Antibiotic Consumption and Healthcare-Associated Infection Surveillance in a Multi-Unit Emergency Hospital in Romania: A Retrospective Observational Study

Background and Objectives: Healthcare-associated infections (HAIs) remain a major challenge in emergency hospital settings, where high patient turnover and empirical antibiotic use may contribute to the emergence and spread of multidrug-resistant organisms. Monitoring antibiotic consumption is essential for antimicrobial stewardship and infection prevention. This study evaluated antibiotic consumption patterns across multiple hospital units and explored their ecological relationship with HAI rates. Materials and Methods: A retrospective observational study was conducted in a tertiary-level emergency hospital in Romania between 1 January 2021 and 31 October 2025. Antibiotic consumption was quantified using Defined Daily Dose per 100 bed-days (DDD/100 bed-days) according to World Health Organization (WHO) methodology and categorized using the WHO Access, Watch, and Reserve (AWaRe) classification. HAI data were collected using standardized surveillance definitions. Statistical analyses were primarily descriptive and exploratory and included graphical trend assessment, simple linear regression for temporal trend description, and Spearman correlation analysis for exploratory ecological co-variation assessment. Results: Antibiotic consumption showed substantial variability across hospital units, without a consistent temporal trend over the study period. The Watch group predominated over the Access group from 2023 onward, while Access antibiotics remained below the WHO-recommended 60% threshold. The highest antibiotic consumption was observed in the Medical Wards, followed by Surgical Wards and the Intensive Care Unit. A total of 27 HAIs were identified (0.27 per 1000 patient-days), with the highest incidence observed in the ICU. The most frequent infections were Clostridioides difficile infections (33.3%) and catheter-associated urinary tract infections (29.6%). Exploratory ecological analyses did not identify robust associations between total antibiotic consumption and HAI rates across hospital units. A numerically elevated co-variation was observed between fluoroquinolone consumption and Clostridioides difficile infection incidence; however, this finding should be interpreted strictly as exploratory and hypothesis-generating. Conclusions: Antibiotic use varied across hospital units, with predominance of broad-spectrum agents and suboptimal adherence to WHO AWaRe targets. Reported HAI incidence remained low and should be interpreted within the limitations of routine surveillance systems and potential under-ascertainment. These findings support the value of continuous institutional surveillance of antibiotic use and HAIs while highlighting the limitations of aggregated ecological analyses.