IDENTIFYING PREDICTORS OF RESPONSE FOLLOWING DISCONTINUATION OF JAKI THERAPY IN PATIENTS WITH RHEUMATOID ARTHRITIS

Abstract

Objectives: This study aimed to explore the use of semi-automated graphical representations in analysing and understanding infection outbreaks caused by multi-drug resistant microorganism within a hospital setting. Methods: Comprehensive data on inpatient interactions and contacts were collected from hospital information systems (HIS). Code for graph displaying was developed to easily perform representations with raw HIS extractions, without needing further data depuration. Three graph types were employed: (1) a patient-patient contact network diagram to study the interaction between patients, (2) a patient-ward contact network diagram to analyse the patient interaction with different wards, and (3) a timescale graph representing the length of stay within a ward of cases, marking their infection onset and the period of coincidence with their contacts in that ward. All graphics were performed using R language. Results: The adapted graphical representations provided a comprehensive visual overview of the patient contact networks within the hospital. The case-patient contact network diagram helped identify clusters of patients , shedding light on potential routes of infection transmission. The case-ward contact network diagram facilitated understanding of the distribution of cases across different wards and connect clusters and the graphical representation of contact duration within a ward show the variations in contact time among patients. These graphical representations significantly contribute to visualizing infection outbreak dynamics and enable targeted interventions for effective outbreak control. Conclusions: The utilization of graphical representations based on real-world data proved effective in visualizing patient contact networks and analysing infection outbreaks within a hospital setting. The graphical representation provided valuable insights into de dynamics of infection transmission, empowering healthcare professionals to implement timely targeted interventions for effective outbreak control. This data-driven approach using case-patient and case-wards networks, along contact duration visualization, enhances the understanding and management of infection outbreaks caused by multi-drug resistant microorganism within hospital.