Precision medicine approaches in the diagnosis of community-acquired pneumonia in children

PROJECT SUMMARY/ABSTRACT Candidate: Sriram Ramgopal, MD is a pediatric emergency medicine physician and early career clinical investigator focused on improving the diagnosis of community-acquired pneumonia (CAP) among children in the acute care setting. Dr. Ramgopal’s long-term career goal is to become an independent physician-

investigator focused on improving the health outcomes of children with common infectious emergencies using predictive analytics, informatics, and implementation science approaches. Research Context/Objective: CAP is the second most common reason for hospitalization and the second most costly pediatric condition to treat

among children in the United States. The diagnosis of CAP is complicated by a lack of data driven tools and overlapping presenting features with other common respiratory conditions, resulting in overuse, and occasionally underuse, of chest radiographs (CXR) and antibiotics. There are a paucity of validated tools to

guide the diagnosis of children with suspected CAP. To address this knowledge and practice gap, the objectives of this proposal are to develop a clinical prediction model for CAP using electronic health record (EHR) data, validate a prediction model for CAP, and develop a prototype clinical decision support (CDS)

system for CAP. Specific Aims: 1) Derive and internally validate a prediction model for the diagnosis of CAP using an EHR registry; 2) determine how a prediction model for pediatric CAP will be implemented as a CDS system in routine clinical practice using implementation science and human-centered design methods; and 3)

prospectively evaluate the performance of clinical and EHR-based prediction models for CAP. Research Plan: Dr. Ramgopal will perform a retrospective cohort study using data from a harmonized, high-resolution pediatric emergency department registry to develop a prediction model to identify the risk of CAP in children. He will

lead co-design sessions to develop a CDS system prototype using implementation science approaches, and externally validate three clinical prediction rules for CAP within a pediatric emergency department. Career Development Plan/Environment: Through a combination of local and national didactic activities, experiential

learning, workshops, seminars, professional activities, and coursework, Dr. Ramgopal will acquire expertise in clinical informatics, predictive analytics, and implementation science methodology. Dr. Ramgopal is supported by an experienced, multidisciplinary team of NIH-funded mentors, advisors, and collaborators with expertise in

lower respiratory tract infections, prediction modeling, machine learning, and implementation science. Dr. Ramgopal will meet regularly with his mentors individually, and quarterly as a full committee, to ensure completion of all research and training benchmarks. The institutional commitment to researchers at

Northwestern University and Lurie Children’s Hospital provide an outstanding infrastructure to enable a successful award and ensure Dr. Ramgopal’s development into an independent clinician-scientist.