Liza's article to ACOR: Publication of Protocols
I have been looking for trials of promising new agents for my Dad. Finding
accurate information is a formidable task.
There is a great need for the publication of protocols for medical research,
in particular the protocols of randomised controlled trials.
The internet provides a means to do this easily and efficiently, and the
only impediment to implementing this now as standard practice is an interest
in withholding information from the public in order to give researchers the
room to manipulate, distort, or deliberately misinterpret data.
The protocol of a study provides a detailed log of the rationale and
methodology of the study, as well as being a plan to which all investigators
of the study or trial must adhere. Protocols are rarely made available to
the public, and once a study is over they are often lost. The abstracts and
full text papers that report on studies cannot and do not ever include the
full protocol - no "hard-copy" journal could possibly provide enough space
to do this. So all that is reported is a very shortened summary of the
study.
Chalmers and Altman discuss what they call "the scandal of poor medical
research" and give reasons why the publishing of protocols would be critical
in preventing such poor research. (citation below). Publishing the protocol
while it is in draft form - "early remedial intervention" they call it -
would allow other researchers to suggest improvements before the study
begins.
Protocol publication would allow comparison between what was the stated
purpose of the study and what was actually done.
It would reduce the practice of "data dredging" - in which "associations are
sought or stumbled upon during data analysis rather than hypothesised a
priori."[6]
It would also reduce the practice of unacknowledged or post-hoc revision of
the study aims or design.
"Such practices are not only detrimental to the advancement of medical
research, they are ethically unsound since they may result in patients
receiving inappropriate care." [6] A randomised controlled trial recently
published in the BMJ was, by coincidence, reviewed for the journal by the
same person as had reviewed the protocol for the funding body. The reviewer
noticed discrepancies in the power calculation that turned an inconclusive
result into a strongly negative one.[2].
But for colon cancer patients and caregivers, and others seeking information
on studies that might result in something that could save life - or just as
importantly endanger life - the most important reason for the publication
of protocols is that it lets the public know what studies and trials are
actually underway.
Proptocol publication would make it possible to find trials. It would reduce
wasteful duplication of research - thus reducing distortion of the evidence
as a result of publication bias (meaning which journals report which
studies, and the credibility or lack of credibility that this conveys). It
would boost recruitment into trials, speeding up evaluation of drugs and
protocols. It would prevent bias due to the practice of deliberate
over-reporting of trials (a Cochrane systematic review of olanzapine found
158 reports of a single trial [5]) as well as deliberate under-reporting of
trials (in which some studies never even reach the public domain).
Widespread protocol publication is now fully possible because of the
internet. The National Library of Medicine and meta trial databases not only
exist but are ready and eager to publish, file and organize such data.
Protocol publication could easily be adopted as the standard, and put into
practice, in a very short time, if unethical concerns did not stand in the
way.
I hope for our sake that ethical considerations will prevail, and soon.
Love Liza
1. DG Altman: The scandal of poor medical research; BMJ 1994, 308: 283-84
2. G Mires, F Williams, P Howie, S Goldbeck-Wood, GD Murray, B-I Nesheim:
Randomised controlled trial of cardiotocography versus Doppler auscultation
of fetal heart at admission in labour in low risk obstetric population
Commentary: changes between protocol and manuscript should be declared at
submission Commentary: research governance must focus on research training
Commentary: Approach to power calculations has to be realistic; BMJ 2001,
322: 1457-1462
3. An overstretched hypothesis? (Editorial) Lancet 2001, 357: 405
4. RJ Simes: Publication Bias: The case for an international registry of
clinical trials; J Clin Oncol 1986, 4: 1529-1541
5. L Duggan, M Fenton, RM Dardennes, A El-Dosoky, S Indran: Olanzapine for
schizophrenia; Cochrane Database Syst Rev 2000, 2: CD001359
6. Fiona Godlee; BMC News and Views 2001 2: 4
accurate information is a formidable task.
