Abstract

BACKGROUND: New tests, such as magnetic resonance imaging (MRI) and electron-beam computed tomography (CT), are being developed for the diagnosis of coronary artery disease.
OBJECTIVE: To determine the conditions that a new test must meet to be a cost-effective alternative to established imaging tests.
DESIGN: Decision model and cost-effectiveness analysis.
DATA SOURCES: Literature review and meta-analysis.
TARGET POPULATION: 55-year-old men and 65-year-old women presenting with chest pain.
TIME HORIZON: Lifetime of the patient.
PERSPECTIVE: Health care policy.
INTERVENTIONS: MRI, electron-beam CT, exercise echocardiography, exercise single-photon emission CT, and coronary angiography.
OUTCOME MEASURES: Target sensitivity and specificity values for a new noninvasive test.
RESULTS OF BASE-CASE ANALYSIS: Assuming that society is willing to pay $75000 per quality-adjusted life-year (QALY) gained, a new test that costs $1000 would need a sensitivity of 94% and a specificity of 90% to be cost-effective.
RESULTS OF SENSITIVITY ANALYSIS: Assuming that society is willing to pay $50000 per QALY gained, a new test that costs $1000 or more would never be cost-effective. For a test that costs $500, the sensitivity and specificity must each be 95%.
CONCLUSIONS: New imaging techniques, such as MRI and electron-beam CT, must be relatively inexpensive and have excellent sensitivity and specificity to be cost-effective compared with other techniques for the diagnosis of coronary artery disease. Similar analyses in other areas of health care may help to focus the development of new diagnostic technology.