Vol.73 No.4 July 2025
Multivariate analysis of long-term sustainability and acceptance of antimicrobial stewardship for bloodstream infections in a Japanese university hospital
1)Division of Clinical Infectious Diseases, Department of Medicine, School of Medicine, Showa Medical University
(Past: Division of Clinical Infectious Diseases, Department of Medicine, School of Medicine, Showa University)
2)Division of Infectious Diseases, Department of Medicine, Showa Medical University Northern Yokohama Hospital, 35-1 Chigasaki Chuo, Tsuzuki-ku, Yokohama, Kanagawa, Japan
(Past: Division of Infectious Diseases, Department of Medicine, Showa University Northern Yokohama Hospital)
3)Division of Infection Control Sciences, Department of Clinical Pharmacy, School of Pharmacy, Showa Medical University
(Past: Division of Infection Control Sciences, Department of Clinical Pharmacy, School of Pharmacy, Showa University)
4)Department of Medicine, Aisenkai Nichinan Hospital
5)Clinical Laboratory, Showa Medical University Hospital
(Past: Clinical Laboratory, Showa University Hospital)
Abstract
As part of antimicrobial stewardship, we have been conducting prospective audit and feedback (PAF) on patients with positive blood cultures to improve the prognosis, with fixed core members and methods since April 2015. Herein, we report on the progress of advice that we have provided over this period, focusing on the advice rates and advice implementation rates.
We analyzed the data of 6,201 cases with initial positive blood cultures seen from April 2015 to March 2024. Trends in the patient background, blood culture-isolated bacteria, causative infection site, medical department, advice rate, and advice implementation rate were evaluated using standard least squares analysis; multivariate analysis was also performed.
Although the number of patients increased slightly, there were no significant changes in the patient background. Multivariate analysis revealed several trends, including a decrease in the number of PAFs to emergency departments (P<0.0001) among medical departments, a decrease in Streptococcus pneumoniae (P=0.0013) among initial blood culture isolates, and an increase in the number of "unknown" (P<0.0001) among sources of infection. When counted by bacterial species, there were biases in the source of infection, and it was thought that identification of the bacterial species would help in guessing the source of the infection. Multivariate analysis adjusted for advice showed a trend toward decreased advice for antibiotic choice (P=0.0116), antibiotic dosage and administration (P=0.0004), diagnosis (P<0.0001), source control (P<0.0001), and others (P=0.0019), although there was no change in the microbiology advice. In terms of the subclass of antibiotics selected, there was a decreasing trend for initiation (P=0.0126) and addition (P=0.0407), but no change for "change" or "discontinuation". In terms of the antibiotic dosage and administration, there was a trend toward a decrease in dosage (P<0.0001) and blood concentration measurement (P=0.0524; not significant), but no change in attempts at simplification, increased use or combined use. Univariate analysis showed that the advice implementation rate increased for "other" (P=0.0354), but there were no other changes.
It seems that as the long-term educational effects of antimicrobial stewardship become apparent, medical professionals are able to use antimicrobials appropriately without advice. However, there are still areas where expert advice is needed, and we believe that a steady continuation of our activities is important.
Key word
antimicrobial stewardship, bacteremia, fungemia, appropriate use, acceptance rate
Received
October 25, 2024
Accepted
February 10, 2025
Jpn. J. Chemother. 73 (4): 309-321, 2025
