Poster (Times are in PDT)

Board 05 - Pitfalls of Using Structured Data in a Pediatric De-identified Database

3:30 PM–4:30 PM May 20, 2020 (Conference Time: US - Pacific)

3:30 PM–4:30 PM May 20, 2020


Abstract: A large database of 10,000 neonates was analyzed for accuracy of ICD 9 and ICD 10 data fields. Using structured data fields resulted in a large amount of missing data for gestational age and birth weight. Structured data fields for congenital infections are highly inaccurate and need to be reviewed by a content expert. Including maternal data into these databases may help to improve diagnostic accuracy.

Describe the new knowledge and additional skills the participant will gain after attending your presentation.: 1. Attendees will learn pitfalls to using only structured data fields when extracting de-identified data for neonates
2. Attendees will learn about the benefits of including structured data fields of maternal data when creating a large de-identified database that includes pediatric patients


Lindsey Knake (Presenter)
Vanderbilt University Medical Center

Leigh Anne Tang, Vanderbilit University Medical Center
Allison McCoy, Vanderbilit University Medical Center