CARDIOLOGY AND VASCULAR/ORIGINAL RESEARCH The Risk of Missed Diagnosis of Acute Myocardial Infarction Associated With Emergency Department Volume Michael J. Schull, MD, MSc Marian J. Vermeulen, MHSc Therese A. Stukel, PhD From the Institute for Clinical Evaluative Sciences (Schull, Vermeulen, Stukel), the Clinical Epidemiology Unit, Sunnybrook Health Sciences Centre (Schull, Stukel), the Department of Emergency Services, Sunnybrook Health Sciences Centre (Schull), and the Department of Medicine (Schull), and the Department of Health Policy, Management and Evaluation (Schull, Vermeulen, Stukel), University of Toronto, Toronto, Ontario, Canada. Study objective: Missed diagnosis of acute myocardial infarction is associated with adverse clinical outcomes and more dollars recovered in malpractice suits than any other condition. The rate of missed diagnosis varies between emergency departments (EDs); we hypothesized that it is associated with the volume of acute myocardial infarction patients treated in an ED and that the association can be explained by other hospital characteristics. Methods: We linked the records of all acute myocardial infarction patients admitted to an Ontario hospital in 2002 to 2003 to their ED visit records in the 7 days preceding admission. Acute myocardial infarctions were defined as missed if the diagnosis on the previous visit matched a list of cardiac symptoms and illnesses. We assessed whether annual volume of admitted acute myocardial infarction patients treated in the ED (grouped as 0 to 49; 50 to 99; 100 to 199; 200 to 299; and 300) was associated with missed acute myocardial infarction, adjusting for age, sex, teaching hospital status, and acute myocardial infarction severity. In a secondary analysis, we used data from a survey of Ontario EDs to assess whether hospital characteristics (triage practices, use of diagnostic tests, and consultant availability) explained the volume association. Results: Of 19,663 acute myocardial infarction patients, mean age (68.3 years), sex (63% men), and predicted 1-year mortality (mean 0.21; SD 0.18) were similar across volume groups. The rate of missed acute myocardial infarction was 2.1% (95% confidence interval [CI] 1.9% to 2.3%) and varied from 0% to 29% across EDs. Compared with very high-volume EDs, the adjusted odds ratio of missed acute myocardial infarction was 2.0 in very low- (95% CI 1.5 to 2.7) and 1.6 in low- (95% CI 1.1 to 2.3) volume EDs. Consultant availability partially explained the volume effect. Conclusion: Lower-volume EDs have up to 2-fold higher odds of missed acute myocardial infarctions compared with highest-volume ones after controlling for patient factors. Many current technologies designed to increase diagnostic sensitivity are feasible only in higher-volume centers. Efforts to reduce overall rates of missed acute myocardial infarctions should instead focus on simpler solutions appropriate for lower-volume EDs, such as telemedicine to improve access to consultant expertise. [Ann Emerg Med. 2006;48:647-655.] 0196-0644/$-see front matter Copyright © 2006 by the American College of Emergency Physicians. doi:10.1016/j.annemergmed.2006.03.025 SEE EDITORIAL, P. 657. INTRODUCTION Background Patients with symptoms suggestive of acute myocardial infarction should promptly seek medical evaluation, yet the diagnosis is missed in about 2% to 3% of acute myocardial infarction patients presenting to emergency departments (EDs) in the United States 1 and Canada 2 and about 6% of patients in the United Kingdom. 3 Failure to accurately diagnose an acute myocardial infarction leads to delays in the initiation of appropriate treatments, may increase mortality, 1,3,4 and is responsible for more dollars recovered in malpractice suits than any other condition. 3,5–7 Importance Studies to identify predictors of missed acute myocardial infarction have focused on patient-level factors. 1,8 However, predictors that have been identified, such as nonwhite race and a normal ECG, 1 are of limited clinical utility and lack specificity. Interventions designed to reduce the risk, such as specialized Volume , .  : December  Annals of Emergency Medicine 647