A New Practical Tool for Assessing the Analytical Performance and Designing Risk-Based Statistical Quality Control Procedures in Laboratories

Abstract

We developed a new practical tool and applied it to assess the performance of 14 biochemical assays and designed risk-based statistical quality control (SQC) procedures. Two graphs were combined to develop the new tool. Data points of assays were plotted on the tool to determine their sigma performance and the risk-based SQC procedures. The quality goal index (QGI) was also calculated for quality improvement. Among 14 assays, total bilirubin, direct bilirubin, alanine aminotransferase, creatine kinase, and gammaglutamyl transferase achieved 6-sigma performance, the recommended SQC procedure was 13s rule (n = 2) with a run size of 1,000 patient samples. Triglycerides was 5-sigma quality and could be controlled with 13s/22s/R4s multi-rule procedure (n = 2) with a run size of 450. Uric acid, creatinine, total cholesterol, and aspartate aminotransferase obtained 4-sigma quality and could be controlled using 13s/22s/R4s/41s multi-rule procedure (n = 4) with run size of 200. The performance for urea, alkaline phosphatase, amylase, and lactate dehydrogenase was 3-sigma and 13s/22s/R4s/41s/6X multi-rule procedure (n = 6) with run size of 45 was recommended. The QGI for assays with sigma quality below 6.0 were all less than 0.8. The developed tool can be used to simplify laboratory practices in assessing analytical performance and designing SQC procedures.