Modeling, analysis and optimization of calibration uncertainty in clinical laboratories

• We build a mathematical model of the serum triglycerides clinical laboratory measurement process.
• We estimate the uncertainty associated with this clinical assay using the Monte Carlo method.
• We estimate the effect of uncertainty in calibration on the measurement result.
• We estimate the effect of correlation between parameters of the calibration function on the measurement result.
• We determine optimal combinations of calibrator concentrations so as to minimize uncertainty at medical decision points.

Uncertainty in the calibration of a clinical laboratory measurement process has a significant effect on the uncertainty of the measurement result. We develop a mathematical model of the analytical stage of the measurement of serum triglyceride concentration in the clinical laboratory, and use the Monte Carlo method to estimate the net uncertainty associated with this model. We then use the model to study the effect of instrument calibration on the uncertainty of the laboratory measurement result. The effect of the correlation between the parameters of the linear calibration function on the measurement result is quantified using the model. In addition, the effect of the choice of calibrator concentration levels on the measurement result distribution is studied using the model, by studying the effect of the value or the position of the calibrator concentration, and the difference or the distance between calibrator concentrations, on the uncertainty of the measurement result.