Impact of a clinical decision rule on hospital triage of patients with suspected acute cardiac ischemia in the emergency department.

CONTEXT: Emergency department (ED) physicians often are uncertain about where in the hospital to triage patients with suspected acute cardiac ischemia. Many patients are triaged unnecessarily to intensive or intermediate cardiac care units.

OBJECTIVE: To determine whether use of a clinical decision rule improves physicians' hospital triage decisions for patients with suspected acute cardiac ischemia.

DESIGN AND SETTING: Prospective before-after impact analysis conducted at a large, urban, US public hospital.

PARTICIPANTS: Consecutive patients admitted from the ED with suspected acute cardiac ischemia during 2 periods: preintervention group (n = 207 patients enrolled in March 1997) and intervention group (n = 1008 patients enrolled in August-November 1999).

INTERVENTION: An adaptation of a previously validated clinical decision rule was adopted as the standard of care in the ED after a 3-month period of pilot testing and training. The rule predicts major cardiac complications within 72 hours after evaluation in the ED and stratifies patients' risk of major complications into 4 groups--high, moderate, low, and very low--according to electrocardiographic findings and presence or absence of 3 clinical predictors in the ED.

MAIN OUTCOME MEASURES: Safety of physicians' triage decisions, defined as the proportion of patients with major cardiac complications who were admitted to inpatient cardiac care beds (coronary care unit or inpatient telemetry unit); efficiency of decisions, defined as the proportion of patients without major complications who were triaged to an ED observation unit or an unmonitored ward.

RESULTS: By intention-to-treat analysis, efficiency was higher in the intervention group (36%) than the preintervention group (21%) (difference, 15%; 95% confidence interval [CI], 8%-21%; P<.001). Safety was not significantly different (94% in the intervention group vs 89%; difference, 5%; 95% CI, -11% to 39%; P =.57). Subgroup analysis of intervention-group patients showed higher efficiency when physicians actually used the decision rule (38% vs 27%; difference, 11%; 95% CI, 3%-18%; P =.01). Improved efficiency was explained solely by different triage decisions for very low-risk patients. Most surveyed physicians (16/19 [84%]) believed that the decision rule improved patient care.

CONCLUSIONS: Use of the clinical decision rule had a favorable impact on physicians' hospital triage decisions. Efficiency improved without compromising safety.