Early reports of the prevalence, diagnosis, and outcomes of coronary artery disease (CAD) in women have led to the acceptance of several myths concerning noninvasive diagnostic studies in women. Many of the myths can be explained by age-related differences in prevalence, methodological errors that exclude women from enrollment, worse clinical baseline risk profiles, comorbid diseases at the time of interventions, and smaller coronary vessels. Awareness of these age-related prevalence factors in women and the potential for delaying the onset of CAD by estrogen treatment in postmenopausal women must be considered in assessing the accuracy of diagnostic studies. The critical role of noninvasive testing in women is to diagnose CAD accurately in a population with a lower prevalence of disease prior to the development of more severe clinical manifestations when therapeutic interventions have a high risk and a poor outcome. Exercise radionuclide ventriculography has lower specificity in women, which may be due to gender differences in the response to exercise. Radionuclide perfusion imaging for the detection of CAD is accurate, there is extensive published literature, and it is used widely. High-dose dipyridamole in combination with echocardiography is not widespread in the United States; most pharmacologic echocardiographic studies are performed using dobutamine, and there are no reports on gender differences using dobutamine echocardiography.(ABSTRACT TRUNCATED AT 250 WORDS)