# Risk Stratification and Decision Support to Improve Care and Outcomes in Children with Pneumonia

> **NIH NIH R01** · VANDERBILT UNIVERSITY MEDICAL CENTER · 2020 · $545,428

## Abstract

PROJECT SUMMARY
Pneumonia is a common reason children present to the emergency department (ED) and one of the top three causes for
pediatric hospitalization in the United States. Pneumonia also accounts for more days of antibiotic therapy in pediatric
hospitals than any other condition. Unfortunately, extensive variation in the delivery of care for children with pneumonia
is evident nationally, including differences in antibiotic selection that are often discordant with national pneumonia
treatment guidelines, and inconsistent site of care decisions not explained by differences in illness severity. National
research priorities for childhood pneumonia emphasize the need for standardized treatment approaches to improve
antibiotic stewardship and the creation of objective risk stratification tools to optimize disposition decisions. Our research
team recently developed a prognostic tool using prospective data collected from more than 2600 children hospitalized
with community-acquired pneumonia that accurately discriminates between children who develop severe in-hospital
outcomes and those who do not. Expanding and recalibrating our prognostic tool for use in the ED, where the disease
burden and spectrum of illness is greatest, is a critical next step in our research. The ED is also an important setting to test
innovative strategies for improving antibiotic prescribing. Few formal antibiotic stewardship programs exist within the
ED, and pediatric studies addressing antibiotic overuse for pneumonia in this environment are limited. In addition to our
demonstrated expertise in pediatric pneumonia, our team also possesses prior success leveraging health information
technology to improve care using clinical decision support that is seamlessly integrated within the electronic health record
(EHR). Use of EHR-based antibiotic and prognostic decision support applications in the ED environment are innovative
approaches to improve care for children with pneumonia and a central focus of the proposed research. Our
multidisciplinary study team, with expertise across the continuum of clinical care as well as in epidemiology, biostatistics,
and biomedical informatics, is uniquely suited to conduct these studies. Within two free-standing children's hospitals and
one general community hospital, we will test the following hypotheses: 1) Our expanded and recalibrated prognostic tool
will accurately discriminate between children with and without high risk for severe outcomes in the ED; 2) EHR-based
antibiotic decision support increases guideline-concordant antibiotic use compared with usual care in the ED; and 3)
Delivery of EHR-based severity information generated by our prognostic tool improves appropriate site of care
disposition in the ED compared to usual care. These hypotheses will be tested within the framework of a prospective,
observational cohort and two pragmatic randomized controlled trials. The overarching goal of this research is to
demonstrate the potential...

## Key facts

- **NIH application ID:** 9854881
- **Project number:** 5R01AI125642-04
- **Recipient organization:** VANDERBILT UNIVERSITY MEDICAL CENTER
- **Principal Investigator:** Derek J. Williams
- **Activity code:** R01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2020
- **Award amount:** $545,428
- **Award type:** 5
- **Project period:** 2017-02-07 → 2022-01-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/9854881

## Citation

> US National Institutes of Health, RePORTER application 9854881, Risk Stratification and Decision Support to Improve Care and Outcomes in Children with Pneumonia (5R01AI125642-04). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/9854881. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*