Patient-based quality control for glucometers: using the moving sum of positive patient results and moving average
View/ Open
File version
Version of Record (VoR)
Author(s)
Lim, Chun Yee
Badrick, Tony
Loh, Tze Ping
Griffith University Author(s)
Year published
2020
Metadata
Show full item recordAbstract
Introduction: The capability of glucometer internal quality control (QC) in detecting varying magnitude of systematic error (bias), and the potential use of moving sum of positive results (MovSum) and moving average (MA) techniques as potential alternatives were evaluated. Materials and methods: The probability of error detection using routine QC and manufacturer’s control limits were investigated using historical data. Moving sum of positive results and MA algorithms were developed and optimized before being evaluated through numerical simulation for false positive rate and probability of error detection. Results: When the ...
View more >Introduction: The capability of glucometer internal quality control (QC) in detecting varying magnitude of systematic error (bias), and the potential use of moving sum of positive results (MovSum) and moving average (MA) techniques as potential alternatives were evaluated. Materials and methods: The probability of error detection using routine QC and manufacturer’s control limits were investigated using historical data. Moving sum of positive results and MA algorithms were developed and optimized before being evaluated through numerical simulation for false positive rate and probability of error detection. Results: When the manufacturer’s default control limits (that are multiple times higher than the running standard deviation (SD) of the glucome-ter) was used, they had 0-75% probability of detecting small errors up to 0.8 mmol/L. However, the error detection capability improved to 20-100% when the running SD of the glucometer was used. At a binarization threshold of 6.2 mmol/L and block sizes of 200 to 400, MovSum has a 100% probability of detecting a bias that is greater than 0.5 mmol/L. Compared to MovSum, the MA technique had lower probability of bias detection, especially for smaller bias magnitudes; MA also had higher false positive rates. Conclusions: The MovSum technique is suited for detecting small, but clinically significant biases. Point of care QC should follow conventional practice by setting the control limits according to the running mean and SD to allow proper error detection. The glucometer manufacturers have an active role to play in liberalizing QC settings and also enhancing the middleware to facility patient-based QC practices.
View less >
View more >Introduction: The capability of glucometer internal quality control (QC) in detecting varying magnitude of systematic error (bias), and the potential use of moving sum of positive results (MovSum) and moving average (MA) techniques as potential alternatives were evaluated. Materials and methods: The probability of error detection using routine QC and manufacturer’s control limits were investigated using historical data. Moving sum of positive results and MA algorithms were developed and optimized before being evaluated through numerical simulation for false positive rate and probability of error detection. Results: When the manufacturer’s default control limits (that are multiple times higher than the running standard deviation (SD) of the glucome-ter) was used, they had 0-75% probability of detecting small errors up to 0.8 mmol/L. However, the error detection capability improved to 20-100% when the running SD of the glucometer was used. At a binarization threshold of 6.2 mmol/L and block sizes of 200 to 400, MovSum has a 100% probability of detecting a bias that is greater than 0.5 mmol/L. Compared to MovSum, the MA technique had lower probability of bias detection, especially for smaller bias magnitudes; MA also had higher false positive rates. Conclusions: The MovSum technique is suited for detecting small, but clinically significant biases. Point of care QC should follow conventional practice by setting the control limits according to the running mean and SD to allow proper error detection. The glucometer manufacturers have an active role to play in liberalizing QC settings and also enhancing the middleware to facility patient-based QC practices.
View less >
Journal Title
Biochemia Medica
Volume
30
Issue
2
Copyright Statement
© Croatian Society of Medical Biochemistry and Laboratory Medicine. This work is licensed under a Creative Commons Attribution 4.0 International License which permits users to read, download, copy, distribute, print, search, or link to the full texts of these articles in any medium or format and to remix, transform and build upon the material, provided the original work is properly cited and any changes properly indicated.
Subject
Medical biochemistry and metabolomics
Science & Technology
Life Sciences & Biomedicine
Medical Laboratory Technology
laboratory management
quality control