papers

Factors associated with 90-day death after emergency department discharge for atrial fibrillation

Ann Emerg Med 61:539-548.e1.

Institute for Clinical Evaluative Sciences, Toronto, ON, Canada

STUDY OBJECTIVE: More than 10% of patients treated in the emergency department (ED) for atrial fibrillation die within a year of the visit. We sought to describe the post-ED care of an older population of atrial fibrillation patients who were discharged home from the ED and to assess patient characteristics and processes of care associated with risk of death within 90 days of discharge.

METHODS: This retrospective cohort analysis included patients aged 65 years or older with a primary ED diagnosis of atrial fibrillation who were treated at all nonpediatric EDs in the province of Ontario, Canada, between April 2007 and March 2010. Only the index emergency visit for each patient was included, and patients admitted to the hospital were excluded. We evaluated the association of postdischarge outpatient care and medications, ED cardioversion, and site volumes of atrial fibrillation patients with adjusted hazard of 90-day death.

RESULTS: Among 12,772 qualifying index ED visits, there were 417 (3.3%; 95% confidence interval [CI] 3.0% to 3.6%) deaths within 90 days of the visit. Patients with no follow-up care had a significantly increased hazard of death (hazard ratio [HR] 2.27; 95% CI 1.50 to 3.43) relative to those who consulted a family physician, as did patients prescribed a calcium-channel blocker (HR 1.55; 95% CI 1.15 to 2.09) relative to a β-blocker. A filled warfarin prescription was associated with a lower hazard of death (HR 0.70; 95% CI 0.51 to 0.95). Higher site volumes (HR 0.66; 95% CI 0.41 to 1.08), cardioversion (HR 0.69; 95% CI 0.42 to 1.15), and follow-up care by a specialist only (HR 0.75; 95% CI 0.51 to 1.12) were not associated with 90-day mortality.

CONCLUSION: Among older atrial fibrillation patients discharged from the ED in the province of Ontario, lack of follow-up care had the strongest association with subsequent mortality. If validated, these results suggest that as proportionately more of these patients are discharged from EDs in future, the focus should go beyond ED care itself to the care subsequent to the emergency visit.

PMID: 23522608