Increased wheezing risk with diesel exposure among children of younger mothers

Community-level characteristics and environmental factors of child respiratory illnesses in Southern Arizona

BACKGROUND

Lower respiratory illnesses (LRIs) and asthma are common diseases in children <5 years of age. Few studies have investigated the relationships between multiple, home-based social and environmental risk factors and asthma and LRIs in children. Of those that have, none have focused exclusively on children <5 years of age, who are more physiologically vulnerable and spend more time at home compared to older children. Further, no studies have done so at the community level.

METHODS

We modeled relationships between emergency department visits and hospitalization rates for asthma and LRIs for children <5 years and geographic risk factors, including socio-economic and housing characteristics, ambient air pollution levels, and population density in Maricopa and Pima Counties, Arizona, from 2005 to 2009. We used a generalized linear model with a negative binomial observation distribution and an offset for the population of very young children in each tract. To reduce multicollinearity among predictors, socio-economic characteristics, and ambient air pollutant levels were combined into unit-less indices using the principal components analysis (PCA). Housing characteristics variables did not exhibit moderate-to-high correlations and thus were not included in PCA. Spatial autocorrelation among regression model residuals was assessed with the Global Moran's I test.

RESULTS

Following the regression analyses, almost all predictors were significantly related to at least one disease outcome. Lower socio-economic status (SES) and reduced population density were associated with asthma hospitalization rates and both LRI outcomes (p values <0.001). After adjusting for differences between counties, Pima County residence was associated with lower asthma and LRI hospitalization rates. No spatial autocorrelation was found among multiple regression model residuals (p values >0.05).

CONCLUSIONS

Our study revealed complex, multi-factorial associations between predictors and outcomes. Findings indicate that many rural areas with lower SES have distinct factors for childhood respiratory diseases that require further investigation. County-wide differences in maternal characteristics or agricultural land uses (not tested here) may also play a role in Pima County residence protecting against hospitalizations, when compared to Maricopa County. By better understanding this and other relationships, more focused public health interventions at the community level could be developed to reduce and better control these diseases in children <5 years, who are more physiologically vulnerable.