Noninvasive testing for coronary artery disease (CAD) is warranted for symptomatic patients with intermediate pretest likelihood of CAD. Accomplishing testing in an emergency department (ED) environment is challenging. We compared two strategies of CAD testing in ED patients: immediate computed tomography coronary angiography (CTCA) versus delayed outpatient stress testing. We conducted a historical control cohort study comparing symptomatic ED patients without an acute coronary syndrome who warranted noninvasive CAD testing. Two cohorts (50 patients each) were defined by CAD testing strategy, immediate CTCA versus delayed stress testing. Outcomes were duration of ED stay, detection of CAD, and 3-month rates of readmission, myocardial infarction, (MI) or death. Median duration of stay was 417.5 minutes (interquartile range [IQR] 359.0-581.0) in the CT cohort and 400.0 minutes (IQR 338.0-471.0) in the control cohort (P = 0.53). CAD was detected in 14 CT cohort patients versus 1 in control (P = 0.0004), due to low follow-up in the control cohort (18 of 50, 36%). Obstructive CAD was diagnosed in 6 CT cohort patients versus 1 in control (P = 0.11). During 3 months of follow-up, four patients in each cohort were reevaluated in the ED for chest pain; no patients suffered MI or death. A strategy of immediate CTCA is superior to a delayed stress testing strategy for detecting CAD in ED patients with chest pain and prompting appropriate referrals for further management. Delayed stress testing was primarily ineffective due to low follow-up. Immediate CTCA can be used safely without altering the ED duration of stay.