Trends and predictors of hospitalization after emergency department asthma visits among U.S. Adults, 2006-2014.

Background: Asthma hospitalizations are an ambulatory care-sensitive condition; a majority originate in emergency departments (EDs).Objective: Describe trends and predictors of adult asthma hospitalizations originating in EDs.Methods: Observational study of ED visits resulting in hospitalization using a nationally representative sample. We tested trend in hospitalization rates from 2006 to 2014 using logistic regression, then assessed the association between hospitalization rates and patient and hospital characteristics using hierarchical multivariable regression accounting for hospital-level clustering.Results: Total ED asthma visits increased 15% from 2006 to 2014, from 1.06 to 1.22 million, while the likelihood of hospitalization decreased (20.9-18.2%, p < 0.01). Adjusting for increased asthma prevalence, ED visit rates and hospitalization rates decreased by 10 and 21%, respectively. Hospitalization was independently associated with older age, female gender (OR = 1.23, 95% CI 1.20-1.26), higher Charlson score (OR = 1.99, 95% CI 1.97-2.01), Medicaid (OR = 1.05, 95% CI 1.01-1.08) and Medicare (OR = 1.26, 95% CI 1.22-1.31) insurance, and trauma centers (OR = 1.34, 95% CI 1.12-1.60). Hospitalization was less likely for uninsured visits (OR = 0.7, 95% CI 0.67-0.73), lower income areas (OR = 0.89, 95% CI 0.85-0.93), non-metropolitan teaching hospitals (OR = 0.83, 95% CI 0.71-0.96), Midwestern (OR = 0.84, 95% CI 0.69-1.01) or Western regions (OR 0.69, 95% CI 0.56-0.83). Unmeasured hospital-specific effects account for 15.8% of variability in hospital admission rates after adjusting for patient and hospital factors.Conclusions: Total asthma ED visits increased, but prevalence-adjusted ED visits, and ED hospitalization rates have declined. Uninsured patients have disproportionately more ED visits but 30% lower odds of hospitalization. Substantial variation implies unmeasured clinical, social and environmental factors accounting for hospital-specific differences in hospitalization.