Transcript here - Association for Research Ethics

Ethical research publication:
who’s responsibility is it?
Liz Wager PhD
Publications Consultant,
Sideview
[email protected]
©Sideview
"A discovery does not exist until
it is safely reviewed
and in print"
EO Wilson
(1929 - )
"Words are, of course,
the most powerful drug
used by mankind"
Rudyard Kipling (1865-1936)
Publication matters
 Trial participants expect findings to
contribute to medical knowledge
 Misrepresenting findings (or not publishing
them at all) is unethical
 But who’s responsible for ensuring
good practice?
©Sideview
Research and publication ethics
are a spectrum
Unethical
research
design
Design
Inappropriate
analysis
Data
falsification
Analysis
Conduct
Lack of
patient
consent
Data
fabrication
Plagiarism
Reporting
Authorship
abuse
Image
manipulation
Selective /
nonpublication
Redundant
publication
Many players are
(or could be) involved:







Researchers (mentors / supervisors)
Journals (editors / publishers)
Funders
Institutions
Professional / academic organizations
RECs
Regulators
©Sideview
Does the system
need to change?
 Half of all clinical trials are never published
©Sideview
Ross et al PLOS Med 2009;e1000144
Does the system
need to change?
 About 50% of studies presented at
conferences never get published in full
(Scherer et al JAMA 1994;272:158-62)
 Cohort of RCTs submitted to Swiss REC (to
1998) 52% published by 2006
(von Elm et al Swiss Med Wkly 2008;138:197-203)
 22% of trials on CT.gov reported results
within 1 year of study end
(Prayle et al BMJ 2012)
©Sideview
Does the system
need to change?
 German drug assessment body found
74% of data on the antidepressant
reboxitine was unpublished
 Including the unpublished data changed
their recommendation
(to “ineffective and potentially harmful”)
Wieseler et al BMJ 2010;341:c4942
Eyding et al BMJ 2010;341:c4737
©Sideview
Does the system
need to change?




585 registered trials with >500 participants
Completed before Jan 2009
29% (171) unpublished in November 2012
Of the unpublished trials 78% did not have results posted
on CT.gov

“The lack of availability of results from these trials … constitutes a failure to
honor the ethical contract that is the basis for exposing study participants to
the risks inherent in trial participation. Additional safeguards are needed to
ensure timely public dissemination of trial data.”
Jones et al BMJ 2013;347:f6104
©Sideview
published
29 Oct 2013!
Does the system
need to change?
 Comparison of protocols cf publications (N=102)
 Found incomplete reporting of:
• 50% of efficacy
• 65% of safety/AE outcomes
 Statistically significant outcomes more likely
to be reported
 62% of trials had at least one primary outcome
changed, introduced or omitted
Chan et al JAMA 2004;291:2457-65
©Sideview
Change probably requires a
‘systems’ approach




Incentives
Guidelines
Training
‘Infrastructure’ / technology
©Sideview
Many players
should be involved:







Researchers (mentors / supervisors)
Journals (editors / publishers)
Funders
Institutions
Professional / academic organizations
RECs
Regulators
©Sideview
Influence on clinical trials
Funding
decision
Ethical
approval
Publication
Registration
Design
Agreement /
contract
Funders
Analysis
Reporting
Conduct
RECs
Institutions
Regulators
Publishers
Opportunities
 Funders could make final payment conditional on report /
publication (eg HTA)
 Funders could make publication of previous projects a
criterion for future funding
 Institutions could reward responsible publication (quality
and quantity)
 Institutions should promote and enforce good authorship
practices
 Alternative venues for disseminating results (registers,
institutional repositories, databanks, websites)*
©Sideview
alternatives to traditional journals
©Sideview
More opportunities
 Institutions and professional bodies could offer training in
RCR / publication ethics
 Journals could require use of reporting guidelines (or
develop more structured reports)
 Funders, RECs, journals could require trial registration
©Sideview
Journal policies can be effective
ICJME
deadline
New trials registered at clinicaltrials.gov May – Oct 2005
Good news! (Sept 2013)
©Sideview
Research ethics committees could
help prevent other serious problems
Unethical
research
design
Design
Inappropriate
analysis
Data
falsification
Analysis
Conduct
Lack of
patient
consent
Data
fabrication
Plagiarism
Reporting
Authorship
abuse
Image
manipulation
Selective /
nonpublication
Redundant
publication
Could RECs extend their remit?
 Require researchers to publish all studies (on
trial register / repository / journal)
 Encourage / check posting on institutional
repositories / company websites
 Make publication of previous projects a
requirement for future approvals
©Sideview
Are RECs behaving unethically?


“The fate of scientific data is no idle issue. Reliable research syntheses depend
on ensuring that as much as possible of the potentially relevant evidence is
taken into account in systematic reviews. .. Contrary to a widely held
assumption, it is investigators and some research funders—and not necessarily
journal editors—who are primarily responsible for under-reporting of
research … Prevention of this form of scientific misconduct must therefore
involve the bodies to which investigators are answerable, in particular
research ethics committees.
The consequences of under-reporting of research are that patients are being
expected to accept the harmful side effects of ineffective forms of care; accept
advice about the effects of health care which is based on evidence that is less
complete than it should be;… and contribute to research … which may not be
published if the results come as a disappointment or an embarrassment to the
investigators or sponsors.
Savulescu et al BMJ 1996;313:1390-3
©Sideview
Conclusions
 Unethical reporting is a ‘systems problem’
 Unethical reporting may harm patients,
harm trial participants, waste resources
 Responsible research reporting is
EVERYBODY’s responsibility
including RECs
©Sideview