Background information concerning the topics of
the Sixth Symposium of the IIfTC
"The Modern Clinical Trial: Is it Compatible
with Common Sense?"
The Modern Clinical Trial: Is it Compatible
with Common Sense?
The idea of the conference arose from the feeling of deepening dissatisfaction
among academic cardiologists about the loss of many valuable approaches
to new treatment caused by the dead-weight of randomised, double-blind,
multicentre clinical trial using orthodox statistical methods and restricted
inclusion and exclusion criteria - the standard trial. The dead-weight
is made up of a mass of rigid dogmata shared between statisticians, the
pharmaceutical industry, and regulatory authorities. There appears to
be little attempt to test the validity of this dogma, or to explore alternative
hypotheses. One obvious disadvantage of the present system is the need
for prospective multicentre trials with (tens of) thousands of patients.
Participation in such trials soon shows that they are flawed by heterogeneity
of diagnosis, patient type, physician management etc.
Our question for this conference is: "What alternative is there
that avoids these problems?" We might also ask:
- "Is it really the case that non-standard ways of doing clinical
trials give us less reliable results?"
- "Can the result of the standard trial be applied to the individual
patient, and if not - is it of any use?"
- "What does the standard trial tell us about the general safety
of a drug?"
- "What role do other sources of knowledge have (eg from pathophysiology,
from clinical experience, etc.)?"
In order to stimulate discussion at the conference we have
prepared some questions and points which we would like to explore. These
are:
- Alvan Feinstein has recently stated that: "Medical historians
two centuries from now will surely marvel at the peculiar scientific
policies of national and international research leaders in the late
twentieth century who invested all their intellectual and often emotional
energy in developing randomized trials, while giving little or no scientific
attention to the many questions that cannot be answered with randomized
trials and that require nonrandomized forms of investigation."
(Drug Res. 1989, 39:982) Is it the case that too much research attention
is being given to clinical trials? Should we rather pursue other forms
of investigation? What might these be, generally and in detail?
- One obvious disadvantage of the present system is the need for prospective
multicentre trials with (tens of) thousands of patients. It has been
claimed that participation in such trials shows that they are flawed
by heterogeneity of diagnosis, patient type, physician management etc.
To what an extent is this a problem for the typical clinical trial?
- Are there non-standard ways of doing clinical trials (such as Bayesianism)
which are capable of giving us swifter as well as more reliable results,
or is this not the case?
- Can the results from the standard trial be applied to patients having
widely varying characteristics, such as suffering from concomitant diseases?
- What role do other sources of knowledge have (eg from pathophysiology,
from clinical experience etc)?
- Should post-marketing surveillance of drug safety perhaps be made
more systematic rather than continuing to insist on strict rules for
initial approval of drugs? This may be used to reassess the effectiveness
of the drug in light of the knowledge accumulated over several years.
|