Bioequivalence and Bioavailability Forum

I continue to explore Phoenix WinNonlin and Bioequivalence output. I have studied two way bioequivalence in textbooks such as Pharmaceutical Statistics by Sanford Bolton. The two way BE calculations are well explained in that book. As far as programs are concerned I like the output from Equivtest best, but EquivTest doe not do three-way crossover studies.

I looked at an old example I have here of a 3 way BE study (6 sequences, 3 periods and 3 treatments, balanced Williams design) run in Phoenix WinNonlin.

I am puzzled by the "sequential" tab of the program, since there is a term there that I am not familiar with. It is an inferential test of "int" without the inverted commas. Please access link below

http://screencast.com/t/9dPhDKPVPZ

As you can see the F test is very high and the p test is,as you may well imagine, is very small. In this program (Phoenix) the two way BE analysis corresponding output does not give this "int" parameter.

I am trying to find out what it means...................int intercept?

and how to arrive at an understanding of the nature of this parameter and the role it plays. I have never seen it in a two way BE output.

Angus

---

Edit: Category changed. [Helmut]