Childhood Cancer Data Initiative Annual Symposium (Abstract Registration): Submission #48

Abstract Submission for Poster Presentation
-------------------------------------------
Abstract Title:: Leveraging the ExtractEHR+ Toolkit to Enhance Medication Data in the Children’s Brain Tumor Network and Childhood Cancer Data Initiative
Abstract::
Background:
The National Cancer Institute (NCI) Childhood Cancer Data Initiative (CCDI) has enabled the distribution of previously unshared data. A critical component to maximize the utility of these data is to have longitudinal clinical data, such as treatment data. Manual abstraction has typically been used to capture treatment data, but this process is time consuming and subject to human error. This study aimed to use automated extraction and processing of electronic health record (EHR) data using the ExtractEHR+ Toolkit to describe chemotherapy exposures for children enrolled in the Children’s Brain Tumor Network (CBTN) who have contributed data to the CCDI.

Methods:
Patients enrolled in CBTN at two hospitals were included. ExtractEHR extracted all medication orders and administrations, including outpatient prescriptions, heights, and weights. Medication order and administration data were merged and MedCleanEHR centrally cleaned these data to identify unique chemotherapy exposures and dose amounts. Height and weight data were used to calculate body surface area and merged with medication data. For prescriptions where a discrete dose field was not included, regular expressions were used to identify doses in free text fields. MedCleanEHR calculated cumulative chemotherapy doses each patient received. 

Results:
The cohort included 1628 patients. ExtractEHR successfully pulled 3,849,099 medication administrations and 839,736 medication orders. Once cleaned, there were 98,877 unique chemotherapy administrations and 8,781 chemotherapy prescriptions.

Conclusions:
The ExtractEHR+ Toolkit can efficiently ascertains chemotherapy exposures and dosing for patients in the CBTN cohort. These data will be transferred to the CCDI to enhance clinical data in the CCDI ecosystem.

Abstract:: {Empty}
Authors::
1. First Name: Tamara
   Middle Initial: P
   Last Name: Miller
   Degree(s): MD, MSCE
   Organization: Emory University/Children's Healthcare of Atlanta
2. First Name: Gregory
   Last Name: Joseph
   Degree(s): MS
   Organization: Emory University
3. First Name: Edward
   Last Name: Krause
   Degree(s): MS
   Organization: Children's Hospital of Philadelphia
4. First Name: Yun Gun
   Last Name: Jo
   Organization: Children's Hospital of Philadelphia
5. First Name: Judy
   Last Name: Lee
   Organization: Children's Healthcare of Atlanta
6. First Name: Hannah
   Last Name: Calkins
   Degree(s): MSIS
   Organization: Children's Hospital of Philadelphia
7. First Name: Adam
   Middle Initial: C
   Last Name: Resnick
   Degree(s): PhD
   Organization: Children's Hospital of Philadelphia
8. First Name: Richard
   Last Name: Aplenc
   Degree(s): MD, PhD
   Organization: Children's Hospital of Philadelphia
9. First Name: Allison
   Last Name: Heath
   Degree(s): PhD
   Organization: Children's Hospital of Philadelphia

Presenting Author:: Tamara P. Miller, MD, MSCE
Institution:: Emory University/Children's Healthcare of Atlanta
Email Address:: tamara.miller@emory.edu