0.41, 1.35)], nearly one-third extra day/two weeks of waking at night [IRD=0.32 (95%CI: 0.01, 0.63)] and nearly one-third extra day/six months of unplanned visits [IRD=0.30 (95%CI: 0.06, 0.54)]. Controllability [RD=6.19 (95%CI: 0.85, 11.54)] and child GH [RD=6.28 (95%CI: 1.22, 11.35)] were moderately associated with HS; parent GH's association was weak. More negative neighborhood perceptions tended to be associated with poor outcomes, though results differed by exposure. Only waking at night was strongly associated with CE [RDlow v. high=16.7 (95%CI: 2.8, 30.6)] and order [RDlow v. high=22.2 (95%CI: 8.6, 35.8)]. Exercise intolerance [RDlow v. high=15.8 (95%CI: 2.1, 29.5)] and controllability [RDmid v. high=12.0 (95%CI: 1.8, 22.3)] were moderately associated with order. Parent GH was strongly associated with CE [RDlow v. high =20.8 (95%CI: 7.8, 33.9)]; child GH's association was weak. Conclusions. Findings add to the conceptualization of stress as a social pollutant that becomes biologically embedded. Interventions must address physical and social dimensions of residence and mitigate individual-level stress while structural solutions to inequities are sought