17th November.
Stefan,
Allowing preferences to enter at all means the trial is not controlled.
Either the innovation is "clinically proven" or it is not; if it is not, then
there is no ethical problem. I have noticed that "clinically proven", as
the term is used by doctors, does not always have a scientific
meaning, and that clinical evidence can be understood by non-
physicians willing to read medical journals. It is possible to disagree
with doctors about medical issues, if you have read the literature!
If it is clinically proven, then we have to measure how much
increase/reduction in costs results from the innovation; this should
initially be done by measuring the innovation's costs, and comparing
them with the costs occurring in a model [that is, definitions, and a set
of assumptions] of the replaced approach, to determine what the
differences in costs are expected to be. If the replaced approach is
much cheaper, then a larger investigation may be called for, using a
controlled study, however objective the outcome measure(s) happen
to be. If the difference is the other way, then the problem has solved
itself.
If the innovation is more expensive, and clinically proven, then there is
a trade-off between the quality of care and the number of patients
treated; the more treated, the (relatively) worse the care. Or is the
choice different? It is important just where the choice arises - are
these people going to get treatment anyway, so that there is in effect
no choice about numbers, and we have to spend more or less on a
fixed number of patients?
I don't think you can allow patients to choose their path from hospital
to home. It matters, because you want to know whether the new
intervention would work with other patients elsewhere, who have no
information on which to choose between the two approaches. That,
after all, is why we DO science, to get results which are universal.
Good luck with it, anyway - an interesting problem.
Regards,
Douglas.
Subject: ethical question regarding RCTs
From: Stefan Hebenstreit <[log in to unmask]>
To: [log in to unmask]
Send reply to: Stefan Hebenstreit <[log in to unmask]>
Dear all,
I would welcome any advice from the members of this list regarding a
question I consider crucial in the design of a planned study.
Together with partners we are planning a study regarding the
cost-effectiveness of a hospital-based transitional care unit.
Our partners believe that the effectiveness of their transitional care
unit is "clinically proven" (which might be questionable, but shall be
omitted for now) and demand a randomized-controlled clinical trial to
prove its cost-effectiveness. In this trial transitional care as
delivered for several years at this hospital (intervention group) would
be compared to usual care defined as discharge from hospital and usual
community care at home (control group) for a group of elderly patients.
The randomization procedure would be conducted according to patients
preferences, i. e. if a person is indifferent between both alternatives,
he/she would be randomized to either of the groups. If a person prefers
one intervention, he/she will not be randomized but instead be assigned
to the preferred intervention.
Given that the clinical effectiveness of transitional care as delivered
for several years in this case is regarded as "clinically proven", is a
RCT ethically feasible in this case to prove the cost-effectiveness of
the intervention ?
This would in my opinion result in assigning some patients (i. e. those
assigned to the control group) to a less effective intervention.
Can we consider patients' indifference regarding the treatment options
as sufficient to enter them into the randomization process given the
fact that clinicians (as health professionals) are regarding
transitional care as the more effective intervention ?
In other words: Do you think we should consider other study designs ?
Thank you in advance for comments !
Best regards,
Stefan Hebenstreit
University of Bielefeld
Faculty of Heath Sciences
School of Public Health - WHO Collaborating Center
Dr. D. McCulloch,
PPEL,
University of Ulster,
Newtownabbey,
Co.Antrim BT37 0QB,
Northern Ireland.
Tel: 00 44 28 90368254
Fax: 00 44 28 90366847
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|