Fungal bloodstream infections are associated with significant patient mortality and health care costs. Nevertheless, the relationship between a delay of the initial empiric antifungal treatment until blood culture results are known and the clinical outcome is not well established. A retrospective cohort analysis with automated patient medical records and the pharmacy database at Barnes-Jewish Hospital was conducted. One hundred fifty-seven patients with a Candida bloodstream infection were identified over a 4-year period (January 2001 through December 2004). Fifty (31.8%) patients died during hospitalization. One hundred thirty-four patients had empiric antifungal treatment begun after the results of fungal cultures were known. From the time that the first blood sample for culture that was positive was drawn, 9 (5.7%) patients received antifungal treatment within 12 h, 10 (6.4%) patients received antifungal treatment between 12 and 24 h, 86 (54.8%) patients received antifungal treatment between 24 and 48 h, and 52 (33.1%) patients received antifungal treatment after 48 h. Multiple logistic regression analysis identified Acute Physiology and Chronic Health Evaluation II scores (one-point increments) (adjusted odds ratio [AOR], 1.24; 95% confidence interval [CI], 1.18 to 1.31; P < 0.001), prior antibiotic treatment (AOR, 4.05; 95% CI, 2.14 to 7.65; P = 0.028), and administration of antifungal treatment 12 h after having the first positive blood sample for culture (AOR, 2.09; 95% CI, 1.53 to 2.84; P = 0.018) as independent determinants of hospital mortality. Administration of empiric antifungal treatment 12 h after a positive blood sample for culture is drawn is common among patients with Candida bloodstream infections and is associated with greater hospital mortality. Delayed treatment of Candida bloodstream infections could be minimized by the development of more rapid diagnostic techniques for the identification of Candida bloodstream infections. Alternatively, increased use of empiric antifungal treatment in selected patients at high risk for fungal bloodstream infection could also reduce delays in treatment.