# PharmPK Discussion - Bioequivalence - parallel design

PharmPK Discussion List Archive Index page
• On 15 Aug 2000 at 10:06:26, ml11439.-a-.goodnet.com sent the message
`Brian,     The rationale behind determining the sample size for a giveneperiment is statistical power. Statistical power is basicallythe likelihood of determining a significant treatment effectwhen one really exists. So that a sufficiently large samplesize (i.e. the number of patients in each group) will allowenough statistical power so that there is a good likelihoodthat a significant treatment effect will be detected (i.e. bya t-test) if one exists, and this provides a good reason forconducting the study.     Since your study design indicates that if the measuredparameter is normally distributed a t-test would be an appropriatestatistical test, a power chart for a t-test could be usedto determine the necessary sample size. If the size of the changeyou expect to detect is about the same size as the standarddeviation of the parameter in the population, then the samplesize to obtain statistical power above 80% is around 20 patientsper group. However, if the expected change or treatment effectis only about 50% of the standard deviation of the parameter inthe population, then the required sample size ot obtain 80%statistical power would be about 50 per group.     The same reasoning applies to analysis of variance but anoncentrality parameter is calculated to determine statistical powerrather than just phi=delta/sigma in the case of the t-test.    Noncentrality parameter (phi):     phi= (delta/sigma)SQRT(n/2k)     eg for delta/sigma= 1 and n=25, k= 3 treatment groups     phi= (1)SQRT(25/6)= (1)(2.04)= 2.04     Consulting a power chart for analysis of variance indicates thata sample size of about 25 per treatment group would provide thesufficient 80% statistical power given a delta/sigma of 1 and and k=3treatment groups.     At any rate, the idea would be to determine the magnitude ofchange you are expecting to find, compare this to the standarddeviation of the parameter in the patient population, and thenconsult a power chart to determine the necessary sample size toachieve statistical power of 80%. The power chart you consultdepends on the statistical test you intend to perform.     One good reference for this would be Glantz's Primer ofBiostatistics from McGraw-Hill.                     Good Luck!!                               Mike Leibold, PharmD, RPh                               ML11439.aaa.goodnet.com`
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 "Bioequivalence - parallel design" as the subject

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