Transcript Ethical Guidelines for Journal publication

Professional Development Workshop:
Ethics in Science and Publishing
•
Overview presentation:
Defining the problem: What is research misconduct and why is it a problem?
•
Panelist Discussion: Current publishing policies and approaches related to ethics
and misconduct
•
Questions from Audience, Discussion
Katja Brose (Editor, Neuron – Cell Press) moderator
I-han Chou (Senior Editor – Nature Publishing Group)
Emilie Marcus (Editor, Cell – CEO, Cell Press)
John Maunsell (Editor in Chief, Journal of Neuroscience)
Shamus O’Reilly (Publisher, Elsevier)
Defining the problem:
What is research misconduct?
Why is it a problem?
Publication ethics—is there a problem?
The view from the press
Retractions on the rise
But is misconduct also on the rise?
Ricahrd Van Noorden, Nature (2011)
summarizing
Carl Zimmer, NYTimes
(2012)summarizing
Steen. Journal of Medical Ethics
Retractions in the scientific literature: is the
incidence of research fraud increasing
Wager & Williams. Journal Medical Ethics
Why and how do journals retract articles?
An analysis of Medline retractions 1988–
2008
“A surprising upsurge in the
number of scientific papers that
have had to be retracted because
they were wrong or even
fraudulent has journal editors
and ethicists wringing their
hands. The retracted papers are
a small fraction of the vast flood
of research published each year,
but they offer a revealing
glimpse of the pressures driving
many scientists to improper
conduct.”
“There are many theories for why retractions
and fraud have increased. A benign view
suggests that because journals are now
published online and more accessible to a wider
audience, it’s easier for experts to spot
erroneous or fraudulent papers. A darker view
suggests that publish-or-perish pressures in the
race to be first with a finding and to place it in a
prestigious journal has driven scientists to make
sloppy mistakes or even falsify data. The
solutions are not obvious, but clearly greater
vigilance by reviewers and editors is needed.”
Is misconduct on the rise?
Some studies and statistics
RETRACTIONS ARE ON THE RISE
Fang, Steen, Casadevall. PNAS Misconduct accounts for the majority of retracted scientific publications (Oct 1 2012)
Comprehensive analysis of 2047 retracted life science/biomedical papers, including reasons for retraction not indicated in
retraction notice. Proposes that fraud underestimated. 10x increase in retractions due to fraud since ‘75. Reasons for
retraction: error (21.3%), misconduct (67.4%) including, fraud or suspected fraud (43.4%), duplicate publication (14.2%), and
plagiarism (9.8%). Issue with transparency of retraction notice.
Van Noorden. Nature The trouble with retractions (Oct 2011)
in last 10 years, Web of Science tracks number of retractions to increase 10 fold, while literature has only expanded 44%
Steen. Journal of Medical Ethics (2010) Retractions in the scientific literature: is the incidence of research fraud increasing
PubMed survey of retracted papers between 2000-2010. “Error” or undisclosed reasons more common (73.5%) than fraud
(26.6%). Issues with notification---31.8% retracted papers not noted/marked as retracted
MORE CASES OF FIGURE MANIPULATION
Journal of Cell Biology--routine screening of accepted papers for figure manipulation, since 2004.
25% of all accepted manuscripts have at least one figure that must be remade because of “inappropriate” manipulation. 1% of
accepted papers have acceptance revoked because of image manipulation that impacts conclusions
Despite a decade of screening and education, the issue remains
SELF-REPORTS FROM RESEARCHERS
Fanelli et al. PLoS One (2009) Meta-analysis of various surveys of scientists on misconduct
self report of own behaviors: 2% data falsification, 34% other questionable practice
Titus et al. Repairing Research Integrity Nature 453 (2005): 980-982
Of 2212 NIH-funded researchers surveyed and asked to report how often they witnessed some form of research misconduct in
their own departments. Also asked to describe the situation and distinguish as misconduct and “grey zone” issues -major fraud
(<2%), minor fraud (5-15%)
suspected misconduct occurred at all levels: senior investigators, students, postdocs
Is this a case of a relativel few, rogue cases?
Or is there a deeper problem?
•
Difficult to measure how much misconduct there is - - -few hard and fast numbers exist. But
number of retracted papers are still small percent of all science published (.02%)
•
More scientists, more papers? But growth in retractions outpacing growth in # papers
•
More activism in lay press and community to “root out” misconduct, blogs like Retraction
Watch, community policing
•
More concerted efforts by journals/publishers to police. Retractions occurring faster and
reaching back deeper into the published literature. Getting better at correcting mistakes
when they happen
•
Digital/electronic makes it easier to detect some types of fraud/plagiarism —various
software systems for detecting image manipulation, plagiarism
•
Some evidence that more minor cases of questionable conduct (sloppiness, “honest errors”)
are coming to light
What is research misconduct?
Isn’t one specific and agreed upon definition for research misconduct
A common definition (as put forth by the US-Public Health Service, Office of Research Integrity) and also
accepted by many international agencies/institutions:
Research misconduct includes, fabrication, falsification or plagiarism, in proposing, performing or
reviewing research or in reporting research results.
