Ahikal
☆

Egypt,
2022-02-11 19:34
(406 d 10:13 ago)

Posting: # 22773
Views: 1,647

## statistical analysis of William's Design [Study As­sess­ment]

Dear all, I did a bioequivalence project using William's design (3 treatment, 3 period, 6 sequence), 2 test products against one reference. I analyzed the data as 3-way, and also separated the test products, each with a reference, and analyzed them. My question: should I get the same result?

Best Regards
Ahmed

Edit: Category changed; see also this post #1[Helmut]
Helmut
★★★

Vienna, Austria,
2022-02-11 20:07
(406 d 09:41 ago)

@ Ahikal
Posting: # 22774
Views: 1,451

## statistical analysis of William's Design

Hi Ahmed,

❝ […] William's design (3 treatment, 3 period, 6 sequence), 2 test products against one reference. I analyzed the data as 3-way, and also separated the test products, each with a reference, and analyzed them. My question: should I get the same result?

Extremely unlikely. In the separate evaluations (as Incomplete Block Designs) you get two estimates of the within-subject variances (one in the comparison of T1 vs R and the other in the comparison of T2 vs R). Even if they would be identical, you have different degrees of freedom in the analyses. In each of the IBDs you have $$\small{df=n-2}$$ and in the pooled ANOVA $$\small{df=2\times n-4}$$, where $$\small{n}$$ is the total number of subjects. Although the point estimates might be the same, the confidence intervals will not.

Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Ahikal
☆

Egypt,
2022-02-11 20:43
(406 d 09:04 ago)

@ Helmut
Posting: # 22775
Views: 1,427

## statistical analysis of William's Design

Thank you much, Helmut, for clarifying the issue.

Sincerely
Ahmed
Ahikal
☆

Egypt,
2022-02-12 15:58
(405 d 13:49 ago)

@ Helmut
Posting: # 22776
Views: 1,384

## statistical analysis of William's Design

Now a question arises, which way should be submitted to regulatory authorities?
Helmut
★★★

Vienna, Austria,
2022-02-12 16:02
(405 d 13:46 ago)

@ Ahikal
Posting: # 22777
Views: 1,379

## statistical analysis of William's Design

Hi Ahmed,

❝ Now a question arises, which way should be submitted to regulatory authorities?

Simple: The one you stated in the protocol.

Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Ahikal
☆

Egypt,
2022-02-12 16:41
(405 d 13:06 ago)

@ Helmut
Posting: # 22778
Views: 1,400

## statistical analysis of William's Design

Thanks again
Ahmed
Ahikal
☆

Egypt,
2022-02-17 23:16
(400 d 06:31 ago)

@ Helmut
Posting: # 22792
Views: 1,292

## statistical analysis of William's Design

Hi Helmut,
One last question, if you separate the two test products, do you maintain three periods or make it as two period?

Best Regards
Ahmed
Helmut
★★★

Vienna, Austria,
2022-02-17 23:33
(400 d 06:14 ago)

@ Ahikal
Posting: # 22794
Views: 1,256

## Two at a Time (Incomplete Blocks)

Hi Ahmed,

❝ One last question, if you separate the two test products, do you maintain three periods or make it as two period?

You have to keep the codes (sequence, period). See here (scroll down to David Brown’s slides) and there.

Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Ahikal
☆

Egypt,
2022-02-18 00:45
(400 d 05:02 ago)

@ Helmut
Posting: # 22796
Views: 1,270

## Two at a Time (Incomplete Blocks)

Hi Helmut,

Thanks a million

Best Regards
Ahmed
dshah
★★

India/United Kingdom,
2022-02-18 11:26
(399 d 18:21 ago)

(edited by dshah on 2022-02-18 16:27)
@ Ahikal
Posting: # 22797
Views: 1,241

## statistical analysis of William's Design

Dear Ahmed!

Particularly for EMA as per the guideline: - "In studies with more than two treatment arms (e.g. a three period study including two references, one from EU and another from USA, or a four period study including test and reference in fed and fasted states), the analysis for each comparison should be conducted excluding the data from the treatments that are not relevant for the comparison in question."
So you will need to show the calculations separately.
Regards,
Dshah
Ahikal
☆

Egypt,
2022-02-25 14:42
(392 d 15:05 ago)

@ dshah
Posting: # 22805
Views: 1,115

## statistical analysis of William's Design

Dear Dshah:
Thank you, this is how our regulatory committee wanted me to do it, and I was fighting them

Best Regards
Ahmed
Ahikal
☆

Egypt,
2022-02-25 15:22
(392 d 14:26 ago)

@ dshah
Posting: # 22806
Views: 1,142

## statistical analysis of William's Design

Dear Dshah;

❝ Particularly for EMA as per the guideline"

Would you please provide me with a reference to this

Best Regards
Ahmed
Helmut
★★★

Vienna, Austria,
2022-02-25 15:37
(392 d 14:10 ago)

@ Ahikal
Posting: # 22807
Views: 1,094

## Do your homework first

Hi Ahikal,

see there #3 and on top of the page.

Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Ahikal
☆

Egypt,
2022-02-26 10:54
(391 d 18:53 ago)

@ Helmut
Posting: # 22808
Views: 1,098

## Do your homework first

Dear Helmut:

Got it, my apology

Sincerely
Ahmed
Admin contact
22,550 posts in 4,724 threads, 1,606 registered users;
15 visitors (0 registered, 15 guests [including 9 identified bots]).
Forum time: 05:48 CET (Europe/Vienna)

If there is an exception to any rule,
and if it can be proved by observation,
that rule is wrong.    Richard Feynman

The Bioequivalence and Bioavailability Forum is hosted by
Ing. Helmut Schütz