Back to the Top
The following message was posted to: PharmPK
Hi, the discussion about applications of non-parametric statistical
methods in the PK area had been highly enriching and motivating.The
responses of the members on this topic have been summarized here
followe by a list of references cited by them. The actively
participating members were Dr.Roger W. Jelliffe,Professor of
Medicine,USC Laboratory of Applied Pharmacokinetics; Dr.Helmut
Schuetz,Head biometrics,Biokinet GmbH;Dr.Alun Bedding,Senior
Statistician, Clinical Pharamacology Regulatory and Scientific
Expert,Eli Lilly ; Dr.Federico Lerner ; Dr.Joga Gobburu ,
Pharmacometrics, CDER, FDA.
Dr.Jelliffe advises that the Non-Parametric statistical methods do
not require assumption of normal or log-normal distribution and hence
can detect unsuspected subpopulations such as fast or slow
metabolizers.They are consistent too.In particular these are best
suited while working with the "multiple model" dosage design and
especially best when the parameter distributions are not normal,but
have genetic polymorphism.
Dr.Schuetz has stated that the Nonparametrics are mandatory for
measurements originating from discrete distributions (e.g., tmax).
This is also mentioned in the current EU-guidance (July 2001) for
the investigation of bioavailability and bioequivalence,which can be
obtained from EMEA:
http://www.emea.eu.int/pdfs/human/ewp/140198en.pdf . A confirmatory
study solely based on nonparametrics can also be planned and
evaluated. Dr.Schuetz's group's sponsors have got approvals for about
300 such studies.
According to Dr.Gobburu,nonparametrics has been used for Statistical
testing to compare two distributions,Population PKPD analyses in the
past.Frequency calculations for complicated scenarios, where the
probability is determined by shear computational force, especially
when the underlying statistical distribution is under question.
Dr.Bedding has said that the most obvious use of non-parametric
methods is in the analysis of
the time to maximum concentration (tmax), which is dependent on the
sampling interval and therefore rarely fufils the assumptions for a
parametric assessment.
Dr.Lerner suggests that Non-parametrics statistics are appliable for
the analyses of non-nermal metabolites Pk profile. This is applicalbe
in cases of different metabilizers phenotype, fluoxetine as example.
This is because at least two different subpopulations are present.
List of references :
Hauschke D, Steinijans VW, and Diletti E (1990) A distribution free
procedure for the statistical analysis of bioequivalence
studiesInternational Journal of Clinical Pharmacology, Therapy and
Toxicology, 28, 72-8
Vuorinen, J. and J. Turunen;
A Simple Three-Step Procedure for Parametric and Nonparametric
Assessment of Bioequivalence
Drug Information Journal 31/1, 167-180 (1997)
Abebe, A., Crimin, K. and J.K. McKean;
Rank-Based Procedures for Linear Models: Applications to
Pharmaceutical Science Data
Drug Information Journal 35/3, 947-971 (2001)
Websites :
http://www.usc.edu/hsc/lab_apk
http://www.stat.wmich.edu/slab/RGLM/ (own data can be analysed).
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)