To assess the true association between appropriate prescription of antibiotics and prognosis in patients with sepsis, a key issue in health care and quality improvement strategies.Prospective cohort study in three university hospitals to determine whether the empirical prescription of antibiotics was adequate or inadequate, and to compare hospital death rates and length of stay according to different classifications of antibiotics prescription. Logistic regression models for risk estimation were fitted.A total of 705 patients with severe sepsis were included. No differences were found in positive-culture patients (n = 545) regarding the risk of death with insufficient spectrum antibiotics, compared to patients who received adequate spectrum antibiotics (OR = 0.90; 95% CI = 0.55-1.48). Delay in initiating antibiotics was not associated with the risk of death in patients with adequate spectrum of antibiotics, either with positive (OR = 1.04; 95% CI = 0.99-1.08) or negative cultures (OR = 0.98; 95% CI = 0.92-1.04). There were no differences in the length of hospital stay, according to the antibiotic prescription (median 11 days, IQR = 7-18 days for the whole cohort).No associations were found between inadequate antibiotic prescription or delay to initiate therapy and mortality or length of stay.