A novel quantitative evaluation method for quality control results

Background: Quality control (QC) procedures using stable control materials are important for preventing systematic errors (SEs). While the current QC methods assess QC results semi-quantitatively, we designed a novel quantitative QC procedure (QQCP). Methods: QC results were expressed as Z-scores to analyze results quantitatively. The decision values were accumulated up to 30, with three decision values per run, and were compared to rejection criteria at each run. The probability for false rejection (P_fr) and error detection (P_ed) for the QQCP and Westgard multirule methods were estimated using simulated QC data with SEs ranging from 0 to 3 standard deviations (SDs). Results: The P_fr of the QQCP was 3.4% at the 10th run. When 2 QC materials with the same SEs (0.5 SD and 1.0 SD) were used, the P_eds were 36.1% and 95.7% at run 10, respectively. When the SE of each material was greater than 1.5 SDs, the P_ed reached 100% at run 10. The QQCP could detect more than 99% of errors in the 6th, 4th, 3rd, and 2nd runs for 2 QC results with 1.5, 2.0, 2.5, and 3.0 SD SEs, respectively. Conclusion: The QQCP exhibited a P_ed value up to 3.3-fold higher than the Westgard method. Implementation of the QQCP would satisfy the high quality goals derived from biological variations.