A trial fibrillation (AF) is a frequent reason for emergency room visits.1 Traditional practice has been to admit patients with new AF for management.2 Despite this widespread practice pattern, the clinical utility of this approach is uncertain, as evidenced by a recent retrospective review of patients with acute, uncomplicated AF.3 A retrospective chart review was performed on all patients presenting with AF to the emergency room at a university hospital from September 1997 to November 1998, to assess the impact of a first episode of uncomplicated AF on clinical outcomes, length of stay, and actual direct costs. • • • Patients meeting the following eligibility criteria had data abstracted: (1) newly diagnosed or new onset AF; (2) uncomplicated clinical status at initial presentation, with no indication for hospital admission other than AF; (3) age between 18 and 75 years; (4) no contraindications to anticoagulation; and (5) no symptomatic congestive heart failure or known left ventricular ejection fraction ,30%. Forty-seven patients were identified. AF was confirmed in each case and the records were abstracted for baseline clinical and demographic information, clinical outcomes, medical management, length of stay, and clinical follow-up after initial presentation. Actual direct costs were calculated using the Transition Systems, Inc., cost allocation system (Eclipsys Corporation, Delray Beach, Florida). The costs of diagnostic testing (echocardiography, stress testing) related to AF after discharge performed on an outpatient basis were included. Subgroup analyses were performed based on the disposition of the patient after emergency room evaluation and on the presence or absence of spontaneous conversion of AF to sinus rhythm. Bivariate comparisons were made using Fisher’s exact test, chi-square test, or Pearson’s chi-square test, and means were compared across groups using Student’s t test. SAS version 6.2 (SAS Institute, Cary, North Carolina) was used for analyses. A p value ,0.05 was considered significant. The study group consisted of 47 patients with a mean age of 57 6 16 years (range 18 to 75). Baseline clinical and demographic characteristics are listed in Table 1. The use of rate control, anticoagulation, and resource utilization after initial presentation are shown in Table 2. The mean length of stay was 1.72 6 0.99 days (range 1 to 5) and the mean actual direct cost was $1,989 6 $1,583 dollars (range $202 to $4,430). The 25th and 75th percentiles for length of stay were 1 and 2 days, and for actual direct costs were $1,171 and $2,443. Among the 47 patients, 39 patients (83%) were admitted to the hospital and 8 patients (17%) were discharged from the emergency room. Mean age and gender ratio did not differ between the admitted paFrom the Cardiovascular Division, Washington University, St. Louis, Missouri; the Heart Care Program, Consortium for Health Outcomes, Innovation, and Cost Effectiveness Studies, and the Department of Emergency Medicine, University of Michigan Health System, Ann Arbor, Michigan. Drs. Kim and Eagle were supported by a research grant from Pharmacia-Upjohn Pharmaceuticals, Kalamazoo, Michigan. Dr. Kim’s address is: Campus Box 8086, Cardiovascular Division,Washington University in St. Louis, 660 S. Euclid Ave., St. Louis, Missouri 63110. E-mail: mkim@im.wustl.edu. Manuscript received September 14, 2000; revised manuscript received and accepted January 24, 2001. TABLE 1 Clinical Features of the Study Group (n 5 47)