Bacteremia is a serious infectious disease, and there is a strong relationship between delayed initiation of effective antimicrobials to treat bacteremia and in-hospital mortality [1]. Blood culture is essential for the rapid detection of causative organisms and the selection of antimicrobial agents, and quality assurance is important to ensure that blood culture samples are properly collected [2]. Several target values are provided in Cumitech 1C: Blood Cultures IV (the Cumitech guideline), which describes the number of blood cultures per 1000 patient-days and the contamination rate as parameters for quality assurance [3]. The German Mikrobiologisch-Infektiologische Qualitätsstandards (MIQ) guideline is 100–200 blood culture sets per 1000 patient-days as a standard, which was established with reference to the United States guideline [2]. However, there are no target values in Japan. A pilot study of blood culture surveillance conducted by Ohmagari et al. [4] in 2012 revealed 25.2 sets of samples per 1000 patient-days and 359.6 sets of samples per 1000 admissions. The number of blood cultures per 1000 patient-days was less than that recommended by the Cumitech guideline (103–188) [3]. Because blood culture monitoring is affected by several factors unique to each country, such as the medical system, type of medical insurance, and typical hospital stay length, original benchmarks that reflect the medical environment in Japan should be established [4,5]. The Japan Infection Prevention and Control Conference for National and Public University Hospitals (JIPCCNPUH) has conducted yearly blood culture surveillance since 2015 to verify the quality assurance of blood culture. In this study, we examined trends in the blood culture quality assurance data from 6 years of JIPCCNPUH surveillance.