Weber EJ, Silverman RA, Callaham ML, et al. Am J Med. 2002;113:371–378
Purpose of the Study.
The authors sought to determine patient characteristics associated with hospital admission after emergency treatment for asthma, and whether disposition guidelines are followed.
Study Population and Methods.
The authors performed a prospective multicenter cohort study involving 64 emergency departments in the United States and Canada. Consecutive adult patients with asthma exacerbations were interviewed, and their charts were reviewed using standardized protocols. Telephone follow-up at 2 weeks determined relapse.
Of 1805 patients, 363 (20%; 95% confidence interval [CI]: 18%–22%) were hospitalized. Among patients with severe exacerbations (final peak flow <50% of predicted), 122 (49%; 95% CI: 43%–55%) were hospitalized. Admission was associated with final peak flow, female sex, non-white race, severity of chronic illness, and severity of exacerbation. Admission predictors were similar regardless of hospital funding, region, or size. Among patients with mild or moderate exacerbations of asthma (peak flow ≥50% predicted), the likelihood of admission was associated significantly with the number of predefined risk factors for death from asthma. Of patients who were discharged from the emergency department, 62 (5%; 95% CI: 4%–6%) relapsed within 72 hours. Relapse was not associated with final peak flow (P = .39).
Associations between patient characteristics and disposition were similar across sites. Despite guidelines to the contrary, half of patients with final peak flow <50% were discharged. After emergency department treatment and discharge, short-term relapse was uncommon among patients with asthma, suggesting that strict peak flow cutoffs may be unnecessary if risk factors in patients with mild or moderate exacerbations are considered.
Results of this and other similar studies suggest that rational criteria for hospital admission can be established for patients with attacks of asthma. It’s good to see that emergency departments have developed admission criteria that extend beyond just peak flow measurements.