RE: [NMusers] Minimum patients number...

Hi All

I think we have had somewhat better luck justifying D optimal design for
just the reasons that Steve states below: the process is definitely faster
than simulation, the optimal designs have fewer samples which generally gets
better compliance with patients and clinical sites, the assay costs and
sample handling/shipping costs are lower and last but definitely not least
the evaluations from such studies tend to be much easier to do since the
study design is informative.

That said, there is no harm in mixing these methods. I have done simulation
evaluations based on D optimal designs. The use of D optimization does cut
down the simulation work a lot.

Diane

> -----Original Message-----
> From: owner-nmusers_at_globomaxnm.com [mailto:owner-nmusers_at_globomaxnm.com]
> On Behalf Of Stephen Duffull
> Sent: Sunday, January 28, 2007 3:31 PM
> To: 'Mark Sale - Next Level Solutions'
> Cc: nmusers_at_globomaxnm.com
> Subject: RE: [NMusers] Minimum patients number...
>
> Mark
>
> Overall - I agree with your conjecture: That designing by simulation is
> hard
> to make a business case for.
>
> But using optimal design algorithms, where runtimes are in the order of
> seconds to minutes (rather than days to weeks) then in my mind it's hard
> to
> justify not having a look to see what you can gain from understanding your
> model and design better.
>
> Cheers
>
> Steve
> --
> Professor Stephen Duffull
> Chair of Clinical Pharmacy
> School of Pharmacy
> University of Otago
> PO Box 913 Dunedin
> New Zealand
> E: stephen.duffull_at_otago.ac.nz
> P: +64 3 479 5044
> F: +64 3 479 7034
>
> Design software: www.winpopt.com
>



Received on Sun Jan 28 2007 - 16:24:59 EST