TABLE 3

Multivariable Analysis Results: Adjusted Relationship between SHS Exposure and Outcome Variables

Outcome VariableOverall CohortCHCOMUSC
LOS, h,a IRR (95% CI)1.1 (1.0–1.2)1.2 (1.1–1.3)1.0 (0.8–1.1)
IV steroids,b OR (95% CI)1.6 (1.1–2.3)1.5 (0.9–2.5)1.8 (0.9–3.6)
PICU,b OR (95% CI)1.5 (1.0–2.1)1.4 (0.8–2.4)1.6 (0.5–2.7)
Magnesium,b OR (95% CI)1.3 (0.9–2.0)1.4 (0.9–2.3)1.0 (0.5–2.2)
  • a The LOS model is a zero-truncated negative binomial regression model and included age, race, insurance, gender, oxygen use, home inhaled corticosteroids, diagnosis of pneumonia, and diagnosis of respiratory syncytial virus as covariates. The overall cohort model had significant interaction between institution and SHS in predicting LOS. An interaction term (institution*SHS) was included for this reason.

  • b Logistic regression models controlled for age, race, insurance, gender, oxygen use, home inhaled corticosteroids, diagnosis of pneumonia, diagnosis of respiratory syncytial virus, and institution (for the overall cohort model).