A basic program to determine outliers in biopharmaceutical analysis using the Nair criterion

In bioanalysis, as in other areas of analysis, difficulties can arise in judging whether to accept or reject individual values from a set of replicate analytical results. There appear to be no definitive procedures described in the literature which are applicable to bioanalytical results. In the aut...

Full description

Saved in:
Bibliographic Details
Published in:Journal of pharmaceutical and biomedical analysis 1988, Vol.6 (3), p.317-319
Main Authors: Clarke, Graham S., Robinson, Malcolm L.
Format: Article
Language:English
Subjects:
Analysis
Biological and medical sciences
biopharmaceutical analysis
General pharmacology
Medical sciences
Nair criterion
Outliers
Pharmacology. Drug treatments
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In bioanalysis, as in other areas of analysis, difficulties can arise in judging whether to accept or reject individual values from a set of replicate analytical results. There appear to be no definitive procedures described in the literature which are applicable to bioanalytical results. In the authors' laboratory, the analysis of a sample can yield a set of triplicate results, two of which are close and the other an apparent outlier. If this is judged to be so, then the outlier is rejected and the mean of the two remaining values is quoted as the analytical result. If it is judged otherwise, the mean of the three results is reported. However, a problem arises when two operators judge the same data. A finite number of instances will occur where there is disagreement. To overcome this problem, a number of statistical tests were examined. Of these, the Nair criterion was found to be the test that most often gave a result agreeing with the authors' own judgement.
ISSN:0731-7085
1873-264X
DOI:10.1016/0731-7085(88)80059-1