Multiple calibrator measurements improve accuracy and stability estimates of automated assays

Neval Akbas, Jeffrey R. Budd, George G. Klee

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Objectives The effects of combining multiple calibrations on assay accuracy (bias) and measurement of calibration stability were investigated for total triiodothyronine (TT3), vitamin B12 and luteinizing hormone (LH) using Beckman Coulters Access 2 analyzer. Methods Three calibration procedures (CC1, CC2 and CC3) combined 12, 34 and 56 calibrator measurements over 1, 2, and 3 days. Bias was calculated between target values and average measured value over 3 consecutive days after calibration. Using regression analysis of calibrator measurements versus measurement date, calibration stability was determined as the maximum number of days before a calibrator measurement exceeded 5% tolerance limits. Results Competitive assays (TT3, vitamin B12) had positive time regression slopes, while sandwich assay (LH) had a negative slope. Bias values for TT3 were -2.49%, 1.49%, and -0.50% using CC1, CC2 and CC3 respectively, with calibrator stability of 32, 20, and 30 days. Bias values for vitamin B12 were 2.44%, 0.91%, and -0.50%, with calibrator stability of 4, 9, and 12 days. Bias values for LH were 2.26%, 1.44% and -0.29% with calibrator stability of >43, 39 and 36 days. Measured stability was more consistent across calibration procedures using percent change rather than difference from target: 26 days for TT3, 12 days for B12 and 31 days for LH. Conclusions Averaging over multiple calibrations produced smaller bias, consistent with improved accuracy. Time regression slopes in percent change were unaffected by number of calibration measurements but calibrator stability measured from the target value was highly affected by the calibrator value at time zero.

Original languageEnglish (US)
Pages (from-to)177-180
Number of pages4
JournalScandinavian Journal of Clinical and Laboratory Investigation
Volume76
Issue number2
DOIs
StatePublished - Feb 17 2016

Fingerprint

Calibration
Assays
Luteinizing Hormone
Vitamin B 12
Triiodothyronine
Regression analysis
Regression Analysis

Keywords

  • accuracy
  • bias
  • Calibration
  • immunoassay
  • stability

ASJC Scopus subject areas

  • Clinical Biochemistry

Cite this

Multiple calibrator measurements improve accuracy and stability estimates of automated assays. / Akbas, Neval; Budd, Jeffrey R.; Klee, George G.

In: Scandinavian Journal of Clinical and Laboratory Investigation, Vol. 76, No. 2, 17.02.2016, p. 177-180.

Research output: Contribution to journalArticle

@article{69c4acaee035452482d846e3557b809b,
title = "Multiple calibrator measurements improve accuracy and stability estimates of automated assays",
abstract = "Objectives The effects of combining multiple calibrations on assay accuracy (bias) and measurement of calibration stability were investigated for total triiodothyronine (TT3), vitamin B12 and luteinizing hormone (LH) using Beckman Coulters Access 2 analyzer. Methods Three calibration procedures (CC1, CC2 and CC3) combined 12, 34 and 56 calibrator measurements over 1, 2, and 3 days. Bias was calculated between target values and average measured value over 3 consecutive days after calibration. Using regression analysis of calibrator measurements versus measurement date, calibration stability was determined as the maximum number of days before a calibrator measurement exceeded 5{\%} tolerance limits. Results Competitive assays (TT3, vitamin B12) had positive time regression slopes, while sandwich assay (LH) had a negative slope. Bias values for TT3 were -2.49{\%}, 1.49{\%}, and -0.50{\%} using CC1, CC2 and CC3 respectively, with calibrator stability of 32, 20, and 30 days. Bias values for vitamin B12 were 2.44{\%}, 0.91{\%}, and -0.50{\%}, with calibrator stability of 4, 9, and 12 days. Bias values for LH were 2.26{\%}, 1.44{\%} and -0.29{\%} with calibrator stability of >43, 39 and 36 days. Measured stability was more consistent across calibration procedures using percent change rather than difference from target: 26 days for TT3, 12 days for B12 and 31 days for LH. Conclusions Averaging over multiple calibrations produced smaller bias, consistent with improved accuracy. Time regression slopes in percent change were unaffected by number of calibration measurements but calibrator stability measured from the target value was highly affected by the calibrator value at time zero.",
keywords = "accuracy, bias, Calibration, immunoassay, stability",
author = "Neval Akbas and Budd, {Jeffrey R.} and Klee, {George G.}",
year = "2016",
month = "2",
day = "17",
doi = "10.3109/00365513.2015.1126760",
language = "English (US)",
volume = "76",
pages = "177--180",
journal = "Scandinavian Journal of Clinical and Laboratory Investigation",
issn = "0036-5513",
publisher = "Informa Healthcare",
number = "2",

}

TY - JOUR

T1 - Multiple calibrator measurements improve accuracy and stability estimates of automated assays

AU - Akbas, Neval

AU - Budd, Jeffrey R.

