AMR VIVLI Multi-Region Data Challenge Project

Expression of Interest Summary: Leveraging ATLAS Data to Address AMR in S.Pneumoniae and S.Aureus Our international research team, comprising working professionals from Singapore, India, Croatia and Ghana, proposes a targeted study utilizing the ATLAS Program dataset. We aim to analyze AMR patterns in two priority pathogens: Streptococcus pneumoniae (Medium Priority) and Staphylococcus aureus (High Priority), as identified in the 2024 WHO Bacterial Priority Pathogens List. Primary Aim: To identify, characterize, and predict AMR patterns in S. pneumoniae and S. aureus across different geographical and economic contexts, with a focus on lower respiratory tract infections (LRTI) and bacterial pneumonia. This research aims to improve public health and clinical practice in Southeast Asia, South Asia, Europe, and Africa. Rationale: 1. S. pneumoniae and S. aureus are significant contributors to LRTI and bacterial pneumonia globally. 2. These pathogens are of particular concern in our regions of focus, presenting unique challenges in AMR. 3. The ATLAS dataset's extensive coverage allows for robust analysis of these pathogens across diverse settings. Methodology: 1. Systematic Literature Review: - Conduct a comprehensive review of AMR studies on S. pneumoniae and S. aureus from 2004-2024, aligning with the ATLAS dataset timeframe. - Focus on regional AMR trends, treatment outcomes, and intervention strategies specific to LRTI. 2. Comparative Analysis: - Compare ATLAS data on S. pneumoniae and S. aureus with other global AMR surveillance programs. - Analyze AMR patterns in relation to antibiotic consumption data, focusing on antibiotics commonly used for LRTI. 3. Advanced Data Analytics: a) Descriptive Statistics: - Summarize AMR rates for S. pneumoniae and S. aureus across regions and over time. - Develop interactive dashboards using R Shiny to visualize AMR trends in LRTI. b) Inferential Statistics: - Apply time series analysis to assess AMR trends in these pathogens. - Implement multilevel mixed-effects models to account for regional variations. c) Machine Learning Approaches: - Develop random forest models to identify key predictors of AMR in S. pneumoniae and S. aureus. - Implement neural network models for AMR pattern recognition and prediction in LRTI. 4. Data Integration: Merge ATLAS data with open-access sources to enhance analysis: - World Bank Open Data: Socioeconomic indicators related to respiratory health - WHO Global Health Observatory: LRTI burden and treatment access metrics - WorldClim: Climate data to explore environmental factors influencing these pathogens 5. Statistical and Clinical Significance: - Calculate population attributable fraction (PAF) for AMR in S. pneumoniae and S. aureus related to LRTI mortality. - Assess the impact of AMR on treatment outcomes and healthcare costs associated with LRTI. Expected Outcomes: Our focused approach will lead to improving public health and clinical practice by: 1. Strengthening stewardship: Identifying region-specific AMR patterns in S. pneumoniae and S. aureus to inform targeted antibiotic use policies for LRTI. 2. Improving patient outcomes: Developing evidence-based guidelines for empiric therapy in LRTI, tailored to local epidemiology of these priority pathogens. 3. Strengthening health systems: Providing data-driven insights to optimize resource allocation for AMR control in LRTI, considering the specific challenges posed by S. pneumoniae and S. aureus. By leveraging the comprehensive ATLAS dataset and focusing on these WHO BPPL priority pathogens, we aim to contribute nuanced, actionable insights to combat AMR in LRTI. Our diverse team's expertise ensures relevance to varied healthcare landscapes, particularly in Southeast Asia, South Asia, Europe, and Africa, where these pathogens pose significant public health challenges.