There is a great need for the publication of protocols for medical research,
in particular the protocols of randomised controlled trials.
The internet provides a means to do this easily and efficiently, and the
only impediment to implementing this now as standard practice is an interest
in withholding information from the public in order to give researchers the
room to manipulate, distort, or deliberately misinterpret data.
The protocol of a study provides a detailed log of the rationale and
methodology of the study, as well as being a plan to which all investigators
of the study or trial must adhere. Protocols are rarely made available to
the public, and once a study is over they are often lost. The abstracts and
full text papers that report on studies cannot and do not ever include the
full protocol - no "hard-copy" journal could possibly provide enough space
to do this. So all that is reported is a very shortened summary of the
study.
Chalmers and Altman discuss what they call "the scandal of poor medical
research" and give reasons why the publishing of protocols would be critical
in preventing such poor research. (citation below). Publishing the protocol
while it is in draft form - "early remedial intervention" they call it -
would allow other researchers to suggest improvements before the study
begins.
Protocol publication would allow comparison between what was the stated
purpose of the study and what was actually done.
It would reduce the practice of "data dredging" - in which "associations are
sought or stumbled upon during data analysis rather than hypothesised a
priori."[6]
It would also reduce the practice of unacknowledged or post-hoc revision of
the study aims or design.
"Such practices are not only detrimental to the advancement of medical
research, they are ethically unsound since they may result in patients
receiving inappropriate care." [6] A randomised controlled trial recently
published in the BMJ was, by coincidence, reviewed for the journal by the
same person as had reviewed the protocol for the funding body. The reviewer
noticed discrepancies in the power calculation that turned an inconclusive
result into a strongly negative one.[2].
But for colon cancer patients and caregivers, and others seeking information
on studies that might result in something that could save life - or just as
importantly endanger life - the most important reason for the publication
of protocols is that it lets the public know what studies and trials are
actually underway.
Proptocol publication would make it possible to find trials. It would reduce
wasteful duplication of research - thus reducing distortion of the evidence
as a result of publication bias (meaning which journals report which
studies, and the credibility or lack of credibility that this conveys). It
would boost recruitment into trials, speeding up evaluation of drugs and
protocols. It would prevent bias due to the practice of deliberate
over-reporting of trials (a Cochrane systematic review of olanzapine found
158 reports of a single trial [5]) as well as deliberate under-reporting of
trials (in which some studies never even reach the public domain).
Widespread protocol publication is now fully possible because of the
internet. The National Library of Medicine and meta trial databases not only
exist but are ready and eager to publish, file and organize such data.
Protocol publication could easily be adopted as the standard, and put into
practice, in a very short time, if unethical concerns did not stand in the
way.
I hope for our sake that ethical considerations will prevail, and soon.
Love Liza
1. DG Altman: The scandal of poor medical research; BMJ 1994, 308: 283-84
2. G Mires, F Williams, P Howie, S Goldbeck-Wood, GD Murray, B-I Nesheim:
Randomised controlled trial of cardiotocography versus Doppler auscultation
of fetal heart at admission in labour in low risk obstetric population
Commentary: changes between protocol and manuscript should be declared at
submission Commentary: research governance must focus on research training
Commentary: Approach to power calculations has to be realistic; BMJ 2001,
322: 1457-1462
3. An overstretched hypothesis? (Editorial) Lancet 2001, 357: 405
4. RJ Simes: Publication Bias: The case for an international registry of
clinical trials; J Clin Oncol 1986, 4: 1529-1541
5. L Duggan, M Fenton, RM Dardennes, A El-Dosoky, S Indran: Olanzapine for
schizophrenia; Cochrane Database Syst Rev 2000, 2: CD001359
6. Fiona Godlee; BMC News and Views 2001 2: 4
posted by Liza May at 4:25 PM
0 Comments:
Post a Comment
<< Home