Investigating the Variable Component of the Systematic Error, a Neglected Error Parameter: Theoretical Reevaluation...

Area of Science:

• Metrology
• Analytical Chemistry
• Quality Control

Background:

• The variable component of systematic error (VCSE) is recognized but lacks formal definition and is often misclassified as random error.
• Existing metrology frameworks do not adequately address the time-varying nature of systematic error.

Purpose of the Study:

• To re-evaluate the significance and role of the VCSE in modern quality control (QC) protocols.
• To propose a new error model that accounts for time-varying systematic errors.

Main Methods:

• Theoretical analysis based on three core principles of measurement and calibration.
• Computer simulations to model bias drift from reagent instability and human interventions.
• Validation using real-world QC data from Roche reagents on Cobas analyzers.

Main Results:

• Bias drift, stemming from reagent instability and human interventions, is identified as the primary source of VCSE.
• Computer simulations and real-world data confirm the causes and cyclic variations of VCSE.
• The VCSE is distinct from random error due to its predictability and corrigibility in the short term.

Conclusions:

• A new error model is proposed, distinguishing between constant and variable systematic errors.
• Current QC practices, including Westgard rules, may be based on flawed assumptions regarding error components.
• The proposed model offers a foundation for a more accurate QC system, addressing distinct bias scenarios post-calibration.