Many patients with acute heart failure are initially managed in emergency departments (EDs) worldwide. Although some require hospitalization for further management, it is likely that a sizeable proportion could be safely discharged either directly from the ED or after a more extended period of management in an observation-type unit. Identification of low-risk patients who are safe for such an approach to management continues to be a global unmet need. This is driven in part by a lack of clarity on postdischarge outcomes for lower risk patients and a nonexistent consensus on what may be acceptable event rates. The current paper reviews previous studies carried out on patients directly discharged from the ED, suggests a general disposition algorithm and focuses on discharge metrics, which are based on both evidence and expert opinion. In addition, we propose that the following variables be considered for future determination of acceptable event rates: (a) baseline characteristics and risk status of the patient; (b) access to follow-up; (c) ED capability to provide an extended period of observation before discharge; (d) the temporal relationship between the event and ED discharge decision; and (e) the type of event experienced.