Prospective multicenter study of relapse following treatment for acute asthma among adults presenting to the emergency department

Study objective: To identify factors associated with relapse following trea tment for acute asthma among adults presenting to the emergency department( ED). Design: Prospective inception cohort study performed during October 1996 to December 1996 and April 1997 to June 1997, as part of the Multicenter Asth ma Research Collaboration. Setting: Thirty-six EDs in 18 states. Patients: ED patients, aged 18 to 54 years, with physician diagnosis of acu te asthma. For the present analysis, we restricted the cohort to patients s ent home from the ED (n = 971), then further excluded patients with comorbi d respiratory conditions (n 32). This left 939 eligible subjects to have fo llow-up data. Interventions: None. Measurements and results: Two weeks after being sent home from the ED, pati ents were contacted by telephone. A relapse was defined as an urgent or uns cheduled visit to any physician for worsening asthma symptoms during the 14 -day follow-up period. Complete follow-up data were available for 641 patie nts, of whom 17% reported relapse (95% confidence interval, 14 to 20). Ther e was no significant difference in peak expiratory flow rate (PEFR) between patients who suffered relapse and those who did not. In a multivariate log istic regression analysis (controlling for age, gender, race, and primary c are provider status), patients who suffered relapse were more Likely to hav e a history of numerous ED (odds ratio [OD] 1.3 per 5 visits) and urgent cl inic visits (OR 1.4 per 5 visits) for asthma in the past year, use a home n ebulizer (OR 2.2), report multiple triggers of their asthma (OR 1.1 per tri gger), and report a longer duration of symptoms (OR 2.5 for 1 to 7 days). Conclusion: Among patients sent home from the ED following acute asthma the rapy, 17% will have a relapse and PEFR does not predict who will develop th is outcome. By contrast, several historical features were associated with i ncreased risk. Further research should focus on ways to decrease the relaps e rate among these high-risk patients. The clinician may wish to consider t hese historical factors when making ED decisions.