Risk Analysis for Quality Control Part 2: Theoretical Foundations for Risk Analysis

Area of Science:

• Clinical diagnostics
• Laboratory medicine
• Risk management

Background:

• Traditional risk analysis for quality control (QC), like the Parvin model, has limitations.
• The Parvin model can yield paradoxical results and underestimate risks in QC settings.

Purpose of the Study:

• To develop an improved framework for risk analysis in laboratory quality control.
• To create a dynamic model that accurately assesses long-term assay behavior under QC monitoring.

Main Methods:

• Developed a dynamic Markov Reward Model to simulate assay behavior over time.
• The model incorporates assay shift frequency, shift distribution, and error impact on patient outcomes.
• Analyzed competing risks, including false detections and mean shifts, leading to out-of-control states.

Main Results:

• The model generates a tradeoff curve balancing patient risk reduction against laboratory costs (false-positive QC).
• It quantifies undetected reported errors and false-positive laboratory results based on QC settings.
• Provides a method to optimize specific QC strategies or compare different QC approaches.

Conclusions:

• A novel method was developed to evaluate the cost of reducing patient risk from unacceptable errors.
• This framework quantifies the trade-off between patient safety and laboratory operational costs.
• The dynamic model offers a more robust approach to risk analysis in quality control settings.