Related Experiment Video
Updated: May 5, 2026

Human Liver Microphysiological System for Assessing Drug-Induced Liver Toxicity In Vitro
Published on: January 31, 2022
Evaluation considerations for EHR-based phenotyping algorithms: A case study for drug-induced liver injury
Casey Lynnette Overby1, Chunhua Weng, Krystl Haerian
1Department of Biomedical Informatics, Columbia University, New York, NY.
This study introduces a new framework for evaluating electronic health record (EHR) phenotyping algorithms, combining measurement and demonstration studies to improve accuracy in drug-induced liver injury (DILI) research.
Area of Science:
- Biomedical Informatics
- Clinical Research Methodology
- Pharmacogenomics
Background:
- Electronic health record (EHR) phenotyping algorithms are crucial for clinical research.
- Current evaluation methods often lack sufficient emphasis on measurement precision and accuracy.
- Refinements to algorithms are frequently needed during development.
Purpose of the Study:
- To develop and evaluate a novel framework for EHR phenotyping algorithm assessment.
- To incorporate both measurement and demonstration studies for comprehensive evaluation.
- To assess a drug-induced liver injury (DILI) phenotyping algorithm within the eMERGE network.
Main Methods:
- Developed an evaluation framework integrating measurement and demonstration studies.
- Conducted a measurement study assessing qualitative perceptions and quantitative inter-rater reliability.
- Performed a demonstration study evaluating qualitative appropriateness and quantitative positive predictive value.
Main Results:
- The measurement study led to framework modifications, including laboratory value visualization and enhanced clinical note review.
- Demonstration study results informed algorithm refinements, such as excluding overdose patients for genetic susceptibility studies.
- The integrated framework improved the evaluation process for EHR phenotyping algorithms.
Conclusions:
- A combined measurement and demonstration study approach enhances EHR phenotyping algorithm evaluation.
- This framework provides a more robust assessment of algorithm precision and accuracy.
- The refined DILI algorithm is better suited for genetic susceptibility research within the eMERGE network.
Related Concept Videos
Effect of Hepatic Disease on Pharmacokinetics: Pathophysiologic Assessment and Liver Function Test
Effect of Hepatic Disease on Pharmacokinetics: Drug Dosing and Hepatic Blood Flow
Drug toxicity: Idiosyncratic Reactions
Pharmacogenetic Phenotypes: Alterations in Pharmacokinetics, Drug Targets and Biologic Milieu
Effect of Hepatic Disease on Pharmacokinetics: Dose Adjustments Due to Hepatic Impairment
Pharmacogenetics of Drug Metabolism: Overview

