Bioequivalence and Bioavailability Forum

Main page Policy/Terms of Use Abbreviations Latest Posts

 Log-in |  Register |  Search

Back to the forum  Query: 2017-03-31 02:36 CEST (UTC+2h)
 
Maryam
Junior

Iran,
2017-01-13 16:28

Posting: # 16946
Views: 2,063
 

 SAS [Software]

HI

I want to run this procedure in sas9.1:

proc ttest data=auc dist=lognormal tost(0.8, 1.25);
   paired TestAUC*RefAUC;
run;


but it had an error::no:

How should I solve this problem?:clap:

thanks:ok:
Mahmoud
Junior

Jordan,
2017-01-15 08:56

@ Maryam
Posting: # 16947
Views: 1,840
 

 SAS

Dear Sir
=====
For BE studies do not use the t-test paired data.if you need any results in BE study please send me your data in the following email

mahmohsen111[at]yahoo.com


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post #5! I activated the e-mail link in your profile. [Helmut]
Helmut
Hero
Homepage
Vienna, Austria,
2017-01-15 12:16

@ Mahmoud
Posting: # 16948
Views: 1,825
 

 Paired vs. crossover

Salam Mahmoud,

» Dear Sir

I assume that Maryam is not a Sir. ;-)

» For BE studies do not use the t-test paired data.

We don’t know the design of Maryam’s study. What if it was not a crossover? Examples where a paired t-test would be correct indeed:
  • Assessment of linear PK. Single dose followed by steady state and AUC0–τ vs. AUC0–∞.
  • 4-period food-effect design where periods 1&2 fed and periods 3&4 fasting. Crossover in fed and fasting state but assessment of the food effect of T and R by a paired t-test (assuming no period effects).
@Maryam: What is the structure of you data? I don’t speak SAS but which error did you get?

[image]All the best,
Helmut Schütz 
[image]

The quality of responses received is directly proportional to the quality of the question asked. ☼
Science Quotes
Maryam
Junior

Iran,
2017-01-15 16:44

@ Helmut
Posting: # 16951
Views: 1,801
 

 Paired vs. crossover

» We don’t know the design of Maryam’s study.
» @Maryam: What is the structure of you data? I don’t speak SAS but which error did you get?

Hi sir
Maryam is not a sir :ok:

thanks for your answer,
my structure is a crossover study! actually I want to calculate 90% CI for BE parameters of test/reference! and I want to evaluate the effect of period, sequence, subject and treatment!

I know how calculate these by formulations but I need a software to calculate these! would you please to guide me in this problem?

thanks


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post #5! [Helmut]
DavidManteigas
Regular

Portugal,
2017-01-17 15:35

@ Maryam
Posting: # 16967
Views: 1,646
 

 Paired vs. crossover

@Maryam

For a 2x2 cross-over, the following code will get you everything you need (ods outputs LSMeandiffCL, LSMeans, AltErrTests and OverallANOVA)

proc glm data=replicate;
class formulation subject period sequence;
model logDATA= sequence subject (sequence) period formulation;
estimate "test-ref" formulation -1+1;
test h=sequence e=subject(sequence);
lsmeans formulation / adjust=t pdiff=control("R") CL alpha=0.10;
run;


Formulation and Period p values come from OverallAnova, while sequence p value from AltErrTests.

This code in principle will fit both EMA and FDA requirements for a 2x2 design. Although FDA requests subject within sequence as random effect, this will not change the result.
Helmut
Hero
Homepage
Vienna, Austria,
2017-01-18 12:57

@ DavidManteigas
Posting: # 16972
Views: 1,483
 

 Picky

Hi David,

» proc glm data=replicate;

<nitpicking>

Why are you calling a data set of a 2×2 crossover replicate? Of course you can baptize it as you like but “replicate” might confuse newbies.

</nitpicking>

» lsmeans formulation / adjust=t pdiff=control("R") CL alpha=0.10;

I would remove adjust=t. There is nothing to adjust in a 2×2 crossover. See this thread.

[image]All the best,
Helmut Schütz 
[image]

The quality of responses received is directly proportional to the quality of the question asked. ☼
Science Quotes
DavidManteigas
Regular

Portugal,
2017-01-27 15:03

@ Helmut
Posting: # 16996
Views: 829
 

 Picky

Hi Helmut,

The reason for "Replicate" as the dataset name is because I took the code from the EMA Q&A Guideline and adapted it - but forgot to remove the misleading name :-D

I agree with you regarding the adjusment question. Nevertheless, as far as I understand the parameter does nothing if you leave it on the code.
Maryam
Junior

Iran,
2017-01-15 16:31

@ Mahmoud
Posting: # 16950
Views: 1,803
 

 SAS

» For BE studies do not use the t-test paired data.

Hi sir
thanks for your answer.
your are right! t-test paired does not use for BE studies. By SAS and ttest I want to calculate 90% CI for pharmacokinetics parameters of test/reference.
would you please to help me to calculate 90% CI for pharmacokinetics parameters of test/reference by any softwares? can I use this calculation by SPSS?

thanks


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post #5! [Helmut]
Mahmoud
Junior

Jordan,
2017-01-15 14:36

@ Maryam
Posting: # 16949
Views: 1,815
 

 SAS

» I want to run this procedure in sas9.1:
»
» proc ttest data=auc dist=lognormal tost(0.8, 1.25);
»    paired TestAUC*RefAUC;
» run;

Dear all
=====
For the codes above you need to use SAS 9.3

Mahmoud Ph.D


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post #5! [Helmut]
Maryam
Junior

Iran,
2017-01-15 16:50

@ Mahmoud
Posting: # 16952
Views: 1,803
 

 SAS

» For the codes above you need to use SAS 9.3

Hi sir
thanks for your answer,
Do you know how I can correct these codes for SAS 9.1? because I have only SAS 9.1!
thanks


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post #5! [Helmut]
Mahmoud
Junior

Jordan,
2017-01-16 10:08

@ Maryam
Posting: # 16954
Views: 1,758
 

 SAS

Dear Dr
Please let me know what is the design for your data


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post #5!
Since you repeatedly did not follow the Policy, this is the first warning. [Helmut]
Back to the forum Activity
 Thread view
Bioequivalence and Bioavailability Forum | Admin contact
16,723 Posts in 3,587 Threads, 1,040 registered users;
6 users online (0 registered, 6 guests).

A big computer, a complex algorithm and a long time
does not equal science.    Robert Gentleman

The BIOEQUIVALENCE / BIOAVAILABILITY FORUM is hosted by
BEBAC Ing. Helmut Schütz
XHTML/CSS RSS Feed