AUTHOR=Yin Donghong , Tang Yang , Wang Song , Wang Shuyun , Hou Ruigang , Duan Jinju TITLE=Use of multiple metrics and clustering analysis to assess antimicrobial use in Shanxi hospitals, China: a cross-sectional study based on 25 general hospitals JOURNAL=Frontiers in Public Health VOLUME=Volume 13 - 2025 YEAR=2025 URL=https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2025.1464613 DOI=10.3389/fpubh.2025.1464613 ISSN=2296-2565 ABSTRACT=ObjectiveTo investigate the current patterns of antimicrobial use among nonsurgical inpatients across 25 general hospitals in Shanxi Province and to evaluate the antimicrobial use rate, antimicrobial use density (AUD), days of therapy (DOT), length of therapy (LOT), and the application of cluster analysis in monitoring antimicrobial prescribing practices.MethodsThis study included 25 general hospitals covering 11 cities in Shanxi Province. In total, 2064 hospitalized nonsurgical patients were evaluated for antimicrobial use between December 1, 2022, and January 31, 2023. Data collected included the proportion of antimicrobial prescriptions, antimicrobial use rate, AUD, DOT, and LOT. Statistical analyses were conducted using IBM SPSS version 21.0. Cluster analysis was employed to categorize the 25 hospitals systematically.ResultsAmong the hospitals, the antimicrobial utilization rate ranged from 43.00 to 83.33%. The intensity of antimicrobial use ranged from 40DDDs/ 100pd to 98.99DDDs/100pd. DOT values ranged from 380/1000pd to 713/1000pd, while LOT ranged from 425/1000pd to 1,014/1000pd. The top three antimicrobial classes by AUD were third-generation cephalosporins (15.38 DDDs/100pd), quinolones (13.60 DDDs/100pd), and cephalosporins (11.54 DDDs/100pd). The ICU had the highest antimicrobial use rate and AUD—91.67% and 133.28 DDDs/100pd, respectively —and the longest DOT (1,230/1000 pd). The infection department recorded the highest LOT (988/1000pd). In pediatrics, the AUD and DOT were 53.77DDDs/ 100pd and 1,106/1000pd, respectively. The 25 hospitals were grouped into three distinct clusters via cluster analysis. Statistically significant differences in some antimicrobial indicators were observed among the groups (p < 0.05).ConclusionAcross the 25 hospitals, the rate and intensity of antimicrobial use were relatively high in institutions and departments. During the study period, the use of cefoperazone/sulbactam and fluoroquinolones increased. Concurrently, the combined use of AUD and DOT provided complementary perspectives for evaluating antimicrobial consumption, allowing for a more comprehensive understanding of exposure levels across hospitals and departments. Cluster analysis provides valuable insights for identifying patterns into antimicrobial management and usage.