Identifying the Clinical Laboratory Tests from Unspecified "Other Lab Test" Data for Secondary Use

Area of Science:

• Biomedical Informatics
• Clinical Data Management
• Health Data Standardization

Background:

• Electronic health records (EHRs) store clinical laboratory results as structured data.
• Laboratory tests performed at outside laboratories are often ambiguously labeled in EHRs, hindering data retrieval.
• Aggregated clinical data repositories face challenges with these ambiguous 'outside test' labels.

Purpose of the Study:

• To develop a general multi-step solution for identifying, standardizing, reconciling, and transforming ambiguous laboratory test results.
• To improve the retrieval and usability of specific laboratory test results from outside laboratories within EHR data.
• To create a reference table for the EHR data extract-transform-load (ETL) process.

Main Methods:

• Applied a multi-step solution to data within the NIH Biomedical Translational Research Information System (BTRIS).
• Developed methods to identify laboratory tests and map free-text comment values to standard codes (LOINC).
• Created a reference table to facilitate the ETL process for EHR data.

Main Results:

• Successfully identified and standardized ambiguous laboratory test results from outside laboratories.
• Mapped free-text laboratory test information to LOINC codes for improved data consistency.
• Developed a reference table to aid in the ETL process for clinical data repositories.

Conclusions:

• The presented solution effectively addresses the challenge of ambiguous laboratory test labels in EHRs.
• Standardization of outside laboratory test data enhances data quality and facilitates research.
• The developed methods and reference table are valuable for improving clinical data management and research data accessibility.