Back to the Top
Dear All,
"We have detected an oulier. Then performed Re-Dose study including
that subject & Another 4 subjects with same sequence frm the same study.
Redose concentration data replaced in original study & Again outlier
test performed.
If we detect the same subject is outlier it means, we can't treat that
subject as an outlier & we have to include that subject in statistical
Analysis.
But If the same subject is not detected outlier it means, we the data
obtained in Original study for that subject are in DOUBT & that
subject is to be excluded from statistical Analysis."
Is it correct Approach to handle the outlier subject data
Thank You,
Regards
Back to the Top
If I take this correctly, you analyzed a batch of samples. The
results for one of the samples did not look "right". Then you re-
analyzed the sample twice again and now the results were more
"acceptable". Is this approach for lab reanalysis covered in your
SOP? Was the analytical batch accepted? Did you investigate
possible causes before risking sample? How are you handling the
deviation?
1) Did you investigate or have the clinic investigate whether there
was a pre-analytical error before you assumed it was the lab? Often
the patient could have been misdosed or missassigned. It could have
been an accidental misdosing. The subject could have been late or
early for dosing or sampling ( day, time).
2) When you re-analyzed didyou go back to the original, take a retain
or use a clin lab sample? To rule out missampling go back to a retain
or to the clin lab sample as long as they have been handled
correctly. Using the original-same sample is not the best way, since
it could have been mislabeled at the clinic or at the lab.
3) it could simply mean that on retest-you did test it twice more-you
sampled correctly and the original was an error. Can you identify the
source of the error? Pre, Lab or Post analytical?
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 "Approach to handle Outlier Data" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)