Costs of an emergency department-based accelerated diagnostic protocol vs hospitalization in patients with chest pain: a randomized controlled trial

Abstract

Context.-More than 3 million patients are hospitalized yearly in the United States for chest pain. The cost is over $3 billion just for those found to be free of acute disease. New rapid diagnostic tests for acute myocardial infarction (AMI) have resulted in the proliferation of accelerated diagnostic protocols (ADPs) and chest pain observation units. Objective.-To determine whether use of an emergency department (ED)-based ADP can reduce hospital admission rate, total cost, and length of stay (LOS) for patients needing admission for evaluation of chest pain. Design.-Prospective randomized controlled trial comparing admission rate, total cost, and LOS for patients treated using ADP vs inpatient controls. Total costs were determined using empirically measured resource utilization and microcosting techniques. Setting.-A large urban public teaching hospital serving a predominantly African American and Hispanic population. Patients.-A sample of 165 patients was randomly selected from a larger consecutive sample of 429 patients with chest pain concurrently enrolled in an ADP diagnostic cohort trial. Eligible patients presented to the ED with clinical findings suggestive of AMI or acute cardiac ischemia (ACI) but at low risk using a validated predictive algorithm. Main Outcome Measures.-Primary outcomes measured for each subject were LOS and total cost of treatment. Results.-The hospital admission rate for ADP vs control patients was 45.2% vs 100% (P<.001). The mean total cost per patient for ADP vs control patients was $1528 vs $2095 (P<.001). The mean LOS measured in hours for ADP vs control patients was 33.1 hours vs 44.8 hours (P<.01). Conclusions.-In this trial, ADP saved $567 in total hospital costs per patient treated. Use of ED-based ADPs can reduce hospitalization rates, LOS, and total cost for low-risk patients with chest pain needing evaluation for possible AMI or ACl.