Current databases on biological variation: pros, cons and progress

Area of Science:

• Clinical Laboratory Medicine
• Analytical Chemistry
• Biostatistics

Background:

• Establishing reliable analytical quality specifications is crucial for clinical laboratory tests.
• Previous efforts lacked a systematic approach correlating biological variation with quality specifications.
• A comprehensive database is needed to support evidence-based quality control in laboratory diagnostics.

Purpose of the Study:

• To develop a database of analytical quality specifications for clinical laboratory tests.
• To utilize biological variation data to derive these specifications.
• To provide a resource for improving the reliability of laboratory test results.

Main Methods:

• Compiled data from published literature (BIOS, CURRENT CONTENTS, EMBASE, MEDLINE) and other sources using "biological variation & laboratory medicine" keywords.
• Correlated descriptive data and observations with biological variation information.
• Calculated within- and between-subject coefficients of variation to derive quality specifications for precision, bias, and total error.

Main Results:

• A database covering 316 quantities was created, reviewing 191 articles.
• Quality specifications for precision, bias, and total error were derived based on biological variation.
• Sex-related stratification was justified for only four quantities; pathological variation sometimes exceeded healthy variation, necessitating stringent specifications.
• Discrepant results and limited data were noted for several quantities, particularly hormones.

Conclusions:

• The developed database provides a robust foundation for analytical quality specifications in laboratory medicine.
• Professionals are encouraged to investigate quantities with discrepant results or limited published data.
• This work highlights the importance of biological variation in setting meaningful quality targets for clinical tests.