(a)
(b)
(c)
(d)
FABRICATION is making up data or results and recording or reporting them
FALSIFICATION is manipulating research materials, equipment or processes, or changing or omitting
data or results that the research is not accurately presented in the research record
PLAGIARISM is the appropriation of another person’s ideas, processes, results or words without
giving appropriate credit
Research misconduct DOES NOT include honest error or differences of opinion or necessarily,
inability to replicate
An easy to remember scientific moral code:
do not lie (fabrication), cheat (falsification) or steal (plagiarism)
Research misconduct vs publishing misconduct---are they the same thing?
• there is much focus on publication fraud and whether the publication system is “broken”
• BUT most research misconduct begins before the paper and is only caught when it is published, which
puts more attention on publication misconduct
Spectrum of misconduct:
Research misconduct vs sloppy science-- Does it matter which it is?
Spectrum of research misconduct - - many grey zones
Questionable
Research
Practices
(QRP)
fabrication of data
distortion of data (including figure manipulation)
plagiarism
failure to achieve ethics approval for animals/subjects
selective use of statistics
ignoring outliers in data set
removing data/hiding or holding back data/date selection
not disclosing conflict of interest
redundant publication
questionable statistics
authorship issues
poor data management/recording
seriousness
Misconduct
Misconduct vs Poor Practice --- INTENT
• Generally, “misconduct” usually entails some level of intent---but this can be hard to prove
• Leaving room for “honest errors” or mistakes but not as justification for “sloppy science”
which can be as damaging
• When and how do “sloppy errors” become misconduct?
Roots of misconduct:
Contributing factors
Does misconduct arise from “a few bad apples” or is it rooted in institutional problems?
Is misconduct an individual problem or does the research environment contribute?
Some factors that have been discussed-•lack of appropriate training and mentorship about good scientific practice
• high pressure/high stakes environments that put more weight on the answers than the
process, high profile publications, high stakes may inadvertantly promote fraud
• competition in science
• nature of the work relationships in labs: PIs have only indirect view of results—”honest
mistakes” amplified to cover up and fraud
• lack of accountability – assumption that it is difficult to get caught and when you do, there
are few penalties
• lack of institutional ethics - - -feeling that “everyone does this,” “cutting corners is OK”
• large collaborations where collaborators are not aware of the details of other contributors
work
• Scientific blindsight: “I am sure that my conclusions are right and so taking shortcuts is OK”
• is it human nature to cheat when you can?
• the relative ease with which electronic data can be altered—photoshop, etc.
None of these factors cause fraud or should be used as an excuse, but may contribute
Publishing policies
related to ethics and misconduct
Common author misconduct situations arising
for publishers
•
Figure manipulation or falsification
•
Data falsification or fraud
•
Plagiarism—copying someone else’s words, ideas, procedures without attribution
•
Duplicate/redundant publication, self-plagiarism--overlap with previous publications or other
submission, “salami slicing”
•
Conflicts of interest (financial, professional, personal)
•
Authorship conflict---missed authors,
•
Failure to provide a published reagent
•
Unethical research (violation of legal/ethical guidelines for use of subjects, materials)
•
Reviewer misconduct NOTE: Author misconduct is not the only ethical challenge that
publishers face---reviewer ethics and editorial ethics are equally important.
What are journals doing about addressing
misconduct in the publishing process?
Editors and publishers have an ethical obligation to
(1) support the quality, ethics of the review process (Pre-publication--weed out, prevent)
(2) insure the correctness of the published literature (Post-publication-correct, communicate)
(3) Educate—(prevent)
Editors/journals have an ethical obligation to respond to and address ethical allegations that may arise
about published papers or papers in review. Systems and procedures for investigating and addressing
situations of misconduct. Cooperation with investigations from institutions or funding bodies.
Most journals have author and reviewer guidelines for appropriate ethical conduct. Authors may be asked to
“sign off” on understanding and acceptance of these policies prior to publication. Co-authorship
statements on contribution
Routine screening:
Journal of Cell Biology: routine figure screening for image manipulation
CrossRef/CrossCheck: systems for detecting plagiarism
Random screening of “certain types of papers”
Science, The Lancet: policies for heavier screening of papers in competitive fields, “hot topics,”
“extraordinary claims”
Contribute to education and development of community standards: editorials, sponsorship of
talks/workshops to discuss issues related to scientific ethics. Work with industry organizations to set
standards
Sources and Resources
•
Committee on Publishing Ethics (COPE) www.publicationethics.org.uk
COPE is a forum for editors and publishers of peer-reviewed journals to discuss all aspects of
publication ethics
**website has good Case Studies, Flow Charts, Guidelines for Best Practices
**COPE runs regular meetings for journal editors to discuss ethics cases
•
Professional societies, like SFN
**Society for Neuroscience Policy on Ethics: Guidelines on Responsible Conduct in Publishing
(www.sfn.org)
•
Publisher resources: resources and tools on publishing ethics
**Elsevier Ethics in Resource and Publication www.elsevier.ethics.com
•
Instructions for Authors, Instructions for Reviewers on journal websites
**define specific journal policies, guidelines, and sometimes other ethics related materials
•
Journal of Cell Biology’s Guidelines on handing digital images www.jcb.org
excellent discussion and overview of issues related to appropriate practice for image
handling/manipulation. A number of other journal’s have adopted guidelines based on JCB’s
•
Funding and governmental agencies’ websites (NIH, NSF, ORI, MRC, etc)
Open Issues:
Questions for Discussion
International standards and agencies to address misconduct and education on these issues. Science is increasingly
international but ethical policies and investigative bodies are often national-- Changing demographics across
geographies.
