[Minutes, Advisory Committee to the Director GP-160 Meeting (1992: Bethesda, Maryland)]

153

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

Dr. Broder. When you design the study and, in fact,
the studies we have designed, these are concerns that
come up.

There are
yourself against that.
there's no reason why
group should get your r
thing is working and t
have a clear result.
different issue.

various ways you protect
One way is if it's blinded,
the active versus the control
new wonderful thing unless one:he other isn't, but then you
If it's an open trial, it's a

The other part of it is your randomization
will be usually by certain strata, which would mean --
let's just say there is a community that has easy
access to the wonderful drug and another one less easy
access. Your randomization is within each of those
strata. So you would be able to deal with that.
I think your points are important and are
taken into account when you designed these trials; for
instance, the dig trial that we designed. Those of
you who are internists and cardiologists would know
heart failure is an incredibly difficult area to do
long-term trials because these people receive multiple
NEAL R. GROSS
COURT REPORTENS AND TRANSCRIBERS
1323 RHODE ISLAND AVENUE, N.W.
[phone] WASHINGTON, D.C. 20005 [phone]