AU - Klee, George G.

PY - 2016/2/17

Y1 - 2016/2/17

N2 - Objectives The effects of combining multiple calibrations on assay accuracy (bias) and measurement of calibration stability were investigated for total triiodothyronine (TT3), vitamin B12 and luteinizing hormone (LH) using Beckman Coulters Access 2 analyzer. Methods Three calibration procedures (CC1, CC2 and CC3) combined 12, 34 and 56 calibrator measurements over 1, 2, and 3 days. Bias was calculated between target values and average measured value over 3 consecutive days after calibration. Using regression analysis of calibrator measurements versus measurement date, calibration stability was determined as the maximum number of days before a calibrator measurement exceeded 5% tolerance limits. Results Competitive assays (TT3, vitamin B12) had positive time regression slopes, while sandwich assay (LH) had a negative slope. Bias values for TT3 were -2.49%, 1.49%, and -0.50% using CC1, CC2 and CC3 respectively, with calibrator stability of 32, 20, and 30 days. Bias values for vitamin B12 were 2.44%, 0.91%, and -0.50%, with calibrator stability of 4, 9, and 12 days. Bias values for LH were 2.26%, 1.44% and -0.29% with calibrator stability of >43, 39 and 36 days. Measured stability was more consistent across calibration procedures using percent change rather than difference from target: 26 days for TT3, 12 days for B12 and 31 days for LH. Conclusions Averaging over multiple calibrations produced smaller bias, consistent with improved accuracy. Time regression slopes in percent change were unaffected by number of calibration measurements but calibrator stability measured from the target value was highly affected by the calibrator value at time zero.

AB - Objectives The effects of combining multiple calibrations on assay accuracy (bias) and measurement of calibration stability were investigated for total triiodothyronine (TT3), vitamin B12 and luteinizing hormone (LH) using Beckman Coulters Access 2 analyzer. Methods Three calibration procedures (CC1, CC2 and CC3) combined 12, 34 and 56 calibrator measurements over 1, 2, and 3 days. Bias was calculated between target values and average measured value over 3 consecutive days after calibration. Using regression analysis of calibrator measurements versus measurement date, calibration stability was determined as the maximum number of days before a calibrator measurement exceeded 5% tolerance limits. Results Competitive assays (TT3, vitamin B12) had positive time regression slopes, while sandwich assay (LH) had a negative slope. Bias values for TT3 were -2.49%, 1.49%, and -0.50% using CC1, CC2 and CC3 respectively, with calibrator stability of 32, 20, and 30 days. Bias values for vitamin B12 were 2.44%, 0.91%, and -0.50%, with calibrator stability of 4, 9, and 12 days. Bias values for LH were 2.26%, 1.44% and -0.29% with calibrator stability of >43, 39 and 36 days. Measured stability was more consistent across calibration procedures using percent change rather than difference from target: 26 days for TT3, 12 days for B12 and 31 days for LH. Conclusions Averaging over multiple calibrations produced smaller bias, consistent with improved accuracy. Time regression slopes in percent change were unaffected by number of calibration measurements but calibrator stability measured from the target value was highly affected by the calibrator value at time zero.

KW - accuracy

KW - bias

KW - Calibration

KW - immunoassay

KW - stability

UR - http://www.scopus.com/inward/record.url?scp=84959540756&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959540756&partnerID=8YFLogxK

U2 - 10.3109/00365513.2015.1126760

DO - 10.3109/00365513.2015.1126760

M3 - Article

C2 - 26785381

AN - SCOPUS:84959540756

VL - 76

SP - 177

EP - 180

JO - Scandinavian Journal of Clinical and Laboratory Investigation

JF - Scandinavian Journal of Clinical and Laboratory Investigation

SN - 0036-5513

IS - 2

ER -