Back to the Top
Does anyone have an outlier test that is industry standard?
Back to the Top
Dear members,
I have an doubt regarding report preparation with statistical
outlier calculation on QC AND STUDY SAMPLES.
1. Statistical Outliers in the Quality control samples cannot be
taken for calculating Accuracy and Precision (During inter and intra
day results reporting in bioanalytical method validation) as per
USFDA guidelines, please clarify weather same to be applied for study
sample analysis?
2. Whether the Inter day and Intra day Accuracy and Precision tables
can be reported with and without QC outliers in the bioanalytical
report?
hope with best,
regards,
Muthuswamy.C
Executive Analyst
wellquest Clinical Research
Nicholas Piramal India ltd.,
Hyderabad.
India.
Back to the Top
The following message was posted to: PharmPK
Dear Stephen,
I do not believe that there is an outlier test that has been accepted
by any regulatory agency or by
wider community as a standard. Agencies prefer monitoring of control
variables as the only valid way
for inclusion/exclusion of objects from some experiment (clinical
study, process control etc.). At
least this is my experience with the FDA. However, I believe that
people use Grubbs test the most
frequently.
Zeljko Debeljak, PhD
Medical Biochemistry Specialist
Clinical Hospital Osijek
CROATIA
Back to the Top
I agree with Zeljko (Grubbs test for outliers)
Dr. Gagandeep Singh, Toronto, Canada
Website: www.gagansworld.com
Back to the Top
Here are two possibilities we use:
Grubb's Test for Detecting Outlier
or
by Box_Whisker plot.
Matthias
Matthias Reule PhD
29 Hoadly Road
Cambridge
CB3 0HX
UK
matthias.reule.-a-.googlemail.com
Back to the Top
This references might help
1. R. Dennis ,B wiliam , Graham . Chromotographic validation
guidelines for primary validation parameters. Journal of liquid
chromotography and related technology 1996 19(5) 737-757
2. R. Dennis ,Jenke . Chromotographic method validation a review of
current practices and procedures. Journal of liquid Chromotography
and related technology 1996 19(12) 1873-1891
s.o.o
Back to the Top
The following message was posted to: PharmPK
dear stephen,
i do agree with zeljko as grubbs test for outliers, but again FDA
doesnt encourage this excerise and if do than only it should be
predefine in protocol or in any of your concern SOP.
with and without data in this case again ask by FDA and all
regulatories.
hope i am near about your answer,
expert can comment.
regards,
paresh
Back to the Top
The following message was posted to: PharmPK
Dear all,
My (biased) view on this is that you should use robust techniques and
specify that in your stat plan. Much better than all these outlier tests
where basically 1) either an observation is in or out, 2) the
asymptotics of a case-deletion and refit approach is dubious (&
non-robust). The FDA is not against robust techniques but I have not
tested their reaction yet!
Cheers
S. Heritier
Dr Stephane Heritier
Head of Statistical Methodology
Senior lecturer in biostatistics
The George Institute
Level 10
King George V Building
Royal Prince Alfred Hospital
Missenden Road
Camperdown NSW 2050 Australia
Postal address:
PO Box M201 Missenden Road
Sydney NSW 2050 Australia
Email: sheritier.aaa.george.org.au
Back to the Top
The following message was posted to: PharmPK
Dear Stephane,
do you assume resampling statistics under the term "robust
techniques"? I do agree that selected
outlier test and its application should be defined in some of the
SOPs at the begining of the
experiment. I think that resampling methods do have some advantages
over methods that assume normal
or other predefined distribution of data, but I have not seen it in
drug master files or similar
documentation. It would be interesting to hear people from FDA or
other regulatory institutions
regarding this matter.
Zeljko Debeljak, PhD
Medical Biochemistry Specialist,
Osijek Clinical Hospital
CROATIA
Back to the Top
The following message was posted to: PharmPK
While the Grubbs Test is the most used it has not been recognized as
an industry standard. Graphic tests such as box-whisker tests have
also been use. A descriptive test for distribution should be done
prior to the selection of any outlier test.
Ed O'Connor, Ph.D.
Laboratory Director
Matrix BioAnalytical Laboratories
25 Science Park at Yale
New Haven, CT 06511
Web: www.matrixbioanalytical.com
Email: eoconnor.at.matrixbioanalytical.com
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Outlier test" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)