Pneumonia is the leading killer of children worldwide, responsible for one in five pediatric deaths annually, including nearly 2 million deaths each year among children less than 5 years of age. In the United States (US), despite low overall mortality, pneumonia remains the most common indication for hospitalization among children, with a significant proportion of children experiencing severe disease requiring intensive cardiorespiratory interventions and prolonged hospitalization. Despite this enormous burden, substantial knowledge gaps exist regarding precise determinants of disease severity and expected outcomes for children with pneumonia. The result is wide variation in management-including site of care decisions, diagnostic testing, and use of antimicrobials-with downstream effects on individual outcomes and societal costs. Thus, improving our understanding of pneumonia disease severity, including the ability to accurately predict outcomes, is a critical area of unmet need. The research detailed in this proposal responds directly to this need and includes the following scientific aims: 1) Develop and rigorously validate a clinical prediction rue using proportional odds logistic regression that predicts risk for severe outcomes among children hospitalized with community-acquired pneumonia (CAP);2) Incorporate the determination of microbiologic data into the model to assess the influence of pneumonia etiology on performance characteristics of the prediction rule;3) Externally validate the clinical prediction rule among a prospective observational cohort of children presenting to the emergency department with CAP. For the past two and one half years, the candidate has worked closely with his mentor and other investigators to conduct prospective, population-based surveillance for CAP hospitalizations among children in three US cities to determine disease incidence and comprehensive microbiologic etiology. This study includes over 2500 children with CAP and will serve as the exclusive data source for the development and initial validation of the proposed prognostic studies (scientific aims 1 and 2).
Aim 3 will enroll 300 children in the emergency department requiring hospital admission for CAP in a prospective validation study to ensure accurate predictive performance of the developed models in new populations and assess their potential clinical applicability. The overarching objective of this mentored career development experience is for the candidate to emerge as an independent clinical investigator leading a multidisciplinary research program to improve care and outcomes for children with pneumonia. To accomplish this goal, the candidate will augment his prior research training with advanced coursework and practical skills development in predictive modeling, clinical research, implementation science, and leadership training. Throughout the award period, the candidate will work closely with a multidisciplinary team of mentors and advisors-including experts in infectious diseases, biostatistics, hospital medicine, and epidemiology-to carry out his stated career and scientific aims. PROJECT NARRATIVE: This proposal seeks to better understand determinants of disease severity for children with pneumonia, the leading killer of children worldwide and the most common reason for childhood hospitalization in the United States. Knowledge of these factors and their often complex interactions will be used to predict risk for severe outcomes among children hospitalized with pneumonia, and could be applied in clinical settings to improve care and outcomes for this population.
This proposal seeks to better understand determinants of disease severity for children with pneumonia, the leading killer of children worldwide and the most common reason for childhood hospitalization in the United States. Knowledge of these factors and their often complex interactions will be used to predict risk for severe outcomes among children hospitalized with pneumonia, and could be applied in clinical settings to improve care and outcomes for this population.
Showing the most recent 10 out of 39 publications
