Related Experiment Video
Updated: Aug 28, 2025
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
Performance of EHR classifiers for patient eligibility in a clinical trial of precision screening
Nicholas V J Alexander1, Charles A Brunette2, Eric T Guardino3
1C.I. Parhon National Institute of Endocrinology, Bucharest, Romania; Veterans Affairs Boston Healthcare System, Boston, MA, USA.
Insights
Validated computable disease classifiers accurately pre-screened patients for clinical trials using real-world data. These genomic medicine eligibility criteria demonstrated high negative predictive value, improving trial efficiency.
Area of Science:
- Genomic Medicine
- Clinical Trial Methodology
- Health Informatics
Background:
- The Genomic Medicine at VA (GenoVA) Study is a pragmatic trial evaluating polygenic risk score testing.
- The trial enrolls patients without pre-existing diagnoses of six common diseases: atrial fibrillation, coronary artery disease, type 2 diabetes, breast cancer, colorectal cancer, and prostate cancer.
- This study focuses on validating computable disease classifiers as eligibility criteria for clinical trials.
Purpose of the Study:
- To validate computable disease classifiers for use as eligibility criteria in the GenoVA Study.
- To assess the performance of these classifiers in real-world clinical trial enrollment.
- To determine the efficiency and accuracy of computable classifiers in pre-screening trial participants.
Main Methods:
- Published computable classifiers for six diseases were identified and validated in the target population via blinded physician review.
- Classifiers were refined iteratively to achieve true positive and true negative rates of at least 80%.
- Optimized classifiers were implemented as pre-screening exclusion criteria, with real-world negative predictive value (NPV-RW) assessed via telephone screens.
Main Results:
- Classifiers for type 2 diabetes, breast cancer, and prostate cancer met performance standards without modification.
- The atrial fibrillation classifier required two refinement rounds to achieve desired performance.
- Across 1077 screened participants, NPV-RW ranged from 98.4% (coronary artery disease) to 99.9% (colorectal cancer), with no performance differences by gender or race/ethnicity.
Conclusions:
- Computable disease classifiers are efficient and accurate tools for pre-screening in clinical trials.
- The performance of these classifiers is dependent on specific trial objectives and the diseases being studied.
- Validated computable eligibility criteria enhance the feasibility and conduct of clinical trials using real-world data.
Background:
Validated computable eligibility criteria use real-world data and facilitate the conduct of clinical trials. The Genomic Medicine at VA (GenoVA) Study is a pragmatic trial of polygenic risk score testing enrolling patients without known diagnoses of 6 common diseases: atrial fibrillation, coronary artery disease, type 2 diabetes, breast cancer, colorectal cancer, and prostate cancer. We describe the validation of computable disease classifiers as eligibility criteria and their performance in the first 16 months of trial enrollment.
Methods:
We identified well-performing published computable classifiers for the 6 target diseases and validated these in the target population using blinded physician review. If needed, classifiers were refined and then underwent a subsequent round of blinded review until true positive and true negative rates ≥80% were achieved. The optimized classifiers were then implemented as pre-screening exclusion criteria; telephone screens enabled an assessment of their real-world negative predictive value (NPV-RW).
Results:
Published classifiers for type 2 diabetes and breast and prostate cancer achieved desired performance in blinded chart review without modification; the classifier for atrial fibrillation required two rounds of refinement before achieving desired performance. Among the 1077 potential participants screened in the first 16 months of enrollment, NPV-RW of the classifiers ranged from 98.4% for coronary artery disease to 99.9% for colorectal cancer. Performance did not differ by gender or race/ethnicity.
Conclusions:
Computable disease classifiers can serve as efficient and accurate pre-screening classifiers for clinical trials, although performance will depend on the trial objectives and diseases under study.
More Related Videos
Related Concept Videos
01:30Methods of Documentation VII: EMR
01:15Receiver Operating Characteristic Plot
01:16Clinical Trials
There are four phases in a clinical trial. A phase one...
01:11Clinical Trials: Overview
01:13Sensitivity, Specificity, and Predicted Value
Sensitivity is the...
01:12Hazard Ratio
For example, in a clinical trial...