Considerable controversy exists regarding whether women are less likely than men to be referred to coronary angiography after an abnormal noninvasive test. This prospective cohort study analyzed consecutive subjects (2,351 men and 1,318 women) with no prior history of invasive cardiac procedures who were referred for treadmill thallium testing at the Cleveland Clinic Foundation. The primary end point was performance of coronary angiography within 90 days of treadmill thallium testing. A secondary end point was all-cause mortality during 1.8 years of follow-up. Women were less likely than men to undergo coronary angiography (6% vs 14%, odds ratio [OR] 0.42, 95% confidence interval [Cl] 0.33 to 0.54, p < 0.001), but were also less likely to have an abnormal thallium scan (8% vs 29%, p < 0.001). In logistic regression analyses with adjustment for thallium result and age, women were as likely as men to be referred for coronary angiography (adjusted OR 1.00, 95% Cl 0.75 to 1.34, p > 0.9). Women were less likely to have severe coronary disease on angiography (15% vs 30%, p = 0.006). During 1.8 years of follow-up there were 26 deaths (2%) among women and 84 deaths (4%) among men. After adjusting for age, thallium abnormalities, and clinical characteristics in Cox regression analyses, women had a lower mortality rate than men (relative risk 0.58, 95% Cl 0.36 to 0.94, p = 0.03). Thus, gender-related differences in referral for coronary angiography after treadmill thallium testing can be explained by a higher rate of abnormal tests in men. No evidence of a post-test gender bias was detected, but a pretest bias affecting referral to nuclear testing cannot be excluded. Furthermore, women have a lower prevalence of severe coronary disease and a lower adjusted mortality rate.