# PharmPK Discussion - Sample Size Calculations and BE Question

PharmPK Discussion List Archive Index page
• On 9 Jun 2006 at 11:02:44, "Laura T. Letendre" (letendrl.-at-.eden.rutgers.edu) sent the message
`The following message was posted to: PharmPKHi all,I have two questions regarding bioequivalence studies:1) WinNonlin outputs Westlake confidence intervals.  When would it beappropriate to use these confidence intervals and would they beaccepted by the FDA?2)  What is the correct way to do a sample size calculations for athree way cross over study.  I have 3 drugs to compare with 3sequences.  I am using nQuery advisory and I get a per group samplesize for a two group design.  Is it correct to just multiply thisnumber by three - the sample size I am getting seems too high.  Doesanyone have any references that would help with this?Thank you in advance.Laura`
Back to the Top

• On 10 Jun 2006 at 10:30:39, "srinivas" (esreddy.at.bioserve.co.in) sent the message
`The following message was posted to: PharmPKDear laura,Sample size calculation depends on coefficient of variation. Ifcoefficientof variation is large, automatically sample size will come large.You can calculate sample size using SAS software. In that you can getsamplesize N per group. So If study is three period, multiply N with three,so youcan get sample size.regardsSrinivasa Reddy.EResearch AnalystBioserve Clinical Research Pvt LtdHyderabad.`
Back to the Top

• On 13 Jun 2006 at 12:51:55, =?ISO-8859-1?Q?Helmut_Sch=FCtz?= (helmut.schuetz.-at-.bebac.at) sent the message
`The following message was posted to: PharmPKDear Laura! >1) WinNonlin outputs Westlake confidence intervals.  When would it be >appropriate to use these confidence intervals and would they be >accepted by the FDA? >Westlake's symetrical confidence interval is of historical interestonly and not accepted in any country...IMHO they should be removed form WinNonlin... >2)  What is the correct way to do a sample size calculations for a >three way cross over study.  I have 3 drugs to compare with 3 >sequences.  I am using nQuery advisory and I get a per group sample >size for a two group design.  Is it correct to just multiply this >number by three - the sample size I am getting seems too high.  Does >anyone have any references that would help with this? >3 sequences are not adequate anyway; to compare 3 treatments youshould apply a 6-sequence William's design.For details (including sample size estimation) look at this posthttp://forum.bebac.at/forum_entry.php?id=50 and the followings.best regards,Helmut--Helmut SchuetzBEBACConsultancy Services for Bioequivalence and Bioavailability StudiesNeubaugasse 36/111070 Vienna/Austriatel/fax +43 1 2311746Web http://BEBAC.atBE/BA Forum http://forum.bebac.athttp://www.goldmark.org/netrants/no-word/attach.html`
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 "Sample Size Calculations and BE Question" as the subject

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