AMROrbit Scorecard: A dynamic phase space model for strategic monitoring and actionable insights on Global AMR trajectories

Background and Objectives: Countries need actionable evidence to address the imminent crisis of antimicrobial resistance. Our previous work validated the hypothesis that associations between resistance patterns can uncover early warning signals and proposed a scorecard, AMR scorecard that combined baseline and rate of change of resistance to provide a visual, actionable tool for surveillance in Urinary Tract Infections (UTIs) and Blood Stream Infections (BSIs). A critical gap in the AMR scorecard was the lack of gene mutations. Here we propose to leverage the data from the Pfizer Atlas hosted on the Vivli platform to generate: 1. an actionable, multi-dimensional AMR scorecard for country-level trends of resistance to Critical Antibiotics used for UTIs and BSIs. 2. Indicator antibiotics for effective surveillance for AMR against Critical Antibiotics for UTIs and BSIs using a time-series lead-lag analysis. 3. Intercorrelations and leading associations between UTIs and BSIs presenting in in-patient and out-patient locations. 4. Mining genomic patterns for emerging trends in cross-resistant organisms in UTIs and BSIs. Methods and Statistical Analysis: The surveillance data will be aggregated yearly to generate resistance time series. Univariate and multivariate time series analyses will uncover temporal trends and associations. Lead-lag analyses will identify early indicators of emerging resistance and to uncover associations between in-patient and out-patient locations. Cluster robust trends will be used to develop a multidimensional AMR scorecard with baseline, rate of change overlaid on gene background. Outcomes: Our research aims to enhance AMR surveillance with a digital tool for targeted interventions. Leveraging the Vivli datasets, we will develop 1. a multi-dimensional AMR Scorecard that integrates genomic data with traditional resistance patterns, 2. uncovering indicator antibiotics and providing actionable intelligence for healthcare providers. 3. Understanding the dynamics of resistance spread from in-patient to out-patient and vice versa 4. Genomic surveillance features to tackle emerging resistance.