Premium
Repeat patient testing‐based quality control shows promise for use in veterinary biochemistry testing
Author(s) -
Flatland Bente,
Freeman Kathleen P.
Publication year - 2020
Publication title -
veterinary clinical pathology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.537
H-Index - 51
eISSN - 1939-165X
pISSN - 0275-6382
DOI - 10.1111/vcp.12921
Subject(s) - control limits , limit (mathematics) , quality (philosophy) , statistics , limits of agreement , detection limit , control (management) , probability of error , control chart , computer science , biomedical engineering , mathematics , reliability engineering , medicine , nuclear medicine , algorithm , engineering , physics , artificial intelligence , mathematical analysis , process (computing) , quantum mechanics , operating system
Background Repeat patient testing‐based quality control (RPT‐QC) is a form of statistical QC and an alternative to commercial quality control materials (QCM). Objective This study investigated the suitability of canine heparinized plasma for use in RPT‐QC and assessed the predicted performance of RPT‐QC for the detection of analytical error in chemistry testing. Methods The stability of canine plasma for RPT‐QC was investigated via storage at two temperatures for three or six time points. Storage data were analyzed using repeated measures ANOVA and by comparing results for stored specimens to baseline data using predetermined criteria. To generate RPT‐QC limit‐setting and ‐validation data, leftover plasma was prospectively measured. Once control limits were established, these were challenged by measuring specimens for which the repeat aliquot had been manipulated to mimic analytical error. Finally, the predicted performance of RPT‐QC and QCM‐QC with four control rules was investigated using Westgard's EZ Rules 3. Results Refrigerated storage of canine plasma for 7 days allowed mild changes facilitating RPT‐QC. RPT‐QC limits for 12 of 17 common measurands were validated. Validated limits successfully flagged differences from manipulated specimen pairs as “error.” The predicted performance of RPT‐QC for analytical error detection (represented by smallest achievable allowable total error, given a probability of error detection ≥ 85% and a probability of false rejection ≤ 5%) for four common control rules is comparable to that of QCM‐QC. Conclusions This study provides evidence that RPT‐QC using canine heparinized plasma refrigerated for 7 days can be used with simple control rules and low numbers of control materials, suggesting RPT‐QC is applicable to both reference and in‐clinic laboratory settings.