Atrial fibrillation (AF) has been established as an independent predictor of long-term mortality after acute myocardial infarction. However, this is less well defined across the whole spectrum of acute coronary syndromes (ACSs). The Acute Coronary Syndrome Prospective Audit is a prospective multicenter registry with 12-month outcome data for 3,393 patients (755 with ST-segment elevation myocardial infarction, 1942 with high-risk non-ST-segment elevation ACS [NSTE-ACS], and 696 with intermediate-risk NSTE-ACS). A total of 149 patients (4.4%) had new-onset AF and 387 (11.4%) had previous AF. New-onset AF was more, and previous AF was less frequent in those with ST-segment elevation myocardial infarction than in those with high-risk NSTE-ACS or intermediate-risk NSTE-ACS (p <0.001). Compared to patients without arrhythmia, patients with new-onset AF and previous AF were significantly older and had more high-risk features at presentation (p <0.004). Patients with new-onset AF more often had left main coronary artery disease, resulting in a greater rate of surgical revascularization (p <0.001). Only new-onset AF resulted in adverse in-hospital outcomes (p <0.001). Only patients with previous AF had greater long-term mortality (hazard ratio 1.42, p <0.05). New-onset AF was only associated with a worse long-term composite outcome (hazard ratio 1.66, p = 0.004). However, the odds ratio for the composite outcome was greatest for patients with new-onset AF with intermediate-risk NSTE-ACS (odds ratio 3.9, p = 0.02) than for those with high-risk NSTE-ACS (odds ratio 2.0, p = 0.01) or ST-segment elevation myocardial infarction (odds ratio 1.4, p = 0.4). In conclusion, new-onset AF was associated with worse short-term outcomes and previous AF was associated with greater mortality even at long-term follow-up. The prognostic burden of new-onset AF differed with the type of ACS presentation.