Blood culture is the gold standard primary test to evaluate patients with sepsis. The need for use of blood culture is increasing, because the incidence of sepsis is rising [1,2]. Several guidelines for optimal blood culture have been proposed including timing and volume of blood collection, number of cultures obtained for each patient, and thoroughness of skin disinfection before collection [3,4]. However, the optimal number of blood cultures per year has not been well documented. Blood culture should be requested adequately, neither overused nor underused [5]. In the Cumitech 1C [4], released in 2005, the number of blood cultures per 1,000 patient days was recommended to be 103-188 in the USA. As the patient population, disease incidence, proportion of acute illnesses, and medical insurance systems in the USA are different from those in Korea, we cannot apply this data directly. The numbers suggested by this documentation need to be evaluated before to be applied to our country. The authors tried to calculate the blood culture numbers per 1,000 patient days, which to our knowledge has not previously been conducted in Korea. The absolute number of blood cultures performed per year could be difficult to compare among hospitals, because the proportion of acute illnesses, hospital size, and turnover rate will be different at each hospital. Using 1,000 patient days could be the denominator to compensate for these variables. The authors compared the number of blood cultures by 1,000 patient days for seven university-affiliated hospitals in 2010. The average stay in the hospital ranged from 6.9 to 11.3 days.