Do we need more consistent standards/procedures across journals/publishers?
Does “minor misconduct” lead to bigger fraud? Are all cases of misconduct the same? Should punishment be the
same for different kinds of error/fraud
How to handle contributions from multiple co-authors?---many papers are now result of collaborative research, in
many cases involving international collaborations. Who is responsible for the standards in the paper?
•Ultimately ALL authors hold some level of responsibility ”With shared credit comes with shared responsibility”
Notification/communication of retractions/errors to the scientific and medical communities, general public?
Alternatives notifications besides retractions; better ways to indicate type of retraction---remove stigma for
self-correct
How much does the research environment (or publication forum---”high profile publication”) contribute to
misconduct?--- is it really about competition, pressure and high stakes?
How best to educate young scientists (and also continue to educate senior scientists)?
• Importance of mentoring and of scientific mentors/PIs modeling good behavior
• Scientific education is most often an apprentice education, with few “rule books” for good practice
• Few studies or data on whether ethics courses are effective in practice
Professional Development Workshop:
Questions and Discussion
1
•Journal practices/policies for insuring ethical compliance prior to publication
2
•Plagiarism—standards, journal and publisher practices for screening
3
•Figure/image manipulation
4
•Conflicts of interest, issues related to authorship
5
•Ethical reviewer practice
6
•Correcting the literature post-publication--corrections, retractions
7
•Ethical misconduct versus poor scientific practice
8
•Who’s responsible for setting and maintaining standards on misconduct/ethics
Curbing Misconduct---What can be done?
Recommendations
What can be done?
Are there ways to prevent misconduct
Education (journals, universities, labs)
Institutions should establish clear guidelines for responsible conduct in research, not only for students but
all scientists in the institution. One or two ethics courses in graduate school is not enough.
Active mentoring —senior investigators and mentors should not only talk to their trainees about the
importance of good scientific practice but also model good behaviour. Appropriate training for mentors
Create a zero tolerance environment: Clear and stringent penalties for violations of guidelines
Clear system by for reporting suspected cases of misconduct-Institutions should have mechanisms by which
scientists can bring to light potential misconduct situations. Protection of informants or “whistle blowers”
Create visible oversight committees at institutions for fair investigation. Findings of committees should be
made public when possible. Or at least appropriate stakeholders (journals, funding sources, collaborators,
future employers) should be notified when appropriate.
Whether at institution or journal, transparency of decision. Clear and appropriate sanctions and penalties .
Clear retraction statements, distinguish between fraud vs “honest errors.” Consider scale of act.
Better mechanisms for linking/updating papers, so retracted papers don’t continue to be referenced and cited
Develop institute-level standards for record keeping and education of researchers at all levels of training.
Centralized storage of data/records? Institutional or lab audits of data records, protocols.
Carefully consider reward systems that may contribute to poor practices or focus on short term gains
How do editors address an allegation of
publication misconduct?
How do editors address an allegation of
publication misconduct?
•
First course, is usually contact and discussion with the authors (and reviewers). In many cases, what may appear to be
intentional fraud, may be a lack of understanding of the guidelines or poor practices, but not necessarily intention to
commit fraud
•
Notification of the author’s university, institution, funding agency, government agencies like ORI
•
Investigation has typically been seen as the responsibility of research institutions and funding agencies, NOT usually
the journal’s domain. Journals/publishers could support institutional investigations and act on recommendations, but
typically did not oversee investigations directly. This is changing.
•
Some countries have governmental agencies which investigate and arbitrate allegations of misconduct (ie in USA,
ORI/Office of Research Integrity, NSF). Internationally situation less clear.
If there is evidence of misconduct/fraud:
•
Prior to publication (during review): manuscript can be withdrawn from review
•
Post-publication (literature correction)---publication of an Retraction, Note of Editorial Concern, Errata/Correction.
Either w/ authors (some or all) signature or editorial. Appropriate statement for situation. Paper is “marked” in the
literature/PubMed.
•
When to retract vs correct. Fraud vs honest mistake. Intent? Extent to which the data is incorrect/misleading
•
Author may be banned from submitting to the journal or other sanctions
•
Editors may, in some cases, provide information for other Editors/Publishers
•
Publication of an editorial in the journal to discuss issue “generally” and raise awareness of the issue