Project Details
Description
Although electronic health records are ubiquitous in U.S. medical practice, using the EHR as a source of data to improve clinical care or identify patients at risk for significant morbidity is still novel, particularly in food allergy. Like many other health conditions, structured data within the EHR are insufficient to identify food-allergy related morbidity and health care use. With the guidance of a team of clinical co-investigators on an existing project (FORWARD), we are developing a clinically informed manual extraction tool to identify physician diagnosed food allergy and related morbidity and health care use. We will use this work as a springboard to develop natural language processing (NLP) tools that can be applied to other clinical populations.
The goal of this project is to build an informatics team to accomplish three aims: 1) Build a data base of electronically extracted EHR data (including narrative notes) of patients with and without physician diagnosed food allergy. Patients will be drawn from general pediatric clinics at Lurie Children’s. Physician diagnosed food allergy will be determined using our clinically informed manual extraction tool developed for the FORWARD study. 2) Develop a NLP system to identify food allergy diagnosis, severity of food allergy reaction, emergency department visits and hospitalizations, and quality of life impacts. 3) Assess the sensitivity, specificity, positive predictive value, and negative predictive value of the NLP by comparing with the determination based on manual chart review data.
Status | Finished |
---|---|
Effective start/end date | 7/9/19 → 1/15/21 |
Funding
- Ann & Robert H. Lurie Children's Hospital of Chicago (Agmt 07/09/19 // Agmt 07/09/19)
- Genentech, Inc (Agmt 07/09/19 // Agmt 07/09/19)
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.