# PharmPK Discussion - Confidence Interval for unbalanced study design - Ln transformed data

PharmPK Discussion List Archive Index page
• On 6 Mar 2007 at 04:13:30, USFDA EMEA (usfda_emea.at.yahoo.com) sent the message
`Dear,I am looking for help on estimation of 90% confidence interval forlog transformed data of AUC (0-inf) for a bioequivalence study. Thisstudy involved two-way crossover, two-period, two-sequence designwith a total of 33 subjects completing the study.In order to estimate the confidence interval, can some one help mewith the formula and a worked out example for the ease of understanding.I believe least-squares mean estimate for each formulation needs tobe used to form the difference, together with the standard error.However, in lack of exact formula i am seeking the help of expertsout there.Thanks in advance for the help.`
Back to the Top

• On 7 Mar 2007 at 19:33:47, "Nand Kishore Rawat" (Nand.KishoreRawat.-at-.ranbaxy.com) sent the message
`The following message was posted to: PharmPKDear Friend,To calculate 90% CI for Log-Transformed data following are the stepsto be followed (Using SAS Software)1.Initially Least square means separately for Test and Reference arecalculated using appropriate SAS codessteps 2- exponential conversion of Earlier calculated Least squaremeans for Test (say my1) and Reference (say my2)is done.step 3- Pooled standard deviation is calculated and taking squareroot gives Standard error as resultant.Finally Lower and upper CI's are calculated using formulalower_ci = exp((my2 - my1 )- (absct*se))*100upper_ci = exp((my2 - my1) + (absct*se))*100However your idea is absolutely fine regarding "least-squares meanestimate for each formulation needs tobe used to form the difference, together with the standard error"I think this will help you.Best Regards,Nand Kishore RawatBio-StatisticianDepartment of Clinical Pharmacology and pharmacokineticsRanbaxy Research LaboratoriesGurgaon-122001.India`
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 "Confidence Interval for unbalanced study design - Ln transformed data" as the subject Support PharmPK by using thelink to buy books etc.from Amazon.com

Copyright 1995-2011 David W. A. Bourne (david@boomer.org)