Bioequivalence and Bioavailability Forum • EMEA CI on treatment, period & sequence effects

qwerty12
☆

2012-11-26 23:11
(4962 d 17:08 ago)

Posting: # 9594
Views: 4,806

EMEA CI on treatment, period & sequence effects [Regulatives / Guidelines]

In the most recent EMEA guideline it asks for the "tests for difference and the respective confidence intervals for the treatment effect, the period effect, and the sequence effect should be reported as descriptive data."

What exactly does this mean? How are they calculated? Can a SAS code be used to calculate the CIs on these effects.

I'm slightly confused as I thought that the confidence intervals for the difference in means between the two products to show bioequivalence would be the same as the CI for the treatment effect.

Thanks for your help in advance.

Helmut
★★★

Vienna, Austria,
2012-11-26 23:39
(4962 d 16:39 ago)

@ qwerty12
Posting: # 9595
Views: 4,295

EMA: no CI on period & sequence effects

Post reply

Hi Qwerty!

❝ In the most recent EMEA guideline it asks for the "tests for difference and the respective confidence intervals for the treatment effect, the period effect, and the sequence effect should be reported as descriptive data."

No it doesn’t. You are referring to the draft GL (p.13, lines 512–514), back in 2008 when EMA was still called EMEA.

❝ What exactly does this mean?

Gibberish; except the CI of the treatment effect – that’s how we assess BE, right?

❝ How are they calculated? Can a SAS code be used to calculate the CIs on these effects.

If you are able to code SAS for BE (the CI of the treatment effect), ask politely for the CI of the others. But: The period effect does not influence the treatment effect anyway (means out), and the sequence effect (or unequal carry-over) cannot be handled properly in a 2×2 cross-over (confounded with treatment-by-period interaction). Therefore: forget it.

❝ I'm slightly confused as I thought that the confidence intervals for the difference in means between the two products to show bioequivalence would be the same as the CI for the treatment effect.

Right.

—
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

qwerty12 ☆ 2012-11-27 00:16 (4962 d 16:03 ago) @ Helmut Posting: # 9596 Views: 4,080	EMA: no CI on period & sequence effects Post reply
	This guideline was made effective in November 2011 and as the guideline requests it I don't want to just forget about it. http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2011/04/WC500105372.pdf Any information on this is greatly appreciated. Edit: GL linked. What about saying “hello” next time? [Helmut]

Helmut
★★★

Vienna, Austria,
2012-11-27 00:59
(4962 d 15:19 ago)

@ qwerty12
Posting: # 9597
Views: 4,090

EMA: Veterinary (!) BE

Post reply

Hi Querty!

Next time please state already in the first place what you are talking about. In the Vet. BE GL the stupid stuff from the human draft BE GL was copypasted 1:1. Obviously nobody commented on this nonsense (like on the BE draft). Now you have two options:

Switch off your brain, fire up SAS, and report all effects and their CIs as requested.
State already in the protocol that you well aware of the GL, but refer also to the human GL – where this nonsensical requirement was removed. Search the forum for some background discussions in order to get some support in giving a justification (period effect, sequence effect, treatment effect). See also this oldie (slides 40–50).

Up to you.

—
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