Acute chest pain is a high volume complaint that creates risk for patients and emergency providers. Approximately 8 million people in the United States visit emergency departments (EDs) each year for chest pain and 1.2 million are diagnosed with acute myocardial infarction (AMI).1 Two-third of patients presenting to the ED with chest pain are admitted, and of these 15% are ultimately diagnosed with acute coronary syndrome (ACS).2-4
However, approximately 2% of US patients with AMI and 2.3% of patients with unstable angina are initially misdiagnosed.5 The mortality rate for missed ACS is 10% to 50%.6-8 Rapidly identifying patients with AMI and risk stratifying remaining chest pain patients can be difficult because initial diagnostic studies are not highly accurate. Initial electrocardiography (ECG) diagnoses only 40% to 65% of patients with AMI and is even less useful in unstable angina.9 Initial serum markers detect up to 66% of patients with AMI.10
Accurately identifying patients with ACS involves more than 1 ECG and 1 set of cardiac enzymes. In 1997, Graff et al11 reported an inverse linear relationship between admission rate of chest pain patients and MI miss rate (Figure 42-1). Simply stated, the higher the discharge rate for patients with chest pain, the higher the risk of missing AMI. Recent studies have evaluated the usefulness of chest pain units (CPUs) in streamlining the diagnosis of ACS while minimizing risk.
Inverse association between admission rate and missed MI rate (Graff et al).
Since 1981, ED CPUs have facilitated rapid, cost-effective evaluation of patients with chest pain.12-14 Accordingly, recent American Heart Association/American College of Cardiology guidelines emphasize the effectiveness of the CPU in evaluating patients with acute chest pain.15 However, segmenting chest pain patients is not necessarily the correct approach for all EDs, depending upon local circumstances. All chest pain patients, however, must receive consistent, evidence-based care.
Is Patient Segmentation the Correct Financial Decision?
The financial decision to develop a CPU depends on (chest pain) admission rate and cost per patient. Since CPUs save cost by reducing admissions, the potential for CPUs to be cost effective is related to the proportion of chest pain patients who are admitted. EDs that admit fewer than 35% of their chest pain patients probably should not develop CPUs because this strategy is less likely to reduce cost.16 ED admission rates for adults with chest pain vary from 20% to 80%.17
The direct cost of CPU care reduces cost savings.18 High cost per patient mitigates the financial benefit of a CPU. Direct cost per patient may be driven by local factors (eg, availability of appropriately trained staff, short stay facilities, chemical pathology services, and exercise treadmill facilities). Unfortunately, an absolute cost threshold is difficult to discern ...