2022
DOI: 10.1016/j.cca.2022.07.006
|View full text |Cite
|
Sign up to set email alerts
|

Lot-to-lot reagent verification: Effect of sample size and replicate measurement on linear regression approaches

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
4
0

Year Published

2022
2022
2023
2023

Publication Types

Select...
5

Relationship

1
4

Authors

Journals

citations
Cited by 6 publications
(4 citation statements)
references
References 14 publications
0
4
0
Order By: Relevance
“…The simulation framework for this study has been published previously [16,17]. This study only involved numerical simulations and was exempted from local institutional ethics board review.…”
Section: Methodsmentioning
confidence: 99%
See 1 more Smart Citation
“…The simulation framework for this study has been published previously [16,17]. This study only involved numerical simulations and was exempted from local institutional ethics board review.…”
Section: Methodsmentioning
confidence: 99%
“…t-statistics of regression slope and intercept, confidence ellipses, and a modified Bland-Altman regression) of different regression models (ordinary and weighted forms of least squares regression as well as Deming regressions), to determine the presence of significant analytical bias [16]. In a follow up study, the impact of the number of samples and replicates on the probability of bias detection using ordinary and weighted forms of least square and Deming regression was examined [17]. This simulation study reported here was undertaken to assess the statistical performance of six commonly used rejection criteria, as summarized in a recent systematic review [1], based on either difference or regression-based approaches for bias detection.…”
Section: Introductionmentioning
confidence: 99%
“…This assessment is commonly performed by paired measurements of remainder clinical samples (ideally) spanning the analytical measurement range assayed with the existing and candidate reagent lots. Linear regression in various forms can then be applied on the paired results, which derive coefficients for slope (proportional difference) and intercept (constant difference) [3,4]. The advantage of using linear regression approaches is that assessment of lot-to-lot changes are investigated across the measurement range and not select concentrations.…”
Section: Dear Editormentioning
confidence: 99%
“…It is hope that this tool will help laboratory practitioners better assess the linear regression coefficients against a priori defined acceptance criteria for lot-to-lot verification. More formal statistical testing of regression-based assessment of lot-to-lot verification has been described elsewhere [3,4].…”
Section: Dear Editormentioning
confidence: 99%