# PharmPK Discussion - Applications of Non-parametric methods

PharmPK Discussion List Archive Index page
• On 13 Oct 2001 at 11:37:53, "Rahul Pradhan" (rahulprd.aaa.hotmail.com) sent the message
`Hi, I am a graduate student in Pharmaceutics. I am studying a coursein Non-parametric statistical methods. I am curious to know ifanybody can suggest me some examples in Pharmacokinetics wherenon-parametric statistics has been applied. Please also suggest anypossible examples in the PK area where these methods may be needed.Regards to all.Rahul Pradhan`
Back to the Top

• On 15 Oct 2001 at 23:39:47, "Gobburu, Jogarao V" (GOBBURUJ.-a-.cder.fda.gov) sent the message
`The following message was posted to: PharmPKDear Rahul,   Nonparametric statistics are employed in pkpd related areas reasonablyfrequently but not as frequently as parametric statistics. I have seentheir use in:1. Statistical testing to compare two distributions.    Example: Kolmogorov-Smirnov test2. Population PKPD analyses.    Example: http://lapk3.hsc.usc.edu/lapk/3. Frequency calculations for complicated scenarios, where the probabilityis determined by shear computational force, especially when the underlyingstatistical distribution is under question.Example: Re-randomization/permutation tests, nonparametric bootstrap, etc.Regards,Joga Gobburu,Pharmacometrics,CDER, FDA`
Back to the Top

• On 16 Oct 2001 at 11:23:14, David Bourne (david.aaa.boomer.org) sent the message
`[Two replies - db]From: BEDDING_ALUN.aaa.LILLY.COMDate: Mon, 15 Oct 2001 08:38:15 +0100To: david.at.boomer.orgSubject: Re: PharmPKDear Rahul,The most obvious use of non-parametric methods is in the analysis ofthe time to maximum concentration (tmax), which is dependent on thesampling interval and therefore rarely fufils the assumptions for aparametric assessment.A useful reference for the non-parametric assessment in abioequivalence study is:Hauschke D, Steinijans VW, and Diletti E (1990) A distribution freeprocedure for the statistical analysis of bioequivalencestudiesInternational Journal of Clinical Pharmacology, Therapy andToxicology, 28, 72-8Kind regards,Alun BeddingSenior Statistician, Clinical Pharamacology Regulatory and Scientific ExpertEli Lilly---From: "Federico Lerner, MD" Date: Mon, 15 Oct 2001 09:55:05 -0300To: david.at.boomer.orgSubject: RE: Applications of Non-parametric methodsThe following message was posted to: PharmPKDear Rahul,Non-parametrics statistics are appliable for the analyses of non-nermalmetabolites Pk profile. This is applicalbe in cases of differentmetabilizers phenotype, fluoxetine as example. This is because at least twodifferent subpopulations are present.RegardsFederico Lerner, MD`
Back to the Top

• On 17 Oct 2001 at 23:02:59, Helmut =?iso-8859-1?Q?Sch=FCtz?= (helmut.schuetz.-at-.chello.at) sent the message
`The following message was posted to: PharmPKDear Rahul,Nonparametrics are mandatory for measurements originating from discretedistributions (e.g., tmax). This is also mentioned in the current EU-guidance(July 2001) for the investigation of bioavailability and bioequivalence, whichcan be obtained from EMEA:  http://www.emea.eu.int/pdfs/human/ewp/140198en.pdfIf you plan and evaluate a confirmatory study solely on nonparametrics (yes,AUC, cmax,...), everything is fine too (our sponsors got approvals with about300 BA/BE-studies).Some more references:Steinijans, V.W. and D. Hauschke;Update on the statistical analysis of bioequivalence studiesInt. J. Clin. Pharm. Ther. Toxicol. 28, 105-110 (1990)Vuorinen, J. and J. Turunen;A Simple Three-Step Procedure for Parametric and Nonparametric Assessment ofBioequivalenceDrug Information Journal 31/1, 167-180 (1997)Nonparametrical methods may also be useful in all areas, where aformal prove ofdistributional assumptions may either not be possible, or lack of statisticalpower (e.g., if sample sizes are below appr. 30). The assymptotic efficacy ofmany nonparametric methods is 3/pi=95% if the underlying distributionis normal.If distributions are 'heavy tailed' (which is actually very often the case inthe 'real world'), nonparametric methods will perform superior to theirparametric analogues.Yet another reference:Abebe, A., Crimin, K. and J.K. McKean;Rank-Based Procedures for Linear Models: Applications to Pharmaceutical ScienceDataDrug Information Journal 35/3, 947-971 (2001)They have also run a very nice web-site, were you can input your own data...  http://www.stat.wmich.edu/slab/RGLM/Best regards,Helmut SchuetzHead biometricsBiokinet GmbHNattergasse 4A-1170 Vienna/Austriatel +43(0)1 4856969-77fax +43(0)1 4856970-90email helmut.schuetz.aaa.chello.at`
Back to the Top

• On 18 Oct 2001 at 13:43:00, Roger Jelliffe (jelliffe.aaa.hsc.usc.edu) sent the message
`Dear Rahul:         Yes, nonparametric (NP) methods are very useful in populationPK/PD modeling, as they are not constrained by any assumptions ofnormality or lognormality in the parameter distributions. Because ofthis, they can detect unsuspected subpopulations such as fast or slowmetabolizers. The methods are also consistent - that is, the more yousample from the population, the more the results approach the trueresults.         NP models are also best suited for acting on the basis of thepopulation information and data, using the new method of "multiplemodel" dosage design, which evaluates the predicted precision withwhich any dosage regimen with achieve a desired target goal, andspecifically designs the regimen which achieves the goal with maximalprecision. This is especially useful then the parameter distributionsare not Gaussian, but have genetic polymorphism, as so many do.         More info is on our web site www.lapk.org, under teachingtopics, and under old workshop sand presentations. You can download alot of stuff there.Very best regards,Roger JelliffeRoger W. Jelliffe, M.D. Professor of Medicine, USCUSC Laboratory of Applied Pharmacokinetics2250 Alcazar St, Los Angeles CA 90033, USAPhone (323)442-1300, fax (323)442-1302, email=  jelliffe.-a-.hsc.usc.eduOur web site=  http://www.usc.edu/hsc/lab_apk`
Back to the Top

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 "Applications of Non-parametric methods" as the subject

Copyright 1995-2010 David W. A. Bourne (david@boomer.org)